SYMBOL INDEX (30522 symbols across 1110 files)

FILE: build_dataset.py
  function build_instruction_dataset (line 19) | def build_instruction_dataset(data_path: Union[List[str],str],
  class DataCollatorForSupervisedDataset (line 89) | class DataCollatorForSupervisedDataset(object):
    method __call__ (line 94) | def __call__(self, instances: Sequence[Dict]) -> Dict[str, torch.Tensor]:

FILE: flash_attn_patch.py
  function forward (line 17) | def forward(
  function _prepare_decoder_attention_mask (line 105) | def _prepare_decoder_attention_mask(
  function replace_llama_attn_with_flash_attn (line 112) | def replace_llama_attn_with_flash_attn():

FILE: peft/mapping.py
  function get_peft_config (line 63) | def get_peft_config(config_dict):
  function _prepare_prompt_learning_config (line 74) | def _prepare_prompt_learning_config(peft_config, model_config):
  function _prepare_lora_config (line 116) | def _prepare_lora_config(peft_config, model_config):
  function get_peft_model (line 129) | def get_peft_model(model, peft_config):

FILE: peft/peft_model.py
  class PeftModel (line 48) | class PeftModel(PushToHubMixin, torch.nn.Module):
    method __init__ (line 72) | def __init__(self, model, peft_config: PeftConfig): # casualLM, LoraCo...
    method save_pretrained (line 87) | def save_pretrained(self, save_directory, **kwargs):
    method from_pretrained (line 120) | def from_pretrained(cls, model, model_id, **kwargs):
    method _setup_prompt_encoder (line 189) | def _setup_prompt_encoder(self):
    method get_prompt_embedding_to_save (line 223) | def get_prompt_embedding_to_save(self):
    method get_prompt (line 234) | def get_prompt(self, batch_size):
    method print_trainable_parameters (line 268) | def print_trainable_parameters(self):
    method __getattr__ (line 290) | def __getattr__(self, name: str):
    method forward (line 297) | def forward(self, *args, **kwargs):  # pylint: disable=E0202
    method disable_adapter (line 304) | def disable_adapter(self):
    method get_base_model (line 319) | def get_base_model(self):
  class PeftModelForSequenceClassification (line 326) | class PeftModelForSequenceClassification(PeftModel):
    method __init__ (line 352) | def __init__(self, model, peft_config: PeftConfig):
    method forward (line 364) | def forward(    # pylint: disable=W0221
    method _prefix_tuning_forward (line 425) | def _prefix_tuning_forward(
  class PeftModelForCausalLM (line 497) | class PeftModelForCausalLM(PeftModel):
    method __init__ (line 518) | def __init__(self, model, peft_config: PeftConfig): # casualLM, LoraCo...
    method forward (line 522) | def forward(# pylint: disable=W0221
    method generate (line 584) | def generate(self, **kwargs):
    method prepare_inputs_for_generation (line 595) | def prepare_inputs_for_generation(self, *args, **kwargs):
  class PeftModelForSeq2SeqLM (line 613) | class PeftModelForSeq2SeqLM(PeftModel):
    method __init__ (line 635) | def __init__(self, model, peft_config: PeftConfig):
    method forward (line 642) | def forward(
    method generate (line 732) | def generate(self, **kwargs):
    method prepare_inputs_for_generation (line 771) | def prepare_inputs_for_generation(self, *args, **kwargs):
  class PeftModelForTokenClassification (line 780) | class PeftModelForTokenClassification(PeftModel):
    method __init__ (line 806) | def __init__(self, model, peft_config: PeftConfig):
    method forward (line 818) | def forward(
    method _prefix_tuning_forward (line 879) | def _prefix_tuning_forward(

FILE: peft/tuners/lora.py
  class LoraConfig (line 32) | class LoraConfig(PeftConfig):
    method __post_init__ (line 82) | def __post_init__(self):
  class LoraModel (line 86) | class LoraModel(torch.nn.Module):
    method __init__ (line 110) | def __init__(self, config, model): # LoraConfig, CasualLM
    method _find_and_replace (line 118) | def _find_and_replace(self):
    method _get_submodules (line 161) | def _get_submodules(self, key):
    method _replace_module (line 167) | def _replace_module(self, parent_module, child_name, new_module, old_m...
    method __getattr__ (line 181) | def __getattr__(self, name: str):
    method modules_to_save (line 189) | def modules_to_save(self):
    method get_peft_config_as_dict (line 192) | def get_peft_config_as_dict(self, inference: bool = False):
    method _set_adapter_layers (line 198) | def _set_adapter_layers(self, enabled=True):
    method enable_adapter_layers (line 203) | def enable_adapter_layers(self):
    method disable_adapter_layers (line 206) | def disable_adapter_layers(self):
  function mark_only_lora_as_trainable (line 221) | def mark_only_lora_as_trainable(model: nn.Module, bias: str = "none") ->...
  class LoraLayer (line 239) | class LoraLayer:
    method __init__ (line 240) | def __init__(
  class Linear (line 260) | class Linear(nn.Linear, LoraLayer):
    method __init__ (line 262) | def __init__(
    method reset_parameters (line 299) | def reset_parameters(self):
    method train (line 309) | def train(self, mode: bool = True):
    method eval (line 316) | def eval(self):
    method cv_squared (line 323) | def cv_squared(self, x):
    method forward (line 338) | def forward(self, x: torch.Tensor, task_types=None):

FILE: peft/tuners/p_tuning.py
  class PromptEncoderReparameterizationType (line 26) | class PromptEncoderReparameterizationType(str, enum.Enum):
  class PromptEncoderConfig (line 32) | class PromptEncoderConfig(PromptLearningConfig):
    method __post_init__ (line 61) | def __post_init__(self):
  class PromptEncoder (line 67) | class PromptEncoder(torch.nn.Module):
    method __init__ (line 103) | def __init__(self, config):
    method forward (line 150) | def forward(self, indices):

FILE: peft/tuners/prefix_tuning.py
  class PrefixTuningConfig (line 25) | class PrefixTuningConfig(PromptLearningConfig):
    method __post_init__ (line 43) | def __post_init__(self):
  class PrefixEncoder (line 49) | class PrefixEncoder(torch.nn.Module):
    method __init__ (line 77) | def __init__(self, config):
    method forward (line 95) | def forward(self, prefix: torch.Tensor):

FILE: peft/tuners/prompt_tuning.py
  class PromptTuningInit (line 26) | class PromptTuningInit(str, enum.Enum):
  class PromptTuningConfig (line 32) | class PromptTuningConfig(PromptLearningConfig):
    method __post_init__ (line 61) | def __post_init__(self):
  class PromptEmbedding (line 65) | class PromptEmbedding(torch.nn.Module):
    method __init__ (line 93) | def __init__(self, config, word_embeddings):
    method forward (line 117) | def forward(self, indices):

FILE: peft/utils/config.py
  class PeftType (line 27) | class PeftType(str, enum.Enum):
  class TaskType (line 34) | class TaskType(str, enum.Enum):
  class PeftConfigMixin (line 41) | class PeftConfigMixin(PushToHubMixin):
    method __dict__ (line 54) | def __dict__(self):
    method to_dict (line 57) | def to_dict(self):
    method save_pretrained (line 60) | def save_pretrained(self, save_directory, **kwargs):
    method from_pretrained (line 84) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method from_json_file (line 113) | def from_json_file(cls, path_json_file, **kwargs):
  class PeftConfig (line 128) | class PeftConfig(PeftConfigMixin):
  class PromptLearningConfig (line 145) | class PromptLearningConfig(PeftConfig):

FILE: peft/utils/other.py
  function bloom_model_postprocess_past_key_value (line 20) | def bloom_model_postprocess_past_key_value(past_key_values):
  function shift_tokens_right (line 39) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _set_trainable (line 60) | def _set_trainable(model):
  function fsdp_auto_wrap_policy (line 67) | def fsdp_auto_wrap_policy(model):
  function transpose (line 102) | def transpose(weight, fan_in_fan_out):

FILE: peft/utils/save_and_load.py
  function get_peft_model_state_dict (line 19) | def get_peft_model_state_dict(model, state_dict=None):
  function set_peft_model_state_dict (line 65) | def set_peft_model_state_dict(model, peft_model_state_dict):

FILE: run_loramoe.py
  class SavePeftModelCallback (line 60) | class SavePeftModelCallback(transformers.TrainerCallback):
    method save_model (line 61) | def save_model(self, args, state, kwargs):
    method on_save (line 71) | def on_save(self, args, state, control, **kwargs):
    method on_train_end (line 75) | def on_train_end(self, args, state, control, **kwargs):
  function prepare_model_for_kbit_training (line 81) | def prepare_model_for_kbit_training(model, use_gradient_checkpointing=Tr...
  class ModelArguments (line 122) | class ModelArguments:
    method __post_init__ (line 191) | def __post_init__(self):
  class DataTrainingArguments (line 199) | class DataTrainingArguments:
  class MyTrainingArguments (line 236) | class MyTrainingArguments(TrainingArguments):
  function main (line 256) | def main():

FILE: transformers/activations.py
  class PytorchGELUTanh (line 28) | class PytorchGELUTanh(nn.Module):
    method __init__ (line 37) | def __init__(self):
    method forward (line 45) | def forward(self, input: Tensor) -> Tensor:
  class NewGELUActivation (line 49) | class NewGELUActivation(nn.Module):
    method forward (line 55) | def forward(self, input: Tensor) -> Tensor:
  class GELUActivation (line 59) | class GELUActivation(nn.Module):
    method __init__ (line 67) | def __init__(self, use_gelu_python: bool = False):
    method _gelu_python (line 74) | def _gelu_python(self, input: Tensor) -> Tensor:
    method forward (line 77) | def forward(self, input: Tensor) -> Tensor:
  class FastGELUActivation (line 81) | class FastGELUActivation(nn.Module):
    method forward (line 86) | def forward(self, input: Tensor) -> Tensor:
  class QuickGELUActivation (line 90) | class QuickGELUActivation(nn.Module):
    method forward (line 95) | def forward(self, input: Tensor) -> Tensor:
  class ClippedGELUActivation (line 99) | class ClippedGELUActivation(nn.Module):
    method __init__ (line 112) | def __init__(self, min: float, max: float):
    method forward (line 120) | def forward(self, x: Tensor) -> Tensor:
  class AccurateGELUActivation (line 124) | class AccurateGELUActivation(nn.Module):
    method __init__ (line 132) | def __init__(self):
    method forward (line 136) | def forward(self, input: Tensor) -> Tensor:
  class SiLUActivation (line 140) | class SiLUActivation(nn.Module):
    method forward (line 149) | def forward(self, input: Tensor) -> Tensor:
  class MishActivation (line 153) | class MishActivation(nn.Module):
    method __init__ (line 159) | def __init__(self):
    method _mish_python (line 166) | def _mish_python(self, input: Tensor) -> Tensor:
    method forward (line 169) | def forward(self, input: Tensor) -> Tensor:
  class LinearActivation (line 173) | class LinearActivation(nn.Module):
    method forward (line 178) | def forward(self, input: Tensor) -> Tensor:
  class LaplaceActivation (line 182) | class LaplaceActivation(nn.Module):
    method forward (line 190) | def forward(self, input, mu=0.707107, sigma=0.282095):
  class ReLUSquaredActivation (line 195) | class ReLUSquaredActivation(nn.Module):
    method forward (line 200) | def forward(self, input):
  class ClassInstantier (line 206) | class ClassInstantier(OrderedDict):
    method __getitem__ (line 207) | def __getitem__(self, key):
  function get_activation (line 236) | def get_activation(activation_string):

FILE: transformers/activations_tf.py
  function _gelu (line 21) | def _gelu(x):
  function _gelu_new (line 34) | def _gelu_new(x):
  function mish (line 52) | def mish(x):
  function gelu_fast (line 58) | def gelu_fast(x):
  function quick_gelu (line 66) | def quick_gelu(x):
  function gelu_10 (line 72) | def gelu_10(x):
  function glu (line 86) | def glu(x, axis=-1):
  function approximate_gelu_wrap (line 104) | def approximate_gelu_wrap(x):
  function get_tf_activation (line 130) | def get_tf_activation(activation_string):

FILE: transformers/audio_utils.py
  function hertz_to_mel (line 25) | def hertz_to_mel(freq: Union[float, np.ndarray], mel_scale: str = "htk")...
  function mel_to_hertz (line 59) | def mel_to_hertz(mels: Union[float, np.ndarray], mel_scale: str = "htk")...
  function _create_triangular_filter_bank (line 93) | def _create_triangular_filter_bank(fft_freqs: np.ndarray, filter_freqs: ...
  function mel_filter_bank (line 115) | def mel_filter_bank(
  function optimal_fft_length (line 193) | def optimal_fft_length(window_length: int) -> int:
  function window_function (line 206) | def window_function(
  function spectrogram (line 267) | def spectrogram(
  function power_to_db (line 455) | def power_to_db(
  function amplitude_to_db (line 506) | def amplitude_to_db(
  function get_mel_filter_banks (line 558) | def get_mel_filter_banks(
  function fram_wave (line 582) | def fram_wave(waveform: np.array, hop_length: int = 160, fft_window_size...
  function stft (line 638) | def stft(frames: np.array, windowing_function: np.array, fft_window_size...

FILE: transformers/benchmark/benchmark.py
  class PyTorchBenchmark (line 50) | class PyTorchBenchmark(Benchmark):
    method framework_version (line 56) | def framework_version(self):
    method _inference_speed (line 59) | def _inference_speed(self, model_name: str, batch_size: int, sequence_...
    method _inference_memory (line 63) | def _inference_memory(
    method _train_speed (line 69) | def _train_speed(self, model_name: str, batch_size: int, sequence_leng...
    method _train_memory (line 73) | def _train_memory(
    method _prepare_inference_func (line 79) | def _prepare_inference_func(self, model_name: str, batch_size: int, se...
    method _prepare_train_func (line 138) | def _prepare_train_func(self, model_name: str, batch_size: int, sequen...
    method _measure_speed (line 198) | def _measure_speed(self, func) -> float:
    method _measure_memory (line 226) | def _measure_memory(self, func: Callable[[], None]) -> [Memory, Memory...

FILE: transformers/benchmark/benchmark_args.py
  class PyTorchBenchmarkArguments (line 35) | class PyTorchBenchmarkArguments(BenchmarkArguments):
    method __init__ (line 46) | def __init__(self, **kwargs):
    method _setup_devices (line 78) | def _setup_devices(self) -> Tuple["torch.device", int]:
    method is_tpu (line 93) | def is_tpu(self):
    method device_idx (line 97) | def device_idx(self) -> int:
    method device (line 103) | def device(self) -> "torch.device":
    method n_gpu (line 108) | def n_gpu(self):
    method is_gpu (line 113) | def is_gpu(self):

FILE: transformers/benchmark/benchmark_args_tf.py
  class TensorFlowBenchmarkArguments (line 32) | class TensorFlowBenchmarkArguments(BenchmarkArguments):
    method __init__ (line 43) | def __init__(self, **kwargs):
    method _setup_tpu (line 79) | def _setup_tpu(self) -> Tuple["tf.distribute.cluster_resolver.TPUClust...
    method _setup_strategy (line 93) | def _setup_strategy(self) -> Tuple["tf.distribute.Strategy", "tf.distr...
    method is_tpu (line 113) | def is_tpu(self) -> bool:
    method strategy (line 118) | def strategy(self) -> "tf.distribute.Strategy":
    method gpu_list (line 123) | def gpu_list(self):
    method n_gpu (line 128) | def n_gpu(self) -> int:
    method is_gpu (line 135) | def is_gpu(self) -> bool:

FILE: transformers/benchmark/benchmark_args_utils.py
  function list_field (line 30) | def list_field(default=None, metadata=None):
  class BenchmarkArguments (line 35) | class BenchmarkArguments:
    method __post_init__ (line 135) | def __post_init__(self):
    method to_json_string (line 143) | def to_json_string(self):
    method model_names (line 150) | def model_names(self):
    method do_multi_processing (line 158) | def do_multi_processing(self):

FILE: transformers/benchmark/benchmark_tf.py
  function run_with_tf_optimizations (line 51) | def run_with_tf_optimizations(do_eager_mode: bool, use_xla: bool):
  function random_input_ids (line 73) | def random_input_ids(batch_size: int, sequence_length: int, vocab_size: ...
  class TensorFlowBenchmark (line 79) | class TensorFlowBenchmark(Benchmark):
    method framework_version (line 85) | def framework_version(self):
    method _inference_speed (line 88) | def _inference_speed(self, model_name: str, batch_size: int, sequence_...
    method _train_speed (line 95) | def _train_speed(self, model_name: str, batch_size: int, sequence_leng...
    method _inference_memory (line 101) | def _inference_memory(
    method _train_memory (line 112) | def _train_memory(
    method _prepare_inference_func (line 123) | def _prepare_inference_func(self, model_name: str, batch_size: int, se...
    method _prepare_train_func (line 164) | def _prepare_train_func(self, model_name: str, batch_size: int, sequen...
    method _measure_speed (line 213) | def _measure_speed(self, func) -> float:
    method _measure_memory (line 232) | def _measure_memory(self, func: Callable[[], None]) -> [Memory, Memory...

FILE: transformers/benchmark/benchmark_utils.py
  function separate_process_wrapper_fn (line 76) | def separate_process_wrapper_fn(func: Callable[[], None], do_multi_proce...
  function is_memory_tracing_enabled (line 112) | def is_memory_tracing_enabled():
  class Frame (line 117) | class Frame(NamedTuple):
  class UsedMemoryState (line 135) | class UsedMemoryState(NamedTuple):
  class Memory (line 151) | class Memory(NamedTuple):
    method __repr__ (line 161) | def __repr__(self) -> str:
  class MemoryState (line 165) | class MemoryState(NamedTuple):
  class MemorySummary (line 181) | class MemorySummary(NamedTuple):
  function measure_peak_memory_cpu (line 204) | def measure_peak_memory_cpu(function: Callable[[], None], interval=0.5, ...
  function start_memory_tracing (line 323) | def start_memory_tracing(
  function stop_memory_tracing (line 475) | def stop_memory_tracing(
  function bytes_to_mega_bytes (line 591) | def bytes_to_mega_bytes(memory_amount: int) -> int:
  class Benchmark (line 596) | class Benchmark(ABC):
    method __init__ (line 606) | def __init__(self, args: BenchmarkArguments = None, configs: Pretraine...
    method print_fn (line 633) | def print_fn(self):
    method framework_version (line 649) | def framework_version(self):
    method _inference_speed (line 653) | def _inference_speed(self, model_name: str, batch_size: int, sequence_...
    method _train_speed (line 657) | def _train_speed(self, model_name: str, batch_size: int, sequence_leng...
    method _inference_memory (line 661) | def _inference_memory(
    method _train_memory (line 667) | def _train_memory(
    method inference_speed (line 672) | def inference_speed(self, *args, **kwargs) -> float:
    method train_speed (line 675) | def train_speed(self, *args, **kwargs) -> float:
    method inference_memory (line 678) | def inference_memory(self, *args, **kwargs) -> [Memory, Optional[Memor...
    method train_memory (line 681) | def train_memory(self, *args, **kwargs) -> [Memory, Optional[MemorySum...
    method run (line 684) | def run(self):
    method environment_info (line 784) | def environment_info(self):
    method print_results (line 842) | def print_results(self, result_dict, type_label):
    method print_memory_trace_statistics (line 864) | def print_memory_trace_statistics(self, summary: MemorySummary):
    method save_to_csv (line 888) | def save_to_csv(self, result_dict, filename):

FILE: transformers/commands/__init__.py
  class BaseTransformersCLICommand (line 19) | class BaseTransformersCLICommand(ABC):
    method register_subcommand (line 22) | def register_subcommand(parser: ArgumentParser):
    method run (line 26) | def run(self):

FILE: transformers/commands/add_new_model.py
  function add_new_model_command_factory (line 37) | def add_new_model_command_factory(args: Namespace):
  class AddNewModelCommand (line 41) | class AddNewModelCommand(BaseTransformersCLICommand):
    method register_subcommand (line 43) | def register_subcommand(parser: ArgumentParser):
    method __init__ (line 52) | def __init__(self, testing: bool, testing_file: str, path=None, *args):
    method run (line 57) | def run(self):

FILE: transformers/commands/add_new_model_like.py
  class ModelPatterns (line 41) | class ModelPatterns:
    method __post_init__ (line 86) | def __post_init__(self):
  function is_empty_line (line 119) | def is_empty_line(line: str) -> bool:
  function find_indent (line 126) | def find_indent(line: str) -> int:
  function parse_module_content (line 136) | def parse_module_content(content: str) -> List[str]:
  function extract_block (line 176) | def extract_block(content: str, indent_level: int = 0) -> str:
  function add_content_to_text (line 226) | def add_content_to_text(
  function add_content_to_file (line 284) | def add_content_to_file(
  function replace_model_patterns (line 322) | def replace_model_patterns(
  function simplify_replacements (line 388) | def simplify_replacements(replacements):
  function get_module_from_file (line 425) | def get_module_from_file(module_file: Union[str, os.PathLike]) -> str:
  function remove_attributes (line 455) | def remove_attributes(obj, target_attr):
  function duplicate_module (line 495) | def duplicate_module(
  function filter_framework_files (line 595) | def filter_framework_files(
  function get_model_files (line 628) | def get_model_files(model_type: str, frameworks: Optional[List[str]] = N...
  function find_base_model_checkpoint (line 673) | def find_base_model_checkpoint(
  function get_default_frameworks (line 707) | def get_default_frameworks():
  function retrieve_model_classes (line 724) | def retrieve_model_classes(model_type: str, frameworks: Optional[List[st...
  function retrieve_info_for_model (line 765) | def retrieve_info_for_model(model_type, frameworks: Optional[List[str]] ...
  function clean_frameworks_in_init (line 850) | def clean_frameworks_in_init(
  function add_model_to_main_init (line 929) | def add_model_to_main_init(
  function insert_tokenizer_in_auto_module (line 1018) | def insert_tokenizer_in_auto_module(old_model_patterns: ModelPatterns, n...
  function add_model_to_auto_classes (line 1080) | def add_model_to_auto_classes(
  function duplicate_doc_file (line 1173) | def duplicate_doc_file(
  function create_new_model_like (line 1271) | def create_new_model_like(
  function add_new_model_like_command_factory (line 1447) | def add_new_model_like_command_factory(args: Namespace):
  class AddNewModelLikeCommand (line 1451) | class AddNewModelLikeCommand(BaseTransformersCLICommand):
    method register_subcommand (line 1453) | def register_subcommand(parser: ArgumentParser):
    method __init__ (line 1463) | def __init__(self, config_file=None, path_to_repo=None, *args):
    method run (line 1483) | def run(self):
  function get_user_field (line 1501) | def get_user_field(
  function convert_to_bool (line 1553) | def convert_to_bool(x: str) -> bool:
  function get_user_input (line 1564) | def get_user_input():

FILE: transformers/commands/convert.py
  function convert_command_factory (line 21) | def convert_command_factory(args: Namespace):
  class ConvertCommand (line 38) | class ConvertCommand(BaseTransformersCLICommand):
    method register_subcommand (line 40) | def register_subcommand(parser: ArgumentParser):
    method __init__ (line 67) | def __init__(
    method run (line 85) | def run(self):

FILE: transformers/commands/download.py
  function download_command_factory (line 20) | def download_command_factory(args):
  class DownloadCommand (line 24) | class DownloadCommand(BaseTransformersCLICommand):
    method register_subcommand (line 26) | def register_subcommand(parser: ArgumentParser):
    method __init__ (line 37) | def __init__(self, model: str, cache: str, force: bool):
    method run (line 42) | def run(self):

FILE: transformers/commands/env.py
  function info_command_factory (line 26) | def info_command_factory(_):
  class EnvironmentCommand (line 30) | class EnvironmentCommand(BaseTransformersCLICommand):
    method register_subcommand (line 32) | def register_subcommand(parser: ArgumentParser):
    method run (line 36) | def run(self):
    method format_dict (line 103) | def format_dict(d):

FILE: transformers/commands/lfs.py
  class LfsCommands (line 36) | class LfsCommands(BaseTransformersCLICommand):
    method register_subcommand (line 55) | def register_subcommand(parser: ArgumentParser):
  class LfsEnableCommand (line 75) | class LfsEnableCommand:
    method __init__ (line 76) | def __init__(self, args):
    method run (line 79) | def run(self):
  function write_msg (line 98) | def write_msg(msg: Dict):
  function read_msg (line 105) | def read_msg() -> Optional[Dict]:
  class FileSlice (line 120) | class FileSlice(AbstractContextManager):
    method __init__ (line 127) | def __init__(self, filepath: str, seek_from: int, read_limit: int):
    method __enter__ (line 133) | def __enter__(self):
    method __len__ (line 138) | def __len__(self):
    method read (line 142) | def read(self, n=-1):
    method __iter__ (line 150) | def __iter__(self):
    method __exit__ (line 153) | def __exit__(self, *args):
  class LfsUploadCommand (line 157) | class LfsUploadCommand:
    method __init__ (line 158) | def __init__(self, args):
    method run (line 161) | def run(self):

FILE: transformers/commands/pt_to_tf.py
  function convert_command_factory (line 57) | def convert_command_factory(args: Namespace):
  class PTtoTFCommand (line 75) | class PTtoTFCommand(BaseTransformersCLICommand):
    method register_subcommand (line 77) | def register_subcommand(parser: ArgumentParser):
    method find_pt_tf_differences (line 140) | def find_pt_tf_differences(pt_outputs, tf_outputs):
    method __init__ (line 177) | def __init__(
    method get_inputs (line 199) | def get_inputs(self, pt_model, tf_dummy_inputs, config):
    method run (line 269) | def run(self):

FILE: transformers/commands/run.py
  function try_infer_format_from_ext (line 25) | def try_infer_format_from_ext(path: str):
  function run_command_factory (line 39) | def run_command_factory(args):
  class RunCommand (line 58) | class RunCommand(BaseTransformersCLICommand):
    method __init__ (line 59) | def __init__(self, nlp: Pipeline, reader: PipelineDataFormat):
    method register_subcommand (line 64) | def register_subcommand(parser: ArgumentParser):
    method run (line 95) | def run(self):

FILE: transformers/commands/serving.py
  function Body (line 34) | def Body(*x, **y):
  function serve_command_factory (line 43) | def serve_command_factory(args: Namespace):
  class ServeModelInfoResult (line 59) | class ServeModelInfoResult(BaseModel):
  class ServeTokenizeResult (line 67) | class ServeTokenizeResult(BaseModel):
  class ServeDeTokenizeResult (line 76) | class ServeDeTokenizeResult(BaseModel):
  class ServeForwardResult (line 84) | class ServeForwardResult(BaseModel):
  class ServeCommand (line 92) | class ServeCommand(BaseTransformersCLICommand):
    method register_subcommand (line 94) | def register_subcommand(parser: ArgumentParser):
    method __init__ (line 124) | def __init__(self, pipeline: Pipeline, host: str, port: int, workers: ...
    method run (line 173) | def run(self):
    method model_info (line 176) | def model_info(self):
    method tokenize (line 179) | def tokenize(self, text_input: str = Body(None, embed=True), return_id...
    method detokenize (line 197) | def detokenize(
    method forward (line 214) | async def forward(self, inputs=Body(None, embed=True)):

FILE: transformers/commands/train.py
  function train_command_factory (line 32) | def train_command_factory(args: Namespace):
  class TrainCommand (line 41) | class TrainCommand(BaseTransformersCLICommand):
    method register_subcommand (line 43) | def register_subcommand(parser: ArgumentParser):
    method __init__ (line 93) | def __init__(self, args: Namespace):
    method run (line 138) | def run(self):
    method run_torch (line 143) | def run_torch(self):
    method run_tf (line 146) | def run_tf(self):

FILE: transformers/commands/transformers_cli.py
  function main (line 30) | def main():

FILE: transformers/commands/user.py
  class UserCommands (line 25) | class UserCommands(BaseTransformersCLICommand):
    method register_subcommand (line 27) | def register_subcommand(parser: ArgumentParser):
  class ANSI (line 56) | class ANSI:
    method bold (line 67) | def bold(cls, s):
    method red (line 71) | def red(cls, s):
    method gray (line 75) | def gray(cls, s):
  function tabulate (line 79) | def tabulate(rows: List[List[Union[str, int]]], headers: List[str]) -> str:
  class BaseUserCommand (line 96) | class BaseUserCommand:
    method __init__ (line 97) | def __init__(self, args):
  class LoginCommand (line 101) | class LoginCommand(BaseUserCommand):
    method run (line 102) | def run(self):
  class WhoamiCommand (line 112) | class WhoamiCommand(BaseUserCommand):
    method run (line 113) | def run(self):
  class LogoutCommand (line 135) | class LogoutCommand(BaseUserCommand):
    method run (line 136) | def run(self):
  class RepoCreateCommand (line 146) | class RepoCreateCommand(BaseUserCommand):
    method run (line 147) | def run(self):

FILE: transformers/configuration_utils.py
  class PretrainedConfig (line 49) | class PretrainedConfig(PushToHubMixin):
    method __setattr__ (line 253) | def __setattr__(self, key, value):
    method __getattribute__ (line 258) | def __getattribute__(self, key):
    method __init__ (line 263) | def __init__(self, **kwargs):
    method name_or_path (line 395) | def name_or_path(self) -> str:
    method name_or_path (line 399) | def name_or_path(self, value):
    method use_return_dict (line 403) | def use_return_dict(self) -> bool:
    method num_labels (line 411) | def num_labels(self) -> int:
    method num_labels (line 418) | def num_labels(self, num_labels: int):
    method save_pretrained (line 423) | def save_pretrained(self, save_directory: Union[str, os.PathLike], pus...
    method from_pretrained (line 470) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
    method get_config_dict (line 557) | def get_config_dict(
    method _get_config_dict (line 588) | def _get_config_dict(
    method from_dict (line 678) | def from_dict(cls, config_dict: Dict[str, Any], **kwargs) -> "Pretrain...
    method from_json_file (line 732) | def from_json_file(cls, json_file: Union[str, os.PathLike]) -> "Pretra...
    method _dict_from_json_file (line 748) | def _dict_from_json_file(cls, json_file: Union[str, os.PathLike]):
    method __eq__ (line 753) | def __eq__(self, other):
    method __repr__ (line 756) | def __repr__(self):
    method to_diff_dict (line 759) | def to_diff_dict(self) -> Dict[str, Any]:
    method to_dict (line 798) | def to_dict(self) -> Dict[str, Any]:
    method to_json_string (line 827) | def to_json_string(self, use_diff: bool = True) -> str:
    method to_json_file (line 845) | def to_json_file(self, json_file_path: Union[str, os.PathLike], use_di...
    method update (line 859) | def update(self, config_dict: Dict[str, Any]):
    method update_from_string (line 869) | def update_from_string(self, update_str: str):
    method dict_torch_dtype_to_str (line 907) | def dict_torch_dtype_to_str(self, d: Dict[str, Any]) -> None:
    method register_for_auto_class (line 920) | def register_for_auto_class(cls, auto_class="AutoConfig"):
  function get_configuration_file (line 946) | def get_configuration_file(configuration_files: List[str]) -> str:

FILE: transformers/convert_graph_to_onnx.py
  class OnnxConverterArgumentParser (line 46) | class OnnxConverterArgumentParser(ArgumentParser):
    method __init__ (line 51) | def __init__(self):
  function generate_identified_filename (line 92) | def generate_identified_filename(filename: Path, identifier: str) -> Path:
  function check_onnxruntime_requirements (line 105) | def check_onnxruntime_requirements(minimum_version: Version):
  function ensure_valid_input (line 134) | def ensure_valid_input(model, tokens, input_names):
  function infer_shapes (line 162) | def infer_shapes(nlp: Pipeline, framework: str) -> Tuple[List[str], List...
  function load_graph_from_args (line 226) | def load_graph_from_args(
  function convert_pytorch (line 257) | def convert_pytorch(nlp: Pipeline, opset: int, output: Path, use_externa...
  function convert_tensorflow (line 312) | def convert_tensorflow(nlp: Pipeline, opset: int, output: Path):
  function convert (line 352) | def convert(
  function optimize (line 402) | def optimize(onnx_model_path: Path) -> Path:
  function quantize (line 427) | def quantize(onnx_model_path: Path) -> Path:
  function verify (line 502) | def verify(path: Path):

FILE: transformers/convert_pytorch_checkpoint_to_tf2.py
  function convert_pt_checkpoint_to_tf (line 304) | def convert_pt_checkpoint_to_tf(
  function convert_all_pt_checkpoints_to_tf (line 351) | def convert_all_pt_checkpoints_to_tf(

FILE: transformers/convert_slow_tokenizer.py
  class SentencePieceExtractor (line 31) | class SentencePieceExtractor:
    method __init__ (line 36) | def __init__(self, model: str):
    method extract (line 43) | def extract(self, vocab_scores=None) -> Tuple[Dict[str, int], List[Tup...
  function check_number_comma (line 68) | def check_number_comma(piece: str) -> bool:
  class Converter (line 72) | class Converter:
    method __init__ (line 73) | def __init__(self, original_tokenizer):
    method converted (line 76) | def converted(self) -> Tokenizer:
  class BertConverter (line 80) | class BertConverter(Converter):
    method converted (line 81) | def converted(self) -> Tokenizer:
  class SplinterConverter (line 119) | class SplinterConverter(Converter):
    method converted (line 120) | def converted(self) -> Tokenizer:
  class FunnelConverter (line 169) | class FunnelConverter(Converter):
    method converted (line 170) | def converted(self) -> Tokenizer:
  class MPNetConverter (line 208) | class MPNetConverter(Converter):
    method converted (line 209) | def converted(self) -> Tokenizer:
  class OpenAIGPTConverter (line 247) | class OpenAIGPTConverter(Converter):
    method converted (line 248) | def converted(self) -> Tokenizer:
  class GPT2Converter (line 274) | class GPT2Converter(Converter):
    method converted (line 275) | def converted(self) -> Tokenizer:
  class HerbertConverter (line 309) | class HerbertConverter(Converter):
    method converted (line 310) | def converted(self) -> Tokenizer:
  class RobertaConverter (line 340) | class RobertaConverter(Converter):
    method converted (line 341) | def converted(self) -> Tokenizer:
  class RoFormerConverter (line 369) | class RoFormerConverter(Converter):
    method converted (line 370) | def converted(self) -> Tokenizer:
  class DebertaConverter (line 408) | class DebertaConverter(Converter):
    method converted (line 409) | def converted(self) -> Tokenizer:
  class SpmConverter (line 439) | class SpmConverter(Converter):
    method __init__ (line 440) | def __init__(self, *args):
    method vocab (line 461) | def vocab(self, proto):
    method unk_id (line 464) | def unk_id(self, proto):
    method tokenizer (line 467) | def tokenizer(self, proto):
    method normalizer (line 492) | def normalizer(self, proto):
    method pre_tokenizer (line 501) | def pre_tokenizer(self, replacement, add_prefix_space):
    method post_processor (line 504) | def post_processor(self):
    method decoder (line 507) | def decoder(self, replacement, add_prefix_space):
    method converted (line 510) | def converted(self) -> Tokenizer:
  class AlbertConverter (line 532) | class AlbertConverter(SpmConverter):
    method vocab (line 533) | def vocab(self, proto):
    method normalizer (line 539) | def normalizer(self, proto):
    method post_processor (line 555) | def post_processor(self):
  class BarthezConverter (line 566) | class BarthezConverter(SpmConverter):
    method unk_id (line 567) | def unk_id(self, proto):
    method post_processor (line 571) | def post_processor(self):
  class CamembertConverter (line 582) | class CamembertConverter(SpmConverter):
    method vocab (line 583) | def vocab(self, proto):
    method unk_id (line 596) | def unk_id(self, proto):
    method post_processor (line 600) | def post_processor(self):
  class DebertaV2Converter (line 611) | class DebertaV2Converter(SpmConverter):
    method pre_tokenizer (line 612) | def pre_tokenizer(self, replacement, add_prefix_space):
    method normalizer (line 619) | def normalizer(self, proto):
    method post_processor (line 632) | def post_processor(self):
  class MBartConverter (line 643) | class MBartConverter(SpmConverter):
    method vocab (line 644) | def vocab(self, proto):
    method unk_id (line 682) | def unk_id(self, proto):
    method post_processor (line 685) | def post_processor(self):
  class MBart50Converter (line 696) | class MBart50Converter(SpmConverter):
    method vocab (line 697) | def vocab(self, proto):
    method unk_id (line 711) | def unk_id(self, proto):
    method post_processor (line 714) | def post_processor(self):
  class NllbConverter (line 725) | class NllbConverter(SpmConverter):
    method vocab (line 726) | def vocab(self, proto):
    method unk_id (line 742) | def unk_id(self, proto):
    method post_processor (line 745) | def post_processor(self):
  class XLMRobertaConverter (line 756) | class XLMRobertaConverter(SpmConverter):
    method vocab (line 757) | def vocab(self, proto):
    method unk_id (line 768) | def unk_id(self, proto):
    method post_processor (line 772) | def post_processor(self):
  class XLNetConverter (line 783) | class XLNetConverter(SpmConverter):
    method vocab (line 784) | def vocab(self, proto):
    method normalizer (line 790) | def normalizer(self, proto):
    method post_processor (line 806) | def post_processor(self):
  class ReformerConverter (line 817) | class ReformerConverter(SpmConverter):
  class RemBertConverter (line 821) | class RemBertConverter(SpmConverter):
    method normalizer (line 823) | def normalizer(self, proto):
    method post_processor (line 839) | def post_processor(self):
  class BertGenerationConverter (line 850) | class BertGenerationConverter(SpmConverter):
  class PegasusConverter (line 854) | class PegasusConverter(SpmConverter):
    method vocab (line 855) | def vocab(self, proto):
    method unk_id (line 874) | def unk_id(self, proto):
    method pre_tokenizer (line 877) | def pre_tokenizer(self, replacement, add_prefix_space):
    method post_processor (line 885) | def post_processor(self):
  class T5Converter (line 893) | class T5Converter(SpmConverter):
    method vocab (line 894) | def vocab(self, proto):
    method post_processor (line 900) | def post_processor(self):
  class WhisperConverter (line 910) | class WhisperConverter(Converter):
    method converted (line 911) | def converted(self) -> Tokenizer:
  class BigBirdConverter (line 946) | class BigBirdConverter(SpmConverter):
    method post_processor (line 947) | def post_processor(self):
  class CLIPConverter (line 958) | class CLIPConverter(Converter):
    method converted (line 959) | def converted(self) -> Tokenizer:
  class LayoutLMv2Converter (line 1001) | class LayoutLMv2Converter(Converter):
    method converted (line 1002) | def converted(self) -> Tokenizer:
  class BlenderbotConverter (line 1040) | class BlenderbotConverter(Converter):
    method converted (line 1041) | def converted(self) -> Tokenizer:
  class XGLMConverter (line 1069) | class XGLMConverter(SpmConverter):
    method vocab (line 1070) | def vocab(self, proto):
    method unk_id (line 1083) | def unk_id(self, proto):
    method post_processor (line 1087) | def post_processor(self):
  class LlamaConverter (line 1098) | class LlamaConverter(SpmConverter):
    method vocab (line 1101) | def vocab(self, proto):
    method unk_id (line 1110) | def unk_id(self, proto):
    method decoder (line 1114) | def decoder(self, replacement, add_prefix_space):
    method tokenizer (line 1124) | def tokenizer(self, proto):
    method normalizer (line 1149) | def normalizer(self, proto):
    method pre_tokenizer (line 1157) | def pre_tokenizer(self, replacement, add_prefix_space):
    method post_processor (line 1160) | def post_processor(self):
  class MarkupLMConverter (line 1189) | class MarkupLMConverter(Converter):
    method converted (line 1190) | def converted(self) -> Tokenizer:
  function convert_slow_tokenizer (line 1282) | def convert_slow_tokenizer(transformer_tokenizer) -> Tokenizer:

FILE: transformers/convert_slow_tokenizers_checkpoints_to_fast.py
  function convert_slow_checkpoint_to_fast (line 34) | def convert_slow_checkpoint_to_fast(tokenizer_name, checkpoint_name, dum...

FILE: transformers/convert_tf_hub_seq_to_seq_bert_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 33) | def convert_tf_checkpoint_to_pytorch(tf_hub_path, pytorch_dump_path, is_...

FILE: transformers/data/data_collator.py
  class DataCollatorMixin (line 38) | class DataCollatorMixin:
    method __call__ (line 39) | def __call__(self, features, return_tensors=None):
  function default_data_collator (line 52) | def default_data_collator(features: List[InputDataClass], return_tensors...
  class DefaultDataCollator (line 78) | class DefaultDataCollator(DataCollatorMixin):
    method __call__ (line 99) | def __call__(self, features: List[Dict[str, Any]], return_tensors=None...
  function torch_default_data_collator (line 105) | def torch_default_data_collator(features: List[InputDataClass]) -> Dict[...
  function tf_default_data_collator (line 141) | def tf_default_data_collator(features: List[InputDataClass]) -> Dict[str...
  function numpy_default_data_collator (line 182) | def numpy_default_data_collator(features: List[InputDataClass]) -> Dict[...
  class DataCollatorWithPadding (line 215) | class DataCollatorWithPadding:
    method __call__ (line 248) | def __call__(self, features: List[Dict[str, Any]]) -> Dict[str, Any]:
  class DataCollatorForTokenClassification (line 266) | class DataCollatorForTokenClassification(DataCollatorMixin):
    method torch_call (line 302) | def torch_call(self, features):
    method tf_call (line 341) | def tf_call(self, features):
    method numpy_call (line 372) | def numpy_call(self, features):
  function _torch_collate_batch (line 402) | def _torch_collate_batch(examples, tokenizer, pad_to_multiple_of: Option...
  function _tf_collate_batch (line 438) | def _tf_collate_batch(examples, tokenizer, pad_to_multiple_of: Optional[...
  function _numpy_collate_batch (line 476) | def _numpy_collate_batch(examples, tokenizer, pad_to_multiple_of: Option...
  function tolist (line 508) | def tolist(x):
  class DataCollatorForSeq2Seq (line 517) | class DataCollatorForSeq2Seq:
    method __call__ (line 559) | def __call__(self, features, return_tensors=None):
  class DataCollatorForLanguageModeling (line 607) | class DataCollatorForLanguageModeling(DataCollatorMixin):
    method __post_init__ (line 641) | def __post_init__(self):
    method tf_bernoulli (line 653) | def tf_bernoulli(shape, probability):
    method tf_mask_tokens (line 659) | def tf_mask_tokens(
    method tf_call (line 690) | def tf_call(self, examples: List[Union[List[int], Any, Dict[str, Any]]...
    method torch_call (line 729) | def torch_call(self, examples: List[Union[List[int], Any, Dict[str, An...
    method torch_mask_tokens (line 751) | def torch_mask_tokens(self, inputs: Any, special_tokens_mask: Optional...
    method numpy_call (line 784) | def numpy_call(self, examples: List[Union[List[int], Any, Dict[str, An...
    method numpy_mask_tokens (line 806) | def numpy_mask_tokens(self, inputs: Any, special_tokens_mask: Optional...
  class DataCollatorForWholeWordMask (line 845) | class DataCollatorForWholeWordMask(DataCollatorForLanguageModeling):
    method torch_call (line 860) | def torch_call(self, examples: List[Union[List[int], Any, Dict[str, An...
    method tf_call (line 888) | def tf_call(self, examples: List[Union[List[int], Any, Dict[str, Any]]...
    method numpy_call (line 918) | def numpy_call(self, examples: List[Union[List[int], Any, Dict[str, An...
    method _whole_word_mask (line 946) | def _whole_word_mask(self, input_tokens: List[str], max_predictions=512):
    method torch_mask_tokens (line 993) | def torch_mask_tokens(self, inputs: Any, mask_labels: Any) -> Tuple[An...
    method tf_mask_tokens (line 1033) | def tf_mask_tokens(self, inputs: Any, mask_labels: Any) -> Tuple[Any, ...
    method numpy_mask_tokens (line 1075) | def numpy_mask_tokens(self, inputs: Any, mask_labels: Any) -> Tuple[An...
  class DataCollatorForSOP (line 1117) | class DataCollatorForSOP(DataCollatorForLanguageModeling):
    method __init__ (line 1125) | def __init__(self, *args, **kwargs):
    method __call__ (line 1132) | def __call__(self, examples: List[Dict[str, Any]]) -> Dict[str, Any]:
    method mask_tokens (line 1155) | def mask_tokens(self, inputs: Any) -> Tuple[Any, Any, Any]:
  class DataCollatorForPermutationLanguageModeling (line 1200) | class DataCollatorForPermutationLanguageModeling(DataCollatorMixin):
    method torch_call (line 1213) | def torch_call(self, examples: List[Union[List[int], Any, Dict[str, An...
    method tf_call (line 1220) | def tf_call(self, examples: List[Union[List[int], Any, Dict[str, Any]]...
    method numpy_call (line 1227) | def numpy_call(self, examples: List[Union[List[int], Any, Dict[str, An...
    method torch_mask_tokens (line 1234) | def torch_mask_tokens(self, inputs: Any) -> Tuple[Any, Any, Any, Any]:
    method tf_mask_tokens (line 1333) | def tf_mask_tokens(self, inputs: Any) -> Tuple[Any, Any, Any, Any]:
    method numpy_mask_tokens (line 1440) | def numpy_mask_tokens(self, inputs: Any) -> Tuple[Any, Any, Any, Any]:

FILE: transformers/data/datasets/glue.py
  class GlueDataTrainingArguments (line 36) | class GlueDataTrainingArguments:
    method __post_init__ (line 61) | def __post_init__(self):
  class Split (line 65) | class Split(Enum):
  class GlueDataset (line 71) | class GlueDataset(Dataset):
    method __init__ (line 80) | def __init__(
    method __len__ (line 154) | def __len__(self):
    method __getitem__ (line 157) | def __getitem__(self, i) -> InputFeatures:
    method get_labels (line 160) | def get_labels(self):

FILE: transformers/data/datasets/language_modeling.py
  class TextDataset (line 40) | class TextDataset(Dataset):
    method __init__ (line 45) | def __init__(
    method __len__ (line 106) | def __len__(self):
    method __getitem__ (line 109) | def __getitem__(self, i) -> torch.Tensor:
  class LineByLineTextDataset (line 113) | class LineByLineTextDataset(Dataset):
    method __init__ (line 118) | def __init__(self, tokenizer: PreTrainedTokenizer, file_path: str, blo...
    method __len__ (line 139) | def __len__(self):
    method __getitem__ (line 142) | def __getitem__(self, i) -> Dict[str, torch.tensor]:
  class LineByLineWithRefDataset (line 146) | class LineByLineWithRefDataset(Dataset):
    method __init__ (line 151) | def __init__(self, tokenizer: PreTrainedTokenizer, file_path: str, blo...
    method __len__ (line 187) | def __len__(self):
    method __getitem__ (line 190) | def __getitem__(self, i) -> Dict[str, torch.tensor]:
  class LineByLineWithSOPTextDataset (line 194) | class LineByLineWithSOPTextDataset(Dataset):
    method __init__ (line 199) | def __init__(self, tokenizer: PreTrainedTokenizer, file_dir: str, bloc...
    method create_examples_from_document (line 240) | def create_examples_from_document(self, document, block_size, tokenize...
    method __len__ (line 339) | def __len__(self):
    method __getitem__ (line 342) | def __getitem__(self, i) -> Dict[str, torch.tensor]:
  class TextDatasetForNextSentencePrediction (line 346) | class TextDatasetForNextSentencePrediction(Dataset):
    method __init__ (line 351) | def __init__(
    method create_examples_from_document (line 436) | def create_examples_from_document(self, document: List[List[int]], doc...
    method __len__ (line 526) | def __len__(self):
    method __getitem__ (line 529) | def __getitem__(self, i):

FILE: transformers/data/datasets/squad.py
  class SquadDataTrainingArguments (line 38) | class SquadDataTrainingArguments:
  class Split (line 104) | class Split(Enum):
  class SquadDataset (line 109) | class SquadDataset(Dataset):
    method __init__ (line 119) | def __init__(
    method __len__ (line 194) | def __len__(self):
    method __getitem__ (line 197) | def __getitem__(self, i) -> Dict[str, torch.Tensor]:

FILE: transformers/data/metrics/__init__.py
  function simple_accuracy (line 30) | def simple_accuracy(preds, labels):
  function acc_and_f1 (line 36) | def acc_and_f1(preds, labels):
  function pearson_and_spearman (line 48) | def pearson_and_spearman(preds, labels):
  function glue_compute_metrics (line 60) | def glue_compute_metrics(task_name, preds, labels):
  function xnli_compute_metrics (line 90) | def xnli_compute_metrics(task_name, preds, labels):

FILE: transformers/data/metrics/squad_metrics.py
  function normalize_answer (line 37) | def normalize_answer(s):
  function get_tokens (line 57) | def get_tokens(s):
  function compute_exact (line 63) | def compute_exact(a_gold, a_pred):
  function compute_f1 (line 67) | def compute_f1(a_gold, a_pred):
  function get_raw_scores (line 83) | def get_raw_scores(examples, preds):
  function apply_no_ans_threshold (line 109) | def apply_no_ans_threshold(scores, na_probs, qid_to_has_ans, na_prob_thr...
  function make_eval_dict (line 120) | def make_eval_dict(exact_scores, f1_scores, qid_list=None):
  function merge_eval (line 141) | def merge_eval(main_eval, new_eval, prefix):
  function find_best_thresh_v2 (line 146) | def find_best_thresh_v2(preds, scores, na_probs, qid_to_has_ans):
  function find_all_best_thresh_v2 (line 180) | def find_all_best_thresh_v2(main_eval, preds, exact_raw, f1_raw, na_prob...
  function find_best_thresh (line 191) | def find_best_thresh(preds, scores, na_probs, qid_to_has_ans):
  function find_all_best_thresh (line 214) | def find_all_best_thresh(main_eval, preds, exact_raw, f1_raw, na_probs, ...
  function squad_evaluate (line 224) | def squad_evaluate(examples, preds, no_answer_probs=None, no_answer_prob...
  function get_final_text (line 255) | def get_final_text(pred_text, orig_text, do_lower_case, verbose_logging=...
  function _get_best_indexes (line 349) | def _get_best_indexes(logits, n_best_size):
  function _compute_softmax (line 361) | def _compute_softmax(scores):
  function compute_predictions_logits (line 384) | def compute_predictions_logits(
  function compute_predictions_log_probs (line 591) | def compute_predictions_log_probs(

FILE: transformers/data/processors/glue.py
  function glue_convert_examples_to_features (line 41) | def glue_convert_examples_to_features(
  function _tf_glue_convert_examples_to_features (line 78) | def _tf_glue_convert_examples_to_features(
  function _glue_convert_examples_to_features (line 109) | def _glue_convert_examples_to_features(
  class OutputMode (line 164) | class OutputMode(Enum):
  class MrpcProcessor (line 169) | class MrpcProcessor(DataProcessor):
    method __init__ (line 172) | def __init__(self, *args, **kwargs):
    method get_example_from_tensor_dict (line 176) | def get_example_from_tensor_dict(self, tensor_dict):
    method get_train_examples (line 185) | def get_train_examples(self, data_dir):
    method get_dev_examples (line 190) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 194) | def get_test_examples(self, data_dir):
    method get_labels (line 198) | def get_labels(self):
    method _create_examples (line 202) | def _create_examples(self, lines, set_type):
  class MnliProcessor (line 216) | class MnliProcessor(DataProcessor):
    method __init__ (line 219) | def __init__(self, *args, **kwargs):
    method get_example_from_tensor_dict (line 223) | def get_example_from_tensor_dict(self, tensor_dict):
    method get_train_examples (line 232) | def get_train_examples(self, data_dir):
    method get_dev_examples (line 236) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 240) | def get_test_examples(self, data_dir):
    method get_labels (line 244) | def get_labels(self):
    method _create_examples (line 248) | def _create_examples(self, lines, set_type):
  class MnliMismatchedProcessor (line 262) | class MnliMismatchedProcessor(MnliProcessor):
    method __init__ (line 265) | def __init__(self, *args, **kwargs):
    method get_dev_examples (line 269) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 273) | def get_test_examples(self, data_dir):
  class ColaProcessor (line 278) | class ColaProcessor(DataProcessor):
    method __init__ (line 281) | def __init__(self, *args, **kwargs):
    method get_example_from_tensor_dict (line 285) | def get_example_from_tensor_dict(self, tensor_dict):
    method get_train_examples (line 294) | def get_train_examples(self, data_dir):
    method get_dev_examples (line 298) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 302) | def get_test_examples(self, data_dir):
    method get_labels (line 306) | def get_labels(self):
    method _create_examples (line 310) | def _create_examples(self, lines, set_type):
  class Sst2Processor (line 325) | class Sst2Processor(DataProcessor):
    method __init__ (line 328) | def __init__(self, *args, **kwargs):
    method get_example_from_tensor_dict (line 332) | def get_example_from_tensor_dict(self, tensor_dict):
    method get_train_examples (line 341) | def get_train_examples(self, data_dir):
    method get_dev_examples (line 345) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 349) | def get_test_examples(self, data_dir):
    method get_labels (line 353) | def get_labels(self):
    method _create_examples (line 357) | def _create_examples(self, lines, set_type):
  class StsbProcessor (line 371) | class StsbProcessor(DataProcessor):
    method __init__ (line 374) | def __init__(self, *args, **kwargs):
    method get_example_from_tensor_dict (line 378) | def get_example_from_tensor_dict(self, tensor_dict):
    method get_train_examples (line 387) | def get_train_examples(self, data_dir):
    method get_dev_examples (line 391) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 395) | def get_test_examples(self, data_dir):
    method get_labels (line 399) | def get_labels(self):
    method _create_examples (line 403) | def _create_examples(self, lines, set_type):
  class QqpProcessor (line 417) | class QqpProcessor(DataProcessor):
    method __init__ (line 420) | def __init__(self, *args, **kwargs):
    method get_example_from_tensor_dict (line 424) | def get_example_from_tensor_dict(self, tensor_dict):
    method get_train_examples (line 433) | def get_train_examples(self, data_dir):
    method get_dev_examples (line 437) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 441) | def get_test_examples(self, data_dir):
    method get_labels (line 445) | def get_labels(self):
    method _create_examples (line 449) | def _create_examples(self, lines, set_type):
  class QnliProcessor (line 469) | class QnliProcessor(DataProcessor):
    method __init__ (line 472) | def __init__(self, *args, **kwargs):
    method get_example_from_tensor_dict (line 476) | def get_example_from_tensor_dict(self, tensor_dict):
    method get_train_examples (line 485) | def get_train_examples(self, data_dir):
    method get_dev_examples (line 489) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 493) | def get_test_examples(self, data_dir):
    method get_labels (line 497) | def get_labels(self):
    method _create_examples (line 501) | def _create_examples(self, lines, set_type):
  class RteProcessor (line 515) | class RteProcessor(DataProcessor):
    method __init__ (line 518) | def __init__(self, *args, **kwargs):
    method get_example_from_tensor_dict (line 522) | def get_example_from_tensor_dict(self, tensor_dict):
    method get_train_examples (line 531) | def get_train_examples(self, data_dir):
    method get_dev_examples (line 535) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 539) | def get_test_examples(self, data_dir):
    method get_labels (line 543) | def get_labels(self):
    method _create_examples (line 547) | def _create_examples(self, lines, set_type):
  class WnliProcessor (line 561) | class WnliProcessor(DataProcessor):
    method __init__ (line 564) | def __init__(self, *args, **kwargs):
    method get_example_from_tensor_dict (line 568) | def get_example_from_tensor_dict(self, tensor_dict):
    method get_train_examples (line 577) | def get_train_examples(self, data_dir):
    method get_dev_examples (line 581) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 585) | def get_test_examples(self, data_dir):
    method get_labels (line 589) | def get_labels(self):
    method _create_examples (line 593) | def _create_examples(self, lines, set_type):

FILE: transformers/data/processors/squad.py
  function _improve_answer_span (line 43) | def _improve_answer_span(doc_tokens, input_start, input_end, tokenizer, ...
  function _check_is_max_context (line 56) | def _check_is_max_context(doc_spans, cur_span_index, position):
  function _new_check_is_max_context (line 76) | def _new_check_is_max_context(doc_spans, cur_span_index, position):
  function _is_whitespace (line 98) | def _is_whitespace(c):
  function squad_convert_example_to_features (line 104) | def squad_convert_example_to_features(
  function squad_convert_example_to_features_init (line 311) | def squad_convert_example_to_features_init(tokenizer_for_convert: PreTra...
  function squad_convert_examples_to_features (line 316) | def squad_convert_examples_to_features(
  class SquadProcessor (line 541) | class SquadProcessor(DataProcessor):
    method _get_example_from_tensor_dict (line 550) | def _get_example_from_tensor_dict(self, tensor_dict, evaluate=False):
    method get_examples_from_dataset (line 574) | def get_examples_from_dataset(self, dataset, evaluate=False):
    method get_train_examples (line 607) | def get_train_examples(self, data_dir, filename=None):
    method get_dev_examples (line 629) | def get_dev_examples(self, data_dir, filename=None):
    method _create_examples (line 650) | def _create_examples(self, input_data, set_type):
  class SquadV1Processor (line 687) | class SquadV1Processor(SquadProcessor):
  class SquadV2Processor (line 692) | class SquadV2Processor(SquadProcessor):
  class SquadExample (line 697) | class SquadExample:
    method __init__ (line 712) | def __init__(
  class SquadFeatures (line 760) | class SquadFeatures:
    method __init__ (line 787) | def __init__(
  class SquadResult (line 827) | class SquadResult:
    method __init__ (line 837) | def __init__(self, unique_id, start_logits, end_logits, start_top_inde...

FILE: transformers/data/processors/utils.py
  class InputExample (line 30) | class InputExample:
    method to_json_string (line 49) | def to_json_string(self):
  class InputFeatures (line 55) | class InputFeatures:
    method to_json_string (line 75) | def to_json_string(self):
  class DataProcessor (line 80) | class DataProcessor:
    method get_example_from_tensor_dict (line 83) | def get_example_from_tensor_dict(self, tensor_dict):
    method get_train_examples (line 93) | def get_train_examples(self, data_dir):
    method get_dev_examples (line 97) | def get_dev_examples(self, data_dir):
    method get_test_examples (line 101) | def get_test_examples(self, data_dir):
    method get_labels (line 105) | def get_labels(self):
    method tfds_map (line 109) | def tfds_map(self, example):
    method _read_tsv (line 119) | def _read_tsv(cls, input_file, quotechar=None):
  class SingleSentenceClassificationProcessor (line 125) | class SingleSentenceClassificationProcessor(DataProcessor):
    method __init__ (line 128) | def __init__(self, labels=None, examples=None, mode="classification", ...
    method __len__ (line 134) | def __len__(self):
    method __getitem__ (line 137) | def __getitem__(self, idx):
    method create_from_csv (line 143) | def create_from_csv(
    method create_from_examples (line 160) | def create_from_examples(cls, texts_or_text_and_labels, labels=None, *...
    method add_examples_from_csv (line 165) | def add_examples_from_csv(
    method add_examples (line 195) | def add_examples(
    method get_features (line 232) | def get_features(

FILE: transformers/data/processors/xnli.py
  class XnliProcessor (line 28) | class XnliProcessor(DataProcessor):
    method __init__ (line 34) | def __init__(self, language, train_language=None):
    method get_train_examples (line 38) | def get_train_examples(self, data_dir):
    method get_test_examples (line 59) | def get_test_examples(self, data_dir):
    method get_labels (line 82) | def get_labels(self):

FILE: transformers/debug_utils.py
  class DebugUnderflowOverflow (line 27) | class DebugUnderflowOverflow:
    method __init__ (line 145) | def __init__(self, model, max_frames_to_save=21, trace_batch_nums=[], ...
    method save_frame (line 162) | def save_frame(self, frame=None):
    method expand_frame (line 168) | def expand_frame(self, line):
    method trace_frames (line 171) | def trace_frames(self):
    method reset_saved_frames (line 175) | def reset_saved_frames(self):
    method dump_saved_frames (line 178) | def dump_saved_frames(self):
    method analyse_model (line 186) | def analyse_model(self):
    method analyse_variable (line 194) | def analyse_variable(self, var, ctx):
    method batch_start_frame (line 204) | def batch_start_frame(self):
    method batch_end_frame (line 208) | def batch_end_frame(self):
    method create_frame (line 211) | def create_frame(self, module, input, output):
    method register_forward_hook (line 239) | def register_forward_hook(self):
    method _register_forward_hook (line 242) | def _register_forward_hook(self, module):
    method forward_hook (line 245) | def forward_hook(self, module, input, output):
  function get_abs_min_max (line 293) | def get_abs_min_max(var, ctx):
  function detect_overflow (line 298) | def detect_overflow(var, ctx):
  class DebugOption (line 344) | class DebugOption(ExplicitEnum):

FILE: transformers/deepspeed.py
  function is_deepspeed_available (line 32) | def is_deepspeed_available():
  class HfDeepSpeedConfig (line 44) | class HfDeepSpeedConfig(DeepSpeedConfig):
    method __init__ (line 61) | def __init__(self, config_file_or_dict):
  class HfTrainerDeepSpeedConfig (line 69) | class HfTrainerDeepSpeedConfig(HfDeepSpeedConfig):
    method __init__ (line 75) | def __init__(self, config_file_or_dict):
    method dtype (line 80) | def dtype(self):
    method is_auto (line 85) | def is_auto(self, ds_key_long):
    method fill_match (line 92) | def fill_match(self, ds_key_long, hf_val, hf_key=None, must_match=True):
    method trainer_config_process (line 120) | def trainer_config_process(self, args):
    method trainer_config_finalize (line 178) | def trainer_config_finalize(self, args, model, num_training_steps):
  function set_hf_deepspeed_config (line 230) | def set_hf_deepspeed_config(hf_deepspeed_config_obj):
  function unset_hf_deepspeed_config (line 238) | def unset_hf_deepspeed_config():
  function is_deepspeed_zero3_enabled (line 244) | def is_deepspeed_zero3_enabled():
  function deepspeed_config (line 251) | def deepspeed_config():
  function deepspeed_optim_sched (line 258) | def deepspeed_optim_sched(trainer, hf_deepspeed_config, args, num_traini...
  function deepspeed_init (line 316) | def deepspeed_init(trainer, num_training_steps, inference=False):
  function deepspeed_load_checkpoint (line 371) | def deepspeed_load_checkpoint(deepspeed_engine, checkpoint_path):

FILE: transformers/dependency_versions_check.py
  function dep_version_check (line 46) | def dep_version_check(pkg, hint=None):

FILE: transformers/dynamic_module_utils.py
  function init_hf_modules (line 40) | def init_hf_modules():
  function create_dynamic_module (line 56) | def create_dynamic_module(name: Union[str, os.PathLike]):
  function get_relative_imports (line 72) | def get_relative_imports(module_file):
  function get_relative_import_files (line 90) | def get_relative_import_files(module_file):
  function get_imports (line 119) | def get_imports(filename):
  function check_imports (line 138) | def check_imports(filename):
  function get_class_in_module (line 159) | def get_class_in_module(class_name, module_path):
  function get_cached_module_file (line 168) | def get_cached_module_file(
  function get_class_from_dynamic_module (line 333) | def get_class_from_dynamic_module(
  function custom_object_save (line 446) | def custom_object_save(obj, folder, config=None):
  function _raise_timeout_error (line 519) | def _raise_timeout_error(signum, frame):
  function resolve_trust_remote_code (line 530) | def resolve_trust_remote_code(trust_remote_code, model_name, has_local_c...

FILE: transformers/feature_extraction_sequence_utils.py
  class SequenceFeatureExtractor (line 29) | class SequenceFeatureExtractor(FeatureExtractionMixin):
    method __init__ (line 42) | def __init__(self, feature_size: int, sampling_rate: int, padding_valu...
    method pad (line 52) | def pad(
    method _pad (line 226) | def _pad(
    method _truncate (line 298) | def _truncate(
    method _get_padding_strategies (line 341) | def _get_padding_strategies(self, padding=False, max_length=None):

FILE: transformers/feature_extraction_utils.py
  class BatchFeature (line 60) | class BatchFeature(UserDict):
    method __init__ (line 75) | def __init__(self, data: Optional[Dict[str, Any]] = None, tensor_type:...
    method __getitem__ (line 79) | def __getitem__(self, item: str) -> Union[Any]:
    method __getattr__ (line 89) | def __getattr__(self, item: str):
    method __getstate__ (line 95) | def __getstate__(self):
    method __setstate__ (line 98) | def __setstate__(self, state):
    method keys (line 103) | def keys(self):
    method values (line 107) | def values(self):
    method items (line 111) | def items(self):
    method convert_to_tensors (line 114) | def convert_to_tensors(self, tensor_type: Optional[Union[str, TensorTy...
    method to (line 187) | def to(self, *args, **kwargs) -> "BatchFeature":
  class FeatureExtractionMixin (line 232) | class FeatureExtractionMixin(PushToHubMixin):
    method __init__ (line 240) | def __init__(self, **kwargs):
    method _set_processor_class (line 252) | def _set_processor_class(self, processor_class: str):
    method from_pretrained (line 257) | def from_pretrained(
    method save_pretrained (line 342) | def save_pretrained(self, save_directory: Union[str, os.PathLike], pus...
    method get_feature_extractor_dict (line 391) | def get_feature_extractor_dict(
    method from_dict (line 489) | def from_dict(cls, feature_extractor_dict: Dict[str, Any], **kwargs) -...
    method to_dict (line 525) | def to_dict(self) -> Dict[str, Any]:
    method from_json_file (line 538) | def from_json_file(cls, json_file: Union[str, os.PathLike]) -> PreTrai...
    method to_json_string (line 556) | def to_json_string(self) -> str:
    method to_json_file (line 577) | def to_json_file(self, json_file_path: Union[str, os.PathLike]):
    method __repr__ (line 588) | def __repr__(self):
    method register_for_auto_class (line 592) | def register_for_auto_class(cls, auto_class="AutoFeatureExtractor"):

FILE: transformers/generation/beam_constraints.py
  class Constraint (line 5) | class Constraint(ABC):
    method __init__ (line 20) | def __init__(self):
    method test (line 24) | def test(self):
    method advance (line 49) | def advance(self):
    method does_advance (line 61) | def does_advance(self, token_id: int):
    method update (line 70) | def update(self, token_id: int):
    method reset (line 95) | def reset(self):
    method remaining (line 105) | def remaining(self):
    method copy (line 114) | def copy(self, stateful=False):
  class PhrasalConstraint (line 129) | class PhrasalConstraint(Constraint):
    method __init__ (line 138) | def __init__(self, token_ids: List[int]):
    method advance (line 152) | def advance(self):
    method does_advance (line 157) | def does_advance(self, token_id: int):
    method update (line 166) | def update(self, token_id: int):
    method reset (line 186) | def reset(self):
    method remaining (line 190) | def remaining(self):
    method copy (line 193) | def copy(self, stateful=False):
  class DisjunctiveTrie (line 204) | class DisjunctiveTrie:
    method __init__ (line 205) | def __init__(self, nested_token_ids: List[List[int]], no_subsets=True):
    method next_tokens (line 228) | def next_tokens(self, current_seq):
    method reached_leaf (line 241) | def reached_leaf(self, current_seq):
    method count_leaves (line 246) | def count_leaves(self, root):
    method has_subsets (line 253) | def has_subsets(self, trie, nested_token_ids):
  class DisjunctiveConstraint (line 261) | class DisjunctiveConstraint(Constraint):
    method __init__ (line 270) | def __init__(self, nested_token_ids: List[List[int]]):
    method advance (line 292) | def advance(self):
    method does_advance (line 300) | def does_advance(self, token_id: int):
    method update (line 308) | def update(self, token_id: int):
    method reset (line 328) | def reset(self):
    method remaining (line 332) | def remaining(self):
    method copy (line 339) | def copy(self, stateful=False):
  class ConstraintListState (line 350) | class ConstraintListState:
    method __init__ (line 359) | def __init__(self, constraints: List[Constraint]):
    method init_state (line 369) | def init_state(self):
    method get_bank (line 374) | def get_bank(self):
    method advance (line 382) | def advance(self):
    method reset (line 417) | def reset(self, token_ids: Optional[List[int]]):
    method add (line 432) | def add(self, token_id: int):
    method copy (line 508) | def copy(self, stateful=True):

FILE: transformers/generation/beam_search.py
  class BeamScorer (line 87) | class BeamScorer(ABC):
    method process (line 95) | def process(
    method finalize (line 107) | def finalize(
  class BeamSearchScorer (line 119) | class BeamSearchScorer(BeamScorer):
    method __init__ (line 158) | def __init__(
    method is_done (line 202) | def is_done(self) -> bool:
    method process (line 205) | def process(
    method finalize (line 302) | def finalize(
  class ConstrainedBeamSearchScorer (line 394) | class ConstrainedBeamSearchScorer(BeamScorer):
    method __init__ (line 431) | def __init__(
    method is_done (line 477) | def is_done(self) -> bool:
    method make_constraint_states (line 480) | def make_constraint_states(self, n):
    method check_completes_constraints (line 483) | def check_completes_constraints(self, sequence):
    method process (line 488) | def process(
    method step_sentence_constraint (line 631) | def step_sentence_constraint(
    method finalize (line 772) | def finalize(
  class BeamHypotheses (line 861) | class BeamHypotheses:
    method __init__ (line 862) | def __init__(self, num_beams: int, length_penalty: float, early_stoppi...
    method __len__ (line 879) | def __len__(self):
    method add (line 885) | def add(self, hyp: torch.LongTensor, sum_logprobs: float, beam_indices...
    method is_done (line 899) | def is_done(self, best_sum_logprobs: float, cur_len: int) -> bool:

FILE: transformers/generation/configuration_utils.py
  class GenerationConfig (line 38) | class GenerationConfig(PushToHubMixin):
    method __init__ (line 224) | def __init__(self, **kwargs):
    method __eq__ (line 302) | def __eq__(self, other):
    method __repr__ (line 314) | def __repr__(self):
    method validate (line 317) | def validate(self):
    method save_pretrained (line 325) | def save_pretrained(
    method from_pretrained (line 376) | def from_pretrained(
    method _dict_from_json_file (line 544) | def _dict_from_json_file(cls, json_file: Union[str, os.PathLike]):
    method from_dict (line 550) | def from_dict(cls, config_dict: Dict[str, Any], **kwargs) -> "Generati...
    method dict_torch_dtype_to_str (line 583) | def dict_torch_dtype_to_str(self, d: Dict[str, Any]) -> None:
    method to_diff_dict (line 595) | def to_diff_dict(self) -> Dict[str, Any]:
    method to_dict (line 618) | def to_dict(self) -> Dict[str, Any]:
    method to_json_string (line 635) | def to_json_string(self, use_diff: bool = True) -> str:
    method to_json_file (line 653) | def to_json_file(self, json_file_path: Union[str, os.PathLike], use_di...
    method from_model_config (line 668) | def from_model_config(cls, model_config: PretrainedConfig) -> "Generat...
    method update (line 696) | def update(self, **kwargs):

FILE: transformers/generation/flax_logits_process.py
  class FlaxLogitsProcessor (line 50) | class FlaxLogitsProcessor:
    method __call__ (line 54) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray) -> jnp...
  class FlaxLogitsWarper (line 61) | class FlaxLogitsWarper:
    method __call__ (line 65) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray) -> jnp...
  class FlaxLogitsProcessorList (line 72) | class FlaxLogitsProcessorList(list):
    method __call__ (line 80) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray, cur_le...
  class FlaxTemperatureLogitsWarper (line 95) | class FlaxTemperatureLogitsWarper(FlaxLogitsWarper):
    method __init__ (line 104) | def __init__(self, temperature: float):
    method __call__ (line 110) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray, cur_le...
  class FlaxTopPLogitsWarper (line 115) | class FlaxTopPLogitsWarper(FlaxLogitsWarper):
    method __init__ (line 129) | def __init__(self, top_p: float, filter_value: float = -float("Inf"), ...
    method __call__ (line 137) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray, cur_le...
  class FlaxTopKLogitsWarper (line 157) | class FlaxTopKLogitsWarper(FlaxLogitsWarper):
    method __init__ (line 170) | def __init__(self, top_k: int, filter_value: float = -float("Inf"), mi...
    method __call__ (line 177) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray, cur_le...
  class FlaxForcedBOSTokenLogitsProcessor (line 192) | class FlaxForcedBOSTokenLogitsProcessor(FlaxLogitsProcessor):
    method __init__ (line 201) | def __init__(self, bos_token_id: int):
    method __call__ (line 204) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray, cur_le...
  class FlaxForcedEOSTokenLogitsProcessor (line 214) | class FlaxForcedEOSTokenLogitsProcessor(FlaxLogitsProcessor):
    method __init__ (line 225) | def __init__(self, max_length: int, eos_token_id: int):
    method __call__ (line 229) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray, cur_le...
  class FlaxMinLengthLogitsProcessor (line 239) | class FlaxMinLengthLogitsProcessor(FlaxLogitsProcessor):
    method __init__ (line 250) | def __init__(self, min_length: int, eos_token_id: int):
    method __call__ (line 260) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray, cur_le...
  class FlaxSuppressTokensAtBeginLogitsProcessor (line 269) | class FlaxSuppressTokensAtBeginLogitsProcessor(FlaxLogitsProcessor):
    method __init__ (line 282) | def __init__(self, begin_suppress_tokens, begin_index):
    method __call__ (line 286) | def __call__(self, input_ids, scores, cur_len: int):
  class FlaxSuppressTokensLogitsProcessor (line 294) | class FlaxSuppressTokensLogitsProcessor(FlaxLogitsProcessor):
    method __init__ (line 304) | def __init__(self, suppress_tokens: list):
    method __call__ (line 307) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray, cur_le...
  class FlaxForceTokensLogitsProcessor (line 313) | class FlaxForceTokensLogitsProcessor(FlaxLogitsProcessor):
    method __init__ (line 324) | def __init__(self, force_token_map):
    method __call__ (line 335) | def __call__(self, input_ids: jnp.ndarray, scores: jnp.ndarray, cur_le...
  class FlaxWhisperTimeStampLogitsProcessor (line 361) | class FlaxWhisperTimeStampLogitsProcessor(FlaxLogitsProcessor):
    method __init__ (line 378) | def __init__(self, generate_config, model_config, decoder_input_length):
    method __call__ (line 395) | def __call__(self, input_ids, scores, cur_len):

FILE: transformers/generation/flax_utils.py
  class FlaxGreedySearchOutput (line 55) | class FlaxGreedySearchOutput(ModelOutput):
  class FlaxSampleOutput (line 69) | class FlaxSampleOutput(ModelOutput):
  class FlaxBeamSearchOutput (line 83) | class FlaxBeamSearchOutput(ModelOutput):
  class GreedyState (line 100) | class GreedyState:
  class SampleState (line 109) | class SampleState:
  class BeamSearchState (line 119) | class BeamSearchState:
  class FlaxGenerationMixin (line 129) | class FlaxGenerationMixin:
    method prepare_inputs_for_generation (line 146) | def prepare_inputs_for_generation(self, *args, **kwargs):
    method _run_loop_in_debug (line 152) | def _run_loop_in_debug(cond_fn, body_fn, init_state):
    method _prepare_encoder_decoder_kwargs_for_generation (line 161) | def _prepare_encoder_decoder_kwargs_for_generation(self, input_ids, pa...
    method _prepare_decoder_input_ids_for_generation (line 170) | def _prepare_decoder_input_ids_for_generation(
    method _get_decoder_start_token_id (line 185) | def _get_decoder_start_token_id(self, decoder_start_token_id: int = No...
    method _expand_to_num_beams (line 215) | def _expand_to_num_beams(tensor, num_beams):
    method _adapt_logits_for_beam_search (line 218) | def _adapt_logits_for_beam_search(self, logits):
    method _validate_model_class (line 225) | def _validate_model_class(self):
    method _validate_model_kwargs (line 249) | def _validate_model_kwargs(self, model_kwargs: Dict[str, Any]):
    method generate (line 267) | def generate(
    method _get_logits_warper (line 472) | def _get_logits_warper(self, generation_config: GenerationConfig) -> F...
    method _get_logits_processor (line 488) | def _get_logits_processor(
    method _merge_criteria_processor_list (line 538) | def _merge_criteria_processor_list(
    method _greedy_search (line 559) | def _greedy_search(
    method _sample (line 645) | def _sample(
    method _beam_search (line 742) | def _beam_search(

FILE: transformers/generation/logits_process.py
  class LogitsProcessor (line 51) | class LogitsProcessor:
    method __call__ (line 55) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class LogitsWarper (line 62) | class LogitsWarper:
    method __call__ (line 66) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class LogitsProcessorList (line 73) | class LogitsProcessorList(list):
    method __call__ (line 81) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class MinLengthLogitsProcessor (line 96) | class MinLengthLogitsProcessor(LogitsProcessor):
    method __init__ (line 107) | def __init__(self, min_length: int, eos_token_id: Union[int, List[int]]):
    method __call__ (line 119) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class MinNewTokensLengthLogitsProcessor (line 127) | class MinNewTokensLengthLogitsProcessor(LogitsProcessor):
    method __init__ (line 140) | def __init__(self, prompt_length_to_skip: int, min_new_tokens: int, eo...
    method __call__ (line 157) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class TemperatureLogitsWarper (line 166) | class TemperatureLogitsWarper(LogitsWarper):
    method __init__ (line 175) | def __init__(self, temperature: float):
    method __call__ (line 181) | def __call__(self, input_ids: torch.Tensor, scores: torch.Tensor) -> t...
  class RepetitionPenaltyLogitsProcessor (line 186) | class RepetitionPenaltyLogitsProcessor(LogitsProcessor):
    method __init__ (line 196) | def __init__(self, penalty: float):
    method __call__ (line 202) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class EncoderRepetitionPenaltyLogitsProcessor (line 212) | class EncoderRepetitionPenaltyLogitsProcessor(LogitsProcessor):
    method __init__ (line 223) | def __init__(self, penalty: float, encoder_input_ids: torch.LongTensor):
    method __call__ (line 230) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class TopPLogitsWarper (line 240) | class TopPLogitsWarper(LogitsWarper):
    method __init__ (line 254) | def __init__(self, top_p: float, filter_value: float = -float("Inf"), ...
    method __call__ (line 263) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class TopKLogitsWarper (line 279) | class TopKLogitsWarper(LogitsWarper):
    method __init__ (line 292) | def __init__(self, top_k: int, filter_value: float = -float("Inf"), mi...
    method __call__ (line 299) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class TypicalLogitsWarper (line 307) | class TypicalLogitsWarper(LogitsWarper):
    method __init__ (line 321) | def __init__(self, mass: float = 0.9, filter_value: float = -float("In...
    method __call__ (line 330) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class EpsilonLogitsWarper (line 355) | class EpsilonLogitsWarper(LogitsWarper):
    method __init__ (line 370) | def __init__(self, epsilon: float, filter_value: float = -float("Inf")...
    method __call__ (line 385) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class EtaLogitsWarper (line 398) | class EtaLogitsWarper(LogitsWarper):
    method __init__ (line 409) | def __init__(self, epsilon: float, filter_value: float = -float("Inf")...
    method __call__ (line 424) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  function _get_ngrams (line 439) | def _get_ngrams(ngram_size: int, prev_input_ids: torch.Tensor, num_hypos...
  function _get_generated_ngrams (line 450) | def _get_generated_ngrams(banned_ngrams, prev_input_ids, ngram_size, cur...
  function _calc_banned_ngram_tokens (line 457) | def _calc_banned_ngram_tokens(
  class NoRepeatNGramLogitsProcessor (line 474) | class NoRepeatNGramLogitsProcessor(LogitsProcessor):
    method __init__ (line 484) | def __init__(self, ngram_size: int):
    method __call__ (line 489) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class EncoderNoRepeatNGramLogitsProcessor (line 500) | class EncoderNoRepeatNGramLogitsProcessor(LogitsProcessor):
    method __init__ (line 512) | def __init__(self, encoder_ngram_size: int, encoder_input_ids: torch.L...
    method __call__ (line 523) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class NoBadWordsLogitsProcessor (line 541) | class NoBadWordsLogitsProcessor(LogitsProcessor):
    method __init__ (line 554) | def __init__(self, bad_words_ids: List[List[int]], eos_token_id: Union...
    method __call__ (line 589) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
    method _calc_static_bad_word_mask (line 598) | def _calc_static_bad_word_mask(self, scores: torch.FloatTensor) -> tor...
    method _tokens_match (line 603) | def _tokens_match(self, prev_tokens: List[int], tokens: List[int]) -> ...
    method _calc_banned_bad_words_ids (line 613) | def _calc_banned_bad_words_ids(self, prev_input_ids: List[List[int]]) ...
    method _set_scores_to_inf_for_banned_tokens (line 625) | def _set_scores_to_inf_for_banned_tokens(
  class PrefixConstrainedLogitsProcessor (line 675) | class PrefixConstrainedLogitsProcessor(LogitsProcessor):
    method __init__ (line 688) | def __init__(self, prefix_allowed_tokens_fn: Callable[[int, torch.Tens...
    method __call__ (line 692) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class HammingDiversityLogitsProcessor (line 701) | class HammingDiversityLogitsProcessor(LogitsProcessor):
    method __init__ (line 719) | def __init__(self, diversity_penalty: float, num_beams: int, num_beam_...
    method __call__ (line 732) | def __call__(
  class ForcedBOSTokenLogitsProcessor (line 761) | class ForcedBOSTokenLogitsProcessor(LogitsProcessor):
    method __init__ (line 770) | def __init__(self, bos_token_id: int):
    method __call__ (line 773) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class ForcedEOSTokenLogitsProcessor (line 782) | class ForcedEOSTokenLogitsProcessor(LogitsProcessor):
    method __init__ (line 794) | def __init__(self, max_length: int, eos_token_id: Union[int, List[int]]):
    method __call__ (line 800) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class InfNanRemoveLogitsProcessor (line 810) | class InfNanRemoveLogitsProcessor(LogitsProcessor):
    method __call__ (line 816) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class ExponentialDecayLengthPenalty (line 826) | class ExponentialDecayLengthPenalty(LogitsProcessor):
    method __init__ (line 841) | def __init__(
    method __call__ (line 853) | def __call__(self, input_ids: torch.Tensor, scores: torch.Tensor) -> t...
  class LogitNormalization (line 861) | class LogitNormalization(LogitsProcessor, LogitsWarper):
    method __call__ (line 869) | def __call__(self, input_ids: torch.Tensor, scores: torch.Tensor) -> t...
  class SuppressTokensAtBeginLogitsProcessor (line 874) | class SuppressTokensAtBeginLogitsProcessor(LogitsProcessor):
    method __init__ (line 881) | def __init__(self, begin_suppress_tokens, begin_index):
    method __call__ (line 885) | def __call__(self, input_ids, scores):
  class SuppressTokensLogitsProcessor (line 892) | class SuppressTokensLogitsProcessor(LogitsProcessor):
    method __init__ (line 896) | def __init__(self, suppress_tokens):
    method __call__ (line 899) | def __call__(self, input_ids, scores):
  class ForceTokensLogitsProcessor (line 904) | class ForceTokensLogitsProcessor(LogitsProcessor):
    method __init__ (line 909) | def __init__(self, force_token_map: List[List[int]]):
    method __call__ (line 912) | def __call__(self, input_ids, scores):
  class WhisperTimeStampLogitsProcessor (line 921) | class WhisperTimeStampLogitsProcessor(LogitsProcessor):
    method __init__ (line 938) | def __init__(self, generate_config):  # support for the kwargs
    method __call__ (line 948) | def __call__(self, input_ids, scores):

FILE: transformers/generation/stopping_criteria.py
  class StoppingCriteria (line 33) | class StoppingCriteria(ABC):
    method __call__ (line 37) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class MaxLengthCriteria (line 41) | class MaxLengthCriteria(StoppingCriteria):
    method __init__ (line 51) | def __init__(self, max_length: int):
    method __call__ (line 55) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class MaxNewTokensCriteria (line 59) | class MaxNewTokensCriteria(StoppingCriteria):
    method __init__ (line 72) | def __init__(self, start_length: int, max_new_tokens: int):
    method __call__ (line 84) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class MaxTimeCriteria (line 88) | class MaxTimeCriteria(StoppingCriteria):
    method __init__ (line 101) | def __init__(self, max_time: float, initial_timestamp: Optional[float]...
    method __call__ (line 106) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
  class StoppingCriteriaList (line 110) | class StoppingCriteriaList(list):
    method __call__ (line 112) | def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTen...
    method max_length (line 116) | def max_length(self) -> Optional[int]:
  function validate_stopping_criteria (line 125) | def validate_stopping_criteria(stopping_criteria: StoppingCriteriaList, ...

FILE: transformers/generation/streamers.py
  class BaseStreamer (line 24) | class BaseStreamer:
    method put (line 29) | def put(self, value):
    method end (line 33) | def end(self):
  class TextStreamer (line 38) | class TextStreamer(BaseStreamer):
    method __init__ (line 72) | def __init__(self, tokenizer: "AutoTokenizer", skip_prompt: bool = Fal...
    method put (line 82) | def put(self, value):
    method end (line 116) | def end(self):
    method on_finalized_text (line 130) | def on_finalized_text(self, text: str, stream_end: bool = False):
    method _is_chinese_char (line 134) | def _is_chinese_char(self, cp):
  class TextIteratorStreamer (line 159) | class TextIteratorStreamer(TextStreamer):
    method __init__ (line 205) | def __init__(
    method on_finalized_text (line 213) | def on_finalized_text(self, text: str, stream_end: bool = False):
    method __iter__ (line 219) | def __iter__(self):
    method __next__ (line 222) | def __next__(self):

FILE: transformers/generation/tf_logits_process.py
  class TFLogitsProcessor (line 53) | class TFLogitsProcessor:
    method __call__ (line 57) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFLogitsWarper (line 64) | class TFLogitsWarper:
    method __call__ (line 68) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFLogitsProcessorList (line 75) | class TFLogitsProcessorList(list):
    method __call__ (line 83) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFTemperatureLogitsWarper (line 98) | class TFTemperatureLogitsWarper(TFLogitsWarper):
    method __init__ (line 107) | def __init__(self, temperature: float):
    method __call__ (line 113) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFTopKLogitsWarper (line 118) | class TFTopKLogitsWarper(TFLogitsWarper):
    method __init__ (line 131) | def __init__(self, top_k: int, filter_value: float = -float("Inf"), mi...
    method __call__ (line 138) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFTopPLogitsWarper (line 146) | class TFTopPLogitsWarper(TFLogitsWarper):
    method __init__ (line 160) | def __init__(self, top_p: float, filter_value: float = -float("Inf"), ...
    method __call__ (line 168) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFMinLengthLogitsProcessor (line 200) | class TFMinLengthLogitsProcessor(TFLogitsProcessor):
    method __init__ (line 211) | def __init__(self, min_length: int, eos_token_id: int):
    method _apply_eos_token_mask (line 221) | def _apply_eos_token_mask(self, scores: tf.Tensor) -> tf.Tensor:
    method __call__ (line 226) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFRepetitionPenaltyLogitsProcessor (line 236) | class TFRepetitionPenaltyLogitsProcessor(TFLogitsProcessor):
    method __init__ (line 246) | def __init__(self, penalty: float):
    method _create_score_penalties (line 252) | def _create_score_penalties(self, input_ids: tf.Tensor, logits: tf.Ten...
    method __call__ (line 278) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFNoBadWordsLogitsProcessor (line 286) | class TFNoBadWordsLogitsProcessor(TFLogitsProcessor):
    method __init__ (line 298) | def __init__(self, bad_words_ids: List[List[int]], eos_token_id: int):
    method _calc_row_banned_bad_tokens (line 322) | def _calc_row_banned_bad_tokens(self, row_input_ids: tf.Tensor) -> tf....
    method __call__ (line 362) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFNoRepeatNGramLogitsProcessor (line 383) | class TFNoRepeatNGramLogitsProcessor(TFLogitsProcessor):
    method __init__ (line 393) | def __init__(self, ngram_size: int):
    method calc_banned_ngram_tokens (line 398) | def calc_banned_ngram_tokens(self, input_ids, num_hypos, cur_len):
    method __call__ (line 422) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFForcedBOSTokenLogitsProcessor (line 443) | class TFForcedBOSTokenLogitsProcessor(TFLogitsProcessor):
    method __init__ (line 452) | def __init__(self, bos_token_id: int):
    method __call__ (line 457) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFForcedEOSTokenLogitsProcessor (line 473) | class TFForcedEOSTokenLogitsProcessor(TFLogitsProcessor):
    method __init__ (line 484) | def __init__(self, max_length: int, eos_token_id: int):
    method __call__ (line 490) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFSuppressTokensAtBeginLogitsProcessor (line 506) | class TFSuppressTokensAtBeginLogitsProcessor(TFLogitsProcessor):
    method __init__ (line 513) | def __init__(self, begin_suppress_tokens, begin_index):
    method __call__ (line 517) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFSuppressTokensLogitsProcessor (line 530) | class TFSuppressTokensLogitsProcessor(TFLogitsProcessor):
    method __init__ (line 534) | def __init__(self, suppress_tokens):
    method __call__ (line 537) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...
  class TFForceTokensLogitsProcessor (line 546) | class TFForceTokensLogitsProcessor(TFLogitsProcessor):
    method __init__ (line 551) | def __init__(self, force_token_map: List[List[int]]):
    method __call__ (line 562) | def __call__(self, input_ids: tf.Tensor, scores: tf.Tensor, cur_len: i...

FILE: transformers/generation/tf_utils.py
  class TFGreedySearchDecoderOnlyOutput (line 58) | class TFGreedySearchDecoderOnlyOutput(ModelOutput):
  class TFGreedySearchEncoderDecoderOutput (line 86) | class TFGreedySearchEncoderDecoderOutput(ModelOutput):
  class TFSampleDecoderOnlyOutput (line 128) | class TFSampleDecoderOnlyOutput(ModelOutput):
  class TFSampleEncoderDecoderOutput (line 156) | class TFSampleEncoderDecoderOutput(ModelOutput):
  class TFBeamSearchDecoderOnlyOutput (line 198) | class TFBeamSearchDecoderOnlyOutput(ModelOutput):
  class TFBeamSearchEncoderDecoderOutput (line 233) | class TFBeamSearchEncoderDecoderOutput(ModelOutput):
  class TFBeamSampleDecoderOnlyOutput (line 283) | class TFBeamSampleDecoderOnlyOutput(ModelOutput):
  class TFBeamSampleEncoderDecoderOutput (line 318) | class TFBeamSampleEncoderDecoderOutput(ModelOutput):
  class TFContrastiveSearchDecoderOnlyOutput (line 367) | class TFContrastiveSearchDecoderOnlyOutput(ModelOutput):
  class TFContrastiveSearchEncoderDecoderOutput (line 394) | class TFContrastiveSearchEncoderDecoderOutput(ModelOutput):
  class TFGenerationMixin (line 444) | class TFGenerationMixin:
    method seed_generator (line 464) | def seed_generator(self):
    method prepare_inputs_for_generation (line 472) | def prepare_inputs_for_generation(self, *args, **kwargs):
    method adjust_logits_during_generation (line 477) | def adjust_logits_during_generation(
    method compute_transition_scores (line 498) | def compute_transition_scores(
    method _validate_model_class (line 618) | def _validate_model_class(self):
    method _validate_model_kwargs (line 643) | def _validate_model_kwargs(self, model_kwargs: Dict[str, Any]):
    method generate (line 666) | def generate(
    method _prepare_attention_mask_for_generation (line 1048) | def _prepare_attention_mask_for_generation(
    method _prepare_encoder_decoder_kwargs_for_generation (line 1064) | def _prepare_encoder_decoder_kwargs_for_generation(
    method _prepare_decoder_input_ids_for_generation (line 1094) | def _prepare_decoder_input_ids_for_generation(
    method _get_decoder_start_token_id (line 1133) | def _get_decoder_start_token_id(self, decoder_start_token_id: int = No...
    method _expand_inputs_for_generation (line 1152) | def _expand_inputs_for_generation(
    method _prepare_model_inputs (line 1190) | def _prepare_model_inputs(
    method _maybe_initialize_input_ids_for_generation (line 1256) | def _maybe_initialize_input_ids_for_generation(
    method _extract_past_from_model_output (line 1285) | def _extract_past_from_model_output(outputs: ModelOutput):
    method _update_model_kwargs_for_generation (line 1295) | def _update_model_kwargs_for_generation(
    method _update_model_kwargs_for_xla_generation (line 1311) | def _update_model_kwargs_for_xla_generation(
    method _get_logits_warper (line 1438) | def _get_logits_warper(
    method _get_logits_processor (line 1460) | def _get_logits_processor(
    method _merge_criteria_processor_list (line 1515) | def _merge_criteria_processor_list(
    method greedy_search (line 1536) | def greedy_search(
    method sample (line 1798) | def sample(
    method _gather_beams (line 2089) | def _gather_beams(nested, beam_indices, batch_axis=0):
    method beam_search (line 2109) | def beam_search(
    method contrastive_search (line 2672) | def contrastive_search(
  function tf_top_k_top_p_filtering (line 3080) | def tf_top_k_top_p_filtering(logits, top_k=0, top_p=1.0, filter_value=-f...
  function scatter_values_on_batch_indices (line 3135) | def scatter_values_on_batch_indices(values, batch_indices):
  function sample_without_replacement (line 3145) | def sample_without_replacement(logits, num_samples):
  function _ranking_fast (line 3155) | def _ranking_fast(

FILE: transformers/generation/utils.py
  class GreedySearchDecoderOnlyOutput (line 83) | class GreedySearchDecoderOnlyOutput(ModelOutput):
  class ContrastiveSearchEncoderDecoderOutput (line 111) | class ContrastiveSearchEncoderDecoderOutput(ModelOutput):
  class ContrastiveSearchDecoderOnlyOutput (line 150) | class ContrastiveSearchDecoderOnlyOutput(ModelOutput):
  class GreedySearchEncoderDecoderOutput (line 179) | class GreedySearchEncoderDecoderOutput(ModelOutput):
  class SampleDecoderOnlyOutput (line 221) | class SampleDecoderOnlyOutput(ModelOutput):
  class SampleEncoderDecoderOutput (line 250) | class SampleEncoderDecoderOutput(ModelOutput):
  class BeamSearchDecoderOnlyOutput (line 293) | class BeamSearchDecoderOnlyOutput(ModelOutput):
  class BeamSearchEncoderDecoderOutput (line 328) | class BeamSearchEncoderDecoderOutput(ModelOutput):
  class BeamSampleDecoderOnlyOutput (line 378) | class BeamSampleDecoderOnlyOutput(ModelOutput):
  class BeamSampleEncoderDecoderOutput (line 413) | class BeamSampleEncoderDecoderOutput(ModelOutput):
  class GenerationMixin (line 469) | class GenerationMixin:
    method prepare_inputs_for_generation (line 493) | def prepare_inputs_for_generation(self, *args, **kwargs):
    method _prepare_model_inputs (line 498) | def _prepare_model_inputs(
    method adjust_logits_during_generation (line 562) | def adjust_logits_during_generation(self, logits: torch.FloatTensor, *...
    method _maybe_initialize_input_ids_for_generation (line 568) | def _maybe_initialize_input_ids_for_generation(
    method _prepare_attention_mask_for_generation (line 596) | def _prepare_attention_mask_for_generation(
    method _prepare_encoder_decoder_kwargs_for_generation (line 614) | def _prepare_encoder_decoder_kwargs_for_generation(
    method _prepare_decoder_input_ids_for_generation (line 646) | def _prepare_decoder_input_ids_for_generation(
    method _get_decoder_start_token_id (line 691) | def _get_decoder_start_token_id(self, decoder_start_token_id: int = No...
    method _expand_inputs_for_generation (line 708) | def _expand_inputs_for_generation(
    method _extract_past_from_model_output (line 734) | def _extract_past_from_model_output(self, outputs: ModelOutput, standa...
    method _update_model_kwargs_for_generation (line 749) | def _update_model_kwargs_for_generation(
    method _reorder_cache (line 786) | def _reorder_cache(self, past_key_values, beam_idx):
    method _get_logits_warper (line 792) | def _get_logits_warper(
    method _get_logits_processor (line 830) | def _get_logits_processor(
    method _get_stopping_criteria (line 947) | def _get_stopping_criteria(
    method _merge_criteria_processor_list (line 958) | def _merge_criteria_processor_list(
    method compute_transition_scores (line 979) | def compute_transition_scores(
    method _validate_model_class (line 1100) | def _validate_model_class(self):
    method _validate_model_kwargs (line 1126) | def _validate_model_kwargs(self, model_kwargs: Dict[str, Any]):
    method generate (line 1150) | def generate(
    method contrastive_search (line 1798) | def contrastive_search(
    method greedy_search (line 2171) | def greedy_search(
    method sample (line 2430) | def sample(
    method beam_search (line 2712) | def beam_search(
    method beam_sample (line 3035) | def beam_sample(
    method group_beam_search (line 3368) | def group_beam_search(
    method constrained_beam_search (line 3748) | def constrained_beam_search(
    method assisted_decoding (line 4073) | def assisted_decoding(
  function _crop_past_key_values (line 4497) | def _crop_past_key_values(model, past_key_values, maximum_length):
  function _split_model_outputs (line 4539) | def _split_model_outputs(outputs, new_outputs, cur_len, added_len, is_de...
  function top_k_top_p_filtering (line 4565) | def top_k_top_p_filtering(
  function _ranking_fast (line 4600) | def _ranking_fast(

FILE: transformers/generation_flax_utils.py
  class FlaxGenerationMixin (line 22) | class FlaxGenerationMixin(FlaxGenerationMixin):

FILE: transformers/generation_tf_utils.py
  class TFGenerationMixin (line 22) | class TFGenerationMixin(TFGenerationMixin):

FILE: transformers/generation_utils.py
  class GenerationMixin (line 22) | class GenerationMixin(GenerationMixin):

FILE: transformers/hf_argparser.py
  function string_to_bool (line 42) | def string_to_bool(v):
  function make_choice_type_function (line 55) | def make_choice_type_function(choices: list) -> Callable[[str], Any]:
  function HfArg (line 70) | def HfArg(
  class HfArgumentParser (line 117) | class HfArgumentParser(ArgumentParser):
    method __init__ (line 128) | def __init__(self, dataclass_types: Union[DataClassType, Iterable[Data...
    method _parse_dataclass_field (line 147) | def _parse_dataclass_field(parser: ArgumentParser, field: dataclasses....
    method _add_dataclass_arguments (line 239) | def _add_dataclass_arguments(self, dtype: DataClassType):
    method parse_args_into_dataclasses (line 273) | def parse_args_into_dataclasses(
    method parse_dict (line 359) | def parse_dict(self, args: Dict[str, Any], allow_extra_keys: bool = Fa...
    method parse_json_file (line 387) | def parse_json_file(self, json_file: str, allow_extra_keys: bool = Fal...
    method parse_yaml_file (line 409) | def parse_yaml_file(self, yaml_file: str, allow_extra_keys: bool = Fal...

FILE: transformers/image_processing_utils.py
  class BatchFeature (line 43) | class BatchFeature(BaseBatchFeature):
  class ImageProcessingMixin (line 59) | class ImageProcessingMixin(PushToHubMixin):
    method __init__ (line 67) | def __init__(self, **kwargs):
    method _set_processor_class (line 79) | def _set_processor_class(self, processor_class: str):
    method from_pretrained (line 84) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
    method save_pretrained (line 169) | def save_pretrained(self, save_directory: Union[str, os.PathLike], pus...
    method get_image_processor_dict (line 218) | def get_image_processor_dict(
    method from_dict (line 321) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method to_dict (line 365) | def to_dict(self) -> Dict[str, Any]:
    method from_json_file (line 378) | def from_json_file(cls, json_file: Union[str, os.PathLike]):
    method to_json_string (line 396) | def to_json_string(self) -> str:
    method to_json_file (line 417) | def to_json_file(self, json_file_path: Union[str, os.PathLike]):
    method __repr__ (line 428) | def __repr__(self):
    method register_for_auto_class (line 432) | def register_for_auto_class(cls, auto_class="AutoImageProcessor"):
  class BaseImageProcessor (line 458) | class BaseImageProcessor(ImageProcessingMixin):
    method __init__ (line 459) | def __init__(self, **kwargs):
    method __call__ (line 462) | def __call__(self, images, **kwargs) -> BatchFeature:
    method preprocess (line 466) | def preprocess(self, images, **kwargs) -> BatchFeature:
  function is_valid_size_dict (line 473) | def is_valid_size_dict(size_dict):
  function convert_to_size_dict (line 484) | def convert_to_size_dict(
  function get_size_dict (line 512) | def get_size_dict(

FILE: transformers/image_transforms.py
  function to_channel_dimension_format (line 54) | def to_channel_dimension_format(
  function rescale (line 91) | def rescale(
  function _rescale_for_pil_conversion (line 121) | def _rescale_for_pil_conversion(image):
  function to_pil_image (line 148) | def to_pil_image(
  function get_resize_output_image_size (line 197) | def get_resize_output_image_size(
  function resize (line 261) | def resize(
  function normalize (line 327) | def normalize(
  function center_crop (line 390) | def center_crop(
  function _center_to_corners_format_torch (line 487) | def _center_to_corners_format_torch(bboxes_center: "torch.Tensor") -> "t...
  function _center_to_corners_format_numpy (line 497) | def _center_to_corners_format_numpy(bboxes_center: np.ndarray) -> np.nda...
  function _center_to_corners_format_tf (line 507) | def _center_to_corners_format_tf(bboxes_center: "tf.Tensor") -> "tf.Tens...
  function center_to_corners_format (line 518) | def center_to_corners_format(bboxes_center: TensorType) -> TensorType:
  function _corners_to_center_format_torch (line 539) | def _corners_to_center_format_torch(bboxes_corners: "torch.Tensor") -> "...
  function _corners_to_center_format_numpy (line 550) | def _corners_to_center_format_numpy(bboxes_corners: np.ndarray) -> np.nd...
  function _corners_to_center_format_tf (line 564) | def _corners_to_center_format_tf(bboxes_corners: "tf.Tensor") -> "tf.Ten...
  function corners_to_center_format (line 578) | def corners_to_center_format(bboxes_corners: TensorType) -> TensorType:
  function rgb_to_id (line 601) | def rgb_to_id(color):
  function id_to_rgb (line 612) | def id_to_rgb(id_map):
  class PaddingMode (line 631) | class PaddingMode(ExplicitEnum):
  function pad (line 642) | def pad(
  function convert_to_rgb (line 729) | def convert_to_rgb(image: ImageInput) -> ImageInput:
  function flip_channel_order (line 747) | def flip_channel_order(image: np.ndarray, data_format: Optional[ChannelD...

FILE: transformers/image_utils.py
  class ChannelDimension (line 62) | class ChannelDimension(ExplicitEnum):
  function is_pil_image (line 67) | def is_pil_image(img):
  function is_valid_image (line 71) | def is_valid_image(img):
  function valid_images (line 81) | def valid_images(imgs):
  function is_batched (line 93) | def is_batched(img):
  function make_list_of_images (line 99) | def make_list_of_images(images, expected_ndims: int = 3) -> List[ImageIn...
  function to_numpy_array (line 138) | def to_numpy_array(img) -> np.ndarray:
  function infer_channel_dimension_format (line 147) | def infer_channel_dimension_format(image: np.ndarray) -> ChannelDimension:
  function get_channel_dimension_axis (line 172) | def get_channel_dimension_axis(image: np.ndarray) -> int:
  function get_image_size (line 191) | def get_image_size(image: np.ndarray, channel_dim: ChannelDimension = No...
  function is_valid_annotation_coco_detection (line 215) | def is_valid_annotation_coco_detection(annotation: Dict[str, Union[List,...
  function is_valid_annotation_coco_panoptic (line 231) | def is_valid_annotation_coco_panoptic(annotation: Dict[str, Union[List, ...
  function valid_coco_detection_annotations (line 248) | def valid_coco_detection_annotations(annotations: Iterable[Dict[str, Uni...
  function valid_coco_panoptic_annotations (line 252) | def valid_coco_panoptic_annotations(annotations: Iterable[Dict[str, Unio...
  function load_image (line 256) | def load_image(image: Union[str, "PIL.Image.Image"]) -> "PIL.Image.Image":
  class ImageFeatureExtractionMixin (line 291) | class ImageFeatureExtractionMixin:
    method _ensure_format_supported (line 296) | def _ensure_format_supported(self, image):
    method to_pil_image (line 303) | def to_pil_image(self, image, rescale=None):
    method convert_rgb (line 333) | def convert_rgb(self, image):
    method rescale (line 347) | def rescale(self, image: np.ndarray, scale: Union[float, int]) -> np.n...
    method to_numpy_array (line 354) | def to_numpy_array(self, image, rescale=None, channel_first=True):
    method expand_dims (line 386) | def expand_dims(self, image):
    method normalize (line 406) | def normalize(self, image, mean, std, rescale=False):
    method resize (line 452) | def resize(self, image, size, resample=None, default_to_square=True, m...
    method center_crop (line 519) | def center_crop(self, image, size):
    method flip_channel_order (line 594) | def flip_channel_order(self, image):
    method rotate (line 611) | def rotate(self, image, angle, resample=None, expand=0, center=None, t...

FILE: transformers/integrations.py
  function is_wandb_available (line 78) | def is_wandb_available():
  function is_clearml_available (line 89) | def is_clearml_available():
  function is_comet_available (line 93) | def is_comet_available():
  function is_tensorboard_available (line 97) | def is_tensorboard_available():
  function is_optuna_available (line 101) | def is_optuna_available():
  function is_ray_available (line 105) | def is_ray_available():
  function is_ray_tune_available (line 109) | def is_ray_tune_available():
  function is_sigopt_available (line 115) | def is_sigopt_available():
  function is_azureml_available (line 119) | def is_azureml_available():
  function is_mlflow_available (line 127) | def is_mlflow_available():
  function is_dagshub_available (line 133) | def is_dagshub_available():
  function is_fairscale_available (line 137) | def is_fairscale_available():
  function is_neptune_available (line 141) | def is_neptune_available():
  function is_codecarbon_available (line 145) | def is_codecarbon_available():
  function is_flytekit_available (line 149) | def is_flytekit_available():
  function is_flyte_deck_standard_available (line 153) | def is_flyte_deck_standard_available():
  function hp_params (line 159) | def hp_params(trial):
  function default_hp_search_backend (line 180) | def default_hp_search_backend():
  function run_hp_search_optuna (line 189) | def run_hp_search_optuna(trainer, n_trials: int, direction: str, **kwarg...
  function run_hp_search_ray (line 240) | def run_hp_search_ray(trainer, n_trials: int, direction: str, **kwargs) ...
  function run_hp_search_sigopt (line 376) | def run_hp_search_sigopt(trainer, n_trials: int, direction: str, **kwarg...
  function run_hp_search_wandb (line 476) | def run_hp_search_wandb(trainer, n_trials: int, direction: str, **kwargs...
  function get_available_reporting_integrations (line 545) | def get_available_reporting_integrations():
  function rewrite_logs (line 568) | def rewrite_logs(d):
  class TensorBoardCallback (line 584) | class TensorBoardCallback(TrainerCallback):
    method __init__ (line 593) | def __init__(self, tb_writer=None):
    method _init_summary_writer (line 616) | def _init_summary_writer(self, args, log_dir=None):
    method on_train_begin (line 621) | def on_train_begin(self, args, state, control, **kwargs):
    method on_log (line 643) | def on_log(self, args, state, control, logs=None, **kwargs):
    method on_train_end (line 664) | def on_train_end(self, args, state, control, **kwargs):
  class WandbCallback (line 670) | class WandbCallback(TrainerCallback):
    method __init__ (line 675) | def __init__(self):
    method setup (line 695) | def setup(self, args, state, model, **kwargs):
    method on_train_begin (line 762) | def on_train_begin(self, args, state, control, model=None, **kwargs):
    method on_train_end (line 773) | def on_train_end(self, args, state, control, model=None, tokenizer=Non...
    method on_log (line 807) | def on_log(self, args, state, control, model=None, logs=None, **kwargs):
    method on_save (line 816) | def on_save(self, args, state, control, **kwargs):
  class CometCallback (line 837) | class CometCallback(TrainerCallback):
    method __init__ (line 842) | def __init__(self):
    method setup (line 848) | def setup(self, args, state, model):
    method on_train_begin (line 890) | def on_train_begin(self, args, state, control, model=None, **kwargs):
    method on_log (line 894) | def on_log(self, args, state, control, model=None, logs=None, **kwargs):
    method on_train_end (line 902) | def on_train_end(self, args, state, control, **kwargs):
  class AzureMLCallback (line 914) | class AzureMLCallback(TrainerCallback):
    method __init__ (line 919) | def __init__(self, azureml_run=None):
    method on_init_end (line 924) | def on_init_end(self, args, state, control, **kwargs):
    method on_log (line 930) | def on_log(self, args, state, control, logs=None, **kwargs):
  class MLflowCallback (line 937) | class MLflowCallback(TrainerCallback):
    method __init__ (line 943) | def __init__(self):
    method setup (line 956) | def setup(self, args, state, model):
    method on_train_begin (line 1026) | def on_train_begin(self, args, state, control, model=None, **kwargs):
    method on_log (line 1030) | def on_log(self, args, state, control, logs, model=None, **kwargs):
    method on_train_end (line 1045) | def on_train_end(self, args, state, control, **kwargs):
    method on_save (line 1050) | def on_save(self, args, state, control, **kwargs):
    method __del__ (line 1061) | def __del__(self):
  class DagsHubCallback (line 1072) | class DagsHubCallback(MLflowCallback):
    method __init__ (line 1077) | def __init__(self):
    method setup (line 1086) | def setup(self, *args, **kwargs):
    method on_train_end (line 1113) | def on_train_end(self, args, state, control, **kwargs):
  class NeptuneMissingConfiguration (line 1121) | class NeptuneMissingConfiguration(Exception):
    method __init__ (line 1122) | def __init__(self):
  class NeptuneCallback (line 1132) | class NeptuneCallback(TrainerCallback):
    method __init__ (line 1168) | def __init__(
    method _stop_run_if_exists (line 1223) | def _stop_run_if_exists(self):
    method _initialize_run (line 1229) | def _initialize_run(self, **additional_neptune_kwargs):
    method _use_initial_run (line 1245) | def _use_initial_run(self):
    method _ensure_run_with_monitoring (line 1251) | def _ensure_run_with_monitoring(self):
    method _ensure_at_least_run_without_monitoring (line 1265) | def _ensure_at_least_run_without_monitoring(self):
    method run (line 1280) | def run(self):
    method _metadata_namespace (line 1286) | def _metadata_namespace(self):
    method _log_integration_version (line 1289) | def _log_integration_version(self):
    method _log_trainer_parameters (line 1292) | def _log_trainer_parameters(self, args):
    method _log_model_parameters (line 1295) | def _log_model_parameters(self, model):
    method _log_hyper_param_search_parameters (line 1299) | def _log_hyper_param_search_parameters(self, state):
    method _log_model_checkpoint (line 1306) | def _log_model_checkpoint(self, source_directory: str, checkpoint: str):
    method on_init_end (line 1330) | def on_init_end(self, args, state, control, **kwargs):
    method on_train_begin (line 1338) | def on_train_begin(self, args, state, control, model=None, **kwargs):
    method on_train_end (line 1353) | def on_train_end(self, args, state, control, **kwargs):
    method __del__ (line 1356) | def __del__(self):
    method on_save (line 1362) | def on_save(self, args, state, control, **kwargs):
    method on_evaluate (line 1366) | def on_evaluate(self, args, state, control, metrics=None, **kwargs):
    method get_run (line 1379) | def get_run(cls, trainer):
    method on_log (line 1386) | def on_log(self, args, state, control, logs: Optional[Dict[str, float]...
  class CodeCarbonCallback (line 1399) | class CodeCarbonCallback(TrainerCallback):
    method __init__ (line 1404) | def __init__(self):
    method on_init_end (line 1414) | def on_init_end(self, args, state, control, **kwargs):
    method on_train_begin (line 1419) | def on_train_begin(self, args, state, control, model=None, **kwargs):
    method on_train_end (line 1423) | def on_train_end(self, args, state, control, **kwargs):
  class ClearMLCallback (line 1428) | class ClearMLCallback(TrainerCallback):
    method __init__ (line 1441) | def __init__(self):
    method setup (line 1454) | def setup(self, args, state, model, tokenizer, **kwargs):
    method on_train_begin (line 1482) | def on_train_begin(self, args, state, control, model=None, tokenizer=N...
    method on_train_end (line 1490) | def on_train_end(self, args, state, control, model=None, tokenizer=Non...
    method on_log (line 1497) | def on_log(self, args, state, control, model=None, tokenizer=None, log...
    method on_save (line 1539) | def on_save(self, args, state, control, **kwargs):
  class FlyteCallback (line 1547) | class FlyteCallback(TrainerCallback):
    method __init__ (line 1574) | def __init__(self, save_log_history: bool = True, sync_checkpoints: bo...
    method on_save (line 1592) | def on_save(self, args, state, control, **kwargs):
    method on_train_end (line 1600) | def on_train_end(self, args, state, control, **kwargs):
  function get_reporting_integration_callbacks (line 1624) | def get_reporting_integration_callbacks(report_to):

FILE: transformers/keras_callbacks.py
  class KerasMetricCallback (line 21) | class KerasMetricCallback(Callback):
    method __init__ (line 79) | def __init__(
    method _concatenate_batches (line 145) | def _concatenate_batches(batches, padding_index=-100):
    method _postprocess_predictions_or_labels (line 163) | def _postprocess_predictions_or_labels(self, inputs):
    method on_epoch_end (line 185) | def on_epoch_end(self, epoch, logs=None):
  class PushToHubCallback (line 268) | class PushToHubCallback(Callback):
    method __init__ (line 315) | def __init__(
    method on_train_begin (line 352) | def on_train_begin(self, logs=None):
    method on_train_batch_end (line 357) | def on_train_batch_end(self, batch, logs=None):
    method on_epoch_end (line 368) | def on_epoch_end(self, epoch, logs=None):
    method on_train_end (line 395) | def on_train_end(self, logs=None):

FILE: transformers/kernels/deformable_detr/cpu/ms_deform_attn_cpu.cpp
  function ms_deform_attn_cpu_forward (line 17) | at::Tensor
  function ms_deform_attn_cpu_backward (line 29) | std::vector<at::Tensor>

FILE: transformers/kernels/deformable_detr/ms_deform_attn.h
  function im2col_step (line 27) | int im2col_step)

FILE: transformers/kernels/deformable_detr/vision.cpp
  function PYBIND11_MODULE (line 13) | PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {

FILE: transformers/kernels/rwkv/wkv_op.cpp
  function forward (line 12) | void forward(torch::Tensor &w, torch::Tensor &u, torch::Tensor &k, torch...
  function forward_bf16 (line 18) | void forward_bf16(torch::Tensor &w, torch::Tensor &u, torch::Tensor &k, ...
  function forward_with_state (line 24) | void forward_with_state(torch::Tensor &w, torch::Tensor &u, torch::Tenso...
  function forward_with_state_bf16 (line 30) | void forward_with_state_bf16(torch::Tensor &w, torch::Tensor &u, torch::...
  function backward (line 36) | void backward(torch::Tensor &w, torch::Tensor &u, torch::Tensor &k, torc...
  function backward_bf16 (line 42) | void backward_bf16(torch::Tensor &w, torch::Tensor &u, torch::Tensor &k,...
  function PYBIND11_MODULE (line 50) | PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
  function TORCH_LIBRARY (line 59) | TORCH_LIBRARY(wkv, m) {

FILE: transformers/kernels/yoso/fast_lsh_cumulation_torch.cpp
  function fast_hash (line 7) | std::vector<at::Tensor> fast_hash(
  function lsh_cumulation (line 28) | at::Tensor lsh_cumulation(
  function lsh_weighted_cumulation (line 49) | at::Tensor lsh_weighted_cumulation(
  function PYBIND11_MODULE (line 124) | PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {

FILE: transformers/modelcard.py
  class ModelCard (line 80) | class ModelCard:
    method __init__ (line 91) | def __init__(self, **kwargs):
    method save_pretrained (line 114) | def save_pretrained(self, save_directory_or_file):
    method from_pretrained (line 126) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method from_dict (line 222) | def from_dict(cls, json_object):
    method from_json_file (line 227) | def from_json_file(cls, json_file):
    method __eq__ (line 234) | def __eq__(self, other):
    method __repr__ (line 237) | def __repr__(self):
    method to_dict (line 240) | def to_dict(self):
    method to_json_string (line 245) | def to_json_string(self):
    method to_json_file (line 249) | def to_json_file(self, json_file_path):
  function _listify (line 300) | def _listify(obj):
  function _insert_values_as_list (line 309) | def _insert_values_as_list(metadata, name, values):
  function infer_metric_tags_from_eval_results (line 321) | def infer_metric_tags_from_eval_results(eval_results):
  function _insert_value (line 333) | def _insert_value(metadata, name, value):
  function is_hf_dataset (line 340) | def is_hf_dataset(dataset):
  function _get_mapping_values (line 349) | def _get_mapping_values(mapping):
  class TrainingSummary (line 360) | class TrainingSummary:
    method __post_init__ (line 376) | def __post_init__(self):
    method create_model_index (line 392) | def create_model_index(self, metric_mapping):
    method create_metadata (line 455) | def create_metadata(self):
    method to_model_card (line 468) | def to_model_card(self):
    method from_trainer (line 549) | def from_trainer(
    method from_keras (line 627) | def from_keras(
  function parse_keras_history (line 702) | def parse_keras_history(logs):
  function parse_log_history (line 737) | def parse_log_history(log_history):
  function extract_hyperparameters_from_keras (line 799) | def extract_hyperparameters_from_keras(model):
  function _maybe_round (line 812) | def _maybe_round(v, decimals=4):
  function _regular_table_line (line 818) | def _regular_table_line(values, col_widths):
  function _second_table_line (line 823) | def _second_table_line(col_widths):
  function make_markdown_table (line 828) | def make_markdown_table(lines):
  function extract_hyperparameters_from_trainer (line 855) | def extract_hyperparameters_from_trainer(trainer):

FILE: transformers/modeling_flax_outputs.py
  class FlaxBaseModelOutput (line 23) | class FlaxBaseModelOutput(ModelOutput):
  class FlaxBaseModelOutputWithNoAttention (line 49) | class FlaxBaseModelOutputWithNoAttention(ModelOutput):
  class FlaxBaseModelOutputWithPoolingAndNoAttention (line 67) | class FlaxBaseModelOutputWithPoolingAndNoAttention(ModelOutput):
  class FlaxImageClassifierOutputWithNoAttention (line 88) | class FlaxImageClassifierOutputWithNoAttention(ModelOutput):
  class FlaxBaseModelOutputWithPast (line 107) | class FlaxBaseModelOutputWithPast(ModelOutput):
  class FlaxBaseModelOutputWithPooling (line 137) | class FlaxBaseModelOutputWithPooling(ModelOutput):
  class FlaxBaseModelOutputWithPoolingAndCrossAttentions (line 168) | class FlaxBaseModelOutputWithPoolingAndCrossAttentions(ModelOutput):
  class FlaxBaseModelOutputWithPastAndCrossAttentions (line 217) | class FlaxBaseModelOutputWithPastAndCrossAttentions(ModelOutput):
  class FlaxSeq2SeqModelOutput (line 263) | class FlaxSeq2SeqModelOutput(ModelOutput):
  class FlaxCausalLMOutputWithCrossAttentions (line 324) | class FlaxCausalLMOutputWithCrossAttentions(ModelOutput):
  class FlaxMaskedLMOutput (line 365) | class FlaxMaskedLMOutput(ModelOutput):
  class FlaxSeq2SeqLMOutput (line 394) | class FlaxSeq2SeqLMOutput(ModelOutput):
  class FlaxNextSentencePredictorOutput (line 451) | class FlaxNextSentencePredictorOutput(ModelOutput):
  class FlaxSequenceClassifierOutput (line 478) | class FlaxSequenceClassifierOutput(ModelOutput):
  class FlaxSeq2SeqSequenceClassifierOutput (line 504) | class FlaxSeq2SeqSequenceClassifierOutput(ModelOutput):
  class FlaxMultipleChoiceModelOutput (line 561) | class FlaxMultipleChoiceModelOutput(ModelOutput):
  class FlaxTokenClassifierOutput (line 589) | class FlaxTokenClassifierOutput(ModelOutput):
  class FlaxQuestionAnsweringModelOutput (line 615) | class FlaxQuestionAnsweringModelOutput(ModelOutput):
  class FlaxSeq2SeqQuestionAnsweringModelOutput (line 644) | class FlaxSeq2SeqQuestionAnsweringModelOutput(ModelOutput):

FILE: transformers/modeling_flax_pytorch_utils.py
  function load_pytorch_checkpoint_in_flax_state_dict (line 41) | def load_pytorch_checkpoint_in_flax_state_dict(
  function rename_key_and_reshape_tensor (line 69) | def rename_key_and_reshape_tensor(
  function convert_pytorch_state_dict_to_flax (line 126) | def convert_pytorch_state_dict_to_flax(pt_state_dict, flax_model):
  function convert_pytorch_sharded_state_dict_to_flax (line 204) | def convert_pytorch_sharded_state_dict_to_flax(shard_filenames, flax_mod...
  function load_flax_checkpoint_in_pytorch_model (line 284) | def load_flax_checkpoint_in_pytorch_model(model, flax_checkpoint_path):
  function load_flax_weights_in_pytorch_model (line 302) | def load_flax_weights_in_pytorch_model(pt_model, flax_state):

FILE: transformers/modeling_flax_utils.py
  function quick_gelu (line 61) | def quick_gelu(x):
  function dtype_byte_size (line 75) | def dtype_byte_size(dtype):
  function flax_shard_checkpoint (line 92) | def flax_shard_checkpoint(params, max_shard_size="10GB"):
  class FlaxPreTrainedModel (line 157) | class FlaxPreTrainedModel(PushToHubMixin, FlaxGenerationMixin):
    method __init__ (line 179) | def __init__(
    method init_weights (line 230) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method enable_gradient_checkpointing (line 233) | def enable_gradient_checkpointing(self):
    method _from_config (line 237) | def _from_config(cls, config, **kwargs):
    method framework (line 244) | def framework(self) -> str:
    method config (line 251) | def config(self) -> PretrainedConfig:
    method module (line 255) | def module(self) -> nn.Module:
    method params (line 259) | def params(self) -> Union[Dict, FrozenDict]:
    method required_params (line 269) | def required_params(self) -> Set:
    method params_shape_tree (line 273) | def params_shape_tree(self) -> Dict:
    method params (line 277) | def params(self, params: Union[Dict, FrozenDict]):
    method _cast_floating_to (line 295) | def _cast_floating_to(self, params: Union[Dict, FrozenDict], dtype: jn...
    method to_bf16 (line 319) | def to_bf16(self, params: Union[Dict, FrozenDict], mask: Any = None):
    method to_fp32 (line 358) | def to_fp32(self, params: Union[Dict, FrozenDict], mask: Any = None):
    method to_fp16 (line 385) | def to_fp16(self, params: Union[Dict, FrozenDict], mask: Any = None):
    method load_flax_sharded_weights (line 425) | def load_flax_sharded_weights(cls, shard_files):
    method can_generate (line 471) | def can_generate(self) -> bool:
    method from_pretrained (line 482) | def from_pretrained(
    method save_pretrained (line 1008) | def save_pretrained(
    method register_for_auto_class (line 1112) | def register_for_auto_class(cls, auto_class="FlaxAutoModel"):
  function overwrite_call_docstring (line 1146) | def overwrite_call_docstring(model_class, docstring):
  function append_call_sample_docstring (line 1155) | def append_call_sample_docstring(model_class, checkpoint, output_type, c...
  function append_replace_return_docstrings (line 1165) | def append_replace_return_docstrings(model_class, output_type, config_cl...

FILE: transformers/modeling_outputs.py
  class BaseModelOutput (line 25) | class BaseModelOutput(ModelOutput):
  class BaseModelOutputWithNoAttention (line 51) | class BaseModelOutputWithNoAttention(ModelOutput):
  class BaseModelOutputWithPooling (line 70) | class BaseModelOutputWithPooling(ModelOutput):
  class BaseModelOutputWithPoolingAndNoAttention (line 102) | class BaseModelOutputWithPoolingAndNoAttention(ModelOutput):
  class BaseModelOutputWithPast (line 124) | class BaseModelOutputWithPast(ModelOutput):
  class BaseModelOutputWithCrossAttentions (line 163) | class BaseModelOutputWithCrossAttentions(ModelOutput):
  class BaseModelOutputWithPoolingAndCrossAttentions (line 196) | class BaseModelOutputWithPoolingAndCrossAttentions(ModelOutput):
  class BaseModelOutputWithPastAndCrossAttentions (line 245) | class BaseModelOutputWithPastAndCrossAttentions(ModelOutput):
  class MoECausalLMOutputWithPast (line 291) | class MoECausalLMOutputWithPast(ModelOutput):
  class MoEModelOutput (line 340) | class MoEModelOutput(ModelOutput):
  class MoEModelOutputWithPastAndCrossAttentions (line 372) | class MoEModelOutputWithPastAndCrossAttentions(ModelOutput):
  class Seq2SeqModelOutput (line 425) | class Seq2SeqModelOutput(ModelOutput):
  class Seq2SeqMoEModelOutput (line 486) | class Seq2SeqMoEModelOutput(ModelOutput):
  class CausalLMOutput (line 558) | class CausalLMOutput(ModelOutput):
  class CausalLMOutputWithPast (line 587) | class CausalLMOutputWithPast(ModelOutput):
  class CausalLMOutputWithCrossAttentions (line 623) | class CausalLMOutputWithCrossAttentions(ModelOutput):
  class SequenceClassifierOutputWithPast (line 667) | class SequenceClassifierOutputWithPast(ModelOutput):
  class MaskedLMOutput (line 703) | class MaskedLMOutput(ModelOutput):
  class Seq2SeqLMOutput (line 732) | class Seq2SeqLMOutput(ModelOutput):
  class Seq2SeqMoEOutput (line 792) | class Seq2SeqMoEOutput(ModelOutput):
  class NextSentencePredictorOutput (line 867) | class NextSentencePredictorOutput(ModelOutput):
  class SequenceClassifierOutput (line 897) | class SequenceClassifierOutput(ModelOutput):
  class Seq2SeqSequenceClassifierOutput (line 926) | class Seq2SeqSequenceClassifierOutput(ModelOutput):
  class MultipleChoiceModelOutput (line 986) | class MultipleChoiceModelOutput(ModelOutput):
  class TokenClassifierOutput (line 1017) | class TokenClassifierOutput(ModelOutput):
  class QuestionAnsweringModelOutput (line 1046) | class QuestionAnsweringModelOutput(ModelOutput):
  class Seq2SeqQuestionAnsweringModelOutput (line 1078) | class Seq2SeqQuestionAnsweringModelOutput(ModelOutput):
  class SemanticSegmenterOutput (line 1141) | class SemanticSegmenterOutput(ModelOutput):
  class ImageClassifierOutput (line 1179) | class ImageClassifierOutput(ModelOutput):
  class ImageClassifierOutputWithNoAttention (line 1207) | class ImageClassifierOutputWithNoAttention(ModelOutput):
  class DepthEstimatorOutput (line 1228) | class DepthEstimatorOutput(ModelOutput):
  class ImageSuperResolutionOutput (line 1258) | class ImageSuperResolutionOutput(ModelOutput):
  class Wav2Vec2BaseModelOutput (line 1286) | class Wav2Vec2BaseModelOutput(ModelOutput):
  class XVectorOutput (line 1315) | class XVectorOutput(ModelOutput):
  class BackboneOutput (line 1347) | class BackboneOutput(ModelOutput):
  class BaseModelOutputWithPoolingAndProjection (line 1374) | class BaseModelOutputWithPoolingAndProjection(ModelOutput):
  class Seq2SeqSpectrogramOutput (line 1411) | class Seq2SeqSpectrogramOutput(ModelOutput):
  class Seq2SeqTSModelOutput (line 1471) | class Seq2SeqTSModelOutput(ModelOutput):
  class Seq2SeqTSPredictionOutput (line 1543) | class Seq2SeqTSPredictionOutput(ModelOutput):
  class SampleTSPredictionOutput (line 1615) | class SampleTSPredictionOutput(ModelOutput):
  class MaskedImageModelingOutput (line 1629) | class MaskedImageModelingOutput(ModelOutput):
    method logits (line 1656) | def logits(self):

FILE: transformers/modeling_tf_outputs.py
  class TFBaseModelOutput (line 27) | class TFBaseModelOutput(ModelOutput):
  class TFBaseModelOutputWithNoAttention (line 53) | class TFBaseModelOutputWithNoAttention(ModelOutput):
  class TFBaseModelOutputWithPooling (line 72) | class TFBaseModelOutputWithPooling(ModelOutput):
  class TFBaseModelOutputWithPoolingAndNoAttention (line 106) | class TFBaseModelOutputWithPoolingAndNoAttention(ModelOutput):
  class TFBaseModelOutputWithPoolingAndCrossAttentions (line 128) | class TFBaseModelOutputWithPoolingAndCrossAttentions(ModelOutput):
  class TFBaseModelOutputWithPast (line 176) | class TFBaseModelOutputWithPast(ModelOutput):
  class TFBaseModelOutputWithCrossAttentions (line 212) | class TFBaseModelOutputWithCrossAttentions(ModelOutput):
  class TFBaseModelOutputWithPastAndCrossAttentions (line 245) | class TFBaseModelOutputWithPastAndCrossAttentions(ModelOutput):
  class TFSeq2SeqModelOutput (line 288) | class TFSeq2SeqModelOutput(ModelOutput):
  class TFCausalLMOutput (line 348) | class TFCausalLMOutput(ModelOutput):
  class TFCausalLMOutputWithPast (line 377) | class TFCausalLMOutputWithPast(ModelOutput):
  class TFCausalLMOutputWithCrossAttentions (line 413) | class TFCausalLMOutputWithCrossAttentions(ModelOutput):
  class TFMaskedLMOutput (line 456) | class TFMaskedLMOutput(ModelOutput):
  class TFSeq2SeqLMOutput (line 485) | class TFSeq2SeqLMOutput(ModelOutput):
  class TFNextSentencePredictorOutput (line 544) | class TFNextSentencePredictorOutput(ModelOutput):
  class TFSequenceClassifierOutput (line 574) | class TFSequenceClassifierOutput(ModelOutput):
  class TFSeq2SeqSequenceClassifierOutput (line 603) | class TFSeq2SeqSequenceClassifierOutput(ModelOutput):
  class TFSemanticSegmenterOutput (line 659) | class TFSemanticSegmenterOutput(ModelOutput):
  class TFSemanticSegmenterOutputWithNoAttention (line 696) | class TFSemanticSegmenterOutputWithNoAttention(ModelOutput):
  class TFImageClassifierOutput (line 727) | class TFImageClassifierOutput(ModelOutput):
  class TFMultipleChoiceModelOutput (line 754) | class TFMultipleChoiceModelOutput(ModelOutput):
  class TFTokenClassifierOutput (line 785) | class TFTokenClassifierOutput(ModelOutput):
  class TFQuestionAnsweringModelOutput (line 814) | class TFQuestionAnsweringModelOutput(ModelOutput):
  class TFSeq2SeqQuestionAnsweringModelOutput (line 846) | class TFSeq2SeqQuestionAnsweringModelOutput(ModelOutput):
  class TFSequenceClassifierOutputWithPast (line 901) | class TFSequenceClassifierOutputWithPast(ModelOutput):
  class TFImageClassifierOutputWithNoAttention (line 937) | class TFImageClassifierOutputWithNoAttention(ModelOutput):
  class TFMaskedImageModelingOutput (line 958) | class TFMaskedImageModelingOutput(ModelOutput):
    method logits (line 985) | def logits(self):

FILE: transformers/modeling_tf_pytorch_utils.py
  class TransposeType (line 31) | class TransposeType(ExplicitEnum):
  function convert_tf_weight_name_to_pt_weight_name (line 42) | def convert_tf_weight_name_to_pt_weight_name(
  function apply_transpose (line 114) | def apply_transpose(transpose: TransposeType, weight, match_shape=None, ...
  function load_pytorch_checkpoint_in_tf2_model (line 156) | def load_pytorch_checkpoint_in_tf2_model(
  function load_pytorch_model_in_tf2_model (line 200) | def load_pytorch_model_in_tf2_model(tf_model, pt_model, tf_inputs=None, ...
  function load_pytorch_weights_in_tf2_model (line 209) | def load_pytorch_weights_in_tf2_model(
  function load_pytorch_state_dict_in_tf2_model (line 241) | def load_pytorch_state_dict_in_tf2_model(
  function load_tf2_checkpoint_in_pytorch_model (line 410) | def load_tf2_checkpoint_in_pytorch_model(
  function load_tf2_model_in_pytorch_model (line 451) | def load_tf2_model_in_pytorch_model(pt_model, tf_model, allow_missing_ke...
  function load_tf2_weights_in_pytorch_model (line 460) | def load_tf2_weights_in_pytorch_model(pt_model, tf_weights, allow_missin...
  function load_tf2_state_dict_in_pytorch_model (line 478) | def load_tf2_state_dict_in_pytorch_model(pt_model, tf_state_dict, allow_...

FILE: transformers/modeling_tf_utils.py
  function dummy_loss (line 113) | def dummy_loss(y_true, y_pred):
  class TFModelUtilsMixin (line 121) | class TFModelUtilsMixin:
    method num_parameters (line 126) | def num_parameters(self, only_trainable: bool = False) -> int:
  function keras_serializable (line 143) | def keras_serializable(cls):
  class TFCausalLanguageModelingLoss (line 208) | class TFCausalLanguageModelingLoss:
    method hf_compute_loss (line 219) | def hf_compute_loss(self, labels, logits):
  class TFQuestionAnsweringLoss (line 239) | class TFQuestionAnsweringLoss:
    method hf_compute_loss (line 244) | def hf_compute_loss(self, labels, logits):
  class TFTokenClassificationLoss (line 254) | class TFTokenClassificationLoss:
    method hf_compute_loss (line 265) | def hf_compute_loss(self, labels, logits):
  class TFSequenceClassificationLoss (line 298) | class TFSequenceClassificationLoss:
    method hf_compute_loss (line 303) | def hf_compute_loss(self, labels, logits):
  class TFMultipleChoiceLoss (line 317) | class TFMultipleChoiceLoss:
    method hf_compute_loss (line 320) | def hf_compute_loss(self, labels, logits):
  class TFMaskedLanguageModelingLoss (line 327) | class TFMaskedLanguageModelingLoss(TFCausalLanguageModelingLoss):
  class TFNextSentencePredictionLoss (line 339) | class TFNextSentencePredictionLoss:
    method hf_compute_loss (line 350) | def hf_compute_loss(self, labels, logits):
  function booleans_processing (line 375) | def booleans_processing(config, **kwargs):
  function unpack_inputs (line 408) | def unpack_inputs(func):
  function input_processing (line 452) | def input_processing(func, config, **kwargs):
  function dtype_byte_size (line 625) | def dtype_byte_size(dtype):
  function format_weight_name (line 645) | def format_weight_name(name, _prefix=None):
  function tf_shard_checkpoint (line 653) | def tf_shard_checkpoint(weights, max_shard_size="10GB"):
  function load_tf_sharded_weights (line 719) | def load_tf_sharded_weights(model, shard_files, ignore_mismatched_sizes=...
  function load_tf_shard (line 786) | def load_tf_shard(model, model_layer_map, resolved_archive_file, ignore_...
  function load_tf_weights (line 874) | def load_tf_weights(model, resolved_archive_file, ignore_mismatched_size...
  function load_tf_weights_from_h5 (line 901) | def load_tf_weights_from_h5(model, resolved_archive_file, ignore_mismatc...
  function load_tf_weights_from_safetensors (line 1001) | def load_tf_weights_from_safetensors(model, resolved_archive_file, ignor...
  function init_copy_embeddings (line 1042) | def init_copy_embeddings(old_embeddings, new_num_tokens):
  class TFPreTrainedModel (line 1082) | class TFPreTrainedModel(tf.keras.Model, TFModelUtilsMixin, TFGenerationM...
    method dummy_inputs (line 1117) | def dummy_inputs(self) -> Dict[str, tf.Tensor]:
    method framework (line 1149) | def framework(self) -> str:
    method build (line 1155) | def build(self, input_shape=None):
    method __init__ (line 1162) | def __init__(self, config, *inputs, **kwargs):
    method get_config (line 1181) | def get_config(self):
    method from_config (line 1185) | def from_config(cls, config, **kwargs):
    method _from_config (line 1191) | def _from_config(cls, config, **kwargs):
    method get_head_mask (line 1197) | def get_head_mask(self, head_mask: tf.Tensor | None, num_hidden_layers...
    method _convert_head_mask_to_5d (line 1218) | def _convert_head_mask_to_5d(self, head_mask, num_hidden_layers):
    method eager_serving (line 1229) | def eager_serving(self, inputs):
    method input_signature (line 1243) | def input_signature(self) -> Dict[str, tf.TensorSpec]:
    method _prune_signature (line 1290) | def _prune_signature(self, signature):
    method serving_output (line 1295) | def serving_output(self, output):
    method can_generate (line 1319) | def can_generate(self) -> bool:
    method get_input_embeddings (line 1331) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method _save_checkpoint (line 1345) | def _save_checkpoint(self, checkpoint_dir, epoch):
    method load_repo_checkpoint (line 1358) | def load_repo_checkpoint(self, repo_path_or_name):
    method prepare_tf_dataset (line 1412) | def prepare_tf_dataset(
    method compile (line 1517) | def compile(
    method compute_loss (line 1571) | def compute_loss(self, *args, **kwargs):
    method get_label_to_output_name_mapping (line 1585) | def get_label_to_output_name_mapping(self):
    method train_step (line 1600) | def train_step(self, data):
    method test_step (line 1708) | def test_step(self, data):
    method create_model_card (line 1812) | def create_model_card(
    method set_input_embeddings (line 1872) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1892) | def get_output_embeddings(self) -> Union[None, tf.keras.layers.Layer]:
    method set_output_embeddings (line 1912) | def set_output_embeddings(self, value):
    method get_output_layer_with_bias (line 1929) | def get_output_layer_with_bias(self) -> Union[None, tf.keras.layers.La...
    method get_prefix_bias_name (line 1942) | def get_prefix_bias_name(self) -> Union[None, str]:
    method get_bias (line 1952) | def get_bias(self) -> Union[None, Dict[str, tf.Variable]]:
    method set_bias (line 1969) | def set_bias(self, value):
    method get_lm_head (line 1985) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method resize_token_embeddings (line 1994) | def resize_token_embeddings(
    method _v2_resized_token_embeddings (line 2027) | def _v2_resized_token_embeddings(self, new_num_tokens: Optional[int] =...
    method _get_word_embedding_weight (line 2050) | def _get_word_embedding_weight(model, embedding_layer):
    method _resize_token_embeddings (line 2081) | def _resize_token_embeddings(self, new_num_tokens):
    method _v2_resize_token_embeddings (line 2104) | def _v2_resize_token_embeddings(self, new_num_tokens):
    method _get_resized_lm_head_bias (line 2125) | def _get_resized_lm_head_bias(self, old_lm_head_bias, new_num_tokens):
    method _v2_get_resized_lm_head_bias (line 2178) | def _v2_get_resized_lm_head_bias(
    method _get_resized_lm_head_decoder (line 2212) | def _get_resized_lm_head_decoder(self, old_lm_head_decoder, new_num_to...
    method _get_resized_embeddings (line 2250) | def _get_resized_embeddings(self, old_embeddings, new_num_tokens=None)...
    method _v2_get_resized_embeddings (line 2285) | def _v2_get_resized_embeddings(
    method prune_heads (line 2332) | def prune_heads(self, heads_to_prune):
    method save_pretrained (line 2344) | def save_pretrained(
    method from_pretrained (line 2501) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
    method push_to_hub (line 3029) | def push_to_hub(
    method register_for_auto_class (line 3126) | def register_for_auto_class(cls, auto_class="TFAutoModel"):
  class TFConv1D (line 3152) | class TFConv1D(tf.keras.layers.Layer):
    method __init__ (line 3169) | def __init__(self, nf, nx, initializer_range=0.02, **kwargs):
    method build (line 3175) | def build(self, input_shape):
    method call (line 3181) | def call(self, x):
  class TFSharedEmbeddings (line 3192) | class TFSharedEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 3212) | def __init__(self, vocab_size: int, hidden_size: int, initializer_rang...
    method build (line 3222) | def build(self, input_shape):
    method get_config (line 3232) | def get_config(self):
    method call (line 3242) | def call(self, inputs: tf.Tensor, mode: str = "embedding") -> tf.Tensor:
    method _embedding (line 3274) | def _embedding(self, input_ids):
    method _linear (line 3278) | def _linear(self, inputs):
  class TFSequenceSummary (line 3295) | class TFSequenceSummary(tf.keras.layers.Layer):
    method __init__ (line 3325) | def __init__(self, config: PretrainedConfig, initializer_range: float ...
    method call (line 3359) | def call(self, inputs, cls_index=None, training=False):
  function get_initializer (line 3411) | def get_initializer(initializer_range: float = 0.02) -> tf.keras.initial...

FILE: transformers/modeling_utils.py
  function no_init_weights (line 127) | def no_init_weights(_enable=True):
  class Identity (line 147) | class Identity(nn.Module):
    method __init__ (line 150) | def __init__(self, *args, **kwargs):
    method forward (line 153) | def forward(self, input):
  function get_parameter_device (line 157) | def get_parameter_device(parameter: Union[nn.Module, GenerationMixin, "M...
  function get_first_parameter_dtype (line 172) | def get_first_parameter_dtype(parameter: Union[nn.Module, GenerationMixi...
  function get_parameter_dtype (line 190) | def get_parameter_dtype(parameter: Union[nn.Module, GenerationMixin, "Mo...
  function get_state_dict_float_dtype (line 239) | def get_state_dict_float_dtype(state_dict):
  function get_state_dict_dtype (line 250) | def get_state_dict_dtype(state_dict):
  function dtype_byte_size (line 263) | def dtype_byte_size(dtype):
  function shard_checkpoint (line 283) | def shard_checkpoint(
  function load_sharded_checkpoint (line 360) | def load_sharded_checkpoint(model, folder, strict=True, prefer_safe=True):
  function load_state_dict (line 444) | def load_state_dict(checkpoint_file: Union[str, os.PathLike]):
  function set_initialized_submodules (line 486) | def set_initialized_submodules(model, state_dict_keys):
  function _load_state_dict_into_model (line 497) | def _load_state_dict_into_model(model_to_load, state_dict, start_prefix):
  function find_submodule_and_param_name (line 558) | def find_submodule_and_param_name(model, long_key, start_prefix):
  function _move_model_to_meta (line 581) | def _move_model_to_meta(model, loaded_state_dict_keys, start_prefix):
  function _load_state_dict_into_meta_model (line 607) | def _load_state_dict_into_meta_model(
  function _add_variant (line 735) | def _add_variant(weights_name: str, variant: Optional[str] = None) -> str:
  class ModuleUtilsMixin (line 744) | class ModuleUtilsMixin:
    method _hook_rss_memory_pre_forward (line 750) | def _hook_rss_memory_pre_forward(module, *args, **kwargs):
    method _hook_rss_memory_post_forward (line 762) | def _hook_rss_memory_post_forward(module, *args, **kwargs):
    method add_memory_hooks (line 775) | def add_memory_hooks(self):
    method reset_memory_hooks_state (line 787) | def reset_memory_hooks_state(self):
    method device (line 797) | def device(self) -> torch.device:
    method dtype (line 805) | def dtype(self) -> torch.dtype:
    method invert_attention_mask (line 811) | def invert_attention_mask(self, encoder_attention_mask: Tensor) -> Ten...
    method create_extended_attention_mask_for_decoder (line 836) | def create_extended_attention_mask_for_decoder(input_shape, attention_...
    method get_extended_attention_mask (line 863) | def get_extended_attention_mask(
    method get_head_mask (line 915) | def get_head_mask(
    method _convert_head_mask_to_5d (line 942) | def _convert_head_mask_to_5d(self, head_mask, num_hidden_layers):
    method num_parameters (line 953) | def num_parameters(self, only_trainable: bool = False, exclude_embeddi...
    method estimate_tokens (line 979) | def estimate_tokens(self, input_dict: Dict[str, Union[torch.Tensor, An...
    method floating_point_ops (line 1000) | def floating_point_ops(
  class PreTrainedModel (line 1027) | class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, Push...
    method dummy_inputs (line 1077) | def dummy_inputs(self) -> Dict[str, torch.Tensor]:
    method framework (line 1084) | def framework(self) -> str:
    method __init__ (line 1090) | def __init__(self, config: PretrainedConfig, *inputs, **kwargs):
    method post_init (line 1104) | def post_init(self):
    method _backward_compatibility_gradient_checkpointing (line 1112) | def _backward_compatibility_gradient_checkpointing(self):
    method _from_config (line 1119) | def _from_config(cls, config, **kwargs):
    method _set_default_torch_dtype (line 1152) | def _set_default_torch_dtype(cls, dtype: torch.dtype) -> torch.dtype:
    method base_model (line 1179) | def base_model(self) -> nn.Module:
    method can_generate (line 1185) | def can_generate(self) -> bool:
    method enable_input_require_grads (line 1197) | def enable_input_require_grads(self):
    method disable_input_require_grads (line 1208) | def disable_input_require_grads(self):
    method get_input_embeddings (line 1214) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 1227) | def set_input_embeddings(self, value: nn.Module):
    method get_output_embeddings (line 1240) | def get_output_embeddings(self) -> nn.Module:
    method _init_weights (line 1249) | def _init_weights(self, module):
    method _initialize_weights (line 1255) | def _initialize_weights(self, module):
    method tie_weights (line 1264) | def tie_weights(self):
    method _tie_encoder_decoder_weights (line 1286) | def _tie_encoder_decoder_weights(encoder: nn.Module, decoder: nn.Modul...
    method _tie_or_clone_weights (line 1360) | def _tie_or_clone_weights(self, output_embeddings, input_embeddings):
    method resize_token_embeddings (line 1380) | def resize_token_embeddings(self, new_num_tokens: Optional[int] = None...
    method _resize_token_embeddings (line 1408) | def _resize_token_embeddings(self, new_num_tokens):
    method _get_resized_embeddings (line 1421) | def _get_resized_embeddings(
    method _get_resized_lm_head (line 1485) | def _get_resized_lm_head(
    method resize_position_embeddings (line 1577) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_position_embeddings (line 1583) | def get_position_embeddings(self) -> Union[nn.Embedding, Tuple[nn.Embe...
    method init_weights (line 1589) | def init_weights(self):
    method prune_heads (line 1606) | def prune_heads(self, heads_to_prune: Dict[int, List[int]]):
    method gradient_checkpointing_enable (line 1623) | def gradient_checkpointing_enable(self):
    method gradient_checkpointing_disable (line 1634) | def gradient_checkpointing_disable(self):
    method is_gradient_checkpointing (line 1645) | def is_gradient_checkpointing(self) -> bool:
    method save_pretrained (line 1654) | def save_pretrained(
    method get_memory_footprint (line 1876) | def get_memory_footprint(self, return_buffers=True):
    method to (line 1894) | def to(self, *args, **kwargs):
    method half (line 1904) | def half(self, *args):
    method float (line 1914) | def float(self, *args):
    method from_pretrained (line 1925) | def from_pretrained(cls, pretrained_model_name_or_path: Optional[Union...
    method _load_pretrained_model (line 2952) | def _load_pretrained_model(
    method retrieve_modules_from_names (line 3325) | def retrieve_modules_from_names(self, names, add_prefix=False, remove_...
    method _load_pretrained_model_low_mem (line 3349) | def _load_pretrained_model_low_mem(model, loaded_state_dict_keys, reso...
    method register_for_auto_class (line 3373) | def register_for_auto_class(cls, auto_class="AutoModel"):
    method to_bettertransformer (line 3398) | def to_bettertransformer(self) -> "PreTrainedModel":
    method reverse_bettertransformer (line 3426) | def reverse_bettertransformer(self):
  class PoolerStartLogits (line 3456) | class PoolerStartLogits(nn.Module):
    method __init__ (line 3465) | def __init__(self, config: PretrainedConfig):
    method forward (line 3469) | def forward(
  class PoolerEndLogits (line 3494) | class PoolerEndLogits(nn.Module):
    method __init__ (line 3504) | def __init__(self, config: PretrainedConfig):
    method forward (line 3511) | def forward(
  class PoolerAnswerClass (line 3563) | class PoolerAnswerClass(nn.Module):
    method __init__ (line 3572) | def __init__(self, config):
    method forward (line 3578) | def forward(
  class SquadHeadOutput (line 3629) | class SquadHeadOutput(ModelOutput):
  class SQuADHead (line 3659) | class SQuADHead(nn.Module):
    method __init__ (line 3669) | def __init__(self, config):
    method forward (line 3679) | def forward(
  class SequenceSummary (line 3776) | class SequenceSummary(nn.Module):
    method __init__ (line 3802) | def __init__(self, config: PretrainedConfig):
    method forward (line 3831) | def forward(
  function unwrap_model (line 3875) | def unwrap_model(model: nn.Module) -> nn.Module:
  function expand_device_map (line 3889) | def expand_device_map(device_map, param_names):
  function get_disk_only_shard_files (line 3899) | def get_disk_only_shard_files(device_map, sharded_metadata):

FILE: transformers/models/albert/configuration_albert.py
  class AlbertConfig (line 36) | class AlbertConfig(PretrainedConfig):
    method __init__ (line 113) | def __init__(
  class AlbertOnnxConfig (line 159) | class AlbertOnnxConfig(OnnxConfig):
    method inputs (line 161) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/albert/convert_albert_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 29) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, albert_config_f...

FILE: transformers/models/albert/modeling_albert.py
  function load_tf_weights_in_albert (line 68) | def load_tf_weights_in_albert(model, config, tf_checkpoint_path):
  class AlbertEmbeddings (line 194) | class AlbertEmbeddings(nn.Module):
    method __init__ (line 199) | def __init__(self, config: AlbertConfig):
    method forward (line 218) | def forward(
  class AlbertAttention (line 260) | class AlbertAttention(nn.Module):
    method __init__ (line 261) | def __init__(self, config: AlbertConfig):
    method transpose_for_scores (line 290) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method prune_heads (line 295) | def prune_heads(self, heads: List[int]) -> None:
    method forward (line 313) | def forward(
  class AlbertLayer (line 372) | class AlbertLayer(nn.Module):
    method __init__ (line 373) | def __init__(self, config: AlbertConfig):
    method forward (line 386) | def forward(
    method ff_chunk (line 406) | def ff_chunk(self, attention_output: torch.Tensor) -> torch.Tensor:
  class AlbertLayerGroup (line 413) | class AlbertLayerGroup(nn.Module):
    method __init__ (line 414) | def __init__(self, config: AlbertConfig):
    method forward (line 419) | def forward(
  class AlbertTransformer (line 448) | class AlbertTransformer(nn.Module):
    method __init__ (line 449) | def __init__(self, config: AlbertConfig):
    method forward (line 456) | def forward(
  class AlbertPreTrainedModel (line 501) | class AlbertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 512) | def _init_weights(self, module):
  class AlbertForPreTrainingOutput (line 530) | class AlbertForPreTrainingOutput(ModelOutput):
  class AlbertModel (line 633) | class AlbertModel(AlbertPreTrainedModel):
    method __init__ (line 637) | def __init__(self, config: AlbertConfig, add_pooling_layer: bool = True):
    method get_input_embeddings (line 653) | def get_input_embeddings(self) -> nn.Embedding:
    method set_input_embeddings (line 656) | def set_input_embeddings(self, value: nn.Embedding) -> None:
    method _prune_heads (line 659) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
    method forward (line 682) | def forward(
  class AlbertForPreTraining (line 761) | class AlbertForPreTraining(AlbertPreTrainedModel):
    method __init__ (line 768) | def __init__(self, config: AlbertConfig):
    method get_output_embeddings (line 778) | def get_output_embeddings(self) -> nn.Linear:
    method set_output_embeddings (line 781) | def set_output_embeddings(self, new_embeddings: nn.Linear) -> None:
    method get_input_embeddings (line 784) | def get_input_embeddings(self) -> nn.Embedding:
    method forward (line 789) | def forward(
  class AlbertMLMHead (line 870) | class AlbertMLMHead(nn.Module):
    method __init__ (line 871) | def __init__(self, config: AlbertConfig):
    method forward (line 881) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method _tie_weights (line 891) | def _tie_weights(self) -> None:
  class AlbertSOPHead (line 896) | class AlbertSOPHead(nn.Module):
    method __init__ (line 897) | def __init__(self, config: AlbertConfig):
    method forward (line 903) | def forward(self, pooled_output: torch.Tensor) -> torch.Tensor:
  class AlbertForMaskedLM (line 913) | class AlbertForMaskedLM(AlbertPreTrainedModel):
    method __init__ (line 921) | def __init__(self, config):
    method get_output_embeddings (line 930) | def get_output_embeddings(self) -> nn.Linear:
    method set_output_embeddings (line 933) | def set_output_embeddings(self, new_embeddings: nn.Linear) -> None:
    method get_input_embeddings (line 936) | def get_input_embeddings(self) -> nn.Embedding:
    method forward (line 941) | def forward(
  class AlbertForSequenceClassification (line 1032) | class AlbertForSequenceClassification(AlbertPreTrainedModel):
    method __init__ (line 1033) | def __init__(self, config: AlbertConfig):
    method forward (line 1053) | def forward(
  class AlbertForTokenClassification (line 1133) | class AlbertForTokenClassification(AlbertPreTrainedModel):
    method __init__ (line 1136) | def __init__(self, config: AlbertConfig):
    method forward (line 1158) | def forward(
  class AlbertForQuestionAnswering (line 1218) | class AlbertForQuestionAnswering(AlbertPreTrainedModel):
    method __init__ (line 1221) | def __init__(self, config: AlbertConfig):
    method forward (line 1241) | def forward(
  class AlbertForMultipleChoice (line 1323) | class AlbertForMultipleChoice(AlbertPreTrainedModel):
    method __init__ (line 1324) | def __init__(self, config: AlbertConfig):
    method forward (line 1340) | def forward(

FILE: transformers/models/albert/modeling_flax_albert.py
  class FlaxAlbertForPreTrainingOutput (line 55) | class FlaxAlbertForPreTrainingOutput(ModelOutput):
  class FlaxAlbertEmbeddings (line 151) | class FlaxAlbertEmbeddings(nn.Module):
    method setup (line 157) | def setup(self):
    method __call__ (line 177) | def __call__(self, input_ids, token_type_ids, position_ids, determinis...
  class FlaxAlbertSelfAttention (line 192) | class FlaxAlbertSelfAttention(nn.Module):
    method setup (line 196) | def setup(self):
    method __call__ (line 226) | def __call__(self, hidden_states, attention_mask, deterministic=True, ...
  class FlaxAlbertLayer (line 277) | class FlaxAlbertLayer(nn.Module):
    method setup (line 281) | def setup(self):
    method __call__ (line 297) | def __call__(
  class FlaxAlbertLayerCollection (line 321) | class FlaxAlbertLayerCollection(nn.Module):
    method setup (line 325) | def setup(self):
    method __call__ (line 330) | def __call__(
  class FlaxAlbertLayerCollections (line 364) | class FlaxAlbertLayerCollections(nn.Module):
    method setup (line 369) | def setup(self):
    method __call__ (line 372) | def __call__(
  class FlaxAlbertLayerGroups (line 390) | class FlaxAlbertLayerGroups(nn.Module):
    method setup (line 394) | def setup(self):
    method __call__ (line 400) | def __call__(
  class FlaxAlbertEncoder (line 437) | class FlaxAlbertEncoder(nn.Module):
    method setup (line 441) | def setup(self):
    method __call__ (line 449) | def __call__(
  class FlaxAlbertOnlyMLMHead (line 468) | class FlaxAlbertOnlyMLMHead(nn.Module):
    method setup (line 473) | def setup(self):
    method __call__ (line 480) | def __call__(self, hidden_states, shared_embedding=None):
  class FlaxAlbertSOPHead (line 494) | class FlaxAlbertSOPHead(nn.Module):
    method setup (line 498) | def setup(self):
    method __call__ (line 502) | def __call__(self, pooled_output, deterministic=True):
  class FlaxAlbertPreTrainedModel (line 508) | class FlaxAlbertPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 518) | def __init__(
    method init_weights (line 530) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 555) | def __call__(
  class FlaxAlbertModule (line 603) | class FlaxAlbertModule(nn.Module):
    method setup (line 608) | def setup(self):
    method __call__ (line 623) | def __call__(
  class FlaxAlbertModel (line 677) | class FlaxAlbertModel(FlaxAlbertPreTrainedModel):
  class FlaxAlbertForPreTrainingModule (line 684) | class FlaxAlbertForPreTrainingModule(nn.Module):
    method setup (line 688) | def setup(self):
    method __call__ (line 693) | def __call__(
  class FlaxAlbertForPreTraining (line 745) | class FlaxAlbertForPreTraining(FlaxAlbertPreTrainedModel):
  class FlaxAlbertForMaskedLMModule (line 777) | class FlaxAlbertForMaskedLMModule(nn.Module):
    method setup (line 781) | def setup(self):
    method __call__ (line 785) | def __call__(
  class FlaxAlbertForMaskedLM (line 828) | class FlaxAlbertForMaskedLM(FlaxAlbertPreTrainedModel):
  class FlaxAlbertForSequenceClassificationModule (line 835) | class FlaxAlbertForSequenceClassificationModule(nn.Module):
    method setup (line 839) | def setup(self):
    method __call__ (line 852) | def __call__(
  class FlaxAlbertForSequenceClassification (line 896) | class FlaxAlbertForSequenceClassification(FlaxAlbertPreTrainedModel):
  class FlaxAlbertForMultipleChoiceModule (line 908) | class FlaxAlbertForMultipleChoiceModule(nn.Module):
    method setup (line 912) | def setup(self):
    method __call__ (line 917) | def __call__(
  class FlaxAlbertForMultipleChoice (line 969) | class FlaxAlbertForMultipleChoice(FlaxAlbertPreTrainedModel):
  class FlaxAlbertForTokenClassificationModule (line 984) | class FlaxAlbertForTokenClassificationModule(nn.Module):
    method setup (line 988) | def setup(self):
    method __call__ (line 998) | def __call__(
  class FlaxAlbertForTokenClassification (line 1042) | class FlaxAlbertForTokenClassification(FlaxAlbertPreTrainedModel):
  class FlaxAlbertForQuestionAnsweringModule (line 1054) | class FlaxAlbertForQuestionAnsweringModule(nn.Module):
    method setup (line 1058) | def setup(self):
    method __call__ (line 1062) | def __call__(
  class FlaxAlbertForQuestionAnswering (line 1110) | class FlaxAlbertForQuestionAnswering(FlaxAlbertPreTrainedModel):

FILE: transformers/models/albert/modeling_tf_albert.py
  class TFAlbertPreTrainingLoss (line 80) | class TFAlbertPreTrainingLoss:
    method hf_compute_loss (line 86) | def hf_compute_loss(self, labels: tf.Tensor, logits: tf.Tensor) -> tf....
  class TFAlbertEmbeddings (line 136) | class TFAlbertEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 139) | def __init__(self, config: AlbertConfig, **kwargs):
    method build (line 149) | def build(self, input_shape: tf.TensorShape):
    method call (line 174) | def call(
  class TFAlbertAttention (line 215) | class TFAlbertAttention(tf.keras.layers.Layer):
    method __init__ (line 218) | def __init__(self, config: AlbertConfig, **kwargs):
    method transpose_for_scores (line 250) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 257) | def call(
  class TFAlbertLayer (line 311) | class TFAlbertLayer(tf.keras.layers.Layer):
    method __init__ (line 312) | def __init__(self, config: AlbertConfig, **kwargs):
    method call (line 333) | def call(
  class TFAlbertLayerGroup (line 360) | class TFAlbertLayerGroup(tf.keras.layers.Layer):
    method __init__ (line 361) | def __init__(self, config: AlbertConfig, **kwargs):
    method call (line 368) | def call(
  class TFAlbertTransformer (line 403) | class TFAlbertTransformer(tf.keras.layers.Layer):
    method __init__ (line 404) | def __init__(self, config: AlbertConfig, **kwargs):
    method call (line 420) | def call(
  class TFAlbertPreTrainedModel (line 461) | class TFAlbertPreTrainedModel(TFPreTrainedModel):
  class TFAlbertMLMHead (line 471) | class TFAlbertMLMHead(tf.keras.layers.Layer):
    method __init__ (line 472) | def __init__(self, config: AlbertConfig, input_embeddings: tf.keras.la...
    method build (line 491) | def build(self, input_shape: tf.TensorShape):
    method get_output_embeddings (line 499) | def get_output_embeddings(self) -> tf.keras.layers.Layer:
    method set_output_embeddings (line 502) | def set_output_embeddings(self, value: tf.Variable):
    method get_bias (line 506) | def get_bias(self) -> Dict[str, tf.Variable]:
    method set_bias (line 509) | def set_bias(self, value: tf.Variable):
    method call (line 514) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFAlbertMainLayer (line 528) | class TFAlbertMainLayer(tf.keras.layers.Layer):
    method __init__ (line 531) | def __init__(self, config: AlbertConfig, add_pooling_layer: bool = Tru...
    method get_input_embeddings (line 549) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 552) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 556) | def _prune_heads(self, heads_to_prune):
    method call (line 564) | def call(
  class TFAlbertForPreTrainingOutput (line 655) | class TFAlbertForPreTrainingOutput(ModelOutput):
  class TFAlbertModel (line 787) | class TFAlbertModel(TFAlbertPreTrainedModel):
    method __init__ (line 788) | def __init__(self, config: AlbertConfig, *inputs, **kwargs):
    method call (line 800) | def call(
  class TFAlbertForPreTraining (line 836) | class TFAlbertForPreTraining(TFAlbertPreTrainedModel, TFAlbertPreTrainin...
    method __init__ (line 840) | def __init__(self, config: AlbertConfig, *inputs, **kwargs):
    method get_lm_head (line 849) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method call (line 855) | def call(
  class TFAlbertSOPHead (line 925) | class TFAlbertSOPHead(tf.keras.layers.Layer):
    method __init__ (line 926) | def __init__(self, config: AlbertConfig, **kwargs):
    method call (line 936) | def call(self, pooled_output: tf.Tensor, training: bool) -> tf.Tensor:
  class TFAlbertForMaskedLM (line 944) | class TFAlbertForMaskedLM(TFAlbertPreTrainedModel, TFMaskedLanguageModel...
    method __init__ (line 948) | def __init__(self, config: AlbertConfig, *inputs, **kwargs):
    method get_lm_head (line 954) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method call (line 960) | def call(
  class TFAlbertForSequenceClassification (line 1046) | class TFAlbertForSequenceClassification(TFAlbertPreTrainedModel, TFSeque...
    method __init__ (line 1051) | def __init__(self, config: AlbertConfig, *inputs, **kwargs):
    method call (line 1071) | def call(
  class TFAlbertForTokenClassification (line 1128) | class TFAlbertForTokenClassification(TFAlbertPreTrainedModel, TFTokenCla...
    method __init__ (line 1133) | def __init__(self, config: AlbertConfig, *inputs, **kwargs):
    method call (line 1156) | def call(
  class TFAlbertForQuestionAnswering (line 1211) | class TFAlbertForQuestionAnswering(TFAlbertPreTrainedModel, TFQuestionAn...
    method __init__ (line 1215) | def __init__(self, config: AlbertConfig, *inputs, **kwargs):
    method call (line 1236) | def call(
  class TFAlbertForMultipleChoice (line 1306) | class TFAlbertForMultipleChoice(TFAlbertPreTrainedModel, TFMultipleChoic...
    method __init__ (line 1311) | def __init__(self, config: AlbertConfig, *inputs, **kwargs):
    method call (line 1327) | def call(

FILE: transformers/models/albert/tokenization_albert.py
  class AlbertTokenizer (line 59) | class AlbertTokenizer(PreTrainedTokenizer):
    method __init__ (line 136) | def __init__(
    method vocab_size (line 186) | def vocab_size(self):
    method get_vocab (line 189) | def get_vocab(self):
    method __getstate__ (line 194) | def __getstate__(self):
    method __setstate__ (line 199) | def __setstate__(self, d):
    method preprocess_text (line 209) | def preprocess_text(self, inputs):
    method _tokenize (line 224) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 244) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 248) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 252) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 271) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 296) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 324) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 354) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/albert/tokenization_albert_fast.py
  class AlbertTokenizerFast (line 72) | class AlbertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 124) | def __init__(
    method build_inputs_with_special_tokens (line 170) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 195) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 225) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/align/configuration_align.py
  class AlignTextConfig (line 36) | class AlignTextConfig(PretrainedConfig):
    method __init__ (line 103) | def __init__(
    method from_pretrained (line 141) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class AlignVisionConfig (line 157) | class AlignVisionConfig(PretrainedConfig):
    method __init__ (line 231) | def __init__(
    method from_pretrained (line 282) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class AlignConfig (line 298) | class AlignConfig(PretrainedConfig):
    method __init__ (line 349) | def __init__(
    method from_text_vision_configs (line 376) | def from_text_vision_configs(cls, text_config: AlignTextConfig, vision...
    method to_dict (line 387) | def to_dict(self):

FILE: transformers/models/align/convert_align_tf_to_hf.py
  function preprocess (line 44) | def preprocess(image):
  function get_align_config (line 50) | def get_align_config():
  function prepare_img (line 66) | def prepare_img():
  function get_processor (line 72) | def get_processor():
  function rename_keys (line 88) | def rename_keys(original_param_names):
  function replace_params (line 260) | def replace_params(hf_params, tf_params, key_mapping):
  function convert_align_checkpoint (line 288) | def convert_align_checkpoint(checkpoint_path, pytorch_dump_folder_path, ...

FILE: transformers/models/align/modeling_align.py
  class AlignVisionModelOutput (line 190) | class AlignVisionModelOutput(ModelOutput):
  class AlignTextModelOutput (line 212) | class AlignTextModelOutput(ModelOutput):
  class AlignOutput (line 241) | class AlignOutput(ModelOutput):
    method to_tuple (line 270) | def to_tuple(self) -> Tuple[Any]:
  function contrastive_loss (line 279) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  function align_loss (line 283) | def align_loss(similarity: torch.Tensor) -> torch.Tensor:
  function round_filters (line 290) | def round_filters(config: AlignVisionConfig, num_channels: int):
  function correct_pad (line 306) | def correct_pad(kernel_size: Union[int, Tuple], adjust: bool = True):
  class AlignVisionEmbeddings (line 327) | class AlignVisionEmbeddings(nn.Module):
    method __init__ (line 332) | def __init__(self, config: AlignVisionConfig):
    method forward (line 343) | def forward(self, pixel_values: torch.Tensor) -> torch.Tensor:
  class AlignVisionDepthwiseConv2d (line 353) | class AlignVisionDepthwiseConv2d(nn.Conv2d):
    method __init__ (line 354) | def __init__(
  class AlignVisionExpansionLayer (line 380) | class AlignVisionExpansionLayer(nn.Module):
    method __init__ (line 385) | def __init__(self, config: AlignVisionConfig, in_dim: int, out_dim: in...
    method forward (line 397) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class AlignVisionDepthwiseLayer (line 407) | class AlignVisionDepthwiseLayer(nn.Module):
    method __init__ (line 412) | def __init__(
    method forward (line 434) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class AlignVisionSqueezeExciteLayer (line 447) | class AlignVisionSqueezeExciteLayer(nn.Module):
    method __init__ (line 452) | def __init__(self, config: AlignVisionConfig, in_dim: int, expand_dim:...
    method forward (line 473) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class AlignVisionFinalBlockLayer (line 486) | class AlignVisionFinalBlockLayer(nn.Module):
    method __init__ (line 491) | def __init__(
    method forward (line 508) | def forward(self, embeddings: torch.FloatTensor, hidden_states: torch....
  class AlignVisionBlock (line 519) | class AlignVisionBlock(nn.Module):
    method __init__ (line 546) | def __init__(
    method forward (line 587) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class AlignVisionEncoder (line 600) | class AlignVisionEncoder(nn.Module):
    method __init__ (line 609) | def __init__(self, config: AlignVisionConfig):
    method forward (line 652) | def forward(
  class AlignTextEmbeddings (line 675) | class AlignTextEmbeddings(nn.Module):
    method __init__ (line 678) | def __init__(self, config):
    method forward (line 695) | def forward(
  class AlignTextSelfAttention (line 738) | class AlignTextSelfAttention(nn.Module):
    method __init__ (line 739) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 765) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 770) | def forward(
  class AlignTextSelfOutput (line 873) | class AlignTextSelfOutput(nn.Module):
    method __init__ (line 874) | def __init__(self, config):
    method forward (line 880) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class AlignTextAttention (line 888) | class AlignTextAttention(nn.Module):
    method __init__ (line 889) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 895) | def prune_heads(self, heads):
    method forward (line 913) | def forward(
  class AlignTextIntermediate (line 938) | class AlignTextIntermediate(nn.Module):
    method __init__ (line 939) | def __init__(self, config):
    method forward (line 947) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class AlignTextOutput (line 954) | class AlignTextOutput(nn.Module):
    method __init__ (line 955) | def __init__(self, config):
    method forward (line 961) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class AlignTextLayer (line 969) | class AlignTextLayer(nn.Module):
    method __init__ (line 970) | def __init__(self, config):
    method forward (line 984) | def forward(
    method feed_forward_chunk (line 1049) | def feed_forward_chunk(self, attention_output):
  class AlignTextEncoder (line 1056) | class AlignTextEncoder(nn.Module):
    method __init__ (line 1057) | def __init__(self, config):
    method forward (line 1063) | def forward(
  class AlignTextPooler (line 1155) | class AlignTextPooler(nn.Module):
    method __init__ (line 1156) | def __init__(self, config):
    method forward (line 1161) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class AlignPreTrainedModel (line 1170) | class AlignPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1181) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1199) | def _set_gradient_checkpointing(self, module, value=False):
  class AlignTextModel (line 1208) | class AlignTextModel(AlignPreTrainedModel):
    method __init__ (line 1211) | def __init__(self, config: AlignTextConfig, add_pooling_layer: bool = ...
    method get_input_embeddings (line 1223) | def get_input_embeddings(self):
    method set_input_embeddings (line 1226) | def set_input_embeddings(self, value):
    method forward (line 1231) | def forward(
  class AlignVisionModel (line 1333) | class AlignVisionModel(AlignPreTrainedModel):
    method __init__ (line 1337) | def __init__(self, config: AlignVisionConfig):
    method get_input_embeddings (line 1354) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 1359) | def forward(
  class AlignModel (line 1418) | class AlignModel(AlignPreTrainedModel):
    method __init__ (line 1421) | def __init__(self, config: AlignConfig):
    method get_text_features (line 1452) | def get_text_features(
    method get_image_features (line 1505) | def get_image_features(
    method forward (line 1551) | def forward(

FILE: transformers/models/align/processing_align.py
  class AlignProcessor (line 24) | class AlignProcessor(ProcessorMixin):
    method __init__ (line 42) | def __init__(self, image_processor, tokenizer):
    method __call__ (line 45) | def __call__(self, text=None, images=None, padding="max_length", max_l...
    method batch_decode (line 104) | def batch_decode(self, *args, **kwargs):
    method decode (line 111) | def decode(self, *args, **kwargs):
    method model_input_names (line 119) | def model_input_names(self):

FILE: transformers/models/altclip/configuration_altclip.py
  class AltCLIPTextConfig (line 32) | class AltCLIPTextConfig(PretrainedConfig):
    method __init__ (line 99) | def __init__(
  class AltCLIPVisionConfig (line 142) | class AltCLIPVisionConfig(PretrainedConfig):
    method __init__ (line 196) | def __init__(
    method from_pretrained (line 230) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class AltCLIPConfig (line 246) | class AltCLIPConfig(PretrainedConfig):
    method __init__ (line 294) | def __init__(
    method from_text_vision_configs (line 383) | def from_text_vision_configs(cls, text_config: AltCLIPTextConfig, visi...
    method to_dict (line 394) | def to_dict(self):

FILE: transformers/models/altclip/modeling_altclip.py
  function contrastive_loss (line 151) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  function clip_loss (line 155) | def clip_loss(similarity: torch.Tensor) -> torch.Tensor:
  class AltCLIPOutput (line 163) | class AltCLIPOutput(ModelOutput):
    method to_tuple (line 193) | def to_tuple(self) -> Tuple[Any]:
  class AltRobertaEmbeddings (line 201) | class AltRobertaEmbeddings(nn.Module):
    method __init__ (line 207) | def __init__(self, config):
    method forward (line 230) | def forward(
    method create_position_ids_from_inputs_embeds (line 270) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class AltRobertaSelfAttention (line 289) | class AltRobertaSelfAttention(nn.Module):
    method __init__ (line 290) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 316) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 321) | def forward(
  class AltRobertaSelfOutput (line 424) | class AltRobertaSelfOutput(nn.Module):
    method __init__ (line 425) | def __init__(self, config):
    method forward (line 431) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class AltRobertaAttention (line 439) | class AltRobertaAttention(nn.Module):
    method __init__ (line 440) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 446) | def prune_heads(self, heads):
    method forward (line 464) | def forward(
  class AltRobertaIntermediate (line 489) | class AltRobertaIntermediate(nn.Module):
    method __init__ (line 490) | def __init__(self, config):
    method forward (line 498) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class AltRobertaOutput (line 505) | class AltRobertaOutput(nn.Module):
    method __init__ (line 506) | def __init__(self, config):
    method forward (line 512) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class AltRobertaLayer (line 520) | class AltRobertaLayer(nn.Module):
    method __init__ (line 521) | def __init__(self, config):
    method forward (line 535) | def forward(
    method feed_forward_chunk (line 600) | def feed_forward_chunk(self, attention_output):
  class AltRobertaEncoder (line 607) | class AltRobertaEncoder(nn.Module):
    method __init__ (line 608) | def __init__(self, config):
    method forward (line 614) | def forward(
  class AltRobertaPooler (line 706) | class AltRobertaPooler(nn.Module):
    method __init__ (line 707) | def __init__(self, config):
    method forward (line 712) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class AltCLIPAttention (line 722) | class AltCLIPAttention(nn.Module):
    method __init__ (line 725) | def __init__(self, config):
    method _shape (line 744) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 747) | def forward(
  class AltCLIPMLP (line 827) | class AltCLIPMLP(nn.Module):
    method __init__ (line 828) | def __init__(self, config):
    method forward (line 835) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class AltCLIPEncoderLayer (line 843) | class AltCLIPEncoderLayer(nn.Module):
    method __init__ (line 844) | def __init__(self, config: AltCLIPConfig):
    method forward (line 852) | def forward(
  class AltCLIPEncoder (line 894) | class AltCLIPEncoder(nn.Module):
    method __init__ (line 903) | def __init__(self, config: AltCLIPConfig):
    method forward (line 909) | def forward(
  class AltCLIPVisionEmbeddings (line 998) | class AltCLIPVisionEmbeddings(nn.Module):
    method __init__ (line 999) | def __init__(self, config: AltCLIPVisionConfig):
    method forward (line 1021) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class AltCLIPPreTrainedModel (line 1032) | class AltCLIPPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1043) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1090) | def _set_gradient_checkpointing(self, module, value=False):
  class AltCLIPVisionTransformer (line 1098) | class AltCLIPVisionTransformer(nn.Module):
    method __init__ (line 1099) | def __init__(self, config: AltCLIPVisionConfig):
    method forward (line 1111) | def forward(
  class AltCLIPVisionModel (line 1156) | class AltCLIPVisionModel(AltCLIPPreTrainedModel):
    method __init__ (line 1160) | def __init__(self, config: AltCLIPVisionConfig):
    method get_input_embeddings (line 1166) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 1171) | def forward(
  class AltRobertaModel (line 1210) | class AltRobertaModel(AltCLIPPreTrainedModel):
    method __init__ (line 1229) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 1241) | def get_input_embeddings(self):
    method set_input_embeddings (line 1244) | def set_input_embeddings(self, value):
    method _prune_heads (line 1247) | def _prune_heads(self, heads_to_prune):
    method forward (line 1256) | def forward(
  class AltCLIPTextModel (line 1386) | class AltCLIPTextModel(AltCLIPPreTrainedModel):
    method __init__ (line 1389) | def __init__(self, config):
    method get_input_embeddings (line 1396) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 1399) | def set_input_embeddings(self, value: nn.Embedding) -> None:
    method resize_token_embeddings (line 1402) | def resize_token_embeddings(self, new_num_tokens: Optional[int] = None...
    method forward (line 1407) | def forward(
  class AltCLIPModel (line 1478) | class AltCLIPModel(AltCLIPPreTrainedModel):
    method __init__ (line 1481) | def __init__(self, config: AltCLIPConfig):
    method get_text_features (line 1513) | def get_text_features(
    method get_image_features (line 1560) | def get_image_features(
    method forward (line 1607) | def forward(
  function create_position_ids_from_input_ids (line 1699) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/altclip/processing_altclip.py
  class AltCLIPProcessor (line 24) | class AltCLIPProcessor(ProcessorMixin):
    method __init__ (line 42) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 59) | def __call__(self, text=None, images=None, return_tensors=None, **kwar...
    method batch_decode (line 112) | def batch_decode(self, *args, **kwargs):
    method decode (line 119) | def decode(self, *args, **kwargs):
    method model_input_names (line 127) | def model_input_names(self):

FILE: transformers/models/audio_spectrogram_transformer/configuration_audio_spectrogram_transformer.py
  class ASTConfig (line 31) | class ASTConfig(PretrainedConfig):
    method __init__ (line 91) | def __init__(

FILE: transformers/models/audio_spectrogram_transformer/convert_audio_spectrogram_transformer_original_to_pytorch.py
  function get_audio_spectrogram_transformer_config (line 35) | def get_audio_spectrogram_transformer_config(model_name):
  function rename_key (line 70) | def rename_key(name):
  function convert_state_dict (line 108) | def convert_state_dict(orig_state_dict, config):
  function remove_keys (line 142) | def remove_keys(state_dict):
  function convert_audio_spectrogram_transformer_checkpoint (line 154) | def convert_audio_spectrogram_transformer_checkpoint(model_name, pytorch...

FILE: transformers/models/audio_spectrogram_transformer/feature_extraction_audio_spectrogram_transformer.py
  class ASTFeatureExtractor (line 33) | class ASTFeatureExtractor(SequenceFeatureExtractor):
    method __init__ (line 65) | def __init__(
    method _extract_fbank_features (line 86) | def _extract_fbank_features(
    method normalize (line 122) | def normalize(self, input_values: np.ndarray) -> np.ndarray:
    method __call__ (line 125) | def __call__(

FILE: transformers/models/audio_spectrogram_transformer/modeling_audio_spectrogram_transformer.py
  class ASTEmbeddings (line 54) | class ASTEmbeddings(nn.Module):
    method __init__ (line 59) | def __init__(self, config: ASTConfig) -> None:
    method get_shape (line 72) | def get_shape(self, config):
    method forward (line 80) | def forward(self, input_values: torch.Tensor) -> torch.Tensor:
  class ASTPatchEmbeddings (line 93) | class ASTPatchEmbeddings(nn.Module):
    method __init__ (line 99) | def __init__(self, config):
    method forward (line 110) | def forward(self, input_values: torch.Tensor) -> torch.Tensor:
  class ASTSelfAttention (line 118) | class ASTSelfAttention(nn.Module):
    method __init__ (line 119) | def __init__(self, config: ASTConfig) -> None:
    method transpose_for_scores (line 137) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 142) | def forward(
  class ASTSelfOutput (line 179) | class ASTSelfOutput(nn.Module):
    method __init__ (line 185) | def __init__(self, config: ASTConfig) -> None:
    method forward (line 190) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ASTAttention (line 198) | class ASTAttention(nn.Module):
    method __init__ (line 199) | def __init__(self, config: ASTConfig) -> None:
    method prune_heads (line 205) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 223) | def forward(
  class ASTIntermediate (line 238) | class ASTIntermediate(nn.Module):
    method __init__ (line 239) | def __init__(self, config: ASTConfig) -> None:
    method forward (line 247) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ASTOutput (line 255) | class ASTOutput(nn.Module):
    method __init__ (line 256) | def __init__(self, config: ASTConfig) -> None:
    method forward (line 261) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ASTLayer (line 271) | class ASTLayer(nn.Module):
    method __init__ (line 274) | def __init__(self, config: ASTConfig) -> None:
    method forward (line 284) | def forward(
  class ASTEncoder (line 314) | class ASTEncoder(nn.Module):
    method __init__ (line 315) | def __init__(self, config: ASTConfig) -> None:
    method forward (line 321) | def forward(
  class ASTPreTrainedModel (line 371) | class ASTPreTrainedModel(PreTrainedModel):
    method _init_weights (line 383) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 398) | def _set_gradient_checkpointing(self, module: ASTEncoder, value: bool ...
  class ASTModel (line 445) | class ASTModel(ASTPreTrainedModel):
    method __init__ (line 446) | def __init__(self, config: ASTConfig):
    method get_input_embeddings (line 458) | def get_input_embeddings(self) -> ASTPatchEmbeddings:
    method _prune_heads (line 461) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
    method forward (line 477) | def forward(
  class ASTMLPHead (line 526) | class ASTMLPHead(nn.Module):
    method __init__ (line 527) | def __init__(self, config: ASTConfig):
    method forward (line 532) | def forward(self, hidden_state):
  class ASTForAudioClassification (line 545) | class ASTForAudioClassification(ASTPreTrainedModel):
    method __init__ (line 546) | def __init__(self, config: ASTConfig) -> None:
    method forward (line 567) | def forward(

FILE: transformers/models/auto/auto_factory.py
  function _get_model_class (line 374) | def _get_model_class(config, model_mapping):
  class _BaseAutoModelClass (line 394) | class _BaseAutoModelClass:
    method __init__ (line 398) | def __init__(self, *args, **kwargs):
    method from_config (line 406) | def from_config(cls, config, **kwargs):
    method from_pretrained (line 433) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
    method register (line 493) | def register(cls, config_class, model_class):
  class _BaseAutoBackboneClass (line 512) | class _BaseAutoBackboneClass(_BaseAutoModelClass):
    method _load_timm_backbone_from_pretrained (line 517) | def _load_timm_backbone_from_pretrained(cls, pretrained_model_name_or_...
    method from_pretrained (line 547) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
  function insert_head_doc (line 554) | def insert_head_doc(docstring, head_doc=""):
  function auto_class_update (line 565) | def auto_class_update(cls, checkpoint_for_example="bert-base-cased", hea...
  function get_values (line 600) | def get_values(model_mapping):
  function getattribute_from_module (line 611) | def getattribute_from_module(module, attr):
  class _LazyAutoMapping (line 631) | class _LazyAutoMapping(OrderedDict):
    method __init__ (line 640) | def __init__(self, config_mapping, model_mapping):
    method __len__ (line 647) | def __len__(self):
    method __getitem__ (line 651) | def __getitem__(self, key):
    method _load_attr_from_module (line 667) | def _load_attr_from_module(self, model_type, attr):
    method keys (line 673) | def keys(self):
    method get (line 681) | def get(self, key, default):
    method __bool__ (line 687) | def __bool__(self):
    method values (line 690) | def values(self):
    method items (line 698) | def items(self):
    method __iter__ (line 709) | def __iter__(self):
    method __contains__ (line 712) | def __contains__(self, item):
    method register (line 720) | def register(self, key, value):

FILE: transformers/models/auto/configuration_auto.py
  function model_type_to_module_name (line 636) | def model_type_to_module_name(key):
  function config_class_to_model_type (line 645) | def config_class_to_model_type(config):
  class _LazyConfigMapping (line 657) | class _LazyConfigMapping(OrderedDict):
    method __init__ (line 662) | def __init__(self, mapping):
    method __getitem__ (line 667) | def __getitem__(self, key):
    method keys (line 684) | def keys(self):
    method values (line 687) | def values(self):
    method items (line 690) | def items(self):
    method __iter__ (line 693) | def __iter__(self):
    method __contains__ (line 696) | def __contains__(self, item):
    method register (line 699) | def register(self, key, value):
  class _LazyLoadAllMappings (line 711) | class _LazyLoadAllMappings(OrderedDict):
    method __init__ (line 720) | def __init__(self, mapping):
    method _initialize (line 725) | def _initialize(self):
    method __getitem__ (line 742) | def __getitem__(self, key):
    method keys (line 746) | def keys(self):
    method values (line 750) | def values(self):
    method items (line 754) | def items(self):
    method __iter__ (line 758) | def __iter__(self):
    method __contains__ (line 762) | def __contains__(self, item):
  function _get_class_name (line 770) | def _get_class_name(model_class: Union[str, List[str]]):
  function _list_model_options (line 776) | def _list_model_options(indent, config_to_class=None, use_model_types=Tr...
  function replace_list_option_in_docstrings (line 809) | def replace_list_option_in_docstrings(config_to_class=None, use_model_ty...
  class AutoConfig (line 833) | class AutoConfig:
    method __init__ (line 841) | def __init__(self):
    method for_model (line 848) | def for_model(cls, model_type: str, *args, **kwargs):
    method from_pretrained (line 858) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method register (line 973) | def register(model_type, config):

FILE: transformers/models/auto/feature_extraction_auto.py
  function feature_extractor_class_from_name (line 107) | def feature_extractor_class_from_name(class_name: str):
  function get_feature_extractor_config (line 131) | def get_feature_extractor_config(
  class AutoFeatureExtractor (line 221) | class AutoFeatureExtractor:
    method __init__ (line 229) | def __init__(self):
    method from_pretrained (line 237) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method register (line 357) | def register(config_class, feature_extractor_class):

FILE: transformers/models/auto/image_processing_auto.py
  function image_processor_class_from_name (line 114) | def image_processor_class_from_name(class_name: str):
  function get_image_processor_config (line 138) | def get_image_processor_config(
  class AutoImageProcessor (line 228) | class AutoImageProcessor:
    method __init__ (line 236) | def __init__(self):
    method from_pretrained (line 244) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method register (line 382) | def register(config_class, image_processor_class):

FILE: transformers/models/auto/modeling_auto.py
  class AutoModelForMaskGeneration (line 1091) | class AutoModelForMaskGeneration(_BaseAutoModelClass):
  class AutoModel (line 1095) | class AutoModel(_BaseAutoModelClass):
  class AutoModelForPreTraining (line 1102) | class AutoModelForPreTraining(_BaseAutoModelClass):
  class _AutoModelWithLMHead (line 1110) | class _AutoModelWithLMHead(_BaseAutoModelClass):
  class AutoModelForCausalLM (line 1117) | class AutoModelForCausalLM(_BaseAutoModelClass):
  class AutoModelForMaskedLM (line 1124) | class AutoModelForMaskedLM(_BaseAutoModelClass):
  class AutoModelForSeq2SeqLM (line 1131) | class AutoModelForSeq2SeqLM(_BaseAutoModelClass):
  class AutoModelForSequenceClassification (line 1140) | class AutoModelForSequenceClassification(_BaseAutoModelClass):
  class AutoModelForQuestionAnswering (line 1149) | class AutoModelForQuestionAnswering(_BaseAutoModelClass):
  class AutoModelForTableQuestionAnswering (line 1156) | class AutoModelForTableQuestionAnswering(_BaseAutoModelClass):
  class AutoModelForVisualQuestionAnswering (line 1167) | class AutoModelForVisualQuestionAnswering(_BaseAutoModelClass):
  class AutoModelForDocumentQuestionAnswering (line 1178) | class AutoModelForDocumentQuestionAnswering(_BaseAutoModelClass):
  class AutoModelForTokenClassification (line 1189) | class AutoModelForTokenClassification(_BaseAutoModelClass):
  class AutoModelForMultipleChoice (line 1196) | class AutoModelForMultipleChoice(_BaseAutoModelClass):
  class AutoModelForNextSentencePrediction (line 1203) | class AutoModelForNextSentencePrediction(_BaseAutoModelClass):
  class AutoModelForImageClassification (line 1212) | class AutoModelForImageClassification(_BaseAutoModelClass):
  class AutoModelForZeroShotImageClassification (line 1219) | class AutoModelForZeroShotImageClassification(_BaseAutoModelClass):
  class AutoModelForImageSegmentation (line 1228) | class AutoModelForImageSegmentation(_BaseAutoModelClass):
  class AutoModelForSemanticSegmentation (line 1235) | class AutoModelForSemanticSegmentation(_BaseAutoModelClass):
  class AutoModelForUniversalSegmentation (line 1244) | class AutoModelForUniversalSegmentation(_BaseAutoModelClass):
  class AutoModelForInstanceSegmentation (line 1253) | class AutoModelForInstanceSegmentation(_BaseAutoModelClass):
  class AutoModelForObjectDetection (line 1262) | class AutoModelForObjectDetection(_BaseAutoModelClass):
  class AutoModelForZeroShotObjectDetection (line 1269) | class AutoModelForZeroShotObjectDetection(_BaseAutoModelClass):
  class AutoModelForDepthEstimation (line 1278) | class AutoModelForDepthEstimation(_BaseAutoModelClass):
  class AutoModelForVideoClassification (line 1285) | class AutoModelForVideoClassification(_BaseAutoModelClass):
  class AutoModelForVision2Seq (line 1292) | class AutoModelForVision2Seq(_BaseAutoModelClass):
  class AutoModelForAudioClassification (line 1299) | class AutoModelForAudioClassification(_BaseAutoModelClass):
  class AutoModelForCTC (line 1306) | class AutoModelForCTC(_BaseAutoModelClass):
  class AutoModelForSpeechSeq2Seq (line 1313) | class AutoModelForSpeechSeq2Seq(_BaseAutoModelClass):
  class AutoModelForAudioFrameClassification (line 1322) | class AutoModelForAudioFrameClassification(_BaseAutoModelClass):
  class AutoModelForAudioXVector (line 1331) | class AutoModelForAudioXVector(_BaseAutoModelClass):
  class AutoBackbone (line 1335) | class AutoBackbone(_BaseAutoBackboneClass):
  class AutoModelForMaskedImageModeling (line 1342) | class AutoModelForMaskedImageModeling(_BaseAutoModelClass):
  class AutoModelWithLMHead (line 1349) | class AutoModelWithLMHead(_AutoModelWithLMHead):
    method from_config (line 1351) | def from_config(cls, config):
    method from_pretrained (line 1361) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...

FILE: transformers/models/auto/modeling_flax_auto.py
  class FlaxAutoModel (line 269) | class FlaxAutoModel(_BaseAutoModelClass):
  class FlaxAutoModelForPreTraining (line 276) | class FlaxAutoModelForPreTraining(_BaseAutoModelClass):
  class FlaxAutoModelForCausalLM (line 283) | class FlaxAutoModelForCausalLM(_BaseAutoModelClass):
  class FlaxAutoModelForMaskedLM (line 290) | class FlaxAutoModelForMaskedLM(_BaseAutoModelClass):
  class FlaxAutoModelForSeq2SeqLM (line 297) | class FlaxAutoModelForSeq2SeqLM(_BaseAutoModelClass):
  class FlaxAutoModelForSequenceClassification (line 306) | class FlaxAutoModelForSequenceClassification(_BaseAutoModelClass):
  class FlaxAutoModelForQuestionAnswering (line 315) | class FlaxAutoModelForQuestionAnswering(_BaseAutoModelClass):
  class FlaxAutoModelForTokenClassification (line 322) | class FlaxAutoModelForTokenClassification(_BaseAutoModelClass):
  class FlaxAutoModelForMultipleChoice (line 331) | class FlaxAutoModelForMultipleChoice(_BaseAutoModelClass):
  class FlaxAutoModelForNextSentencePrediction (line 338) | class FlaxAutoModelForNextSentencePrediction(_BaseAutoModelClass):
  class FlaxAutoModelForImageClassification (line 347) | class FlaxAutoModelForImageClassification(_BaseAutoModelClass):
  class FlaxAutoModelForVision2Seq (line 356) | class FlaxAutoModelForVision2Seq(_BaseAutoModelClass):
  class FlaxAutoModelForSpeechSeq2Seq (line 363) | class FlaxAutoModelForSpeechSeq2Seq(_BaseAutoModelClass):

FILE: transformers/models/auto/modeling_tf_auto.py
  class TFAutoModelForMaskGeneration (line 495) | class TFAutoModelForMaskGeneration(_BaseAutoModelClass):
  class TFAutoModel (line 499) | class TFAutoModel(_BaseAutoModelClass):
  class TFAutoModelForAudioClassification (line 506) | class TFAutoModelForAudioClassification(_BaseAutoModelClass):
  class TFAutoModelForPreTraining (line 515) | class TFAutoModelForPreTraining(_BaseAutoModelClass):
  class _TFAutoModelWithLMHead (line 523) | class _TFAutoModelWithLMHead(_BaseAutoModelClass):
  class TFAutoModelForCausalLM (line 530) | class TFAutoModelForCausalLM(_BaseAutoModelClass):
  class TFAutoModelForMaskedImageModeling (line 537) | class TFAutoModelForMaskedImageModeling(_BaseAutoModelClass):
  class TFAutoModelForImageClassification (line 546) | class TFAutoModelForImageClassification(_BaseAutoModelClass):
  class TFAutoModelForZeroShotImageClassification (line 555) | class TFAutoModelForZeroShotImageClassification(_BaseAutoModelClass):
  class TFAutoModelForSemanticSegmentation (line 564) | class TFAutoModelForSemanticSegmentation(_BaseAutoModelClass):
  class TFAutoModelForVision2Seq (line 573) | class TFAutoModelForVision2Seq(_BaseAutoModelClass):
  class TFAutoModelForMaskedLM (line 580) | class TFAutoModelForMaskedLM(_BaseAutoModelClass):
  class TFAutoModelForSeq2SeqLM (line 587) | class TFAutoModelForSeq2SeqLM(_BaseAutoModelClass):
  class TFAutoModelForSequenceClassification (line 596) | class TFAutoModelForSequenceClassification(_BaseAutoModelClass):
  class TFAutoModelForQuestionAnswering (line 605) | class TFAutoModelForQuestionAnswering(_BaseAutoModelClass):
  class TFAutoModelForDocumentQuestionAnswering (line 612) | class TFAutoModelForDocumentQuestionAnswering(_BaseAutoModelClass):
  class TFAutoModelForTableQuestionAnswering (line 623) | class TFAutoModelForTableQuestionAnswering(_BaseAutoModelClass):
  class TFAutoModelForTokenClassification (line 634) | class TFAutoModelForTokenClassification(_BaseAutoModelClass):
  class TFAutoModelForMultipleChoice (line 643) | class TFAutoModelForMultipleChoice(_BaseAutoModelClass):
  class TFAutoModelForNextSentencePrediction (line 650) | class TFAutoModelForNextSentencePrediction(_BaseAutoModelClass):
  class TFAutoModelForSpeechSeq2Seq (line 659) | class TFAutoModelForSpeechSeq2Seq(_BaseAutoModelClass):
  class TFAutoModelWithLMHead (line 668) | class TFAutoModelWithLMHead(_TFAutoModelWithLMHead):
    method from_config (line 670) | def from_config(cls, config):
    method from_pretrained (line 680) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...

FILE: transformers/models/auto/processing_auto.py
  function processor_class_from_name (line 88) | def processor_class_from_name(class_name: str):
  class AutoProcessor (line 112) | class AutoProcessor:
    method __init__ (line 120) | def __init__(self):
    method from_pretrained (line 128) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method register (line 303) | def register(config_class, processor_class):

FILE: transformers/models/auto/tokenization_auto.py
  function tokenizer_class_from_name (line 385) | def tokenizer_class_from_name(class_name: str):
  function get_tokenizer_config (line 413) | def get_tokenizer_config(
  class AutoTokenizer (line 513) | class AutoTokenizer:
    method __init__ (line 521) | def __init__(self):
    method from_pretrained (line 529) | def from_pretrained(cls, pretrained_model_name_or_path, *inputs, **kwa...
    method register (line 724) | def register(config_class, slow_tokenizer_class=None, fast_tokenizer_c...

FILE: transformers/models/autoformer/configuration_autoformer.py
  class AutoformerConfig (line 30) | class AutoformerConfig(PretrainedConfig):
    method __init__ (line 141) | def __init__(
    method _number_of_features (line 238) | def _number_of_features(self) -> int:

FILE: transformers/models/autoformer/modeling_autoformer.py
  class AutoFormerDecoderOutput (line 47) | class AutoFormerDecoderOutput(ModelOutput):
  class AutoformerModelOutput (line 96) | class AutoformerModelOutput(ModelOutput):
  class AutoformerFeatureEmbedder (line 176) | class AutoformerFeatureEmbedder(nn.Module):
    method __init__ (line 187) | def __init__(self, cardinalities: List[int], embedding_dims: List[int]...
    method forward (line 193) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class AutoformerStdScaler (line 211) | class AutoformerStdScaler(nn.Module):
    method __init__ (line 225) | def __init__(self, dim: int, keepdim: bool = False, minimum_scale: flo...
    method forward (line 234) | def forward(self, data: torch.Tensor, weights: torch.Tensor) -> Tuple[...
  class AutoformerMeanScaler (line 245) | class AutoformerMeanScaler(nn.Module):
    method __init__ (line 261) | def __init__(
    method forward (line 271) | def forward(
  class AutoformerNOPScaler (line 303) | class AutoformerNOPScaler(nn.Module):
    method __init__ (line 314) | def __init__(self, dim: int, keepdim: bool = False):
    method forward (line 319) | def forward(
  function weighted_average (line 328) | def weighted_average(input_tensor: torch.Tensor, weights: Optional[torch...
  function nll (line 353) | def nll(input: torch.distributions.Distribution, target: torch.Tensor) -...
  function _make_causal_mask (line 361) | def _make_causal_mask(
  function _expand_mask (line 379) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class AutoformerSinusoidalPositionalEmbedding (line 394) | class AutoformerSinusoidalPositionalEmbedding(nn.Embedding):
    method __init__ (line 397) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method _init_weight (line 402) | def _init_weight(out: nn.Parameter) -> nn.Parameter:
    method forward (line 419) | def forward(self, input_ids_shape: torch.Size, past_key_values_length:...
  class AutoformerValueEmbedding (line 429) | class AutoformerValueEmbedding(nn.Module):
    method __init__ (line 430) | def __init__(self, feature_size, d_model):
    method forward (line 434) | def forward(self, x):
  class AutoformerSeriesDecompositionLayer (line 441) | class AutoformerSeriesDecompositionLayer(nn.Module):
    method __init__ (line 448) | def __init__(self, config: AutoformerConfig):
    method forward (line 453) | def forward(self, x):
  class AutoformerLayernorm (line 470) | class AutoformerLayernorm(nn.Module):
    method __init__ (line 476) | def __init__(self, config: AutoformerConfig):
    method forward (line 480) | def forward(self, x):
  class AutoformerAttention (line 486) | class AutoformerAttention(nn.Module):
    method __init__ (line 493) | def __init__(
    method _shape (line 523) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 526) | def forward(
  class AutoformerEncoderLayer (line 709) | class AutoformerEncoderLayer(nn.Module):
    method __init__ (line 710) | def __init__(self, config: AutoformerConfig):
    method forward (line 729) | def forward(
  class AutoformerDecoderLayer (line 783) | class AutoformerDecoderLayer(nn.Module):
    method __init__ (line 784) | def __init__(self, config: AutoformerConfig):
    method forward (line 827) | def forward(
  class AutoformerPreTrainedModel (line 930) | class AutoformerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 936) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 949) | def _set_gradient_checkpointing(self, module, value=False):
  class AutoformerEncoder (line 1103) | class AutoformerEncoder(AutoformerPreTrainedModel):
    method __init__ (line 1112) | def __init__(self, config: AutoformerConfig):
    method forward (line 1131) | def forward(
  class AutoformerDecoder (line 1241) | class AutoformerDecoder(AutoformerPreTrainedModel):
    method __init__ (line 1249) | def __init__(self, config: AutoformerConfig):
    method _prepare_decoder_attention_mask (line 1270) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 1293) | def forward(
  class AutoformerModel (line 1494) | class AutoformerModel(AutoformerPreTrainedModel):
    method __init__ (line 1495) | def __init__(self, config: AutoformerConfig):
    method _past_length (line 1521) | def _past_length(self) -> int:
    method get_lagged_subsequences (line 1524) | def get_lagged_subsequences(
    method create_network_inputs (line 1562) | def create_network_inputs(
    method get_encoder (line 1662) | def get_encoder(self):
    method get_decoder (line 1665) | def get_decoder(self):
    method forward (line 1670) | def forward(
  class AutoformerForPrediction (line 1830) | class AutoformerForPrediction(AutoformerPreTrainedModel):
    method __init__ (line 1831) | def __init__(self, config: AutoformerConfig):
    method output_params (line 1854) | def output_params(self, decoder_output):
    method get_encoder (line 1857) | def get_encoder(self):
    method get_decoder (line 1860) | def get_decoder(self):
    method output_distribution (line 1864) | def output_distribution(self, params, loc=None, scale=None, trailing_n...
    method forward (line 1872) | def forward(
    method generate (line 2004) | def generate(

FILE: transformers/models/bart/configuration_bart.py
  class BartConfig (line 35) | class BartConfig(PretrainedConfig):
    method __init__ (line 114) | def __init__(
  class BartOnnxConfig (line 185) | class BartOnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 187) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 231) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method _generate_dummy_inputs_for_default_and_seq2seq_lm (line 243) | def _generate_dummy_inputs_for_default_and_seq2seq_lm(
    method _generate_dummy_inputs_for_causal_lm (line 311) | def _generate_dummy_inputs_for_causal_lm(
    method _generate_dummy_inputs_for_sequence_classification_and_question_answering (line 349) | def _generate_dummy_inputs_for_sequence_classification_and_question_an...
    method generate_dummy_inputs (line 375) | def generate_dummy_inputs(
    method _flatten_past_key_values_ (line 399) | def _flatten_past_key_values_(self, flattened_output, name, idx, t):

FILE: transformers/models/bart/convert_bart_original_pytorch_checkpoint_to_pytorch.py
  function remove_ignore_keys_ (line 56) | def remove_ignore_keys_(state_dict):
  function rename_key (line 68) | def rename_key(dct, old, new):
  function load_xsum_checkpoint (line 73) | def load_xsum_checkpoint(checkpoint_path):
  function make_linear_from_emb (line 81) | def make_linear_from_emb(emb):
  function convert_bart_checkpoint (line 89) | def convert_bart_checkpoint(checkpoint_path, pytorch_dump_folder_path, h...

FILE: transformers/models/bart/modeling_bart.py
  function shift_tokens_right (line 74) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 90) | def _make_causal_mask(
  function _expand_mask (line 107) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class BartLearnedPositionalEmbedding (line 121) | class BartLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 126) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 132) | def forward(self, input_ids: torch.Tensor, past_key_values_length: int...
  class BartAttention (line 143) | class BartAttention(nn.Module):
    method __init__ (line 146) | def __init__(
    method _shape (line 173) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 176) | def forward(
  class BartEncoderLayer (line 297) | class BartEncoderLayer(nn.Module):
    method __init__ (line 298) | def __init__(self, config: BartConfig):
    method forward (line 314) | def forward(
  class BartDecoderLayer (line 365) | class BartDecoderLayer(nn.Module):
    method __init__ (line 366) | def __init__(self, config: BartConfig):
    method forward (line 392) | def forward(
  class BartClassificationHead (line 482) | class BartClassificationHead(nn.Module):
    method __init__ (line 485) | def __init__(
    method forward (line 497) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BartPretrainedModel (line 506) | class BartPretrainedModel(PreTrainedModel):
    method _init_weights (line 514) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 525) | def _set_gradient_checkpointing(self, module, value=False):
    method dummy_inputs (line 530) | def dummy_inputs(self):
  class PretrainedBartModel (line 540) | class PretrainedBartModel(BartPretrainedModel):
    method __init_subclass__ (line 541) | def __init_subclass__(self):
  class BartEncoder (line 704) | class BartEncoder(BartPretrainedModel):
    method __init__ (line 714) | def __init__(self, config: BartConfig, embed_tokens: Optional[nn.Embed...
    method get_input_embeddings (line 741) | def get_input_embeddings(self):
    method set_input_embeddings (line 744) | def set_input_embeddings(self, value):
    method forward (line 747) | def forward(
  class BartDecoder (line 881) | class BartDecoder(BartPretrainedModel):
    method __init__ (line 890) | def __init__(self, config: BartConfig, embed_tokens: Optional[nn.Embed...
    method get_input_embeddings (line 914) | def get_input_embeddings(self):
    method set_input_embeddings (line 917) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 920) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 943) | def forward(
  class BartModel (line 1167) | class BartModel(BartPretrainedModel):
    method __init__ (line 1170) | def __init__(self, config: BartConfig):
    method get_input_embeddings (line 1182) | def get_input_embeddings(self):
    method set_input_embeddings (line 1185) | def set_input_embeddings(self, value):
    method get_encoder (line 1190) | def get_encoder(self):
    method get_decoder (line 1193) | def get_decoder(self):
    method forward (line 1203) | def forward(
  class BartForConditionalGeneration (line 1294) | class BartForConditionalGeneration(BartPretrainedModel):
    method __init__ (line 1303) | def __init__(self, config: BartConfig):
    method get_encoder (line 1312) | def get_encoder(self):
    method get_decoder (line 1315) | def get_decoder(self):
    method resize_token_embeddings (line 1318) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_final_logits_bias (line 1323) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_output_embeddings (line 1332) | def get_output_embeddings(self):
    method set_output_embeddings (line 1335) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1341) | def forward(
    method prepare_inputs_for_generation (line 1422) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1452) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 1456) | def _reorder_cache(past_key_values, beam_idx):
  class BartForSequenceClassification (line 1473) | class BartForSequenceClassification(BartPretrainedModel):
    method __init__ (line 1476) | def __init__(self, config: BartConfig, **kwargs):
    method forward (line 1497) | def forward(
  class BartForQuestionAnswering (line 1603) | class BartForQuestionAnswering(BartPretrainedModel):
    method __init__ (line 1606) | def __init__(self, config):
    method forward (line 1626) | def forward(
  class BartDecoderWrapper (line 1721) | class BartDecoderWrapper(BartPretrainedModel):
    method __init__ (line 1727) | def __init__(self, config):
    method forward (line 1731) | def forward(self, *args, **kwargs):
  class BartForCausalLM (line 1741) | class BartForCausalLM(BartPretrainedModel):
    method __init__ (line 1744) | def __init__(self, config):
    method get_input_embeddings (line 1756) | def get_input_embeddings(self):
    method set_input_embeddings (line 1759) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1762) | def get_output_embeddings(self):
    method set_output_embeddings (line 1765) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 1768) | def set_decoder(self, decoder):
    method get_decoder (line 1771) | def get_decoder(self):
    method forward (line 1775) | def forward(
    method prepare_inputs_for_generation (line 1919) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1937) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/bart/modeling_flax_bart.py
  function shift_tokens_right (line 220) | def shift_tokens_right(input_ids: jnp.array, pad_token_id: int, decoder_...
  class FlaxBartAttention (line 232) | class FlaxBartAttention(nn.Module):
    method setup (line 241) | def setup(self) -> None:
    method _split_heads (line 267) | def _split_heads(self, hidden_states):
    method _merge_heads (line 270) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 274) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 305) | def __call__(
  class FlaxBartEncoderLayer (line 399) | class FlaxBartEncoderLayer(nn.Module):
    method setup (line 403) | def setup(self) -> None:
    method __call__ (line 426) | def __call__(
  class FlaxBartEncoderLayerCollection (line 456) | class FlaxBartEncoderLayerCollection(nn.Module):
    method setup (line 460) | def setup(self):
    method __call__ (line 466) | def __call__(
  class FlaxBartDecoderLayer (line 509) | class FlaxBartDecoderLayer(nn.Module):
    method setup (line 513) | def setup(self) -> None:
    method __call__ (line 546) | def __call__(
  class FlaxBartDecoderLayerCollection (line 597) | class FlaxBartDecoderLayerCollection(nn.Module):
    method setup (line 601) | def setup(self):
    method __call__ (line 607) | def __call__(
  class FlaxBartClassificationHead (line 666) | class FlaxBartClassificationHead(nn.Module):
    method setup (line 675) | def setup(self):
    method __call__ (line 686) | def __call__(self, hidden_states: jnp.ndarray, deterministic: bool):
  class FlaxBartEncoder (line 695) | class FlaxBartEncoder(nn.Module):
    method setup (line 700) | def setup(self):
    method __call__ (line 720) | def __call__(
  class FlaxBartDecoder (line 760) | class FlaxBartDecoder(nn.Module):
    method setup (line 765) | def setup(self):
    method __call__ (line 786) | def __call__(
  class FlaxBartModule (line 835) | class FlaxBartModule(nn.Module):
    method setup (line 839) | def setup(self):
    method _get_encoder_module (line 850) | def _get_encoder_module(self):
    method _get_decoder_module (line 853) | def _get_decoder_module(self):
    method __call__ (line 856) | def __call__(
  class FlaxBartPreTrainedModel (line 905) | class FlaxBartPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 910) | def __init__(
    method init_weights (line 922) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 958) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 1001) | def encode(
    method decode (line 1064) | def decode(
    method __call__ (line 1177) | def __call__(
  class FlaxBartModel (line 1241) | class FlaxBartModel(FlaxBartPreTrainedModel):
  class FlaxBartForConditionalGenerationModule (line 1250) | class FlaxBartForConditionalGenerationModule(nn.Module):
    method setup (line 1255) | def setup(self):
    method _get_encoder_module (line 1265) | def _get_encoder_module(self):
    method _get_decoder_module (line 1268) | def _get_decoder_module(self):
    method __call__ (line 1271) | def __call__(
  class FlaxBartForConditionalGeneration (line 1325) | class FlaxBartForConditionalGeneration(FlaxBartPreTrainedModel):
    method decode (line 1331) | def decode(
    method prepare_inputs_for_generation (line 1466) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 1497) | def update_inputs_for_generation(self, model_outputs, model_kwargs):
  class FlaxBartForSequenceClassificationModule (line 1551) | class FlaxBartForSequenceClassificationModule(nn.Module):
    method setup (line 1556) | def setup(self):
    method _get_encoder_module (line 1565) | def _get_encoder_module(self):
    method _get_decoder_module (line 1568) | def _get_decoder_module(self):
    method __call__ (line 1571) | def __call__(
  class FlaxBartForSequenceClassification (line 1638) | class FlaxBartForSequenceClassification(FlaxBartPreTrainedModel):
  class FlaxBartForQuestionAnsweringModule (line 1651) | class FlaxBartForQuestionAnsweringModule(nn.Module):
    method setup (line 1656) | def setup(self):
    method _get_encoder_module (line 1662) | def _get_encoder_module(self):
    method _get_decoder_module (line 1665) | def _get_decoder_module(self):
    method __call__ (line 1668) | def __call__(
  class FlaxBartForQuestionAnswering (line 1724) | class FlaxBartForQuestionAnswering(FlaxBartPreTrainedModel):
  class FlaxBartDecoderPreTrainedModel (line 1737) | class FlaxBartDecoderPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 1742) | def __init__(
    method init_weights (line 1756) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 1779) | def init_cache(self, batch_size, max_length):
    method __call__ (line 1799) | def __call__(
  class FlaxBartDecoderWrapper (line 1872) | class FlaxBartDecoderWrapper(nn.Module):
    method setup (line 1881) | def setup(self):
    method __call__ (line 1891) | def __call__(self, *args, **kwargs):
  class FlaxBartForCausalLMModule (line 1895) | class FlaxBartForCausalLMModule(nn.Module):
    method setup (line 1899) | def setup(self):
    method __call__ (line 1908) | def __call__(
  class FlaxBartForCausalLM (line 1960) | class FlaxBartForCausalLM(FlaxBartDecoderPreTrainedModel):
    method prepare_inputs_for_generation (line 1963) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 1984) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/bart/modeling_tf_bart.py
  function shift_tokens_right (line 66) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int, decoder_...
  function _make_causal_mask (line 90) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 107) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFBartLearnedPositionalEmbedding (line 120) | class TFBartLearnedPositionalEmbedding(tf.keras.layers.Embedding):
    method __init__ (line 125) | def __init__(self, num_embeddings: int, embedding_dim: int, **kwargs):
    method call (line 131) | def call(
  class TFBartAttention (line 147) | class TFBartAttention(tf.keras.layers.Layer):
    method __init__ (line 150) | def __init__(
    method _shape (line 178) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 181) | def call(
  class TFBartEncoderLayer (line 300) | class TFBartEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 301) | def __init__(self, config: BartConfig, **kwargs):
    method call (line 315) | def call(
  class TFBartDecoderLayer (line 356) | class TFBartDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 357) | def __init__(self, config: BartConfig, **kwargs):
    method call (line 384) | def call(
  class TFBartClassificationHead (line 465) | class TFBartClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 468) | def __init__(self, inner_dim: int, num_classes: int, pooler_dropout: f...
    method call (line 474) | def call(self, inputs):
  class TFBartPretrainedModel (line 483) | class TFBartPretrainedModel(TFPreTrainedModel):
    method dummy_inputs (line 488) | def dummy_inputs(self):
  class TFBartEncoder (line 656) | class TFBartEncoder(tf.keras.layers.Layer):
    method __init__ (line 666) | def __init__(self, config: BartConfig, embed_tokens: Optional[tf.keras...
    method call (line 685) | def call(
  class TFBartDecoder (line 808) | class TFBartDecoder(tf.keras.layers.Layer):
    method __init__ (line 818) | def __init__(self, config: BartConfig, embed_tokens: Optional[tf.keras...
    method call (line 836) | def call(
  class TFBartMainLayer (line 1031) | class TFBartMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1034) | def __init__(self, config: BartConfig, load_weight_prefix=None, **kwar...
    method get_input_embeddings (line 1049) | def get_input_embeddings(self):
    method set_input_embeddings (line 1052) | def set_input_embeddings(self, new_embeddings):
    method call (line 1058) | def call(
  class TFBartModel (line 1151) | class TFBartModel(TFBartPretrainedModel):
    method __init__ (line 1154) | def __init__(self, config: BartConfig, load_weight_prefix=None, *input...
    method get_encoder (line 1159) | def get_encoder(self):
    method get_decoder (line 1162) | def get_decoder(self):
    method call (line 1172) | def call(
    method serving_output (line 1215) | def serving_output(self, output):
  class BiasLayer (line 1235) | class BiasLayer(tf.keras.layers.Layer):
    method __init__ (line 1241) | def __init__(self, shape, initializer, trainable, name, **kwargs):
    method call (line 1248) | def call(self, x):
  class TFBartForConditionalGeneration (line 1256) | class TFBartForConditionalGeneration(TFBartPretrainedModel, TFCausalLang...
    method __init__ (line 1260) | def __init__(self, config, load_weight_prefix=None, *inputs, **kwargs):
    method get_decoder (line 1269) | def get_decoder(self):
    method get_encoder (line 1272) | def get_encoder(self):
    method get_output_embeddings (line 1275) | def get_output_embeddings(self):
    method set_output_embeddings (line 1278) | def set_output_embeddings(self, value):
    method get_bias (line 1281) | def get_bias(self):
    method set_bias (line 1284) | def set_bias(self, value):
    method call (line 1296) | def call(
    method serving_output (line 1377) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 1396) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1434) | def prepare_decoder_input_ids_from_labels(self, labels: tf.Tensor):
  class TFBartForSequenceClassification (line 1445) | class TFBartForSequenceClassification(TFBartPretrainedModel, TFSequenceC...
    method __init__ (line 1446) | def __init__(self, config: BartConfig, load_weight_prefix=None, *input...
    method call (line 1456) | def call(
    method serving_output (line 1545) | def serving_output(self, output):

FILE: transformers/models/bart/tokenization_bart.py
  function bytes_to_unicode (line 63) | def bytes_to_unicode():
  function get_pairs (line 87) | def get_pairs(word):
  class BartTokenizer (line 101) | class BartTokenizer(PreTrainedTokenizer):
    method __init__ (line 184) | def __init__(
    method vocab_size (line 239) | def vocab_size(self):
    method get_vocab (line 242) | def get_vocab(self):
    method bpe (line 245) | def bpe(self, token):
    method _tokenize (line 287) | def _tokenize(self, text):
    method _convert_token_to_id (line 297) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 301) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 305) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 311) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 340) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 365) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 392) | def create_token_type_ids_from_sequences(
    method prepare_for_tokenization (line 415) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...

FILE: transformers/models/bart/tokenization_bart_fast.py
  class BartTokenizerFast (line 70) | class BartTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 156) | def __init__(
    method mask_token (line 226) | def mask_token(self) -> str:
    method mask_token (line 241) | def mask_token(self, value):
    method _batch_encode_plus (line 252) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 263) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method save_vocabulary (line 274) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 278) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 285) | def create_token_type_ids_from_sequences(

FILE: transformers/models/barthez/tokenization_barthez.py
  class BarthezTokenizer (line 51) | class BarthezTokenizer(PreTrainedTokenizer):
    method __init__ (line 126) | def __init__(
    method build_inputs_with_special_tokens (line 165) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 191) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 218) | def create_token_type_ids_from_sequences(
    method vocab_size (line 241) | def vocab_size(self):
    method get_vocab (line 244) | def get_vocab(self):
    method _tokenize (line 249) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 252) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 260) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 266) | def convert_tokens_to_string(self, tokens):
    method __getstate__ (line 285) | def __getstate__(self):
    method __setstate__ (line 290) | def __setstate__(self, d):
    method save_vocabulary (line 300) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/barthez/tokenization_barthez_fast.py
  class BarthezTokenizerFast (line 62) | class BarthezTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 119) | def __init__(
    method build_inputs_with_special_tokens (line 151) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 177) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 199) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/bartpho/tokenization_bartpho.py
  class BartphoTokenizer (line 46) | class BartphoTokenizer(PreTrainedTokenizer):
    method __init__ (line 123) | def __init__(
    method __getstate__ (line 177) | def __getstate__(self):
    method __setstate__ (line 183) | def __setstate__(self, d):
    method build_inputs_with_special_tokens (line 193) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 219) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 247) | def create_token_type_ids_from_sequences(
    method vocab_size (line 273) | def vocab_size(self):
    method get_vocab (line 276) | def get_vocab(self):
    method _tokenize (line 281) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 284) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 291) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 295) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 300) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/beit/configuration_beit.py
  class BeitConfig (line 36) | class BeitConfig(PretrainedConfig):
    method __init__ (line 120) | def __init__(
  class BeitOnnxConfig (line 188) | class BeitOnnxConfig(OnnxConfig):
    method inputs (line 192) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 200) | def atol_for_validation(self) -> float:

FILE: transformers/models/beit/convert_beit_unilm_to_pytorch.py
  function create_rename_keys (line 44) | def create_rename_keys(config, has_lm_head=False, is_semantic=False):
  function read_in_q_k_v (line 116) | def read_in_q_k_v(state_dict, config, has_lm_head=False, is_semantic=Fal...
  function rename_key (line 158) | def rename_key(dct, old, new):
  function prepare_img (line 164) | def prepare_img():
  function convert_beit_checkpoint (line 171) | def convert_beit_checkpoint(checkpoint_url, pytorch_dump_folder_path):

FILE: transformers/models/beit/feature_extraction_beit.py
  class BeitFeatureExtractor (line 26) | class BeitFeatureExtractor(BeitImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/beit/image_processing_beit.py
  class BeitImageProcessor (line 47) | class BeitImageProcessor(BaseImageProcessor):
    method __init__ (line 92) | def __init__(
    method reduce_labels (line 132) | def reduce_labels(self) -> bool:
    method from_dict (line 141) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method resize (line 151) | def resize(
    method center_crop (line 179) | def center_crop(
    method rescale (line 201) | def rescale(
    method normalize (line 221) | def normalize(
    method reduce_label (line 244) | def reduce_label(self, label: ImageInput) -> np.ndarray:
    method _preprocess (line 252) | def _preprocess(
    method _preprocess_image (line 284) | def _preprocess_image(
    method _preprocess_segmentation_map (line 320) | def _preprocess_segmentation_map(
    method __call__ (line 356) | def __call__(self, images, segmentation_maps=None, **kwargs):
    method preprocess (line 361) | def preprocess(
    method post_process_semantic_segmentation (line 503) | def post_process_semantic_segmentation(self, outputs, target_sizes: Li...

FILE: transformers/models/beit/modeling_beit.py
  class BeitModelOutputWithPooling (line 68) | class BeitModelOutputWithPooling(BaseModelOutputWithPooling):
  function drop_path (line 93) | def drop_path(input: torch.Tensor, drop_prob: float = 0.0, training: boo...
  class BeitDropPath (line 113) | class BeitDropPath(nn.Module):
    method __init__ (line 116) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 120) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 123) | def extra_repr(self) -> str:
  class BeitEmbeddings (line 129) | class BeitEmbeddings(nn.Module):
    method __init__ (line 135) | def __init__(self, config: BeitConfig) -> None:
    method forward (line 151) | def forward(self, pixel_values: torch.Tensor, bool_masked_pos: Optiona...
  class BeitPatchEmbeddings (line 170) | class BeitPatchEmbeddings(nn.Module):
    method __init__ (line 177) | def __init__(self, config):
    method forward (line 194) | def forward(self, pixel_values: torch.Tensor) -> torch.Tensor:
  class BeitSelfAttention (line 209) | class BeitSelfAttention(nn.Module):
    method __init__ (line 210) | def __init__(self, config: BeitConfig, window_size: Optional[tuple] = ...
    method transpose_for_scores (line 233) | def transpose_for_scores(self, x):
    method forward (line 238) | def forward(
  class BeitSelfOutput (line 286) | class BeitSelfOutput(nn.Module):
    method __init__ (line 292) | def __init__(self, config: BeitConfig) -> None:
    method forward (line 297) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BeitAttention (line 304) | class BeitAttention(nn.Module):
    method __init__ (line 305) | def __init__(self, config: BeitConfig, window_size: Optional[tuple] = ...
    method prune_heads (line 311) | def prune_heads(self, heads):
    method forward (line 329) | def forward(
  class BeitIntermediate (line 344) | class BeitIntermediate(nn.Module):
    method __init__ (line 345) | def __init__(self, config: BeitConfig) -> None:
    method forward (line 353) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BeitOutput (line 360) | class BeitOutput(nn.Module):
    method __init__ (line 361) | def __init__(self, config: BeitConfig) -> None:
    method forward (line 366) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BeitLayer (line 373) | class BeitLayer(nn.Module):
    method __init__ (line 376) | def __init__(self, config: BeitConfig, window_size: Optional[tuple] = ...
    method forward (line 394) | def forward(
  class BeitRelativePositionBias (line 434) | class BeitRelativePositionBias(nn.Module):
    method __init__ (line 435) | def __init__(self, config: BeitConfig, window_size: tuple) -> None:
    method forward (line 464) | def forward(self) -> torch.Tensor:
  class BeitEncoder (line 472) | class BeitEncoder(nn.Module):
    method __init__ (line 473) | def __init__(self, config: BeitConfig, window_size: Optional[tuple] = ...
    method forward (line 495) | def forward(
  class BeitPreTrainedModel (line 548) | class BeitPreTrainedModel(PreTrainedModel):
    method _init_weights (line 559) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 575) | def _set_gradient_checkpointing(self, module, value=False):
  class BeitModel (line 618) | class BeitModel(BeitPreTrainedModel):
    method __init__ (line 619) | def __init__(self, config: BeitConfig, add_pooling_layer: bool = True)...
    method get_input_embeddings (line 634) | def get_input_embeddings(self):
    method _prune_heads (line 637) | def _prune_heads(self, heads_to_prune):
    method forward (line 653) | def forward(
  class BeitPooler (line 707) | class BeitPooler(nn.Module):
    method __init__ (line 708) | def __init__(self, config: BeitConfig) -> None:
    method forward (line 714) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BeitForMaskedImageModeling (line 733) | class BeitForMaskedImageModeling(BeitPreTrainedModel):
    method __init__ (line 734) | def __init__(self, config: BeitConfig) -> None:
    method forward (line 749) | def forward(
  class BeitForImageClassification (line 833) | class BeitForImageClassification(BeitPreTrainedModel):
    method __init__ (line 834) | def __init__(self, config: BeitConfig) -> None:
    method forward (line 853) | def forward(
  class BeitConvModule (line 915) | class BeitConvModule(nn.Module):
    method __init__ (line 923) | def __init__(
    method forward (line 944) | def forward(self, input: torch.Tensor) -> torch.Tensor:
  class BeitPyramidPoolingBlock (line 952) | class BeitPyramidPoolingBlock(nn.Module):
    method __init__ (line 953) | def __init__(self, pool_scale: int, in_channels: int, channels: int) -...
    method forward (line 962) | def forward(self, input: torch.Tensor) -> torch.Tensor:
  class BeitPyramidPoolingModule (line 969) | class BeitPyramidPoolingModule(nn.Module):
    method __init__ (line 983) | def __init__(self, pool_scales: Tuple[int, ...], in_channels: int, cha...
    method forward (line 995) | def forward(self, x: torch.Tensor) -> List[torch.Tensor]:
  class BeitUperHead (line 1006) | class BeitUperHead(nn.Module):
    method __init__ (line 1014) | def __init__(self, config: BeitConfig) -> None:
    method psp_forward (line 1052) | def psp_forward(self, inputs):
    method forward (line 1061) | def forward(self, encoder_hidden_states: torch.Tensor) -> torch.Tensor:
  class BeitFCNHead (line 1091) | class BeitFCNHead(nn.Module):
    method __init__ (line 1106) | def __init__(
    method forward (line 1140) | def forward(self, encoder_hidden_states: torch.Tensor) -> torch.Tensor:
  class BeitForSemanticSegmentation (line 1156) | class BeitForSemanticSegmentation(BeitPreTrainedModel):
    method __init__ (line 1157) | def __init__(self, config: BeitConfig) -> None:
    method compute_loss (line 1183) | def compute_loss(self, logits, auxiliary_logits, labels):
    method forward (line 1204) | def forward(

FILE: transformers/models/beit/modeling_flax_beit.py
  class FlaxBeitModelOutputWithPooling (line 45) | class FlaxBeitModelOutputWithPooling(FlaxBaseModelOutputWithPooling):
  function relative_position_index_init (line 118) | def relative_position_index_init(window_size: Tuple[int, int]) -> jnp.nd...
  function ones_with_scale (line 142) | def ones_with_scale(key, shape, scale, dtype=jnp.float32):
  class FlaxBeitDropPath (line 146) | class FlaxBeitDropPath(nn.Module):
    method __call__ (line 152) | def __call__(self, inputs, deterministic: Optional[bool] = True):
  class FlaxBeitPatchEmbeddings (line 167) | class FlaxBeitPatchEmbeddings(nn.Module):
    method setup (line 171) | def setup(self):
    method __call__ (line 188) | def __call__(self, pixel_values):
  class FlaxBeitEmbeddings (line 199) | class FlaxBeitEmbeddings(nn.Module):
    method setup (line 205) | def setup(self):
    method __call__ (line 217) | def __call__(self, pixel_values, bool_masked_pos=None, deterministic=T...
  class FlaxBeitRelativePositionBias (line 240) | class FlaxBeitRelativePositionBias(nn.Module):
    method setup (line 245) | def setup(self):
    method __call__ (line 256) | def __call__(self):
  class FlaxBeitSelfAttention (line 263) | class FlaxBeitSelfAttention(nn.Module):
    method setup (line 268) | def setup(self):
    method __call__ (line 300) | def __call__(
  class FlaxBeitSelfOutput (line 348) | class FlaxBeitSelfOutput(nn.Module):
    method setup (line 352) | def setup(self):
    method __call__ (line 360) | def __call__(self, hidden_states, deterministic: bool = True):
  class FlaxBeitAttention (line 366) | class FlaxBeitAttention(nn.Module):
    method setup (line 371) | def setup(self):
    method __call__ (line 375) | def __call__(
  class FlaxBeitIntermediate (line 392) | class FlaxBeitIntermediate(nn.Module):
    method setup (line 396) | def setup(self):
    method __call__ (line 404) | def __call__(self, hidden_states):
  class FlaxBeitOutput (line 411) | class FlaxBeitOutput(nn.Module):
    method setup (line 415) | def setup(self):
    method __call__ (line 423) | def __call__(self, hidden_states, deterministic: bool = True):
  class FlaxBeitLayer (line 430) | class FlaxBeitLayer(nn.Module):
    method setup (line 436) | def setup(self):
    method __call__ (line 452) | def __call__(
  class FlaxBeitLayerCollection (line 491) | class FlaxBeitLayerCollection(nn.Module):
    method setup (line 498) | def setup(self):
    method __call__ (line 510) | def __call__(
  class FlaxBeitEncoder (line 546) | class FlaxBeitEncoder(nn.Module):
    method setup (line 551) | def setup(self):
    method __call__ (line 569) | def __call__(
  class FlaxBeitPreTrainedModel (line 586) | class FlaxBeitPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 597) | def __init__(
    method init_weights (line 611) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 632) | def __call__(
  class FlaxBeitPooler (line 669) | class FlaxBeitPooler(nn.Module):
    method setup (line 673) | def setup(self):
    method __call__ (line 677) | def __call__(self, hidden_states):
  class FlaxBeitModule (line 689) | class FlaxBeitModule(nn.Module):
    method setup (line 694) | def setup(self):
    method __call__ (line 703) | def __call__(
  class FlaxBeitModel (line 744) | class FlaxBeitModel(FlaxBeitPreTrainedModel):
  class FlaxBeitForMaskedImageModelingModule (line 774) | class FlaxBeitForMaskedImageModelingModule(nn.Module):
    method setup (line 778) | def setup(self):
    method __call__ (line 789) | def __call__(
  class FlaxBeitForMaskedImageModeling (line 828) | class FlaxBeitForMaskedImageModeling(FlaxBeitPreTrainedModel):
  class FlaxBeitForImageClassificationModule (line 863) | class FlaxBeitForImageClassificationModule(nn.Module):
    method setup (line 867) | def setup(self):
    method __call__ (line 875) | def __call__(
  class FlaxBeitForImageClassification (line 915) | class FlaxBeitForImageClassification(FlaxBeitPreTrainedModel):

FILE: transformers/models/bert/configuration_bert.py
  class BertConfig (line 72) | class BertConfig(PretrainedConfig):
    method __init__ (line 141) | def __init__(
  class BertOnnxConfig (line 180) | class BertOnnxConfig(OnnxConfig):
    method inputs (line 182) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/bert/convert_bert_original_tf2_checkpoint_to_pytorch.py
  function load_tf2_weights_in_bert (line 42) | def load_tf2_weights_in_bert(model, tf_checkpoint_path, config):
  function convert_tf2_checkpoint_to_pytorch (line 212) | def convert_tf2_checkpoint_to_pytorch(tf_checkpoint_path, config_path, p...

FILE: transformers/models/bert/convert_bert_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 29) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, bert_config_fil...

FILE: transformers/models/bert/convert_bert_pytorch_checkpoint_to_original_tf.py
  function convert_pytorch_checkpoint_to_tf (line 28) | def convert_pytorch_checkpoint_to_tf(model: BertModel, ckpt_dir: str, mo...
  function main (line 92) | def main(raw_args=None):

FILE: transformers/models/bert/convert_bert_token_dropping_original_tf2_checkpoint_to_pytorch.py
  function convert_checkpoint_to_pytorch (line 41) | def convert_checkpoint_to_pytorch(tf_checkpoint_path: str, config_path: ...

FILE: transformers/models/bert/modeling_bert.py
  function load_tf_weights_in_bert (line 107) | def load_tf_weights_in_bert(model, config, tf_checkpoint_path):
  class BertEmbeddings (line 180) | class BertEmbeddings(nn.Module):
    method __init__ (line 183) | def __init__(self, config):
    method forward (line 200) | def forward(
  class BertSelfAttention (line 242) | class BertSelfAttention(nn.Module):
    method __init__ (line 243) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 269) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 274) | def forward(
  class BertSelfOutput (line 376) | class BertSelfOutput(nn.Module):
    method __init__ (line 377) | def __init__(self, config):
    method forward (line 383) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BertAttention (line 390) | class BertAttention(nn.Module):
    method __init__ (line 391) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 397) | def prune_heads(self, heads):
    method forward (line 415) | def forward(
  class BertIntermediate (line 439) | class BertIntermediate(nn.Module):
    method __init__ (line 440) | def __init__(self, config):
    method forward (line 448) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BertOutput (line 454) | class BertOutput(nn.Module):
    method __init__ (line 455) | def __init__(self, config):
    method forward (line 461) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BertLayer (line 468) | class BertLayer(nn.Module):
    method __init__ (line 469) | def __init__(self, config):
    method forward (line 483) | def forward(
    method feed_forward_chunk (line 548) | def feed_forward_chunk(self, attention_output):
  class BertEncoder (line 554) | class BertEncoder(nn.Module):
    method __init__ (line 555) | def __init__(self, config):
    method forward (line 561) | def forward(
  class BertPooler (line 652) | class BertPooler(nn.Module):
    method __init__ (line 653) | def __init__(self, config):
    method forward (line 658) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BertPredictionHeadTransform (line 667) | class BertPredictionHeadTransform(nn.Module):
    method __init__ (line 668) | def __init__(self, config):
    method forward (line 677) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BertLMPredictionHead (line 684) | class BertLMPredictionHead(nn.Module):
    method __init__ (line 685) | def __init__(self, config):
    method forward (line 698) | def forward(self, hidden_states):
  class BertOnlyMLMHead (line 704) | class BertOnlyMLMHead(nn.Module):
    method __init__ (line 705) | def __init__(self, config):
    method forward (line 709) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class BertOnlyNSPHead (line 714) | class BertOnlyNSPHead(nn.Module):
    method __init__ (line 715) | def __init__(self, config):
    method forward (line 719) | def forward(self, pooled_output):
  class BertPreTrainingHeads (line 724) | class BertPreTrainingHeads(nn.Module):
    method __init__ (line 725) | def __init__(self, config):
    method forward (line 730) | def forward(self, sequence_output, pooled_output):
  class BertPreTrainedModel (line 736) | class BertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 748) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 764) | def _set_gradient_checkpointing(self, module, value=False):
  class BertForPreTrainingOutput (line 770) | class BertForPreTrainingOutput(ModelOutput):
  class BertModel (line 873) | class BertModel(BertPreTrainedModel):
    method __init__ (line 886) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 898) | def get_input_embeddings(self):
    method set_input_embeddings (line 901) | def set_input_embeddings(self, value):
    method _prune_heads (line 904) | def _prune_heads(self, heads_to_prune):
    method forward (line 918) | def forward(
  class BertForPreTraining (line 1055) | class BertForPreTraining(BertPreTrainedModel):
    method __init__ (line 1058) | def __init__(self, config):
    method get_output_embeddings (line 1067) | def get_output_embeddings(self):
    method set_output_embeddings (line 1070) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1075) | def forward(
  class BertLMHeadModel (line 1161) | class BertLMHeadModel(BertPreTrainedModel):
    method __init__ (line 1165) | def __init__(self, config):
    method get_output_embeddings (line 1177) | def get_output_embeddings(self):
    method set_output_embeddings (line 1180) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1189) | def forward(
    method prepare_inputs_for_generation (line 1274) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1293) | def _reorder_cache(self, past_key_values, beam_idx):
  class BertForMaskedLM (line 1301) | class BertForMaskedLM(BertPreTrainedModel):
    method __init__ (line 1305) | def __init__(self, config):
    method get_output_embeddings (line 1320) | def get_output_embeddings(self):
    method set_output_embeddings (line 1323) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1334) | def forward(
    method prepare_inputs_for_generation (line 1391) | def prepare_inputs_for_generation(self, input_ids, attention_mask=None...
  class BertForNextSentencePrediction (line 1412) | class BertForNextSentencePrediction(BertPreTrainedModel):
    method __init__ (line 1413) | def __init__(self, config):
    method forward (line 1424) | def forward(
  class BertForSequenceClassification (line 1517) | class BertForSequenceClassification(BertPreTrainedModel):
    method __init__ (line 1518) | def __init__(self, config):
    method forward (line 1541) | def forward(
  class BertForMultipleChoice (line 1620) | class BertForMultipleChoice(BertPreTrainedModel):
    method __init__ (line 1621) | def __init__(self, config):
    method forward (line 1640) | def forward(
  class BertForTokenClassification (line 1714) | class BertForTokenClassification(BertPreTrainedModel):
    method __init__ (line 1717) | def __init__(self, config):
    method forward (line 1739) | def forward(
  class BertForQuestionAnswering (line 1799) | class BertForQuestionAnswering(BertPreTrainedModel):
    method __init__ (line 1802) | def __init__(self, config):
    method forward (line 1822) | def forward(

FILE: transformers/models/bert/modeling_flax_bert.py
  class FlaxBertForPreTrainingOutput (line 62) | class FlaxBertForPreTrainingOutput(ModelOutput):
  class FlaxBertEmbeddings (line 177) | class FlaxBertEmbeddings(nn.Module):
    method setup (line 183) | def setup(self):
    method __call__ (line 205) | def __call__(self, input_ids, token_type_ids, position_ids, attention_...
  class FlaxBertSelfAttention (line 220) | class FlaxBertSelfAttention(nn.Module):
    method setup (line 225) | def setup(self):
    method _split_heads (line 254) | def _split_heads(self, hidden_states):
    method _merge_heads (line 257) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 262) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 293) | def __call__(
  class FlaxBertSelfOutput (line 391) | class FlaxBertSelfOutput(nn.Module):
    method setup (line 395) | def setup(self):
    method __call__ (line 404) | def __call__(self, hidden_states, input_tensor, deterministic: bool = ...
  class FlaxBertAttention (line 411) | class FlaxBertAttention(nn.Module):
    method setup (line 416) | def setup(self):
    method __call__ (line 420) | def __call__(
  class FlaxBertIntermediate (line 453) | class FlaxBertIntermediate(nn.Module):
    method setup (line 457) | def setup(self):
    method __call__ (line 465) | def __call__(self, hidden_states):
  class FlaxBertOutput (line 471) | class FlaxBertOutput(nn.Module):
    method setup (line 475) | def setup(self):
    method __call__ (line 484) | def __call__(self, hidden_states, attention_output, deterministic: boo...
  class FlaxBertLayer (line 491) | class FlaxBertLayer(nn.Module):
    method setup (line 495) | def setup(self):
    method __call__ (line 502) | def __call__(
  class FlaxBertLayerCollection (line 548) | class FlaxBertLayerCollection(nn.Module):
    method setup (line 553) | def setup(self):
    method __call__ (line 565) | def __call__(
  class FlaxBertEncoder (line 629) | class FlaxBertEncoder(nn.Module):
    method setup (line 634) | def setup(self):
    method __call__ (line 641) | def __call__(
  class FlaxBertPooler (line 668) | class FlaxBertPooler(nn.Module):
    method setup (line 672) | def setup(self):
    method __call__ (line 679) | def __call__(self, hidden_states):
  class FlaxBertPredictionHeadTransform (line 685) | class FlaxBertPredictionHeadTransform(nn.Module):
    method setup (line 689) | def setup(self):
    method __call__ (line 694) | def __call__(self, hidden_states):
  class FlaxBertLMPredictionHead (line 700) | class FlaxBertLMPredictionHead(nn.Module):
    method setup (line 705) | def setup(self):
    method __call__ (line 710) | def __call__(self, hidden_states, shared_embedding=None):
  class FlaxBertOnlyMLMHead (line 723) | class FlaxBertOnlyMLMHead(nn.Module):
    method setup (line 727) | def setup(self):
    method __call__ (line 730) | def __call__(self, hidden_states, shared_embedding=None):
  class FlaxBertOnlyNSPHead (line 735) | class FlaxBertOnlyNSPHead(nn.Module):
    method setup (line 738) | def setup(self):
    method __call__ (line 741) | def __call__(self, pooled_output):
  class FlaxBertPreTrainingHeads (line 745) | class FlaxBertPreTrainingHeads(nn.Module):
    method setup (line 749) | def setup(self):
    method __call__ (line 753) | def __call__(self, hidden_states, pooled_output, shared_embedding=None):
  class FlaxBertPreTrainedModel (line 759) | class FlaxBertPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 769) | def __init__(
    method enable_gradient_checkpointing (line 787) | def enable_gradient_checkpointing(self):
    method init_weights (line 794) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 837) | def init_cache(self, batch_size, max_length):
    method __call__ (line 857) | def __call__(
  class FlaxBertModule (line 954) | class FlaxBertModule(nn.Module):
    method setup (line 960) | def setup(self):
    method __call__ (line 969) | def __call__(
  class FlaxBertModel (line 1029) | class FlaxBertModel(FlaxBertPreTrainedModel):
  class FlaxBertForPreTrainingModule (line 1036) | class FlaxBertForPreTrainingModule(nn.Module):
    method setup (line 1041) | def setup(self):
    method __call__ (line 1049) | def __call__(
  class FlaxBertForPreTraining (line 1104) | class FlaxBertForPreTraining(FlaxBertPreTrainedModel):
  class FlaxBertForMaskedLMModule (line 1136) | class FlaxBertForMaskedLMModule(nn.Module):
    method setup (line 1141) | def setup(self):
    method __call__ (line 1150) | def __call__(
  class FlaxBertForMaskedLM (line 1195) | class FlaxBertForMaskedLM(FlaxBertPreTrainedModel):
  class FlaxBertForNextSentencePredictionModule (line 1202) | class FlaxBertForNextSentencePredictionModule(nn.Module):
    method setup (line 1207) | def setup(self):
    method __call__ (line 1215) | def __call__(
  class FlaxBertForNextSentencePrediction (line 1259) | class FlaxBertForNextSentencePrediction(FlaxBertPreTrainedModel):
  class FlaxBertForSequenceClassificationModule (line 1294) | class FlaxBertForSequenceClassificationModule(nn.Module):
    method setup (line 1299) | def setup(self):
    method __call__ (line 1316) | def __call__(
  class FlaxBertForSequenceClassification (line 1362) | class FlaxBertForSequenceClassification(FlaxBertPreTrainedModel):
  class FlaxBertForMultipleChoiceModule (line 1374) | class FlaxBertForMultipleChoiceModule(nn.Module):
    method setup (line 1379) | def setup(self):
    method __call__ (line 1388) | def __call__(
  class FlaxBertForMultipleChoice (line 1442) | class FlaxBertForMultipleChoice(FlaxBertPreTrainedModel):
  class FlaxBertForTokenClassificationModule (line 1454) | class FlaxBertForTokenClassificationModule(nn.Module):
    method setup (line 1459) | def setup(self):
    method __call__ (line 1474) | def __call__(
  class FlaxBertForTokenClassification (line 1520) | class FlaxBertForTokenClassification(FlaxBertPreTrainedModel):
  class FlaxBertForQuestionAnsweringModule (line 1529) | class FlaxBertForQuestionAnsweringModule(nn.Module):
    method setup (line 1534) | def setup(self):
    method __call__ (line 1543) | def __call__(
  class FlaxBertForQuestionAnswering (line 1593) | class FlaxBertForQuestionAnswering(FlaxBertPreTrainedModel):
  class FlaxBertForCausalLMModule (line 1605) | class FlaxBertForCausalLMModule(nn.Module):
    method setup (line 1610) | def setup(self):
    method __call__ (line 1619) | def __call__(
  class FlaxBertForCausalLM (line 1677) | class FlaxBertForCausalLM(FlaxBertPreTrainedModel):
    method prepare_inputs_for_generation (line 1680) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 1701) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/bert/modeling_tf_bert.py
  class TFBertPreTrainingLoss (line 116) | class TFBertPreTrainingLoss:
    method hf_compute_loss (line 123) | def hf_compute_loss(self, labels: tf.Tensor, logits: tf.Tensor) -> tf....
  class TFBertEmbeddings (line 146) | class TFBertEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 149) | def __init__(self, config: BertConfig, **kwargs):
    method build (line 159) | def build(self, input_shape: tf.TensorShape):
    method call (line 183) | def call(
  class TFBertSelfAttention (line 224) | class TFBertSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 225) | def __init__(self, config: BertConfig, **kwargs):
    method transpose_for_scores (line 252) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 259) | def call(
  class TFBertSelfOutput (line 341) | class TFBertSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 342) | def __init__(self, config: BertConfig, **kwargs):
    method call (line 351) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFBertAttention (line 359) | class TFBertAttention(tf.keras.layers.Layer):
    method __init__ (line 360) | def __init__(self, config: BertConfig, **kwargs):
    method prune_heads (line 366) | def prune_heads(self, heads):
    method call (line 369) | def call(
  class TFBertIntermediate (line 399) | class TFBertIntermediate(tf.keras.layers.Layer):
    method __init__ (line 400) | def __init__(self, config: BertConfig, **kwargs):
    method call (line 412) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFBertOutput (line 419) | class TFBertOutput(tf.keras.layers.Layer):
    method __init__ (line 420) | def __init__(self, config: BertConfig, **kwargs):
    method call (line 429) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFBertLayer (line 437) | class TFBertLayer(tf.keras.layers.Layer):
    method __init__ (line 438) | def __init__(self, config: BertConfig, **kwargs):
    method call (line 451) | def call(
  class TFBertEncoder (line 523) | class TFBertEncoder(tf.keras.layers.Layer):
    method __init__ (line 524) | def __init__(self, config: BertConfig, **kwargs):
    method call (line 529) | def call(
  class TFBertPooler (line 592) | class TFBertPooler(tf.keras.layers.Layer):
    method __init__ (line 593) | def __init__(self, config: BertConfig, **kwargs):
    method call (line 603) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFBertPredictionHeadTransform (line 612) | class TFBertPredictionHeadTransform(tf.keras.layers.Layer):
    method __init__ (line 613) | def __init__(self, config: BertConfig, **kwargs):
    method call (line 629) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFBertLMPredictionHead (line 637) | class TFBertLMPredictionHead(tf.keras.layers.Layer):
    method __init__ (line 638) | def __init__(self, config: BertConfig, input_embeddings: tf.keras.laye...
    method build (line 650) | def build(self, input_shape: tf.TensorShape):
    method get_output_embeddings (line 655) | def get_output_embeddings(self) -> tf.keras.layers.Layer:
    method set_output_embeddings (line 658) | def set_output_embeddings(self, value: tf.Variable):
    method get_bias (line 662) | def get_bias(self) -> Dict[str, tf.Variable]:
    method set_bias (line 665) | def set_bias(self, value: tf.Variable):
    method call (line 669) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFBertMLMHead (line 680) | class TFBertMLMHead(tf.keras.layers.Layer):
    method __init__ (line 681) | def __init__(self, config: BertConfig, input_embeddings: tf.keras.laye...
    method call (line 686) | def call(self, sequence_output: tf.Tensor) -> tf.Tensor:
  class TFBertNSPHead (line 692) | class TFBertNSPHead(tf.keras.layers.Layer):
    method __init__ (line 693) | def __init__(self, config: BertConfig, **kwargs):
    method call (line 702) | def call(self, pooled_output: tf.Tensor) -> tf.Tensor:
  class TFBertMainLayer (line 709) | class TFBertMainLayer(tf.keras.layers.Layer):
    method __init__ (line 712) | def __init__(self, config: BertConfig, add_pooling_layer: bool = True,...
    method get_input_embeddings (line 722) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 725) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 729) | def _prune_heads(self, heads_to_prune):
    method call (line 737) | def call(
  class TFBertPreTrainedModel (line 895) | class TFBertPreTrainedModel(TFPreTrainedModel):
  class TFBertForPreTrainingOutput (line 906) | class TFBertForPreTrainingOutput(ModelOutput):
  class TFBertModel (line 1038) | class TFBertModel(TFBertPreTrainedModel):
    method __init__ (line 1039) | def __init__(self, config: BertConfig, *inputs, **kwargs):
    method call (line 1051) | def call(
  class TFBertForPreTraining (line 1114) | class TFBertForPreTraining(TFBertPreTrainedModel, TFBertPreTrainingLoss):
    method __init__ (line 1122) | def __init__(self, config: BertConfig, *inputs, **kwargs):
    method get_lm_head (line 1129) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method get_prefix_bias_name (line 1132) | def get_prefix_bias_name(self) -> str:
    method call (line 1139) | def call(
  class TFBertForMaskedLM (line 1220) | class TFBertForMaskedLM(TFBertPreTrainedModel, TFMaskedLanguageModelingL...
    method __init__ (line 1229) | def __init__(self, config: BertConfig, *inputs, **kwargs):
    method get_lm_head (line 1241) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method get_prefix_bias_name (line 1244) | def get_prefix_bias_name(self) -> str:
    method call (line 1257) | def call(
  class TFBertLMHeadModel (line 1305) | class TFBertLMHeadModel(TFBertPreTrainedModel, TFCausalLanguageModelingL...
    method __init__ (line 1314) | def __init__(self, config: BertConfig, *inputs, **kwargs):
    method get_lm_head (line 1323) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method get_prefix_bias_name (line 1326) | def get_prefix_bias_name(self) -> str:
    method prepare_inputs_for_generation (line 1330) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method call (line 1348) | def call(
  class TFBertForNextSentencePrediction (line 1434) | class TFBertForNextSentencePrediction(TFBertPreTrainedModel, TFNextSente...
    method __init__ (line 1438) | def __init__(self, config: BertConfig, *inputs, **kwargs):
    method call (line 1447) | def call(
  class TFBertForSequenceClassification (line 1519) | class TFBertForSequenceClassification(TFBertPreTrainedModel, TFSequenceC...
    method __init__ (line 1524) | def __init__(self, config: BertConfig, *inputs, **kwargs):
    method call (line 1549) | def call(
  class TFBertForMultipleChoice (line 1605) | class TFBertForMultipleChoice(TFBertPreTrainedModel, TFMultipleChoiceLoss):
    method __init__ (line 1610) | def __init__(self, config: BertConfig, *inputs, **kwargs):
    method call (line 1626) | def call(
  class TFBertForTokenClassification (line 1704) | class TFBertForTokenClassification(TFBertPreTrainedModel, TFTokenClassif...
    method __init__ (line 1715) | def __init__(self, config: BertConfig, *inputs, **kwargs):
    method call (line 1740) | def call(
  class TFBertForQuestionAnswering (line 1794) | class TFBertForQuestionAnswering(TFBertPreTrainedModel, TFQuestionAnswer...
    method __init__ (line 1804) | def __init__(self, config: BertConfig, *inputs, **kwargs):
    method call (line 1827) | def call(

FILE: transformers/models/bert/tokenization_bert.py
  function load_vocab (line 117) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 128) | def whitespace_tokenize(text):
  class BertTokenizer (line 137) | class BertTokenizer(PreTrainedTokenizer):
    method __init__ (line 184) | def __init__(
    method do_lower_case (line 231) | def do_lower_case(self):
    method vocab_size (line 235) | def vocab_size(self):
    method get_vocab (line 238) | def get_vocab(self):
    method _tokenize (line 241) | def _tokenize(self, text):
    method _convert_token_to_id (line 254) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 258) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 262) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 267) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 292) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 320) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 349) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 370) | class BasicTokenizer(object):
    method __init__ (line 390) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 398) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 435) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 446) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 468) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 481) | def _is_chinese_char(self, cp):
    method _clean_text (line 505) | def _clean_text(self, text):
  class WordpieceTokenizer (line 519) | class WordpieceTokenizer(object):
    method __init__ (line 522) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 527) | def tokenize(self, text):

FILE: transformers/models/bert/tokenization_bert_fast.py
  class BertTokenizerFast (line 161) | class BertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 207) | def __init__(
    method build_inputs_with_special_tokens (line 249) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 273) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 302) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/bert/tokenization_bert_tf.py
  class TFBertTokenizer (line 11) | class TFBertTokenizer(tf.keras.layers.Layer):
    method __init__ (line 54) | def __init__(
    method from_tokenizer (line 100) | def from_tokenizer(cls, tokenizer: "PreTrainedTokenizerBase", **kwargs...
    method from_pretrained (line 139) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
    method unpaired_tokenize (line 163) | def unpaired_tokenize(self, texts):
    method call (line 169) | def call(
    method get_config (line 241) | def get_config(self):

FILE: transformers/models/bert_generation/configuration_bert_generation.py
  class BertGenerationConfig (line 20) | class BertGenerationConfig(PretrainedConfig):
    method __init__ (line 85) | def __init__(

FILE: transformers/models/bert_generation/modeling_bert_generation.py
  class BertGenerationSelfOutput (line 46) | class BertGenerationSelfOutput(nn.Module):
    method __init__ (line 47) | def __init__(self, config):
    method forward (line 53) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BertGenerationSelfAttention (line 61) | class BertGenerationSelfAttention(nn.Module):
    method __init__ (line 62) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 88) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 93) | def forward(
  class BertGenerationAttention (line 196) | class BertGenerationAttention(nn.Module):
    method __init__ (line 197) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 203) | def prune_heads(self, heads):
    method forward (line 221) | def forward(
  class BertGenerationIntermediate (line 246) | class BertGenerationIntermediate(nn.Module):
    method __init__ (line 247) | def __init__(self, config):
    method forward (line 255) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BertGenerationOutput (line 262) | class BertGenerationOutput(nn.Module):
    method __init__ (line 263) | def __init__(self, config):
    method forward (line 269) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BertGenerationLayer (line 277) | class BertGenerationLayer(nn.Module):
    method __init__ (line 278) | def __init__(self, config):
    method forward (line 292) | def forward(
    method feed_forward_chunk (line 357) | def feed_forward_chunk(self, attention_output):
  class BertEncoder (line 364) | class BertEncoder(nn.Module):
    method __init__ (line 365) | def __init__(self, config):
    method forward (line 371) | def forward(
  function load_tf_weights_in_bert_generation (line 462) | def load_tf_weights_in_bert_generation(
  class BertGenerationEmbeddings (line 546) | class BertGenerationEmbeddings(nn.Module):
    method __init__ (line 549) | def __init__(self, config):
    method forward (line 561) | def forward(self, input_ids=None, position_ids=None, inputs_embeds=Non...
  class BertGenerationPreTrainedModel (line 582) | class BertGenerationPreTrainedModel(PreTrainedModel):
    method _init_weights (line 593) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 609) | def _set_gradient_checkpointing(self, module, value=False):
  class BertGenerationEncoder (line 676) | class BertGenerationEncoder(BertGenerationPreTrainedModel):
    method __init__ (line 693) | def __init__(self, config):
    method get_input_embeddings (line 703) | def get_input_embeddings(self):
    method set_input_embeddings (line 706) | def set_input_embeddings(self, value):
    method _prune_heads (line 709) | def _prune_heads(self, heads_to_prune):
    method forward (line 723) | def forward(
  class BertGenerationOnlyLMHead (line 842) | class BertGenerationOnlyLMHead(nn.Module):
    method __init__ (line 843) | def __init__(self, config):
    method forward (line 849) | def forward(self, hidden_states):
    method _tie_weights (line 853) | def _tie_weights(self):
  class BertGenerationDecoder (line 862) | class BertGenerationDecoder(BertGenerationPreTrainedModel):
    method __init__ (line 865) | def __init__(self, config):
    method get_output_embeddings (line 877) | def get_output_embeddings(self):
    method set_output_embeddings (line 880) | def set_output_embeddings(self, new_embeddings):
    method forward (line 885) | def forward(
    method prepare_inputs_for_generation (line 988) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1000) | def _reorder_cache(self, past_key_values, beam_idx):

FILE: transformers/models/bert_generation/tokenization_bert_generation.py
  class BertGenerationTokenizer (line 43) | class BertGenerationTokenizer(PreTrainedTokenizer):
    method __init__ (line 86) | def __init__(
    method vocab_size (line 116) | def vocab_size(self):
    method get_vocab (line 119) | def get_vocab(self):
    method __getstate__ (line 124) | def __getstate__(self):
    method __setstate__ (line 129) | def __setstate__(self, d):
    method _tokenize (line 139) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 143) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 147) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 152) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 166) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/bert_japanese/tokenization_bert_japanese.py
  function load_vocab (line 86) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 98) | def whitespace_tokenize(text):
  class BertJapaneseTokenizer (line 107) | class BertJapaneseTokenizer(PreTrainedTokenizer):
    method __init__ (line 143) | def __init__(
    method do_lower_case (line 238) | def do_lower_case(self):
    method __getstate__ (line 241) | def __getstate__(self):
    method __setstate__ (line 247) | def __setstate__(self, state):
    method _tokenize (line 262) | def _tokenize(self, text):
    method vocab_size (line 276) | def vocab_size(self):
    method get_vocab (line 281) | def get_vocab(self):
    method _convert_token_to_id (line 288) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 294) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 300) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 308) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 334) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 363) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 392) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class MecabTokenizer (line 425) | class MecabTokenizer:
    method __init__ (line 428) | def __init__(
    method tokenize (line 514) | def tokenize(self, text, never_split=None, **kwargs):
  class SudachiTokenizer (line 533) | class SudachiTokenizer:
    method __init__ (line 536) | def __init__(
    method tokenize (line 594) | def tokenize(self, text, never_split=None, **kwargs):
  class JumanppTokenizer (line 619) | class JumanppTokenizer:
    method __init__ (line 622) | def __init__(
    method tokenize (line 659) | def tokenize(self, text, never_split=None, **kwargs):
  class CharacterTokenizer (line 686) | class CharacterTokenizer:
    method __init__ (line 689) | def __init__(self, vocab, unk_token, normalize_text=True):
    method tokenize (line 705) | def tokenize(self, text):
  class BasicTokenizer (line 733) | class BasicTokenizer(object):
    method __init__ (line 753) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 761) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 798) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 809) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 831) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 844) | def _is_chinese_char(self, cp):
    method _clean_text (line 868) | def _clean_text(self, text):
  class WordpieceTokenizer (line 883) | class WordpieceTokenizer(object):
    method __init__ (line 886) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 891) | def tokenize(self, text):
  class SentencepieceTokenizer (line 940) | class SentencepieceTokenizer(object):
    method __init__ (line 945) | def __init__(
    method preprocess_text (line 964) | def preprocess_text(self, inputs):
    method tokenize (line 979) | def tokenize(self, text):

FILE: transformers/models/bertweet/tokenization_bertweet.py
  function get_pairs (line 52) | def get_pairs(word):
  class BertweetTokenizer (line 68) | class BertweetTokenizer(PreTrainedTokenizer):
    method __init__ (line 123) | def __init__(
    method build_inputs_with_special_tokens (line 184) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 210) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 238) | def create_token_type_ids_from_sequences(
    method vocab_size (line 263) | def vocab_size(self):
    method get_vocab (line 266) | def get_vocab(self):
    method bpe (line 269) | def bpe(self, token):
    method _tokenize (line 313) | def _tokenize(self, text):
    method normalizeTweet (line 324) | def normalizeTweet(self, tweet):
    method normalizeToken (line 358) | def normalizeToken(self, token):
    method _convert_token_to_id (line 377) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 381) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 385) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 390) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method add_from_file (line 415) | def add_from_file(self, f):
  function _str_to_unicode (line 617) | def _str_to_unicode(text, encoding=None, errors="strict"):
  function _replace_html_entities (line 625) | def _replace_html_entities(text, keep=(), remove_illegal=True, encoding=...
  class TweetTokenizer (line 690) | class TweetTokenizer:
    method __init__ (line 710) | def __init__(self, preserve_case=True, reduce_len=False, strip_handles...
    method tokenize (line 715) | def tokenize(self, text):
  function reduce_lengthening (line 746) | def reduce_lengthening(text):
  function remove_handles (line 754) | def remove_handles(text):
  function casual_tokenize (line 770) | def casual_tokenize(text, preserve_case=True, reduce_len=False, strip_ha...

FILE: transformers/models/big_bird/configuration_big_bird.py
  class BigBirdConfig (line 34) | class BigBirdConfig(PretrainedConfig):
    method __init__ (line 109) | def __init__(
  class BigBirdOnnxConfig (line 166) | class BigBirdOnnxConfig(OnnxConfig):
    method inputs (line 168) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/big_bird/convert_bigbird_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 27) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, big_bird_config...

FILE: transformers/models/big_bird/modeling_big_bird.py
  function load_tf_weights_in_big_bird (line 83) | def load_tf_weights_in_big_bird(model, tf_checkpoint_path, is_trivia_qa=...
  class BigBirdEmbeddings (line 244) | class BigBirdEmbeddings(nn.Module):
    method __init__ (line 248) | def __init__(self, config):
    method forward (line 269) | def forward(
  class BigBirdSelfAttention (line 311) | class BigBirdSelfAttention(nn.Module):
    method __init__ (line 312) | def __init__(self, config):
    method transpose_for_scores (line 331) | def transpose_for_scores(self, x):
    method forward (line 336) | def forward(
  class BigBirdBlockSparseAttention (line 415) | class BigBirdBlockSparseAttention(nn.Module):
    method __init__ (line 416) | def __init__(self, config, seed=None):
    method transpose_for_scores (line 439) | def transpose_for_scores(self, x):
    method forward (line 444) | def forward(
    method torch_bmm_nd (line 499) | def torch_bmm_nd(inp_1, inp_2, ndim=None):
    method torch_bmm_nd_transpose (line 507) | def torch_bmm_nd_transpose(inp_1, inp_2, ndim=None):
    method bigbird_block_sparse_attention (line 514) | def bigbird_block_sparse_attention(
    method torch_gather_b2 (line 963) | def torch_gather_b2(params, indices):
    method _create_rand_mask_from_inputs (line 986) | def _create_rand_mask_from_inputs(
    method _get_rand_attn_plan (line 1023) | def _get_rand_attn_plan(from_seq_length, from_block_size, num_rand_blo...
    method _bigbird_block_rand_mask (line 1055) | def _bigbird_block_rand_mask(
    method _bigbird_block_rand_mask_with_head (line 1113) | def _bigbird_block_rand_mask_with_head(
    method _get_single_block_row_attention (line 1253) | def _get_single_block_row_attention(
  class BigBirdSelfOutput (line 1310) | class BigBirdSelfOutput(nn.Module):
    method __init__ (line 1311) | def __init__(self, config):
    method forward (line 1317) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BigBirdAttention (line 1324) | class BigBirdAttention(nn.Module):
    method __init__ (line 1325) | def __init__(self, config, seed=None):
    method set_attention_type (line 1342) | def set_attention_type(self, value: str):
    method forward (line 1367) | def forward(
  class BigBirdIntermediate (line 1413) | class BigBirdIntermediate(nn.Module):
    method __init__ (line 1414) | def __init__(self, config):
    method forward (line 1422) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BigBirdOutput (line 1429) | class BigBirdOutput(nn.Module):
    method __init__ (line 1430) | def __init__(self, config):
    method forward (line 1436) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BigBirdLayer (line 1443) | class BigBirdLayer(nn.Module):
    method __init__ (line 1444) | def __init__(self, config, seed=None):
    method set_attention_type (line 1460) | def set_attention_type(self, value: str):
    method forward (line 1474) | def forward(
    method feed_forward_chunk (line 1551) | def feed_forward_chunk(self, attention_output):
  class BigBirdEncoder (line 1557) | class BigBirdEncoder(nn.Module):
    method __init__ (line 1558) | def __init__(self, config):
    method set_attention_type (line 1568) | def set_attention_type(self, value: str):
    method forward (line 1580) | def forward(
  class BigBirdPredictionHeadTransform (line 1685) | class BigBirdPredictionHeadTransform(nn.Module):
    method __init__ (line 1686) | def __init__(self, config):
    method forward (line 1695) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BigBirdLMPredictionHead (line 1703) | class BigBirdLMPredictionHead(nn.Module):
    method __init__ (line 1704) | def __init__(self, config):
    method forward (line 1717) | def forward(self, hidden_states):
  class BigBirdOnlyMLMHead (line 1724) | class BigBirdOnlyMLMHead(nn.Module):
    method __init__ (line 1725) | def __init__(self, config):
    method forward (line 1729) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class BigBirdOnlyNSPHead (line 1735) | class BigBirdOnlyNSPHead(nn.Module):
    method __init__ (line 1736) | def __init__(self, config):
    method forward (line 1740) | def forward(self, pooled_output):
  class BigBirdPreTrainingHeads (line 1746) | class BigBirdPreTrainingHeads(nn.Module):
    method __init__ (line 1747) | def __init__(self, config):
    method forward (line 1752) | def forward(self, sequence_output, pooled_output):
  class BigBirdPreTrainedModel (line 1758) | class BigBirdPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1770) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1786) | def _set_gradient_checkpointing(self, module, value=False):
  class BigBirdForPreTrainingOutput (line 1853) | class BigBirdForPreTrainingOutput(ModelOutput):
  class BigBirdForQuestionAnsweringModelOutput (line 1887) | class BigBirdForQuestionAnsweringModelOutput(ModelOutput):
  class BigBirdModel (line 1925) | class BigBirdModel(BigBirdPreTrainedModel):
    method __init__ (line 1938) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 1965) | def get_input_embeddings(self):
    method set_input_embeddings (line 1968) | def set_input_embeddings(self, value):
    method set_attention_type (line 1971) | def set_attention_type(self, value: str):
    method forward (line 1988) | def forward(
    method create_masks_for_block_sparse_attn (line 2182) | def create_masks_for_block_sparse_attn(attention_mask: torch.Tensor, b...
    method _pad_to_block_size (line 2219) | def _pad_to_block_size(
  class BigBirdForPreTraining (line 2263) | class BigBirdForPreTraining(BigBirdPreTrainedModel):
    method __init__ (line 2266) | def __init__(self, config):
    method get_output_embeddings (line 2275) | def get_output_embeddings(self):
    method set_output_embeddings (line 2278) | def set_output_embeddings(self, new_embeddings):
    method forward (line 2283) | def forward(
  class BigBirdForMaskedLM (line 2369) | class BigBirdForMaskedLM(BigBirdPreTrainedModel):
    method __init__ (line 2372) | def __init__(self, config):
    method get_output_embeddings (line 2387) | def get_output_embeddings(self):
    method set_output_embeddings (line 2390) | def set_output_embeddings(self, new_embeddings):
    method forward (line 2395) | def forward(
    method prepare_inputs_for_generation (line 2494) | def prepare_inputs_for_generation(self, input_ids, attention_mask=None...
  class BigBirdForCausalLM (line 2513) | class BigBirdForCausalLM(BigBirdPreTrainedModel):
    method __init__ (line 2521) | def __init__(self, config):
    method get_output_embeddings (line 2533) | def get_output_embeddings(self):
    method set_output_embeddings (line 2536) | def set_output_embeddings(self, new_embeddings):
    method forward (line 2545) | def forward(
    method prepare_inputs_for_generation (line 2627) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 2640) | def _reorder_cache(self, past_key_values, beam_idx):
  class BigBirdClassificationHead (line 2649) | class BigBirdClassificationHead(nn.Module):
    method __init__ (line 2652) | def __init__(self, config):
    method forward (line 2663) | def forward(self, features, **kwargs):
  class BigBirdForSequenceClassification (line 2680) | class BigBirdForSequenceClassification(BigBirdPreTrainedModel):
    method __init__ (line 2681) | def __init__(self, config):
    method forward (line 2693) | def forward(
  class BigBirdForMultipleChoice (line 2808) | class BigBirdForMultipleChoice(BigBirdPreTrainedModel):
    method __init__ (line 2809) | def __init__(self, config):
    method forward (line 2827) | def forward(
  class BigBirdForTokenClassification (line 2901) | class BigBirdForTokenClassification(BigBirdPreTrainedModel):
    method __init__ (line 2902) | def __init__(self, config):
    method forward (line 2922) | def forward(
  class BigBirdForQuestionAnsweringHead (line 2975) | class BigBirdForQuestionAnsweringHead(nn.Module):
    method __init__ (line 2978) | def __init__(self, config):
    method forward (line 2985) | def forward(self, encoder_output):
  class BigBirdForQuestionAnswering (line 3000) | class BigBirdForQuestionAnswering(BigBirdPreTrainedModel):
    method __init__ (line 3001) | def __init__(self, config, add_pooling_layer=False):
    method forward (line 3016) | def forward(
    method prepare_question_mask (line 3153) | def prepare_question_mask(q_lengths: torch.Tensor, maxlen: int):

FILE: transformers/models/big_bird/modeling_flax_big_bird.py
  class FlaxBigBirdForPreTrainingOutput (line 59) | class FlaxBigBirdForPreTrainingOutput(ModelOutput):
  class FlaxBigBirdForQuestionAnsweringModelOutput (line 89) | class FlaxBigBirdForQuestionAnsweringModelOutput(ModelOutput):
  class FlaxBigBirdEmbeddings (line 193) | class FlaxBigBirdEmbeddings(nn.Module):
    method setup (line 200) | def setup(self):
    method __call__ (line 222) | def __call__(self, input_ids, token_type_ids, position_ids, attention_...
  class FlaxBigBirdSelfAttention (line 241) | class FlaxBigBirdSelfAttention(nn.Module):
    method setup (line 246) | def setup(self):
    method _split_heads (line 275) | def _split_heads(self, hidden_states):
    method _merge_heads (line 278) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 283) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 314) | def __call__(
  class FlaxBigBirdBlockSparseAttention (line 412) | class FlaxBigBirdBlockSparseAttention(nn.Module):
    method setup (line 417) | def setup(self):
    method transpose_for_scores (line 438) | def transpose_for_scores(x, n_heads, head_size):
    method __call__ (line 443) | def __call__(
    method create_masks_for_block_sparse_attn (line 487) | def create_masks_for_block_sparse_attn(attention_mask, block_size: int):
    method bigbird_block_sparse_attention (line 524) | def bigbird_block_sparse_attention(
    method jax_gather (line 869) | def jax_gather(params, indices, batch_dims=2):
    method _create_rand_mask_from_inputs (line 886) | def _create_rand_mask_from_inputs(
    method _get_rand_attn_plan (line 924) | def _get_rand_attn_plan(from_seq_length, from_block_size, num_rand_blo...
    method _bigbird_block_rand_mask (line 957) | def _bigbird_block_rand_mask(
    method _bigbird_block_rand_mask_with_head (line 1030) | def _bigbird_block_rand_mask_with_head(
    method _get_single_block_row_attention (line 1183) | def _get_single_block_row_attention(
  class FlaxBigBirdSelfOutput (line 1242) | class FlaxBigBirdSelfOutput(nn.Module):
    method setup (line 1246) | def setup(self):
    method __call__ (line 1255) | def __call__(self, hidden_states, input_tensor, deterministic: bool = ...
  class FlaxBigBirdAttention (line 1262) | class FlaxBigBirdAttention(nn.Module):
    method setup (line 1268) | def setup(self):
    method __call__ (line 1281) | def __call__(
  class FlaxBigBirdIntermediate (line 1323) | class FlaxBigBirdIntermediate(nn.Module):
    method setup (line 1327) | def setup(self):
    method __call__ (line 1335) | def __call__(self, hidden_states):
  class FlaxBigBirdOutput (line 1342) | class FlaxBigBirdOutput(nn.Module):
    method setup (line 1346) | def setup(self):
    method __call__ (line 1355) | def __call__(self, hidden_states, attention_output, deterministic: boo...
  class FlaxBigBirdLayer (line 1362) | class FlaxBigBirdLayer(nn.Module):
    method setup (line 1367) | def setup(self):
    method __call__ (line 1377) | def __call__(
  class FlaxBigBirdLayerCollection (line 1423) | class FlaxBigBirdLayerCollection(nn.Module):
    method setup (line 1428) | def setup(self):
    method __call__ (line 1442) | def __call__(
  class FlaxBigBirdEncoder (line 1507) | class FlaxBigBirdEncoder(nn.Module):
    method setup (line 1512) | def setup(self):
    method __call__ (line 1519) | def __call__(
  class FlaxBigBirdPredictionHeadTransform (line 1547) | class FlaxBigBirdPredictionHeadTransform(nn.Module):
    method setup (line 1551) | def setup(self):
    method __call__ (line 1556) | def __call__(self, hidden_states):
  class FlaxBigBirdLMPredictionHead (line 1563) | class FlaxBigBirdLMPredictionHead(nn.Module):
    method setup (line 1568) | def setup(self):
    method __call__ (line 1573) | def __call__(self, hidden_states, shared_embedding=None):
  class FlaxBigBirdOnlyMLMHead (line 1587) | class FlaxBigBirdOnlyMLMHead(nn.Module):
    method setup (line 1591) | def setup(self):
    method __call__ (line 1594) | def __call__(self, hidden_states, shared_embedding=None):
  class FlaxBigBirdPreTrainingHeads (line 1599) | class FlaxBigBirdPreTrainingHeads(nn.Module):
    method setup (line 1603) | def setup(self):
    method __call__ (line 1607) | def __call__(self, hidden_states, pooled_output, shared_embedding=None):
  class FlaxBigBirdPreTrainedModel (line 1613) | class FlaxBigBirdPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 1623) | def __init__(
    method enable_gradient_checkpointing (line 1642) | def enable_gradient_checkpointing(self):
    method init_weights (line 1649) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 1698) | def init_cache(self, batch_size, max_length):
    method __call__ (line 1718) | def __call__(
  class FlaxBigBirdModule (line 1819) | class FlaxBigBirdModule(nn.Module):
    method setup (line 1825) | def setup(self):
    method __call__ (line 1836) | def __call__(
  class FlaxBigBirdModel (line 1890) | class FlaxBigBirdModel(FlaxBigBirdPreTrainedModel):
  class FlaxBigBirdForPreTrainingModule (line 1898) | class FlaxBigBirdForPreTrainingModule(nn.Module):
    method setup (line 1903) | def setup(self):
    method __call__ (line 1911) | def __call__(
  class FlaxBigBirdForPreTraining (line 1967) | class FlaxBigBirdForPreTraining(FlaxBigBirdPreTrainedModel):
  class FlaxBigBirdForMaskedLMModule (line 2000) | class FlaxBigBirdForMaskedLMModule(nn.Module):
    method setup (line 2005) | def setup(self):
    method __call__ (line 2014) | def __call__(
  class FlaxBigBirdForMaskedLM (line 2060) | class FlaxBigBirdForMaskedLM(FlaxBigBirdPreTrainedModel):
  class FlaxBigBirdClassificationHead (line 2067) | class FlaxBigBirdClassificationHead(nn.Module):
    method setup (line 2073) | def setup(self):
    method __call__ (line 2083) | def __call__(self, features, deterministic=True):
  class FlaxBigBirdForSequenceClassificationModule (line 2093) | class FlaxBigBirdForSequenceClassificationModule(nn.Module):
    method setup (line 2098) | def setup(self):
    method __call__ (line 2104) | def __call__(
  class FlaxBigBirdForSequenceClassification (line 2150) | class FlaxBigBirdForSequenceClassification(FlaxBigBirdPreTrainedModel):
  class FlaxBigBirdForMultipleChoiceModule (line 2163) | class FlaxBigBirdForMultipleChoiceModule(nn.Module):
    method setup (line 2168) | def setup(self):
    method __call__ (line 2177) | def __call__(
  class FlaxBigBirdForMultipleChoice (line 2231) | class FlaxBigBirdForMultipleChoice(FlaxBigBirdPreTrainedModel):
    method __init__ (line 2234) | def __init__(
  class FlaxBigBirdForTokenClassificationModule (line 2262) | class FlaxBigBirdForTokenClassificationModule(nn.Module):
    method setup (line 2267) | def setup(self):
    method __call__ (line 2282) | def __call__(
  class FlaxBigBirdForTokenClassification (line 2329) | class FlaxBigBirdForTokenClassification(FlaxBigBirdPreTrainedModel):
  class FlaxBigBirdForQuestionAnsweringHead (line 2341) | class FlaxBigBirdForQuestionAnsweringHead(nn.Module):
    method setup (line 2345) | def setup(self):
    method __call__ (line 2351) | def __call__(self, encoder_output, deterministic=True):
  class FlaxBigBirdForQuestionAnsweringModule (line 2359) | class FlaxBigBirdForQuestionAnsweringModule(nn.Module):
    method setup (line 2365) | def setup(self):
    method __call__ (line 2375) | def __call__(
  class FlaxBigBirdForQuestionAnswering (line 2432) | class FlaxBigBirdForQuestionAnswering(FlaxBigBirdPreTrainedModel):
    method __call__ (line 2436) | def __call__(
    method prepare_question_mask (line 2511) | def prepare_question_mask(q_lengths, maxlen: int):
  class FlaxBigBirdForCausalLMModule (line 2526) | class FlaxBigBirdForCausalLMModule(nn.Module):
    method setup (line 2531) | def setup(self):
    method __call__ (line 2540) | def __call__(
  class FlaxBigBirdForCausalLM (line 2599) | class FlaxBigBirdForCausalLM(FlaxBigBirdPreTrainedModel):
    method prepare_inputs_for_generation (line 2602) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 2623) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/big_bird/tokenization_big_bird.py
  class BigBirdTokenizer (line 52) | class BigBirdTokenizer(PreTrainedTokenizer):
    method __init__ (line 105) | def __init__(
    method vocab_size (line 148) | def vocab_size(self):
    method get_vocab (line 151) | def get_vocab(self):
    method __getstate__ (line 156) | def __getstate__(self):
    method __setstate__ (line 161) | def __setstate__(self, d):
    method _tokenize (line 171) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 175) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 179) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 184) | def convert_tokens_to_string(self, tokens):
    method _decode (line 203) | def _decode(
    method save_vocabulary (line 251) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 268) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 293) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 320) | def create_token_type_ids_from_sequences(

FILE: transformers/models/big_bird/tokenization_big_bird_fast.py
  class BigBirdTokenizerFast (line 68) | class BigBirdTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 116) | def __init__(
    method build_inputs_with_special_tokens (line 155) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 180) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 211) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 241) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/bigbird_pegasus/configuration_bigbird_pegasus.py
  class BigBirdPegasusConfig (line 43) | class BigBirdPegasusConfig(PretrainedConfig):
    method __init__ (line 131) | def __init__(
  class BigBirdPegasusOnnxConfig (line 201) | class BigBirdPegasusOnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 203) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 247) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method _generate_dummy_inputs_for_default_and_seq2seq_lm (line 259) | def _generate_dummy_inputs_for_default_and_seq2seq_lm(
    method _generate_dummy_inputs_for_causal_lm (line 327) | def _generate_dummy_inputs_for_causal_lm(
    method _generate_dummy_inputs_for_sequence_classification_and_question_answering (line 365) | def _generate_dummy_inputs_for_sequence_classification_and_question_an...
    method generate_dummy_inputs (line 391) | def generate_dummy_inputs(
    method _flatten_past_key_values_ (line 415) | def _flatten_past_key_values_(self, flattened_output, name, idx, t):

FILE: transformers/models/bigbird_pegasus/convert_bigbird_pegasus_tf_to_pytorch.py
  function rename_state_dict_key (line 83) | def rename_state_dict_key(k, patterns):
  function convert_bigbird_pegasus (line 89) | def convert_bigbird_pegasus(tf_weights: dict, config_update: dict) -> Bi...
  function get_tf_weights_as_numpy (line 145) | def get_tf_weights_as_numpy(path) -> Dict:
  function convert_bigbird_pegasus_ckpt_to_pytorch (line 158) | def convert_bigbird_pegasus_ckpt_to_pytorch(ckpt_path: str, save_dir: st...

FILE: transformers/models/bigbird_pegasus/modeling_bigbird_pegasus.py
  function shift_tokens_right (line 65) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 82) | def _make_causal_mask(
  function _expand_mask (line 99) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class BigBirdPegasusLearnedPositionalEmbedding (line 113) | class BigBirdPegasusLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 118) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 121) | def forward(self, input_ids_shape: torch.Size, past_key_values_length:...
  class BigBirdPegasusSelfAttention (line 131) | class BigBirdPegasusSelfAttention(nn.Module):
    method __init__ (line 132) | def __init__(self, config):
    method transpose_for_scores (line 151) | def transpose_for_scores(self, x):
    method forward (line 156) | def forward(
  class BigBirdPegasusBlockSparseAttention (line 236) | class BigBirdPegasusBlockSparseAttention(nn.Module):
    method __init__ (line 237) | def __init__(self, config, seed=None):
    method transpose_for_scores (line 260) | def transpose_for_scores(self, x):
    method forward (line 265) | def forward(
    method torch_bmm_nd (line 320) | def torch_bmm_nd(inp_1, inp_2, ndim=None):
    method torch_bmm_nd_transpose (line 328) | def torch_bmm_nd_transpose(inp_1, inp_2, ndim=None):
    method bigbird_block_sparse_attention (line 335) | def bigbird_block_sparse_attention(
    method torch_gather_b2 (line 784) | def torch_gather_b2(params, indices):
    method _create_rand_mask_from_inputs (line 807) | def _create_rand_mask_from_inputs(
    method _get_rand_attn_plan (line 844) | def _get_rand_attn_plan(from_seq_length, from_block_size, num_rand_blo...
    method _bigbird_block_rand_mask (line 876) | def _bigbird_block_rand_mask(
    method _bigbird_block_rand_mask_with_head (line 934) | def _bigbird_block_rand_mask_with_head(
    method _get_single_block_row_attention (line 1074) | def _get_single_block_row_attention(
  class BigBirdPegasusEncoderAttention (line 1130) | class BigBirdPegasusEncoderAttention(nn.Module):
    method __init__ (line 1131) | def __init__(self, config, seed=None):
    method set_attention_type (line 1149) | def set_attention_type(self, value: str):
    method forward (line 1175) | def forward(
  class BigBirdPegasusDecoderAttention (line 1210) | class BigBirdPegasusDecoderAttention(nn.Module):
    method __init__ (line 1213) | def __init__(
    method _shape (line 1240) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 1243) | def forward(
  class BigBirdPegasusEncoderLayer (line 1364) | class BigBirdPegasusEncoderLayer(nn.Module):
    method __init__ (line 1365) | def __init__(self, config: BigBirdPegasusConfig, seed=None):
    method forward (line 1378) | def forward(
    method set_attention_type (line 1439) | def set_attention_type(self, value: str):
  class BigBirdPegasusDecoderLayer (line 1451) | class BigBirdPegasusDecoderLayer(nn.Module):
    method __init__ (line 1452) | def __init__(self, config: BigBirdPegasusConfig):
    method forward (line 1480) | def forward(
  class BigBirdPegasusClassificationHead (line 1571) | class BigBirdPegasusClassificationHead(nn.Module):
    method __init__ (line 1574) | def __init__(
    method forward (line 1586) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BigBirdPegasusPreTrainedModel (line 1595) | class BigBirdPegasusPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1602) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1613) | def _set_gradient_checkpointing(self, module, value=False):
    method dummy_inputs (line 1618) | def dummy_inputs(self):
  class BigBirdPegasusEncoder (line 1770) | class BigBirdPegasusEncoder(BigBirdPegasusPreTrainedModel):
    method __init__ (line 1780) | def __init__(self, config: BigBirdPegasusConfig, embed_tokens: Optiona...
    method forward (line 1810) | def forward(
    method set_attention_type (line 1995) | def set_attention_type(self, value: str):
    method create_masks_for_block_sparse_attn (line 2008) | def create_masks_for_block_sparse_attn(attention_mask: torch.Tensor, b...
    method _pad_to_block_size (line 2045) | def _pad_to_block_size(self, hidden_states: torch.Tensor, attention_ma...
  class BigBirdPegasusDecoder (line 2070) | class BigBirdPegasusDecoder(BigBirdPegasusPreTrainedModel):
    method __init__ (line 2079) | def __init__(self, config: BigBirdPegasusConfig, embed_tokens: Optiona...
    method get_input_embeddings (line 2103) | def get_input_embeddings(self):
    method set_input_embeddings (line 2106) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 2110) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 2133) | def forward(
  class BigBirdPegasusModel (line 2355) | class BigBirdPegasusModel(BigBirdPegasusPreTrainedModel):
    method __init__ (line 2358) | def __init__(self, config: BigBirdPegasusConfig):
    method get_input_embeddings (line 2370) | def get_input_embeddings(self):
    method set_input_embeddings (line 2373) | def set_input_embeddings(self, value):
    method get_encoder (line 2378) | def get_encoder(self):
    method get_decoder (line 2381) | def get_decoder(self):
    method forward (line 2392) | def forward(
  class BigBirdPegasusForConditionalGeneration (line 2485) | class BigBirdPegasusForConditionalGeneration(BigBirdPegasusPreTrainedMod...
    method __init__ (line 2494) | def __init__(self, config: BigBirdPegasusConfig):
    method get_encoder (line 2503) | def get_encoder(self):
    method get_decoder (line 2506) | def get_decoder(self):
    method resize_token_embeddings (line 2509) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_final_logits_bias (line 2514) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_output_embeddings (line 2523) | def get_output_embeddings(self):
    method set_output_embeddings (line 2526) | def set_output_embeddings(self, new_embeddings):
    method forward (line 2532) | def forward(
    method prepare_inputs_for_generation (line 2613) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 2643) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 2647) | def _reorder_cache(past_key_values, beam_idx):
  class BigBirdPegasusForSequenceClassification (line 2664) | class BigBirdPegasusForSequenceClassification(BigBirdPegasusPreTrainedMo...
    method __init__ (line 2667) | def __init__(self, config: BigBirdPegasusConfig, **kwargs):
    method forward (line 2687) | def forward(
  class BigBirdPegasusForQuestionAnswering (line 2793) | class BigBirdPegasusForQuestionAnswering(BigBirdPegasusPreTrainedModel):
    method __init__ (line 2796) | def __init__(self, config):
    method forward (line 2815) | def forward(
  class BigBirdPegasusDecoderWrapper (line 2911) | class BigBirdPegasusDecoderWrapper(BigBirdPegasusPreTrainedModel):
    method __init__ (line 2917) | def __init__(self, config):
    method forward (line 2921) | def forward(self, *args, **kwargs):
  class BigBirdPegasusForCausalLM (line 2925) | class BigBirdPegasusForCausalLM(BigBirdPegasusPreTrainedModel):
    method __init__ (line 2928) | def __init__(self, config):
    method get_input_embeddings (line 2940) | def get_input_embeddings(self):
    method set_input_embeddings (line 2943) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 2946) | def get_output_embeddings(self):
    method set_output_embeddings (line 2949) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 2952) | def set_decoder(self, decoder):
    method get_decoder (line 2955) | def get_decoder(self):
    method forward (line 2959) | def forward(
    method prepare_inputs_for_generation (line 3101) | def prepare_inputs_for_generation(
    method _reorder_cache (line 3119) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/biogpt/configuration_biogpt.py
  class BioGptConfig (line 29) | class BioGptConfig(PretrainedConfig):
    method __init__ (line 97) | def __init__(

FILE: transformers/models/biogpt/convert_biogpt_original_pytorch_checkpoint_to_pytorch.py
  class Dictionary (line 37) | class Dictionary:
    method __init__ (line 40) | def __init__(
    method __eq__ (line 62) | def __eq__(self, other):
    method __getitem__ (line 65) | def __getitem__(self, idx):
    method __len__ (line 70) | def __len__(self):
    method __contains__ (line 74) | def __contains__(self, sym):
    method load (line 78) | def load(cls, f):
    method add_symbol (line 91) | def add_symbol(self, word, n=1, overwrite=False):
    method _load_meta (line 104) | def _load_meta(self, lines):
    method add_from_file (line 107) | def add_from_file(self, f):
  function rewrite_dict_keys (line 147) | def rewrite_dict_keys(d):
  function convert_biogpt_checkpoint_to_pytorch (line 159) | def convert_biogpt_checkpoint_to_pytorch(biogpt_checkpoint_path, pytorch...

FILE: transformers/models/biogpt/modeling_biogpt.py
  function _make_causal_mask (line 58) | def _make_causal_mask(
  function _expand_mask (line 76) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class BioGptLearnedPositionalEmbedding (line 91) | class BioGptLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 96) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 102) | def forward(self, attention_mask: torch.LongTensor, past_key_values_le...
  class BioGptAttention (line 116) | class BioGptAttention(nn.Module):
    method __init__ (line 119) | def __init__(
    method _shape (line 146) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 149) | def forward(
  class BioGptDecoderLayer (line 270) | class BioGptDecoderLayer(nn.Module):
    method __init__ (line 271) | def __init__(self, config: BioGptConfig):
    method forward (line 291) | def forward(
  class BioGptPreTrainedModel (line 354) | class BioGptPreTrainedModel(PreTrainedModel):
    method _init_weights (line 364) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 380) | def _set_gradient_checkpointing(self, module, value=False):
  class BioGptModel (line 454) | class BioGptModel(BioGptPreTrainedModel):
    method __init__ (line 455) | def __init__(self, config: BioGptConfig):
    method get_input_embeddings (line 474) | def get_input_embeddings(self):
    method set_input_embeddings (line 477) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 481) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 510) | def forward(
  class BioGptForCausalLM (line 648) | class BioGptForCausalLM(BioGptPreTrainedModel):
    method __init__ (line 651) | def __init__(self, config):
    method get_output_embeddings (line 660) | def get_output_embeddings(self):
    method set_output_embeddings (line 663) | def set_output_embeddings(self, new_embeddings):
    method forward (line 672) | def forward(
    method prepare_inputs_for_generation (line 729) | def prepare_inputs_for_generation(
    method _reorder_cache (line 752) | def _reorder_cache(past_key_values, beam_idx):
  class BioGptForTokenClassification (line 766) | class BioGptForTokenClassification(BioGptPreTrainedModel):
    method __init__ (line 767) | def __init__(self, config):
    method forward (line 787) | def forward(
  class BioGptForSequenceClassification (line 866) | class BioGptForSequenceClassification(BioGptPreTrainedModel):
    method __init__ (line 867) | def __init__(self, config: BioGptConfig):
    method forward (line 882) | def forward(
    method get_input_embeddings (line 970) | def get_input_embeddings(self):
    method set_input_embeddings (line 973) | def set_input_embeddings(self, value):

FILE: transformers/models/biogpt/tokenization_biogpt.py
  function get_pairs (line 43) | def get_pairs(word):
  class BioGptTokenizer (line 56) | class BioGptTokenizer(PreTrainedTokenizer):
    method __init__ (line 104) | def __init__(
    method vocab_size (line 149) | def vocab_size(self):
    method get_vocab (line 153) | def get_vocab(self):
    method moses_tokenize (line 156) | def moses_tokenize(self, text, lang):
    method moses_detokenize (line 164) | def moses_detokenize(self, tokens, lang):
    method bpe (line 170) | def bpe(self, token):
    method _tokenize (line 214) | def _tokenize(self, text, bypass_tokenizer=False):
    method _convert_token_to_id (line 228) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 232) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 236) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 245) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 269) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 296) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 326) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method __getstate__ (line 354) | def __getstate__(self):
    method __setstate__ (line 359) | def __setstate__(self, d):

FILE: transformers/models/bit/configuration_bit.py
  class BitConfig (line 29) | class BitConfig(BackboneConfigMixin, PretrainedConfig):
    method __init__ (line 92) | def __init__(

FILE: transformers/models/bit/convert_bit_to_pytorch.py
  function get_config (line 39) | def get_config(model_name):
  function rename_key (line 61) | def rename_key(name):
  function prepare_img (line 77) | def prepare_img():
  function convert_bit_checkpoint (line 84) | def convert_bit_checkpoint(model_name, pytorch_dump_folder_path, push_to...

FILE: transformers/models/bit/image_processing_bit.py
  class BitImageProcessor (line 51) | class BitImageProcessor(BaseImageProcessor):
    method __init__ (line 91) | def __init__(
    method resize (line 124) | def resize(
    method center_crop (line 152) | def center_crop(
    method rescale (line 176) | def rescale(
    method normalize (line 196) | def normalize(
    method preprocess (line 219) | def preprocess(

FILE: transformers/models/bit/modeling_bit.py
  function get_padding_value (line 65) | def get_padding_value(padding=None, kernel_size=7, stride=1, dilation=1)...
  class WeightStandardizedConv2d (line 106) | class WeightStandardizedConv2d(nn.Conv2d):
    method __init__ (line 113) | def __init__(
    method forward (line 142) | def forward(self, hidden_state):
  class BitGroupNormActivation (line 154) | class BitGroupNormActivation(nn.GroupNorm):
    method __init__ (line 159) | def __init__(self, config, num_channels, eps=1e-5, affine=True, apply_...
    method forward (line 166) | def forward(self, hidden_state):
  class DynamicPad2d (line 172) | class DynamicPad2d(nn.Module):
    method __init__ (line 178) | def __init__(self, kernel_size, stride, dilation, value=0):
    method __call__ (line 200) | def __call__(self, input):
  class BitMaxPool2d (line 223) | class BitMaxPool2d(nn.MaxPool2d):
    method __init__ (line 226) | def __init__(
    method forward (line 245) | def forward(self, hidden_states):
  class BitEmbeddings (line 252) | class BitEmbeddings(nn.Module):
    method __init__ (line 257) | def __init__(self, config: BitConfig):
    method forward (line 284) | def forward(self, pixel_values: Tensor) -> Tensor:
  function drop_path (line 303) | def drop_path(input, drop_prob: float = 0.0, training: bool = False):
  class BitDropPath (line 324) | class BitDropPath(nn.Module):
    method __init__ (line 327) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 331) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 334) | def extra_repr(self) -> str:
  function make_div (line 338) | def make_div(value, divisor=8):
  class BitPreActivationBottleneckLayer (line 346) | class BitPreActivationBottleneckLayer(nn.Module):
    method __init__ (line 354) | def __init__(
    method forward (line 398) | def forward(self, hidden_states):
  class BitBottleneckLayer (line 414) | class BitBottleneckLayer(nn.Module):
    method __init__ (line 417) | def __init__(
    method forward (line 466) | def forward(self, hidden_states):
  class BitDownsampleConv (line 487) | class BitDownsampleConv(nn.Module):
    method __init__ (line 488) | def __init__(
    method forward (line 506) | def forward(self, x):
  class BitStage (line 510) | class BitStage(nn.Module):
    method __init__ (line 515) | def __init__(
    method _get_updated_hyperparameters (line 561) | def _get_updated_hyperparameters(self, layer_idx, stride, layer_dropout):
    method forward (line 577) | def forward(self, input: Tensor) -> Tensor:
  class BitEncoder (line 584) | class BitEncoder(nn.Module):
    method __init__ (line 585) | def __init__(self, config: BitConfig):
    method _get_updated_hyperparameters (line 623) | def _get_updated_hyperparameters(self, stage_idx, current_stride, curr...
    method forward (line 631) | def forward(
  class BitPreTrainedModel (line 654) | class BitPreTrainedModel(PreTrainedModel):
    method _init_weights (line 665) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 672) | def _set_gradient_checkpointing(self, module, value=False):
  class BitModel (line 706) | class BitModel(BitPreTrainedModel):
    method __init__ (line 707) | def __init__(self, config):
    method forward (line 732) | def forward(
  class BitForImageClassification (line 769) | class BitForImageClassification(BitPreTrainedModel):
    method __init__ (line 770) | def __init__(self, config):
    method forward (line 789) | def forward(
  class BitBackbone (line 845) | class BitBackbone(BitPreTrainedModel, BackboneMixin):
    method __init__ (line 846) | def __init__(self, config):
    method forward (line 858) | def forward(

FILE: transformers/models/blenderbot/configuration_blenderbot.py
  class BlenderbotConfig (line 36) | class BlenderbotConfig(PretrainedConfig):
    method __init__ (line 111) | def __init__(
  class BlenderbotOnnxConfig (line 172) | class BlenderbotOnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 174) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 216) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method _generate_dummy_inputs_for_default_and_seq2seq_lm (line 228) | def _generate_dummy_inputs_for_default_and_seq2seq_lm(
    method _generate_dummy_inputs_for_causal_lm (line 285) | def _generate_dummy_inputs_for_causal_lm(
    method _generate_dummy_inputs_for_sequence_classification_and_question_answering (line 322) | def _generate_dummy_inputs_for_sequence_classification_and_question_an...
    method generate_dummy_inputs (line 349) | def generate_dummy_inputs(
    method _flatten_past_key_values_ (line 374) | def _flatten_past_key_values_(self, flattened_output, name, idx, t):
    method fill_with_past_key_values_ (line 382) | def fill_with_past_key_values_(self, inputs_or_outputs: Mapping[str, M...

FILE: transformers/models/blenderbot/convert_blenderbot_original_pytorch_checkpoint_to_pytorch.py
  function rename_state_dict_key (line 42) | def rename_state_dict_key(k):
  function rename_layernorm_keys (line 60) | def rename_layernorm_keys(sd):
  function convert_parlai_checkpoint (line 78) | def convert_parlai_checkpoint(checkpoint_path, pytorch_dump_folder_path,...

FILE: transformers/models/blenderbot/modeling_blenderbot.py
  function shift_tokens_right (line 63) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 80) | def _make_causal_mask(
  function _expand_mask (line 98) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class BlenderbotLearnedPositionalEmbedding (line 112) | class BlenderbotLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 117) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 120) | def forward(self, input_ids_shape: torch.Size, past_key_values_length:...
  class BlenderbotAttention (line 130) | class BlenderbotAttention(nn.Module):
    method __init__ (line 133) | def __init__(
    method _shape (line 160) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 163) | def forward(
  class BlenderbotEncoderLayer (line 285) | class BlenderbotEncoderLayer(nn.Module):
    method __init__ (line 286) | def __init__(self, config: BlenderbotConfig):
    method forward (line 302) | def forward(
  class BlenderbotDecoderLayer (line 354) | class BlenderbotDecoderLayer(nn.Module):
    method __init__ (line 355) | def __init__(self, config: BlenderbotConfig):
    method forward (line 381) | def forward(
  class BlenderbotPreTrainedModel (line 471) | class BlenderbotPreTrainedModel(PreTrainedModel):
    method _init_weights (line 476) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 487) | def _set_gradient_checkpointing(self, module, value=False):
    method dummy_inputs (line 492) | def dummy_inputs(self):
  class BlenderbotEncoder (line 643) | class BlenderbotEncoder(BlenderbotPreTrainedModel):
    method __init__ (line 653) | def __init__(self, config: BlenderbotConfig, embed_tokens: Optional[nn...
    method forward (line 680) | def forward(
  class BlenderbotDecoder (line 814) | class BlenderbotDecoder(BlenderbotPreTrainedModel):
    method __init__ (line 823) | def __init__(self, config: BlenderbotConfig, embed_tokens: Optional[nn...
    method get_input_embeddings (line 847) | def get_input_embeddings(self):
    method set_input_embeddings (line 850) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 854) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 877) | def forward(
  class BlenderbotModel (line 1099) | class BlenderbotModel(BlenderbotPreTrainedModel):
    method __init__ (line 1102) | def __init__(self, config: BlenderbotConfig):
    method from_pretrained (line 1115) | def from_pretrained(cls, pretrained_model_name_or_path: Optional[Union...
    method get_input_embeddings (line 1127) | def get_input_embeddings(self):
    method set_input_embeddings (line 1130) | def set_input_embeddings(self, value):
    method get_encoder (line 1135) | def get_encoder(self):
    method get_decoder (line 1138) | def get_decoder(self):
    method forward (line 1143) | def forward(
  class BlenderbotForConditionalGeneration (line 1239) | class BlenderbotForConditionalGeneration(BlenderbotPreTrainedModel):
    method __init__ (line 1250) | def __init__(self, config: BlenderbotConfig):
    method from_pretrained (line 1260) | def from_pretrained(cls, pretrained_model_name_or_path: Optional[Union...
    method get_encoder (line 1274) | def get_encoder(self):
    method get_decoder (line 1277) | def get_decoder(self):
    method resize_token_embeddings (line 1280) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_final_logits_bias (line 1285) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_output_embeddings (line 1294) | def get_output_embeddings(self):
    method set_output_embeddings (line 1297) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1303) | def forward(
    method prepare_inputs_for_generation (line 1381) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1410) | def _reorder_cache(past_key_values, beam_idx):
  class BlenderbotDecoderWrapper (line 1421) | class BlenderbotDecoderWrapper(BlenderbotPreTrainedModel):
    method __init__ (line 1427) | def __init__(self, config):
    method forward (line 1431) | def forward(self, *args, **kwargs):
  class BlenderbotForCausalLM (line 1436) | class BlenderbotForCausalLM(BlenderbotPreTrainedModel):
    method __init__ (line 1439) | def __init__(self, config):
    method get_input_embeddings (line 1451) | def get_input_embeddings(self):
    method set_input_embeddings (line 1454) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1457) | def get_output_embeddings(self):
    method set_output_embeddings (line 1460) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 1463) | def set_decoder(self, decoder):
    method get_decoder (line 1466) | def get_decoder(self):
    method forward (line 1470) | def forward(
    method prepare_inputs_for_generation (line 1616) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1634) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/blenderbot/modeling_flax_blenderbot.py
  function shift_tokens_right (line 208) | def shift_tokens_right(input_ids: np.array, pad_token_id: int, decoder_s...
  class FlaxBlenderbotAttention (line 221) | class FlaxBlenderbotAttention(nn.Module):
    method setup (line 230) | def setup(self) -> None:
    method _split_heads (line 256) | def _split_heads(self, hidden_states):
    method _merge_heads (line 259) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 263) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 294) | def __call__(
  class FlaxBlenderbotEncoderLayer (line 389) | class FlaxBlenderbotEncoderLayer(nn.Module):
    method setup (line 393) | def setup(self) -> None:
    method __call__ (line 416) | def __call__(
  class FlaxBlenderbotEncoderLayerCollection (line 446) | class FlaxBlenderbotEncoderLayerCollection(nn.Module):
    method setup (line 450) | def setup(self):
    method __call__ (line 457) | def __call__(
  class FlaxBlenderbotDecoderLayer (line 501) | class FlaxBlenderbotDecoderLayer(nn.Module):
    method setup (line 505) | def setup(self) -> None:
    method __call__ (line 538) | def __call__(
  class FlaxBlenderbotDecoderLayerCollection (line 590) | class FlaxBlenderbotDecoderLayerCollection(nn.Module):
    method setup (line 594) | def setup(self):
    method __call__ (line 601) | def __call__(
  class FlaxBlenderbotEncoder (line 660) | class FlaxBlenderbotEncoder(nn.Module):
    method setup (line 665) | def setup(self):
    method __call__ (line 681) | def __call__(
  class FlaxBlenderbotDecoder (line 729) | class FlaxBlenderbotDecoder(nn.Module):
    method setup (line 734) | def setup(self):
    method __call__ (line 751) | def __call__(
  class FlaxBlenderbotModule (line 809) | class FlaxBlenderbotModule(nn.Module):
    method setup (line 813) | def setup(self):
    method _get_encoder_module (line 824) | def _get_encoder_module(self):
    method _get_decoder_module (line 827) | def _get_decoder_module(self):
    method __call__ (line 830) | def __call__(
  class FlaxBlenderbotPreTrainedModel (line 879) | class FlaxBlenderbotPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 884) | def __init__(
    method init_weights (line 896) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 932) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 975) | def encode(
    method decode (line 1040) | def decode(
    method __call__ (line 1153) | def __call__(
  class FlaxBlenderbotModel (line 1217) | class FlaxBlenderbotModel(FlaxBlenderbotPreTrainedModel):
  class FlaxBlenderbotForConditionalGenerationModule (line 1227) | class FlaxBlenderbotForConditionalGenerationModule(nn.Module):
    method setup (line 1232) | def setup(self):
    method _get_encoder_module (line 1242) | def _get_encoder_module(self):
    method _get_decoder_module (line 1245) | def _get_decoder_module(self):
    method __call__ (line 1248) | def __call__(
  class FlaxBlenderbotForConditionalGeneration (line 1302) | class FlaxBlenderbotForConditionalGeneration(FlaxBlenderbotPreTrainedMod...
    method decode (line 1308) | def decode(
    method prepare_inputs_for_generation (line 1443) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 1474) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/blenderbot/modeling_tf_blenderbot.py
  function shift_tokens_right (line 65) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int, decoder_...
  function _make_causal_mask (line 90) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 108) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFBlenderbotLearnedPositionalEmbedding (line 121) | class TFBlenderbotLearnedPositionalEmbedding(tf.keras.layers.Embedding):
    method __init__ (line 126) | def __init__(self, num_embeddings: int, embedding_dim: int, **kwargs):
    method call (line 129) | def call(
  class TFBlenderbotAttention (line 142) | class TFBlenderbotAttention(tf.keras.layers.Layer):
    method __init__ (line 145) | def __init__(
    method _shape (line 173) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 176) | def call(
  class TFBlenderbotEncoderLayer (line 296) | class TFBlenderbotEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 297) | def __init__(self, config: BlenderbotConfig, **kwargs):
    method call (line 311) | def call(
  class TFBlenderbotDecoderLayer (line 353) | class TFBlenderbotDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 354) | def __init__(self, config: BlenderbotConfig, **kwargs):
    method call (line 381) | def call(
  class TFBlenderbotPreTrainedModel (line 462) | class TFBlenderbotPreTrainedModel(TFPreTrainedModel):
  class TFBlenderbotEncoder (line 616) | class TFBlenderbotEncoder(tf.keras.layers.Layer):
    method __init__ (line 626) | def __init__(self, config: BlenderbotConfig, embed_tokens: Optional[tf...
    method get_embed_tokens (line 644) | def get_embed_tokens(self):
    method set_embed_tokens (line 647) | def set_embed_tokens(self, embed_tokens):
    method call (line 651) | def call(
  class TFBlenderbotDecoder (line 781) | class TFBlenderbotDecoder(tf.keras.layers.Layer):
    method __init__ (line 791) | def __init__(self, config: BlenderbotConfig, embed_tokens: Optional[tf...
    method get_embed_tokens (line 808) | def get_embed_tokens(self):
    method set_embed_tokens (line 811) | def set_embed_tokens(self, embed_tokens):
    method call (line 815) | def call(
  class TFBlenderbotMainLayer (line 1011) | class TFBlenderbotMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1014) | def __init__(self, config: BlenderbotConfig, **kwargs):
    method get_input_embeddings (line 1030) | def get_input_embeddings(self):
    method set_input_embeddings (line 1033) | def set_input_embeddings(self, new_embeddings):
    method call (line 1039) | def call(
  class TFBlenderbotModel (line 1122) | class TFBlenderbotModel(TFBlenderbotPreTrainedModel):
    method __init__ (line 1123) | def __init__(self, config: BlenderbotConfig, *inputs, **kwargs):
    method get_encoder (line 1128) | def get_encoder(self):
    method get_decoder (line 1131) | def get_decoder(self):
    method from_pretrained (line 1135) | def from_pretrained(cls, pretrained_model_name_or_path: Optional[Union...
    method call (line 1157) | def call(
    method serving_output (line 1201) | def serving_output(self, output):
  class BiasLayer (line 1222) | class BiasLayer(tf.keras.layers.Layer):
    method __init__ (line 1228) | def __init__(self, shape, initializer, trainable, name, **kwargs):
    method call (line 1235) | def call(self, x):
  class TFBlenderbotForConditionalGeneration (line 1243) | class TFBlenderbotForConditionalGeneration(TFBlenderbotPreTrainedModel, ...
    method __init__ (line 1249) | def __init__(self, config, *inputs, **kwargs):
    method get_decoder (line 1258) | def get_decoder(self):
    method get_encoder (line 1261) | def get_encoder(self):
    method get_output_embeddings (line 1264) | def get_output_embeddings(self):
    method set_output_embeddings (line 1267) | def set_output_embeddings(self, value):
    method get_bias (line 1270) | def get_bias(self):
    method set_bias (line 1273) | def set_bias(self, value):
    method from_pretrained (line 1282) | def from_pretrained(cls, pretrained_model_name_or_path: Optional[Union...
    method call (line 1301) | def call(
    method serving_output (line 1382) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 1402) | def prepare_inputs_for_generation(

FILE: transformers/models/blenderbot/tokenization_blenderbot.py
  function bytes_to_unicode (line 53) | def bytes_to_unicode():
  function get_pairs (line 78) | def get_pairs(word):
  class BlenderbotTokenizer (line 92) | class BlenderbotTokenizer(PreTrainedTokenizer):
    method __init__ (line 177) | def __init__(
    method vocab_size (line 233) | def vocab_size(self):
    method get_vocab (line 237) | def get_vocab(self):
    method bpe (line 241) | def bpe(self, token):
    method _tokenize (line 284) | def _tokenize(self, text):
    method _convert_token_to_id (line 295) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 300) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 305) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 312) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method get_special_tokens_mask (line 342) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 370) | def create_token_type_ids_from_sequences(
    method prepare_for_tokenization (line 394) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method build_inputs_with_special_tokens (line 400) | def build_inputs_with_special_tokens(self, token_ids_0: List[int], tok...
    method _build_conversation_input_ids (line 416) | def _build_conversation_input_ids(self, conversation: "Conversation") ...

FILE: transformers/models/blenderbot/tokenization_blenderbot_fast.py
  class BlenderbotTokenizerFast (line 50) | class BlenderbotTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 138) | def __init__(
    method mask_token (line 208) | def mask_token(self) -> str:
    method mask_token (line 223) | def mask_token(self, value):
    method _batch_encode_plus (line 235) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 245) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method save_vocabulary (line 256) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method create_token_type_ids_from_sequences (line 261) | def create_token_type_ids_from_sequences(
    method build_inputs_with_special_tokens (line 284) | def build_inputs_with_special_tokens(self, token_ids_0: List[int], tok...
    method _build_conversation_input_ids (line 300) | def _build_conversation_input_ids(self, conversation: "Conversation") ...

FILE: transformers/models/blenderbot_small/configuration_blenderbot_small.py
  class BlenderbotSmallConfig (line 36) | class BlenderbotSmallConfig(PretrainedConfig):
    method __init__ (line 111) | def __init__(
  class BlenderbotSmallOnnxConfig (line 171) | class BlenderbotSmallOnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 173) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 217) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method _generate_dummy_inputs_for_default_and_seq2seq_lm (line 229) | def _generate_dummy_inputs_for_default_and_seq2seq_lm(
    method _generate_dummy_inputs_for_causal_lm (line 297) | def _generate_dummy_inputs_for_causal_lm(
    method _generate_dummy_inputs_for_sequence_classification_and_question_answering (line 335) | def _generate_dummy_inputs_for_sequence_classification_and_question_an...
    method generate_dummy_inputs (line 361) | def generate_dummy_inputs(
    method _flatten_past_key_values_ (line 385) | def _flatten_past_key_values_(self, flattened_output, name, idx, t):

FILE: transformers/models/blenderbot_small/modeling_blenderbot_small.py
  function shift_tokens_right (line 59) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 76) | def _make_causal_mask(
  function _expand_mask (line 94) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class BlenderbotSmallLearnedPositionalEmbedding (line 109) | class BlenderbotSmallLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 114) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 117) | def forward(self, input_ids_shape: torch.Size, past_key_values_length:...
  class BlenderbotSmallAttention (line 127) | class BlenderbotSmallAttention(nn.Module):
    method __init__ (line 130) | def __init__(
    method _shape (line 157) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 160) | def forward(
  class BlenderbotSmallEncoderLayer (line 282) | class BlenderbotSmallEncoderLayer(nn.Module):
    method __init__ (line 283) | def __init__(self, config: BlenderbotSmallConfig):
    method forward (line 299) | def forward(
  class BlenderbotSmallDecoderLayer (line 351) | class BlenderbotSmallDecoderLayer(nn.Module):
    method __init__ (line 352) | def __init__(self, config: BlenderbotSmallConfig):
    method forward (line 378) | def forward(
  class BlenderbotSmallPreTrainedModel (line 468) | class BlenderbotSmallPreTrainedModel(PreTrainedModel):
    method _init_weights (line 473) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 484) | def _set_gradient_checkpointing(self, module, value=False):
    method dummy_inputs (line 489) | def dummy_inputs(self):
  class BlenderbotSmallEncoder (line 640) | class BlenderbotSmallEncoder(BlenderbotSmallPreTrainedModel):
    method __init__ (line 650) | def __init__(self, config: BlenderbotSmallConfig, embed_tokens: Option...
    method forward (line 677) | def forward(
  class BlenderbotSmallDecoder (line 809) | class BlenderbotSmallDecoder(BlenderbotSmallPreTrainedModel):
    method __init__ (line 818) | def __init__(self, config: BlenderbotSmallConfig, embed_tokens: Option...
    method get_input_embeddings (line 842) | def get_input_embeddings(self):
    method set_input_embeddings (line 845) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 849) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 872) | def forward(
  class BlenderbotSmallModel (line 1093) | class BlenderbotSmallModel(BlenderbotSmallPreTrainedModel):
    method __init__ (line 1096) | def __init__(self, config: BlenderbotSmallConfig):
    method get_input_embeddings (line 1108) | def get_input_embeddings(self):
    method set_input_embeddings (line 1111) | def set_input_embeddings(self, value):
    method get_encoder (line 1116) | def get_encoder(self):
    method get_decoder (line 1119) | def get_decoder(self):
    method forward (line 1124) | def forward(
  class BlenderbotSmallForConditionalGeneration (line 1221) | class BlenderbotSmallForConditionalGeneration(BlenderbotSmallPreTrainedM...
    method __init__ (line 1232) | def __init__(self, config: BlenderbotSmallConfig):
    method get_encoder (line 1241) | def get_encoder(self):
    method get_decoder (line 1244) | def get_decoder(self):
    method resize_token_embeddings (line 1247) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_final_logits_bias (line 1252) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_output_embeddings (line 1261) | def get_output_embeddings(self):
    method set_output_embeddings (line 1264) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1270) | def forward(
    method prepare_inputs_for_generation (line 1348) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1377) | def _reorder_cache(past_key_values, beam_idx):
  class BlenderbotSmallDecoderWrapper (line 1388) | class BlenderbotSmallDecoderWrapper(BlenderbotSmallPreTrainedModel):
    method __init__ (line 1394) | def __init__(self, config):
    method forward (line 1398) | def forward(self, *args, **kwargs):
  class BlenderbotSmallForCausalLM (line 1403) | class BlenderbotSmallForCausalLM(BlenderbotSmallPreTrainedModel):
    method __init__ (line 1406) | def __init__(self, config):
    method get_input_embeddings (line 1418) | def get_input_embeddings(self):
    method set_input_embeddings (line 1421) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1424) | def get_output_embeddings(self):
    method set_output_embeddings (line 1427) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 1430) | def set_decoder(self, decoder):
    method get_decoder (line 1433) | def get_decoder(self):
    method forward (line 1437) | def forward(
    method prepare_inputs_for_generation (line 1583) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1601) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/blenderbot_small/modeling_flax_blenderbot_small.py
  function shift_tokens_right (line 219) | def shift_tokens_right(input_ids: jnp.ndarray, pad_token_id: int, decode...
  class FlaxBlenderbotSmallAttention (line 232) | class FlaxBlenderbotSmallAttention(nn.Module):
    method setup (line 241) | def setup(self) -> None:
    method _split_heads (line 267) | def _split_heads(self, hidden_states):
    method _merge_heads (line 270) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 274) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 305) | def __call__(
  class FlaxBlenderbotSmallEncoderLayer (line 400) | class FlaxBlenderbotSmallEncoderLayer(nn.Module):
    method setup (line 404) | def setup(self) -> None:
    method __call__ (line 427) | def __call__(
  class FlaxBlenderbotSmallEncoderLayerCollection (line 458) | class FlaxBlenderbotSmallEncoderLayerCollection(nn.Module):
    method setup (line 462) | def setup(self):
    method __call__ (line 469) | def __call__(
  class FlaxBlenderbotSmallDecoderLayer (line 513) | class FlaxBlenderbotSmallDecoderLayer(nn.Module):
    method setup (line 517) | def setup(self) -> None:
    method __call__ (line 550) | def __call__(
  class FlaxBlenderbotSmallDecoderLayerCollection (line 602) | class FlaxBlenderbotSmallDecoderLayerCollection(nn.Module):
    method setup (line 606) | def setup(self):
    method __call__ (line 613) | def __call__(
  class FlaxBlenderbotSmallEncoder (line 672) | class FlaxBlenderbotSmallEncoder(nn.Module):
    method setup (line 677) | def setup(self):
    method __call__ (line 693) | def __call__(
  class FlaxBlenderbotSmallDecoder (line 733) | class FlaxBlenderbotSmallDecoder(nn.Module):
    method setup (line 738) | def setup(self):
    method __call__ (line 755) | def __call__(
  class FlaxBlenderbotSmallModule (line 806) | class FlaxBlenderbotSmallModule(nn.Module):
    method setup (line 810) | def setup(self):
    method _get_encoder_module (line 821) | def _get_encoder_module(self):
    method _get_decoder_module (line 824) | def _get_decoder_module(self):
    method __call__ (line 827) | def __call__(
  class FlaxBlenderbotSmallPreTrainedModel (line 876) | class FlaxBlenderbotSmallPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 881) | def __init__(
    method init_weights (line 893) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 929) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 972) | def encode(
    method decode (line 1037) | def decode(
    method __call__ (line 1149) | def __call__(
  class FlaxBlenderbotSmallModel (line 1213) | class FlaxBlenderbotSmallModel(FlaxBlenderbotSmallPreTrainedModel):
  class FlaxBlenderbotSmallForConditionalGenerationModule (line 1223) | class FlaxBlenderbotSmallForConditionalGenerationModule(nn.Module):
    method setup (line 1228) | def setup(self):
    method _get_encoder_module (line 1238) | def _get_encoder_module(self):
    method _get_decoder_module (line 1241) | def _get_decoder_module(self):
    method __call__ (line 1244) | def __call__(
  class FlaxBlenderbotSmallForConditionalGeneration (line 1299) | class FlaxBlenderbotSmallForConditionalGeneration(FlaxBlenderbotSmallPre...
    method decode (line 1305) | def decode(
    method prepare_inputs_for_generation (line 1440) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 1471) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/blenderbot_small/modeling_tf_blenderbot_small.py
  function shift_tokens_right (line 64) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int, decoder_...
  function _make_causal_mask (line 89) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 107) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFBlenderbotSmallLearnedPositionalEmbedding (line 121) | class TFBlenderbotSmallLearnedPositionalEmbedding(tf.keras.layers.Embedd...
    method __init__ (line 126) | def __init__(self, num_embeddings: int, embedding_dim: int, **kwargs):
    method call (line 129) | def call(
  class TFBlenderbotSmallAttention (line 142) | class TFBlenderbotSmallAttention(tf.keras.layers.Layer):
    method __init__ (line 145) | def __init__(
    method _shape (line 173) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 176) | def call(
  class TFBlenderbotSmallEncoderLayer (line 296) | class TFBlenderbotSmallEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 297) | def __init__(self, config: BlenderbotSmallConfig, **kwargs):
    method call (line 311) | def call(
  class TFBlenderbotSmallDecoderLayer (line 353) | class TFBlenderbotSmallDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 354) | def __init__(self, config: BlenderbotSmallConfig, **kwargs):
    method call (line 381) | def call(
  class TFBlenderbotSmallPreTrainedModel (line 462) | class TFBlenderbotSmallPreTrainedModel(TFPreTrainedModel):
  class TFBlenderbotSmallEncoder (line 620) | class TFBlenderbotSmallEncoder(tf.keras.layers.Layer):
    method __init__ (line 630) | def __init__(
    method get_embed_tokens (line 650) | def get_embed_tokens(self):
    method set_embed_tokens (line 653) | def set_embed_tokens(self, embed_tokens):
    method call (line 657) | def call(
  class TFBlenderbotSmallDecoder (line 786) | class TFBlenderbotSmallDecoder(tf.keras.layers.Layer):
    method __init__ (line 796) | def __init__(
    method get_embed_tokens (line 815) | def get_embed_tokens(self):
    method set_embed_tokens (line 818) | def set_embed_tokens(self, embed_tokens):
    method call (line 822) | def call(
  class TFBlenderbotSmallMainLayer (line 1019) | class TFBlenderbotSmallMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1022) | def __init__(self, config: BlenderbotSmallConfig, **kwargs):
    method get_input_embeddings (line 1038) | def get_input_embeddings(self):
    method set_input_embeddings (line 1041) | def set_input_embeddings(self, new_embeddings):
    method call (line 1047) | def call(
  class TFBlenderbotSmallModel (line 1130) | class TFBlenderbotSmallModel(TFBlenderbotSmallPreTrainedModel):
    method __init__ (line 1131) | def __init__(self, config: BlenderbotSmallConfig, *inputs, **kwargs):
    method get_encoder (line 1136) | def get_encoder(self):
    method get_decoder (line 1139) | def get_decoder(self):
    method call (line 1149) | def call(
    method serving_output (line 1193) | def serving_output(self, output):
  class BiasLayer (line 1214) | class BiasLayer(tf.keras.layers.Layer):
    method __init__ (line 1220) | def __init__(self, shape, initializer, trainable, name, **kwargs):
    method call (line 1227) | def call(self, x):
  class TFBlenderbotSmallForConditionalGeneration (line 1235) | class TFBlenderbotSmallForConditionalGeneration(TFBlenderbotSmallPreTrai...
    method __init__ (line 1241) | def __init__(self, config, *inputs, **kwargs):
    method get_decoder (line 1250) | def get_decoder(self):
    method get_encoder (line 1253) | def get_encoder(self):
    method get_output_embeddings (line 1256) | def get_output_embeddings(self):
    method set_output_embeddings (line 1259) | def set_output_embeddings(self, value):
    method get_bias (line 1262) | def get_bias(self):
    method set_bias (line 1265) | def set_bias(self, value):
    method call (line 1277) | def call(
    method serving_output (line 1359) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 1379) | def prepare_inputs_for_generation(

FILE: transformers/models/blenderbot_small/tokenization_blenderbot_small.py
  function get_pairs (line 53) | def get_pairs(word):
  class BlenderbotSmallTokenizer (line 69) | class BlenderbotSmallTokenizer(PreTrainedTokenizer):
    method __init__ (line 99) | def __init__(
    method vocab_size (line 121) | def vocab_size(self) -> int:
    method get_vocab (line 124) | def get_vocab(self) -> Dict:
    method bpe (line 127) | def bpe(self, token: str) -> str:
    method _tokenize (line 187) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 197) | def _convert_token_to_id(self, token: str) -> int:
    method _convert_id_to_token (line 202) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 206) | def convert_tokens_to_string(self, tokens: List[str]) -> str:
    method save_vocabulary (line 211) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/blenderbot_small/tokenization_blenderbot_small_fast.py
  class BlenderbotSmallTokenizerFast (line 52) | class BlenderbotSmallTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 66) | def __init__(
    method build_inputs_with_special_tokens (line 91) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 98) | def create_token_type_ids_from_sequences(

FILE: transformers/models/blip/configuration_blip.py
  class BlipTextConfig (line 47) | class BlipTextConfig(PretrainedConfig):
    method __init__ (line 115) | def __init__(
    method from_pretrained (line 163) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class BlipVisionConfig (line 179) | class BlipVisionConfig(PretrainedConfig):
    method __init__ (line 230) | def __init__(
    method from_pretrained (line 262) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class BlipConfig (line 278) | class BlipConfig(PretrainedConfig):
    method __init__ (line 328) | def __init__(
    method from_text_vision_configs (line 359) | def from_text_vision_configs(cls, text_config: BlipTextConfig, vision_...
    method to_dict (line 370) | def to_dict(self):

FILE: transformers/models/blip/convert_blip_original_pytorch_to_hf.py
  function load_demo_image (line 39) | def load_demo_image(image_size, device):
  function rename_key (line 54) | def rename_key(key):
  function convert_blip_checkpoint (line 82) | def convert_blip_checkpoint(pytorch_dump_folder_path, config_path=None):

FILE: transformers/models/blip/image_processing_blip.py
  class BlipImageProcessor (line 43) | class BlipImageProcessor(BaseImageProcessor):
    method __init__ (line 80) | def __init__(
    method resize (line 107) | def resize(
    method rescale (line 140) | def rescale(
    method normalize (line 160) | def normalize(
    method preprocess (line 183) | def preprocess(

FILE: transformers/models/blip/modeling_blip.py
  function contrastive_loss (line 57) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  function blip_loss (line 62) | def blip_loss(similarity: torch.Tensor) -> torch.Tensor:
  class BlipForConditionalGenerationModelOutput (line 69) | class BlipForConditionalGenerationModelOutput(ModelOutput):
  class BlipTextVisionModelOutput (line 105) | class BlipTextVisionModelOutput(ModelOutput):
  class BlipImageTextMatchingModelOutput (line 138) | class BlipImageTextMatchingModelOutput(ModelOutput):
  class BlipOutput (line 181) | class BlipOutput(ModelOutput):
    method to_tuple (line 210) | def to_tuple(self) -> Tuple[Any]:
  class BlipVisionEmbeddings (line 217) | class BlipVisionEmbeddings(nn.Module):
    method __init__ (line 218) | def __init__(self, config: BlipVisionConfig):
    method forward (line 238) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class BlipTextEmbeddings (line 251) | class BlipTextEmbeddings(nn.Module):
    method __init__ (line 252) | def __init__(self, config: BlipTextConfig):
    method forward (line 262) | def forward(
  class BlipAttention (line 282) | class BlipAttention(nn.Module):
    method __init__ (line 285) | def __init__(self, config):
    method _shape (line 303) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 306) | def forward(
  class BlipMLP (line 352) | class BlipMLP(nn.Module):
    method __init__ (line 353) | def __init__(self, config):
    method forward (line 360) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BlipEncoderLayer (line 367) | class BlipEncoderLayer(nn.Module):
    method __init__ (line 368) | def __init__(self, config: BlipConfig):
    method forward (line 376) | def forward(
  class BlipPreTrainedModel (line 415) | class BlipPreTrainedModel(PreTrainedModel):
    method _init_weights (line 426) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 455) | def _set_gradient_checkpointing(self, module, value=False):
  class BlipEncoder (line 558) | class BlipEncoder(nn.Module):
    method __init__ (line 568) | def __init__(self, config: BlipConfig):
    method forward (line 574) | def forward(
  class BlipVisionModel (line 650) | class BlipVisionModel(BlipPreTrainedModel):
    method __init__ (line 654) | def __init__(self, config: BlipVisionConfig):
    method forward (line 667) | def forward(
    method get_input_embeddings (line 712) | def get_input_embeddings(self):
  class BlipModel (line 717) | class BlipModel(BlipPreTrainedModel):
    method __init__ (line 720) | def __init__(self, config: BlipConfig):
    method get_text_features (line 753) | def get_text_features(
    method get_image_features (line 791) | def get_image_features(
    method forward (line 829) | def forward(
  class BlipForConditionalGeneration (line 930) | class BlipForConditionalGeneration(BlipPreTrainedModel):
    method __init__ (line 935) | def __init__(self, config: BlipConfig):
    method get_input_embeddings (line 948) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 953) | def forward(
    method generate (line 1023) | def generate(
  class BlipForQuestionAnswering (line 1102) | class BlipForQuestionAnswering(BlipPreTrainedModel):
    method __init__ (line 1106) | def __init__(self, config: BlipConfig):
    method get_input_embeddings (line 1121) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 1126) | def forward(
    method generate (line 1237) | def generate(
  class BlipForImageTextRetrieval (line 1324) | class BlipForImageTextRetrieval(BlipPreTrainedModel):
    method __init__ (line 1327) | def __init__(self, config: BlipConfig):
    method get_input_embeddings (line 1357) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 1362) | def forward(

FILE: transformers/models/blip/modeling_blip_text.py
  class BlipTextEmbeddings (line 45) | class BlipTextEmbeddings(nn.Module):
    method __init__ (line 48) | def __init__(self, config):
    method forward (line 64) | def forward(
  class BlipTextSelfAttention (line 96) | class BlipTextSelfAttention(nn.Module):
    method __init__ (line 97) | def __init__(self, config, is_cross_attention):
    method save_attn_gradients (line 124) | def save_attn_gradients(self, attn_gradients):
    method get_attn_gradients (line 127) | def get_attn_gradients(self):
    method save_attention_map (line 130) | def save_attention_map(self, attention_map):
    method get_attention_map (line 133) | def get_attention_map(self):
    method transpose_for_scores (line 136) | def transpose_for_scores(self, x):
    method forward (line 141) | def forward(
  class BlipTextSelfOutput (line 223) | class BlipTextSelfOutput(nn.Module):
    method __init__ (line 224) | def __init__(self, config):
    method forward (line 230) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BlipTextAttention (line 238) | class BlipTextAttention(nn.Module):
    method __init__ (line 239) | def __init__(self, config, is_cross_attention=False):
    method prune_heads (line 245) | def prune_heads(self, heads):
    method forward (line 263) | def forward(
  class BlipTextIntermediate (line 288) | class BlipTextIntermediate(nn.Module):
    method __init__ (line 289) | def __init__(self, config):
    method forward (line 297) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BlipTextOutput (line 304) | class BlipTextOutput(nn.Module):
    method __init__ (line 305) | def __init__(self, config):
    method forward (line 311) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BlipTextLayer (line 318) | class BlipTextLayer(nn.Module):
    method __init__ (line 319) | def __init__(self, config, layer_num):
    method forward (line 331) | def forward(
    method feed_forward_chunk (line 375) | def feed_forward_chunk(self, attention_output):
  class BlipTextEncoder (line 382) | class BlipTextEncoder(nn.Module):
    method __init__ (line 383) | def __init__(self, config):
    method forward (line 389) | def forward(
  class BlipTextPooler (line 481) | class BlipTextPooler(nn.Module):
    method __init__ (line 482) | def __init__(self, config):
    method forward (line 487) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BlipTextPredictionHeadTransform (line 497) | class BlipTextPredictionHeadTransform(nn.Module):
    method __init__ (line 498) | def __init__(self, config):
    method forward (line 507) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BlipTextLMPredictionHead (line 515) | class BlipTextLMPredictionHead(nn.Module):
    method __init__ (line 516) | def __init__(self, config):
    method forward (line 529) | def forward(self, hidden_states):
  class BlipTextOnlyMLMHead (line 536) | class BlipTextOnlyMLMHead(nn.Module):
    method __init__ (line 537) | def __init__(self, config):
    method forward (line 541) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class BlipTextPreTrainedModel (line 547) | class BlipTextPreTrainedModel(PreTrainedModel):
    method _init_weights (line 557) | def _init_weights(self, module):
  class BlipTextModel (line 571) | class BlipTextModel(BlipTextPreTrainedModel):
    method __init__ (line 580) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 590) | def get_input_embeddings(self):
    method set_input_embeddings (line 593) | def set_input_embeddings(self, value):
    method _prune_heads (line 597) | def _prune_heads(self, heads_to_prune):
    method get_extended_attention_mask (line 605) | def get_extended_attention_mask(
    method forward (line 670) | def forward(
  class BlipTextLMHeadModel (line 810) | class BlipTextLMHeadModel(BlipTextPreTrainedModel):
    method __init__ (line 814) | def __init__(self, config):
    method get_output_embeddings (line 820) | def get_output_embeddings(self):
    method set_output_embeddings (line 823) | def set_output_embeddings(self, new_embeddings):
    method forward (line 826) | def forward(
    method prepare_inputs_for_generation (line 916) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 935) | def _reorder_cache(self, past_key_values, beam_idx):

FILE: transformers/models/blip/modeling_tf_blip.py
  function contrastive_loss (line 63) | def contrastive_loss(logits: tf.Tensor) -> tf.Tensor:
  function blip_loss (line 72) | def blip_loss(similarity: tf.Tensor) -> tf.Tensor:
  class TFBlipForConditionalGenerationModelOutput (line 79) | class TFBlipForConditionalGenerationModelOutput(ModelOutput):
  class TFBlipTextVisionModelOutput (line 115) | class TFBlipTextVisionModelOutput(ModelOutput):
  class TFBlipImageTextMatchingModelOutput (line 148) | class TFBlipImageTextMatchingModelOutput(ModelOutput):
  class TFBlipOutput (line 191) | class TFBlipOutput(ModelOutput):
    method to_tuple (line 220) | def to_tuple(self) -> Tuple[Any]:
  class TFBlipVisionEmbeddings (line 227) | class TFBlipVisionEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 228) | def __init__(self, config: BlipVisionConfig, **kwargs):
    method build (line 247) | def build(self, input_shape):
    method call (line 263) | def call(self, pixel_values: tf.Tensor) -> tf.Tensor:
  class TFBlipTextEmbeddings (line 278) | class TFBlipTextEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 279) | def __init__(self, config: BlipTextConfig, **kwargs):
    method build (line 286) | def build(self, input_shape: tf.TensorShape = None):
    method call (line 305) | def call(
  class TFBlipAttention (line 336) | class TFBlipAttention(tf.keras.layers.Layer):
    method __init__ (line 339) | def __init__(self, config, **kwargs):
    method call (line 361) | def call(
  class TFBlipMLP (line 406) | class TFBlipMLP(tf.keras.layers.Layer):
    method __init__ (line 407) | def __init__(self, config: BlipConfig, **kwargs):
    method call (line 422) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFBlipEncoderLayer (line 429) | class TFBlipEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 430) | def __init__(self, config: BlipConfig, **kwargs):
    method call (line 438) | def call(
  class TFBlipPreTrainedModel (line 479) | class TFBlipPreTrainedModel(TFPreTrainedModel):
  class TFBlipEncoder (line 558) | class TFBlipEncoder(tf.keras.layers.Layer):
    method __init__ (line 569) | def __init__(self, config: BlipConfig, **kwargs):
    method call (line 575) | def call(
  class TFBlipVisionModel (line 639) | class TFBlipVisionModel(TFBlipPreTrainedModel):
    method __init__ (line 643) | def __init__(self, config: BlipVisionConfig, *args, **kwargs):
    method serving_output (line 651) | def serving_output(self, output: TFBaseModelOutputWithPooling) -> TFBa...
    method call (line 665) | def call(
    method get_input_embeddings (line 714) | def get_input_embeddings(self):
  class TFBlipMainLayer (line 718) | class TFBlipMainLayer(tf.keras.layers.Layer):
    method __init__ (line 721) | def __init__(self, config: BlipConfig, *args, **kwargs):
    method build (line 761) | def build(self, input_shape=None):
    method call (line 771) | def call(
  class TFBlipModel (line 843) | class TFBlipModel(TFBlipPreTrainedModel):
    method __init__ (line 848) | def __init__(self, config: BlipConfig, *inputs, **kwargs):
    method serving_output (line 853) | def serving_output(self, output: TFBlipOutput) -> TFBlipOutput:
    method call (line 864) | def call(
    method get_text_features (line 914) | def get_text_features(
    method get_image_features (line 952) | def get_image_features(
  class TFBlipForConditionalGeneration (line 998) | class TFBlipForConditionalGeneration(TFBlipPreTrainedModel):
    method __init__ (line 1003) | def __init__(self, config: BlipConfig, *args, **kwargs):
    method get_input_embeddings (line 1013) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method call (line 1019) | def call(
    method generate (line 1088) | def generate(
  class TFBlipForQuestionAnswering (line 1170) | class TFBlipForQuestionAnswering(TFBlipPreTrainedModel):
    method __init__ (line 1174) | def __init__(self, config: BlipConfig, *args, **kwargs):
    method get_input_embeddings (line 1186) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method _shift_right (line 1190) | def _shift_right(self, input_ids):
    method call (line 1216) | def call(
    method generate (line 1325) | def generate(
  class TFBlipForImageTextRetrieval (line 1412) | class TFBlipForImageTextRetrieval(TFBlipPreTrainedModel):
    method __init__ (line 1415) | def __init__(self, config: BlipConfig, *args, **kwargs):
    method get_input_embeddings (line 1452) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method call (line 1458) | def call(

FILE: transformers/models/blip/modeling_tf_blip_text.py
  class TFBlipTextEmbeddings (line 77) | class TFBlipTextEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 80) | def __init__(self, config, **kwargs):
    method call (line 105) | def call(self, input_ids=None, position_ids=None, inputs_embeds=None, ...
  class TFBlipTextSelfAttention (line 131) | class TFBlipTextSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 132) | def __init__(self, config, is_cross_attention, **kwargs):
    method transpose_for_scores (line 163) | def transpose_for_scores(self, x):
    method call (line 171) | def call(
  class TFBlipTextSelfOutput (line 253) | class TFBlipTextSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 254) | def __init__(self, config: BlipTextConfig, **kwargs):
    method call (line 263) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFBlipTextAttention (line 272) | class TFBlipTextAttention(tf.keras.layers.Layer):
    method __init__ (line 273) | def __init__(self, config, is_cross_attention=False, **kwargs):
    method call (line 279) | def call(
  class TFBlipTextIntermediate (line 306) | class TFBlipTextIntermediate(tf.keras.layers.Layer):
    method __init__ (line 307) | def __init__(self, config: BlipTextConfig, **kwargs):
    method call (line 319) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFBlipTextOutput (line 326) | class TFBlipTextOutput(tf.keras.layers.Layer):
    method __init__ (line 327) | def __init__(self, config: BlipTextConfig, **kwargs):
    method call (line 336) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFBlipTextLayer (line 344) | class TFBlipTextLayer(tf.keras.layers.Layer):
    method __init__ (line 345) | def __init__(self, config, **kwargs):
    method call (line 356) | def call(
  class TFBlipTextEncoder (line 405) | class TFBlipTextEncoder(tf.keras.layers.Layer):
    method __init__ (line 408) | def __init__(self, config, name=None, **kwargs):
    method call (line 414) | def call(
  class TFBlipTextPooler (line 485) | class TFBlipTextPooler(tf.keras.layers.Layer):
    method __init__ (line 486) | def __init__(self, config: BlipTextConfig, **kwargs):
    method call (line 496) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFBlipTextPredictionHeadTransform (line 506) | class TFBlipTextPredictionHeadTransform(tf.keras.layers.Layer):
    method __init__ (line 507) | def __init__(self, config: BlipTextConfig, **kwargs):
    method call (line 523) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFBlipTextLMPredictionHead (line 531) | class TFBlipTextLMPredictionHead(tf.keras.layers.Layer):
    method __init__ (line 532) | def __init__(self, config, **kwargs):
    method build (line 546) | def build(self, input_shape=None):
    method call (line 550) | def call(self, hidden_states):
  class TFBlipTextOnlyMLMHead (line 556) | class TFBlipTextOnlyMLMHead(tf.keras.layers.Layer):
    method __init__ (line 557) | def __init__(self, config, **kwargs):
    method call (line 561) | def call(self, sequence_output: tf.Tensor) -> tf.Tensor:
  class TFBlipTextPreTrainedModel (line 567) | class TFBlipTextPreTrainedModel(TFPreTrainedModel):
  class TFBlipTextModel (line 579) | class TFBlipTextModel(TFBlipTextPreTrainedModel):
    method __init__ (line 588) | def __init__(self, config, add_pooling_layer=True, name=None, **kwargs):
    method get_input_embeddings (line 596) | def get_input_embeddings(self):
    method set_input_embeddings (line 599) | def set_input_embeddings(self, value):
    method get_extended_attention_mask (line 603) | def get_extended_attention_mask(
    method call (line 669) | def call(
  class TFBlipTextLMHeadModel (line 806) | class TFBlipTextLMHeadModel(TFBlipTextPreTrainedModel):
    method __init__ (line 810) | def __init__(self, config, **kwargs):
    method get_output_embeddings (line 816) | def get_output_embeddings(self):
    method set_output_embeddings (line 819) | def set_output_embeddings(self, new_embeddings):
    method call (line 824) | def call(
    method prepare_inputs_for_generation (line 920) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 939) | def _reorder_cache(self, past_key_values, beam_idx):

FILE: transformers/models/blip/processing_blip.py
  class BlipProcessor (line 26) | class BlipProcessor(ProcessorMixin):
    method __init__ (line 43) | def __init__(self, image_processor, tokenizer):
    method __call__ (line 48) | def __call__(
    method batch_decode (line 130) | def batch_decode(self, *args, **kwargs):
    method decode (line 137) | def decode(self, *args, **kwargs):
    method model_input_names (line 145) | def model_input_names(self):

FILE: transformers/models/blip_2/configuration_blip_2.py
  class Blip2VisionConfig (line 34) | class Blip2VisionConfig(PretrainedConfig):
    method __init__ (line 90) | def __init__(
    method from_pretrained (line 128) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class Blip2QFormerConfig (line 144) | class Blip2QFormerConfig(PretrainedConfig):
    method __init__ (line 209) | def __init__(
    method from_pretrained (line 248) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class Blip2Config (line 264) | class Blip2Config(PretrainedConfig):
    method __init__ (line 320) | def __init__(self, vision_config=None, qformer_config=None, text_confi...
    method from_vision_qformer_text_configs (line 350) | def from_vision_qformer_text_configs(
    method to_dict (line 372) | def to_dict(self):

FILE: transformers/models/blip_2/convert_blip_2_original_to_pytorch.py
  function load_demo_image (line 44) | def load_demo_image():
  function create_rename_keys (line 52) | def create_rename_keys(config):
  function rename_key (line 85) | def rename_key(dct, old, new):
  function read_in_q_v_bias (line 90) | def read_in_q_v_bias(state_dict, config):
  function get_blip2_config (line 101) | def get_blip2_config(model_name, eos_token_id):
  function convert_blip2_checkpoint (line 122) | def convert_blip2_checkpoint(model_name, pytorch_dump_folder_path=None, ...

FILE: transformers/models/blip_2/modeling_blip_2.py
  class Blip2ForConditionalGenerationModelOutput (line 57) | class Blip2ForConditionalGenerationModelOutput(ModelOutput):
    method to_tuple (line 80) | def to_tuple(self) -> Tuple[Any]:
  class Blip2VisionEmbeddings (line 90) | class Blip2VisionEmbeddings(nn.Module):
    method __init__ (line 91) | def __init__(self, config: Blip2VisionConfig):
    method forward (line 111) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class Blip2Attention (line 123) | class Blip2Attention(nn.Module):
    method __init__ (line 126) | def __init__(self, config):
    method _shape (line 156) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 159) | def forward(
  class Blip2MLP (line 209) | class Blip2MLP(nn.Module):
    method __init__ (line 210) | def __init__(self, config):
    method forward (line 217) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Blip2EncoderLayer (line 225) | class Blip2EncoderLayer(nn.Module):
    method __init__ (line 226) | def __init__(self, config: Blip2Config):
    method forward (line 234) | def forward(
  class Blip2PreTrainedModel (line 273) | class Blip2PreTrainedModel(PreTrainedModel):
    method _init_weights (line 292) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 312) | def _set_gradient_checkpointing(self, module, value=False):
  class Blip2Encoder (line 430) | class Blip2Encoder(nn.Module):
    method __init__ (line 440) | def __init__(self, config: Blip2Config):
    method forward (line 446) | def forward(
  class Blip2VisionModel (line 523) | class Blip2VisionModel(Blip2PreTrainedModel):
    method __init__ (line 527) | def __init__(self, config: Blip2VisionConfig):
    method forward (line 540) | def forward(
    method get_input_embeddings (line 585) | def get_input_embeddings(self):
  class Blip2QFormerMultiHeadAttention (line 589) | class Blip2QFormerMultiHeadAttention(nn.Module):
    method __init__ (line 590) | def __init__(self, config, is_cross_attention=False):
    method save_attn_gradients (line 618) | def save_attn_gradients(self, attn_gradients):
    method get_attn_gradients (line 621) | def get_attn_gradients(self):
    method save_attention_map (line 624) | def save_attention_map(self, attention_map):
    method get_attention_map (line 627) | def get_attention_map(self):
    method transpose_for_scores (line 630) | def transpose_for_scores(self, x):
    method forward (line 635) | def forward(
  class Blip2QFormerSelfOutput (line 722) | class Blip2QFormerSelfOutput(nn.Module):
    method __init__ (line 723) | def __init__(self, config):
    method forward (line 729) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class Blip2QFormerAttention (line 736) | class Blip2QFormerAttention(nn.Module):
    method __init__ (line 737) | def __init__(self, config, is_cross_attention=False):
    method prune_heads (line 743) | def prune_heads(self, heads):
    method forward (line 761) | def forward(
  class Blip2QFormerIntermediate (line 786) | class Blip2QFormerIntermediate(nn.Module):
    method __init__ (line 787) | def __init__(self, config):
    method forward (line 795) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Blip2QFormerOutput (line 802) | class Blip2QFormerOutput(nn.Module):
    method __init__ (line 803) | def __init__(self, config):
    method forward (line 809) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class Blip2QFormerLayer (line 816) | class Blip2QFormerLayer(nn.Module):
    method __init__ (line 817) | def __init__(self, config, layer_idx):
    method forward (line 834) | def forward(
    method feed_forward_chunk (line 905) | def feed_forward_chunk(self, attention_output):
    method feed_forward_chunk_query (line 910) | def feed_forward_chunk_query(self, attention_output):
  class Blip2QFormerEncoder (line 916) | class Blip2QFormerEncoder(nn.Module):
    method __init__ (line 917) | def __init__(self, config):
    method forward (line 925) | def forward(
  class Blip2QFormerModel (line 1018) | class Blip2QFormerModel(Blip2PreTrainedModel):
    method __init__ (line 1023) | def __init__(self, config: Blip2QFormerConfig):
    method get_input_embeddings (line 1034) | def get_input_embeddings(self):
    method set_input_embeddings (line 1037) | def set_input_embeddings(self, value):
    method _prune_heads (line 1040) | def _prune_heads(self, heads_to_prune):
    method get_extended_attention_mask (line 1048) | def get_extended_attention_mask(
    method forward (line 1093) | def forward(
  class Blip2Model (line 1218) | class Blip2Model(Blip2PreTrainedModel):
    method __init__ (line 1222) | def __init__(self, config: Blip2Config):
    method get_input_embeddings (line 1240) | def get_input_embeddings(self):
    method set_input_embeddings (line 1243) | def set_input_embeddings(self, value):
    method set_output_embeddings (line 1246) | def set_output_embeddings(self, new_embeddings):
    method get_output_embeddings (line 1249) | def get_output_embeddings(self) -> nn.Module:
    method get_encoder (line 1252) | def get_encoder(self):
    method get_decoder (line 1255) | def get_decoder(self):
    method _tie_weights (line 1258) | def _tie_weights(self):
    method get_text_features (line 1264) | def get_text_features(
    method get_image_features (line 1327) | def get_image_features(
    method get_qformer_features (line 1375) | def get_qformer_features(
    method forward (line 1438) | def forward(
  class Blip2ForConditionalGeneration (line 1573) | class Blip2ForConditionalGeneration(Blip2PreTrainedModel):
    method __init__ (line 1577) | def __init__(self, config: Blip2Config):
    method get_input_embeddings (line 1595) | def get_input_embeddings(self):
    method set_input_embeddings (line 1598) | def set_input_embeddings(self, value):
    method set_output_embeddings (line 1601) | def set_output_embeddings(self, new_embeddings):
    method get_output_embeddings (line 1604) | def get_output_embeddings(self) -> nn.Module:
    method get_encoder (line 1607) | def get_encoder(self):
    method get_decoder (line 1610) | def get_decoder(self):
    method _tie_weights (line 1613) | def _tie_weights(self):
    method _preprocess_accelerate (line 1618) | def _preprocess_accelerate(self):
    method forward (line 1640) | def forward(
    method generate (line 1799) | def generate(

FILE: transformers/models/blip_2/processing_blip_2.py
  class Blip2Processor (line 26) | class Blip2Processor(ProcessorMixin):
    method __init__ (line 44) | def __init__(self, image_processor, tokenizer):
    method __call__ (line 50) | def __call__(
    method batch_decode (line 133) | def batch_decode(self, *args, **kwargs):
    method decode (line 141) | def decode(self, *args, **kwargs):
    method model_input_names (line 150) | def model_input_names(self):

FILE: transformers/models/bloom/configuration_bloom.py
  class BloomConfig (line 42) | class BloomConfig(PretrainedConfig):
    method __init__ (line 113) | def __init__(
  class BloomOnnxConfig (line 152) | class BloomOnnxConfig(OnnxConfigWithPast):
    method __init__ (line 155) | def __init__(
    method inputs (line 168) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method num_layers (line 180) | def num_layers(self) -> int:
    method num_attention_heads (line 184) | def num_attention_heads(self) -> int:
    method atol_for_validation (line 188) | def atol_for_validation(self) -> float:
    method generate_dummy_inputs (line 191) | def generate_dummy_inputs(
    method default_onnx_opset (line 241) | def default_onnx_opset(self) -> int:

FILE: transformers/models/bloom/convert_bloom_original_checkpoint_to_pytorch.py
  function layer_name_mapping (line 51) | def layer_name_mapping(key, file):
  function get_dtype_size (line 71) | def get_dtype_size(dtype):
  function convert_bloom_checkpoint_to_pytorch (line 81) | def convert_bloom_checkpoint_to_pytorch(

FILE: transformers/models/bloom/modeling_bloom.py
  function _make_causal_mask (line 56) | def _make_causal_mask(
  function _expand_mask (line 75) | def _expand_mask(mask: torch.Tensor, tgt_length: int) -> torch.BoolTensor:
  function build_alibi_tensor (line 86) | def build_alibi_tensor(attention_mask: torch.Tensor, num_heads: int, dty...
  function dropout_add (line 130) | def dropout_add(x: torch.Tensor, residual: torch.Tensor, prob: float, tr...
  function bloom_gelu_forward (line 149) | def bloom_gelu_forward(x: torch.Tensor) -> torch.Tensor:
  function bloom_gelu_back (line 161) | def bloom_gelu_back(g: torch.Tensor, x: torch.Tensor) -> torch.Tensor:
  class GeLUFunction (line 179) | class GeLUFunction(torch.autograd.Function):
    method forward (line 181) | def forward(ctx, input: torch.Tensor) -> torch.Tensor:
    method backward (line 186) | def backward(ctx, grad_output: torch.Tensor) -> torch.Tensor:
  class BloomGelu (line 192) | class BloomGelu(nn.Module):
    method __init__ (line 201) | def __init__(self):
    method forward (line 204) | def forward(self, x: torch.Tensor) -> torch.Tensor:
  class BloomAttention (line 211) | class BloomAttention(nn.Module):
    method __init__ (line 212) | def __init__(self, config: BloomConfig):
    method _split_heads (line 238) | def _split_heads(self, fused_qkv: torch.Tensor) -> Tuple[torch.Tensor,...
    method _merge_heads (line 254) | def _merge_heads(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 279) | def forward(
  class BloomMLP (line 371) | class BloomMLP(nn.Module):
    method __init__ (line 372) | def __init__(self, config: BloomConfig):
    method forward (line 383) | def forward(self, hidden_states: torch.Tensor, residual: torch.Tensor)...
  class BloomBlock (line 402) | class BloomBlock(nn.Module):
    method __init__ (line 403) | def __init__(self, config: BloomConfig):
    method forward (line 417) | def forward(
  class BloomPreTrainedModel (line 473) | class BloomPreTrainedModel(PreTrainedModel):
    method __init__ (line 486) | def __init__(self, *inputs, **kwargs):
    method _init_weights (line 489) | def _init_weights(self, module: nn.Module):
    method _set_gradient_checkpointing (line 505) | def _set_gradient_checkpointing(self, module: nn.Module, value: bool =...
    method _convert_to_standard_cache (line 510) | def _convert_to_standard_cache(
    method _convert_to_bloom_cache (line 530) | def _convert_to_bloom_cache(
  class BloomModel (line 623) | class BloomModel(BloomPreTrainedModel):
    method __init__ (line 624) | def __init__(self, config: BloomConfig):
    method build_alibi_tensor (line 645) | def build_alibi_tensor(self, attention_mask: torch.Tensor, num_heads: ...
    method get_input_embeddings (line 648) | def get_input_embeddings(self):
    method _prepare_attn_mask (line 651) | def _prepare_attn_mask(
    method set_input_embeddings (line 673) | def set_input_embeddings(self, new_embeddings: torch.Tensor):
    method forward (line 682) | def forward(
  class BloomForCausalLM (line 828) | class BloomForCausalLM(BloomPreTrainedModel):
    method __init__ (line 831) | def __init__(self, config: BloomConfig):
    method get_output_embeddings (line 839) | def get_output_embeddings(self):
    method set_output_embeddings (line 842) | def set_output_embeddings(self, new_embeddings: torch.Tensor):
    method prepare_inputs_for_generation (line 845) | def prepare_inputs_for_generation(
    method forward (line 882) | def forward(
    method _reorder_cache (line 955) | def _reorder_cache(
  class BloomForSequenceClassification (line 996) | class BloomForSequenceClassification(BloomPreTrainedModel):
    method __init__ (line 999) | def __init__(self, config: BloomConfig):
    method forward (line 1014) | def forward(
  class BloomForTokenClassification (line 1124) | class BloomForTokenClassification(BloomPreTrainedModel):
    method __init__ (line 1127) | def __init__(self, config: BloomConfig):
    method forward (line 1150) | def forward(
  class BloomForQuestionAnswering (line 1227) | class BloomForQuestionAnswering(BloomPreTrainedModel):
    method __init__ (line 1230) | def __init__(self, config):
    method forward (line 1239) | def forward(

FILE: transformers/models/bloom/tokenization_bloom_fast.py
  class BloomTokenizerFast (line 49) | class BloomTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 108) | def __init__(
    method _batch_encode_plus (line 141) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 151) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method save_vocabulary (line 162) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method _build_conversation_input_ids (line 166) | def _build_conversation_input_ids(self, conversation: "Conversation") ...

FILE: transformers/models/bort/convert_bort_original_gluonnlp_checkpoint_to_pytorch.py
  function convert_bort_checkpoint_to_pytorch (line 55) | def convert_bort_checkpoint_to_pytorch(bort_checkpoint_path: str, pytorc...

FILE: transformers/models/bridgetower/configuration_bridgetower.py
  class BridgeTowerVisionConfig (line 35) | class BridgeTowerVisionConfig(PretrainedConfig):
    method __init__ (line 79) | def __init__(
    method from_pretrained (line 106) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class BridgeTowerTextConfig (line 121) | class BridgeTowerTextConfig(PretrainedConfig):
    method __init__ (line 189) | def __init__(
    method from_pretrained (line 235) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class BridgeTowerConfig (line 250) | class BridgeTowerConfig(PretrainedConfig):
    method __init__ (line 305) | def __init__(
    method from_text_vision_configs (line 351) | def from_text_vision_configs(
    method to_dict (line 361) | def to_dict(self):

FILE: transformers/models/bridgetower/image_processing_bridgetower.py
  function max_across_indices (line 46) | def max_across_indices(values: Iterable[Any]) -> List[Any]:
  function make_pixel_mask (line 54) | def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> ...
  function get_max_height_width (line 71) | def get_max_height_width(images: List[np.ndarray]) -> List[int]:
  function get_resize_output_image_size (line 87) | def get_resize_output_image_size(
  class BridgeTowerImageProcessor (line 114) | class BridgeTowerImageProcessor(BaseImageProcessor):
    method __init__ (line 159) | def __init__(
    method resize (line 194) | def resize(
    method rescale (line 231) | def rescale(
    method center_crop (line 251) | def center_crop(
    method normalize (line 274) | def normalize(
    method _pad_image (line 297) | def _pad_image(
    method pad (line 318) | def pad(
    method pad_and_create_pixel_mask (line 356) | def pad_and_create_pixel_mask(
    method preprocess (line 391) | def preprocess(

FILE: transformers/models/bridgetower/modeling_bridgetower.py
  class BridgeTowerModelOutput (line 132) | class BridgeTowerModelOutput(ModelOutput):
  class BridgeTowerContrastiveOutput (line 164) | class BridgeTowerContrastiveOutput(ModelOutput):
  class BridgeTowerResidualAttention (line 197) | class BridgeTowerResidualAttention(nn.Module):
    method __init__ (line 198) | def __init__(self, config):
    method attention (line 215) | def attention(self, hidden_state: torch.Tensor, attention_mask: torch....
    method forward (line 232) | def forward(self, hidden_state: torch.Tensor, attention_mask: torch.Te...
  class BridgeTowerTransformer (line 241) | class BridgeTowerTransformer(nn.Module):
    method __init__ (line 242) | def __init__(self, config):
    method forward (line 256) | def forward(self, hidden_state: torch.Tensor, attention_mask: Optional...
  class BridgeTowerVisionEmbeddings (line 268) | class BridgeTowerVisionEmbeddings(nn.Module):
    method __init__ (line 269) | def __init__(self, config: BridgeTowerVisionConfig):
    method forward (line 291) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class BridgeTowerVisionTransformer (line 302) | class BridgeTowerVisionTransformer(nn.Module):
    method __init__ (line 303) | def __init__(self, config):
    method forward (line 316) | def forward(self, pixel_values: torch.Tensor, attention_mask):
    method forward_pre (line 338) | def forward_pre(self, pixel_values: torch.Tensor):
    method forward_post (line 345) | def forward_post(self, hidden_state: torch.Tensor):
  class BridgeTowerLinkTower (line 351) | class BridgeTowerLinkTower(nn.Module):
    method __init__ (line 352) | def __init__(self, config):
    method forward (line 365) | def forward(self, hidden_states, cross_modal_hidden_states, attention_...
  class BridgeTowerSelfOutput (line 377) | class BridgeTowerSelfOutput(nn.Module):
    method __init__ (line 378) | def __init__(self, config):
    method forward (line 384) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BridgeTowerIntermediate (line 392) | class BridgeTowerIntermediate(nn.Module):
    method __init__ (line 393) | def __init__(self, config):
    method forward (line 401) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BridgeTowerOutput (line 408) | class BridgeTowerOutput(nn.Module):
    method __init__ (line 409) | def __init__(self, config):
    method forward (line 415) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class BridgeTowerPooler (line 423) | class BridgeTowerPooler(nn.Module):
    method __init__ (line 424) | def __init__(self, config):
    method forward (line 429) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class BridgeTowerSelfAttention (line 439) | class BridgeTowerSelfAttention(nn.Module):
    method __init__ (line 440) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 466) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 471) | def forward(
  class BridgeTowerAttention (line 574) | class BridgeTowerAttention(nn.Module):
    method __init__ (line 575) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 581) | def prune_heads(self, heads):
    method forward (line 599) | def forward(
  class BridgeTowerBertCrossLayer (line 623) | class BridgeTowerBertCrossLayer(nn.Module):
    method __init__ (line 624) | def __init__(self, config):
    method forward (line 635) | def forward(
    method feed_forward_chunk (line 679) | def feed_forward_chunk(self, attention_output):
  class BridgeTowerTextLayer (line 685) | class BridgeTowerTextLayer(nn.Module):
    method __init__ (line 686) | def __init__(self, config):
    method forward (line 700) | def forward(
    method feed_forward_chunk (line 765) | def feed_forward_chunk(self, attention_output):
  class BridgeTowerTextEncoder (line 772) | class BridgeTowerTextEncoder(nn.Module):
    method __init__ (line 773) | def __init__(self, config):
    method forward (line 779) | def forward(
  class BridgeTowerTextEmbeddings (line 871) | class BridgeTowerTextEmbeddings(nn.Module):
    method __init__ (line 877) | def __init__(self, config):
    method forward (line 900) | def forward(
    method create_position_ids_from_inputs_embeds (line 940) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  function create_position_ids_from_input_ids (line 959) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...
  class BridgeTowerPreTrainedModel (line 975) | class BridgeTowerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 987) | def _init_weights(self, module):
  class BridgeTowerVisionModel (line 1014) | class BridgeTowerVisionModel(BridgeTowerPreTrainedModel):
    method __init__ (line 1017) | def __init__(self, config):
    method dtype (line 1022) | def dtype(self):
    method forward (line 1025) | def forward(self, image, image_mask=None):
  class BridgeTowerTextModel (line 1029) | class BridgeTowerTextModel(BridgeTowerPreTrainedModel):
    method __init__ (line 1049) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 1061) | def get_input_embeddings(self):
    method set_input_embeddings (line 1064) | def set_input_embeddings(self, value):
    method _prune_heads (line 1067) | def _prune_heads(self, heads_to_prune):
    method forward (line 1076) | def forward(
  class BridgeTowerModel (line 1211) | class BridgeTowerModel(BridgeTowerPreTrainedModel):
    method __init__ (line 1212) | def __init__(self, config):
    method forward (line 1270) | def forward(
    method get_cls_features (line 1494) | def get_cls_features(self, text_features, image_features):
  class BridgeTowerPredictionHeadTransform (line 1501) | class BridgeTowerPredictionHeadTransform(nn.Module):
    method __init__ (line 1502) | def __init__(self, config):
    method forward (line 1511) | def forward(self, hidden_states):
  class BridgeTowerMLMHead (line 1518) | class BridgeTowerMLMHead(nn.Module):
    method __init__ (line 1519) | def __init__(self, config, weight=None):
    method forward (line 1528) | def forward(self, x):
  class BridgeTowerITMHead (line 1534) | class BridgeTowerITMHead(nn.Module):
    method __init__ (line 1535) | def __init__(self, hidden_size):
    method forward (line 1539) | def forward(self, x):
  class BridgeTowerForMaskedLM (line 1550) | class BridgeTowerForMaskedLM(BridgeTowerPreTrainedModel):
    method __init__ (line 1551) | def __init__(self, config):
    method get_output_embeddings (line 1560) | def get_output_embeddings(self):
    method set_output_embeddings (line 1563) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1568) | def forward(
  class BridgeTowerForImageAndTextRetrieval (line 1657) | class BridgeTowerForImageAndTextRetrieval(BridgeTowerPreTrainedModel):
    method __init__ (line 1658) | def __init__(self, config):
    method forward (line 1670) | def forward(
  class BridgeTowerContrastiveHead (line 1752) | class BridgeTowerContrastiveHead(nn.Module):
    method __init__ (line 1753) | def __init__(self, hidden_size, embed_size):
    method forward (line 1757) | def forward(self, x):
  class BridgeTowerForContrastiveLearning (line 1768) | class BridgeTowerForContrastiveLearning(BridgeTowerPreTrainedModel):
    method __init__ (line 1769) | def __init__(self, config):
    method forward (line 1784) | def forward(

FILE: transformers/models/bridgetower/processing_bridgetower.py
  class BridgeTowerProcessor (line 26) | class BridgeTowerProcessor(ProcessorMixin):
    method __init__ (line 45) | def __init__(self, image_processor, tokenizer):
    method __call__ (line 48) | def __call__(
    method batch_decode (line 100) | def batch_decode(self, *args, **kwargs):
    method decode (line 107) | def decode(self, *args, **kwargs):
    method model_input_names (line 115) | def model_input_names(self):

FILE: transformers/models/byt5/convert_byt5_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 27) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, config_file, py...

FILE: transformers/models/byt5/tokenization_byt5.py
  class ByT5Tokenizer (line 28) | class ByT5Tokenizer(PreTrainedTokenizer):
    method __init__ (line 63) | def __init__(
    method vocab_size (line 115) | def vocab_size(self):
    method get_special_tokens_mask (line 118) | def get_special_tokens_mask(
    method _add_eos_if_not_present (line 146) | def _add_eos_if_not_present(self, token_ids: List[int]) -> List[int]:
    method create_token_type_ids_from_sequences (line 157) | def create_token_type_ids_from_sequences(
    method build_inputs_with_special_tokens (line 179) | def build_inputs_with_special_tokens(
    method _tokenize (line 205) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 210) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 222) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 230) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 249) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/camembert/configuration_camembert.py
  class CamembertConfig (line 39) | class CamembertConfig(PretrainedConfig):
    method __init__ (line 109) | def __init__(
  class CamembertOnnxConfig (line 150) | class CamembertOnnxConfig(OnnxConfig):
    method inputs (line 152) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/camembert/modeling_camembert.py
  class CamembertEmbeddings (line 79) | class CamembertEmbeddings(nn.Module):
    method __init__ (line 85) | def __init__(self, config):
    method forward (line 108) | def forward(
    method create_position_ids_from_inputs_embeds (line 148) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class CamembertSelfAttention (line 167) | class CamembertSelfAttention(nn.Module):
    method __init__ (line 168) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 194) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 199) | def forward(
  class CamembertSelfOutput (line 302) | class CamembertSelfOutput(nn.Module):
    method __init__ (line 303) | def __init__(self, config):
    method forward (line 309) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class CamembertAttention (line 317) | class CamembertAttention(nn.Module):
    method __init__ (line 318) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 324) | def prune_heads(self, heads):
    method forward (line 342) | def forward(
  class CamembertIntermediate (line 367) | class CamembertIntermediate(nn.Module):
    method __init__ (line 368) | def __init__(self, config):
    method forward (line 376) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class CamembertOutput (line 383) | class CamembertOutput(nn.Module):
    method __init__ (line 384) | def __init__(self, config):
    method forward (line 390) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class CamembertLayer (line 398) | class CamembertLayer(nn.Module):
    method __init__ (line 399) | def __init__(self, config):
    method forward (line 413) | def forward(
    method feed_forward_chunk (line 478) | def feed_forward_chunk(self, attention_output):
  class CamembertEncoder (line 485) | class CamembertEncoder(nn.Module):
    method __init__ (line 486) | def __init__(self, config):
    method forward (line 492) | def forward(
  class CamembertPooler (line 584) | class CamembertPooler(nn.Module):
    method __init__ (line 585) | def __init__(self, config):
    method forward (line 590) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class CamembertPreTrainedModel (line 599) | class CamembertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 610) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 626) | def _set_gradient_checkpointing(self, module, value=False):
    method update_keys_to_ignore (line 630) | def update_keys_to_ignore(self, config, del_keys_to_ignore):
  class CamembertClassificationHead (line 691) | class CamembertClassificationHead(nn.Module):
    method __init__ (line 694) | def __init__(self, config):
    method forward (line 703) | def forward(self, features, **kwargs):
  class CamembertLMHead (line 714) | class CamembertLMHead(nn.Module):
    method __init__ (line 717) | def __init__(self, config):
    method forward (line 726) | def forward(self, features, **kwargs):
    method _tie_weights (line 736) | def _tie_weights(self):
  class CamembertModel (line 749) | class CamembertModel(CamembertPreTrainedModel):
    method __init__ (line 769) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 781) | def get_input_embeddings(self):
    method set_input_embeddings (line 784) | def set_input_embeddings(self, value):
    method _prune_heads (line 787) | def _prune_heads(self, heads_to_prune):
    method forward (line 802) | def forward(
  class CamembertForMaskedLM (line 937) | class CamembertForMaskedLM(CamembertPreTrainedModel):
    method __init__ (line 942) | def __init__(self, config):
    method get_output_embeddings (line 960) | def get_output_embeddings(self):
    method set_output_embeddings (line 963) | def set_output_embeddings(self, new_embeddings):
    method forward (line 975) | def forward(
  class CamembertForSequenceClassification (line 1043) | class CamembertForSequenceClassification(CamembertPreTrainedModel):
    method __init__ (line 1046) | def __init__(self, config):
    method forward (line 1065) | def forward(
  class CamembertForMultipleChoice (line 1145) | class CamembertForMultipleChoice(CamembertPreTrainedModel):
    method __init__ (line 1148) | def __init__(self, config):
    method forward (line 1166) | def forward(
  class CamembertForTokenClassification (line 1242) | class CamembertForTokenClassification(CamembertPreTrainedModel):
    method __init__ (line 1246) | def __init__(self, config):
    method forward (line 1268) | def forward(
  class CamembertForQuestionAnswering (line 1331) | class CamembertForQuestionAnswering(CamembertPreTrainedModel):
    method __init__ (line 1335) | def __init__(self, config):
    method forward (line 1353) | def forward(
  class CamembertForCausalLM (line 1432) | class CamembertForCausalLM(CamembertPreTrainedModel):
    method __init__ (line 1437) | def __init__(self, config):
    method get_output_embeddings (line 1452) | def get_output_embeddings(self):
    method set_output_embeddings (line 1455) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1460) | def forward(
    method prepare_inputs_for_generation (line 1566) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1578) | def _reorder_cache(self, past_key_values, beam_idx):
  function create_position_ids_from_input_ids (line 1586) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/camembert/modeling_tf_camembert.py
  class TFCamembertEmbeddings (line 171) | class TFCamembertEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 176) | def __init__(self, config, **kwargs):
    method build (line 187) | def build(self, input_shape: tf.TensorShape):
    method create_position_ids_from_input_ids (line 211) | def create_position_ids_from_input_ids(self, input_ids, past_key_value...
    method call (line 225) | def call(
  class TFCamembertPooler (line 272) | class TFCamembertPooler(tf.keras.layers.Layer):
    method __init__ (line 273) | def __init__(self, config: CamembertConfig, **kwargs):
    method call (line 283) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFCamembertSelfAttention (line 293) | class TFCamembertSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 294) | def __init__(self, config: CamembertConfig, **kwargs):
    method transpose_for_scores (line 321) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 328) | def call(
  class TFCamembertSelfOutput (line 411) | class TFCamembertSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 412) | def __init__(self, config: CamembertConfig, **kwargs):
    method call (line 421) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFCamembertAttention (line 430) | class TFCamembertAttention(tf.keras.layers.Layer):
    method __init__ (line 431) | def __init__(self, config: CamembertConfig, **kwargs):
    method prune_heads (line 437) | def prune_heads(self, heads):
    method call (line 440) | def call(
  class TFCamembertIntermediate (line 471) | class TFCamembertIntermediate(tf.keras.layers.Layer):
    method __init__ (line 472) | def __init__(self, config: CamembertConfig, **kwargs):
    method call (line 484) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFCamembertOutput (line 492) | class TFCamembertOutput(tf.keras.layers.Layer):
    method __init__ (line 493) | def __init__(self, config: CamembertConfig, **kwargs):
    method call (line 502) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFCamembertLayer (line 511) | class TFCamembertLayer(tf.keras.layers.Layer):
    method __init__ (line 512) | def __init__(self, config: CamembertConfig, **kwargs):
    method call (line 525) | def call(
  class TFCamembertEncoder (line 598) | class TFCamembertEncoder(tf.keras.layers.Layer):
    method __init__ (line 599) | def __init__(self, config: CamembertConfig, **kwargs):
    method call (line 604) | def call(
  class TFCamembertMainLayer (line 669) | class TFCamembertMainLayer(tf.keras.layers.Layer):
    method __init__ (line 672) | def __init__(self, config, add_pooling_layer=True, **kwargs):
    method get_input_embeddings (line 689) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 693) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 698) | def _prune_heads(self, heads_to_prune):
    method call (line 707) | def call(
  class TFCamembertPreTrainedModel (line 865) | class TFCamembertPreTrainedModel(TFPreTrainedModel):
  class TFCamembertModel (line 880) | class TFCamembertModel(TFCamembertPreTrainedModel):
    method __init__ (line 881) | def __init__(self, config, *inputs, **kwargs):
    method call (line 892) | def call(
  class TFCamembertLMHead (line 950) | class TFCamembertLMHead(tf.keras.layers.Layer):
    method __init__ (line 953) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 968) | def build(self, input_shape):
    method get_output_embeddings (line 973) | def get_output_embeddings(self):
    method set_output_embeddings (line 976) | def set_output_embeddings(self, value):
    method get_bias (line 980) | def get_bias(self):
    method set_bias (line 983) | def set_bias(self, value):
    method call (line 987) | def call(self, hidden_states):
  class TFCamembertForMaskedLM (line 1007) | class TFCamembertForMaskedLM(TFCamembertPreTrainedModel, TFMaskedLanguag...
    method __init__ (line 1011) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 1017) | def get_lm_head(self):
    method get_prefix_bias_name (line 1020) | def get_prefix_bias_name(self):
    method call (line 1034) | def call(
  class TFCamembertClassificationHead (line 1085) | class TFCamembertClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 1088) | def __init__(self, config, **kwargs):
    method call (line 1104) | def call(self, features, training=False):
  class TFCamembertForSequenceClassification (line 1121) | class TFCamembertForSequenceClassification(TFCamembertPreTrainedModel, T...
    method __init__ (line 1125) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1141) | def call(
  class TFCamembertForTokenClassification (line 1198) | class TFCamembertForTokenClassification(TFCamembertPreTrainedModel, TFTo...
    method __init__ (line 1203) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1225) | def call(
  class TFCamembertForMultipleChoice (line 1282) | class TFCamembertForMultipleChoice(TFCamembertPreTrainedModel, TFMultipl...
    method __init__ (line 1287) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1305) | def call(
  class TFCamembertForQuestionAnswering (line 1375) | class TFCamembertForQuestionAnswering(TFCamembertPreTrainedModel, TFQues...
    method __init__ (line 1379) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1397) | def call(
  class TFCamembertForCausalLM (line 1464) | class TFCamembertForCausalLM(TFCamembertPreTrainedModel, TFCausalLanguag...
    method __init__ (line 1468) | def __init__(self, config: CamembertConfig, *inputs, **kwargs):
    method get_lm_head (line 1477) | def get_lm_head(self):
    method get_prefix_bias_name (line 1480) | def get_prefix_bias_name(self):
    method prepare_inputs_for_generation (line 1485) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method call (line 1504) | def call(

FILE: transformers/models/camembert/tokenization_camembert.py
  class CamembertTokenizer (line 45) | class CamembertTokenizer(PreTrainedTokenizer):
    method __init__ (line 120) | def __init__(
    method build_inputs_with_special_tokens (line 161) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 187) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 214) | def create_token_type_ids_from_sequences(
    method vocab_size (line 238) | def vocab_size(self):
    method get_vocab (line 241) | def get_vocab(self):
    method _tokenize (line 246) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 249) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 258) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 264) | def convert_tokens_to_string(self, tokens):
    method __getstate__ (line 283) | def __getstate__(self):
    method __setstate__ (line 288) | def __setstate__(self, d):
    method save_vocabulary (line 298) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/camembert/tokenization_camembert_fast.py
  class CamembertTokenizerFast (line 53) | class CamembertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 111) | def __init__(
    method build_inputs_with_special_tokens (line 145) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 171) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 194) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/canine/configuration_canine.py
  class CanineConfig (line 29) | class CanineConfig(PretrainedConfig):
    method __init__ (line 94) | def __init__(

FILE: transformers/models/canine/convert_canine_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 27) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, pytorch_dump_pa...

FILE: transformers/models/canine/modeling_canine.py
  class CanineModelOutputWithPooling (line 66) | class CanineModelOutputWithPooling(ModelOutput):
  function load_tf_weights_in_canine (line 100) | def load_tf_weights_in_canine(model, config, tf_checkpoint_path):
  class CanineEmbeddings (line 197) | class CanineEmbeddings(nn.Module):
    method __init__ (line 200) | def __init__(self, config):
    method _hash_bucket_tensors (line 222) | def _hash_bucket_tensors(self, input_ids, num_hashes: int, num_buckets...
    method _embed_hash_buckets (line 245) | def _embed_hash_buckets(self, input_ids, embedding_size: int, num_hash...
    method forward (line 259) | def forward(
  class CharactersToMolecules (line 296) | class CharactersToMolecules(nn.Module):
    method __init__ (line 299) | def __init__(self, config):
    method forward (line 314) | def forward(self, char_encoding: torch.Tensor) -> torch.Tensor:
  class ConvProjection (line 342) | class ConvProjection(nn.Module):
    method __init__ (line 348) | def __init__(self, config):
    method forward (line 363) | def forward(
  class CanineSelfAttention (line 400) | class CanineSelfAttention(nn.Module):
    method __init__ (line 401) | def __init__(self, config):
    method transpose_for_scores (line 423) | def transpose_for_scores(self, x):
    method forward (line 428) | def forward(
  class CanineSelfOutput (line 500) | class CanineSelfOutput(nn.Module):
    method __init__ (line 501) | def __init__(self, config):
    method forward (line 507) | def forward(
  class CanineAttention (line 516) | class CanineAttention(nn.Module):
    method __init__ (line 533) | def __init__(
    method prune_heads (line 566) | def prune_heads(self, heads):
    method forward (line 584) | def forward(
  class CanineIntermediate (line 659) | class CanineIntermediate(nn.Module):
    method __init__ (line 660) | def __init__(self, config):
    method forward (line 668) | def forward(self, hidden_states: torch.FloatTensor) -> torch.FloatTensor:
  class CanineOutput (line 674) | class CanineOutput(nn.Module):
    method __init__ (line 675) | def __init__(self, config):
    method forward (line 681) | def forward(self, hidden_states: Tuple[torch.FloatTensor], input_tenso...
  class CanineLayer (line 688) | class CanineLayer(nn.Module):
    method __init__ (line 689) | def __init__(
    method forward (line 716) | def forward(
    method feed_forward_chunk (line 740) | def feed_forward_chunk(self, attention_output):
  class CanineEncoder (line 746) | class CanineEncoder(nn.Module):
    method __init__ (line 747) | def __init__(
    method forward (line 777) | def forward(
  class CaninePooler (line 828) | class CaninePooler(nn.Module):
    method __init__ (line 829) | def __init__(self, config):
    method forward (line 834) | def forward(self, hidden_states: Tuple[torch.FloatTensor]) -> torch.Fl...
  class CaninePredictionHeadTransform (line 843) | class CaninePredictionHeadTransform(nn.Module):
    method __init__ (line 844) | def __init__(self, config):
    method forward (line 853) | def forward(self, hidden_states: Tuple[torch.FloatTensor]) -> torch.Fl...
  class CanineLMPredictionHead (line 860) | class CanineLMPredictionHead(nn.Module):
    method __init__ (line 861) | def __init__(self, config):
    method forward (line 874) | def forward(self, hidden_states: Tuple[torch.FloatTensor]) -> torch.Fl...
  class CanineOnlyMLMHead (line 880) | class CanineOnlyMLMHead(nn.Module):
    method __init__ (line 881) | def __init__(self, config):
    method forward (line 885) | def forward(
  class CaninePreTrainedModel (line 893) | class CaninePreTrainedModel(PreTrainedModel):
    method _init_weights (line 905) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 921) | def _set_gradient_checkpointing(self, module, value=False):
  class CanineModel (line 991) | class CanineModel(CaninePreTrainedModel):
    method __init__ (line 992) | def __init__(self, config, add_pooling_layer=True):
    method _prune_heads (line 1022) | def _prune_heads(self, heads_to_prune):
    method _create_3d_attention_mask_from_input_mask (line 1030) | def _create_3d_attention_mask_from_input_mask(self, from_tensor, to_ma...
    method _downsample_attention_mask (line 1057) | def _downsample_attention_mask(self, char_attention_mask: torch.Tensor...
    method _repeat_molecules (line 1074) | def _repeat_molecules(self, molecules: torch.Tensor, char_seq_length: ...
    method forward (line 1105) | def forward(
  class CanineForSequenceClassification (line 1268) | class CanineForSequenceClassification(CaninePreTrainedModel):
    method __init__ (line 1269) | def __init__(self, config):
    method forward (line 1286) | def forward(
  class CanineForMultipleChoice (line 1365) | class CanineForMultipleChoice(CaninePreTrainedModel):
    method __init__ (line 1366) | def __init__(self, config):
    method forward (line 1382) | def forward(
  class CanineForTokenClassification (line 1456) | class CanineForTokenClassification(CaninePreTrainedModel):
    method __init__ (line 1457) | def __init__(self, config):
    method forward (line 1470) | def forward(
  class CanineForQuestionAnswering (line 1562) | class CanineForQuestionAnswering(CaninePreTrainedModel):
    method __init__ (line 1563) | def __init__(self, config):
    method forward (line 1581) | def forward(

FILE: transformers/models/canine/tokenization_canine.py
  class CanineTokenizer (line 63) | class CanineTokenizer(PreTrainedTokenizer):
    method __init__ (line 79) | def __init__(
    method vocab_size (line 126) | def vocab_size(self) -> int:
    method _tokenize (line 129) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 133) | def _convert_token_to_id(self, token: str) -> int:
    method _convert_id_to_token (line 140) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 152) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 155) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 182) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 210) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 242) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/chinese_clip/configuration_chinese_clip.py
  class ChineseCLIPTextConfig (line 41) | class ChineseCLIPTextConfig(PretrainedConfig):
    method __init__ (line 107) | def __init__(
    method from_pretrained (line 146) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class ChineseCLIPVisionConfig (line 162) | class ChineseCLIPVisionConfig(PretrainedConfig):
    method __init__ (line 215) | def __init__(
    method from_pretrained (line 249) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class ChineseCLIPConfig (line 265) | class ChineseCLIPConfig(PretrainedConfig):
    method __init__ (line 315) | def __init__(
    method from_text_vision_configs (line 405) | def from_text_vision_configs(
    method to_dict (line 416) | def to_dict(self):
  class ChineseCLIPOnnxConfig (line 430) | class ChineseCLIPOnnxConfig(OnnxConfig):
    method inputs (line 432) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 442) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 453) | def atol_for_validation(self) -> float:
    method generate_dummy_inputs (line 456) | def generate_dummy_inputs(
    method default_onnx_opset (line 472) | def default_onnx_opset(self) -> int:

FILE: transformers/models/chinese_clip/convert_chinese_clip_original_pytorch_to_hf.py
  function copy_attn_layer (line 23) | def copy_attn_layer(hf_attn_layer, pt_weights, prefix):
  function copy_mlp (line 43) | def copy_mlp(hf_mlp, pt_weights, prefix):
  function copy_linear (line 48) | def copy_linear(hf_linear, pt_weights, prefix):
  function copy_layer (line 53) | def copy_layer(hf_layer, pt_weights, prefix):
  function copy_layers (line 65) | def copy_layers(hf_layers, pt_weights, prefix):
  function copy_text_model_and_projection (line 70) | def copy_text_model_and_projection(hf_model, pt_weights):
  function copy_vision_model_and_projection (line 79) | def copy_vision_model_and_projection(hf_model, pt_weights):
  function convert_chinese_clip_checkpoint (line 97) | def convert_chinese_clip_checkpoint(checkpoint_path, pytorch_dump_folder...

FILE: transformers/models/chinese_clip/feature_extraction_chinese_clip.py
  class ChineseCLIPFeatureExtractor (line 26) | class ChineseCLIPFeatureExtractor(ChineseCLIPImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/chinese_clip/image_processing_chinese_clip.py
  class ChineseCLIPImageProcessor (line 51) | class ChineseCLIPImageProcessor(BaseImageProcessor):
    method __init__ (line 91) | def __init__(
    method resize (line 124) | def resize(
    method center_crop (line 152) | def center_crop(
    method rescale (line 174) | def rescale(
    method normalize (line 194) | def normalize(
    method preprocess (line 217) | def preprocess(

FILE: transformers/models/chinese_clip/modeling_chinese_clip.py
  function contrastive_loss (line 59) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  function chinese_clip_loss (line 63) | def chinese_clip_loss(similarity: torch.Tensor) -> torch.Tensor:
  class ChineseCLIPOutput (line 70) | class ChineseCLIPOutput(ModelOutput):
    method to_tuple (line 101) | def to_tuple(self) -> Tuple[Any]:
  class ChineseCLIPTextEmbeddings (line 109) | class ChineseCLIPTextEmbeddings(nn.Module):
    method __init__ (line 112) | def __init__(self, config):
    method forward (line 129) | def forward(
  class ChineseCLIPVisionEmbeddings (line 172) | class ChineseCLIPVisionEmbeddings(nn.Module):
    method __init__ (line 173) | def __init__(self, config: ChineseCLIPVisionConfig):
    method forward (line 195) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class ChineseCLIPTextSelfAttention (line 207) | class ChineseCLIPTextSelfAttention(nn.Module):
    method __init__ (line 208) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 234) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 239) | def forward(
  class ChineseCLIPTextSelfOutput (line 342) | class ChineseCLIPTextSelfOutput(nn.Module):
    method __init__ (line 343) | def __init__(self, config):
    method forward (line 349) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ChineseCLIPTextAttention (line 357) | class ChineseCLIPTextAttention(nn.Module):
    method __init__ (line 358) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 364) | def prune_heads(self, heads):
    method forward (line 382) | def forward(
  class ChineseCLIPVisionAttention (line 406) | class ChineseCLIPVisionAttention(nn.Module):
    method __init__ (line 409) | def __init__(self, config):
    method _shape (line 428) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 431) | def forward(
  class ChineseCLIPTextIntermediate (line 491) | class ChineseCLIPTextIntermediate(nn.Module):
    method __init__ (line 492) | def __init__(self, config):
    method forward (line 500) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ChineseCLIPTextOutput (line 507) | class ChineseCLIPTextOutput(nn.Module):
    method __init__ (line 508) | def __init__(self, config):
    method forward (line 514) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ChineseCLIPVisionMLP (line 522) | class ChineseCLIPVisionMLP(nn.Module):
    method __init__ (line 523) | def __init__(self, config):
    method forward (line 530) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ChineseCLIPTextLayer (line 538) | class ChineseCLIPTextLayer(nn.Module):
    method __init__ (line 539) | def __init__(self, config):
    method forward (line 553) | def forward(
    method feed_forward_chunk (line 618) | def feed_forward_chunk(self, attention_output):
  class ChineseCLIPVisionLayer (line 624) | class ChineseCLIPVisionLayer(nn.Module):
    method __init__ (line 625) | def __init__(self, config: ChineseCLIPConfig):
    method forward (line 633) | def forward(
  class ChineseCLIPTextPooler (line 668) | class ChineseCLIPTextPooler(nn.Module):
    method __init__ (line 669) | def __init__(self, config):
    method forward (line 674) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ChineseCLIPPreTrainedModel (line 683) | class ChineseCLIPPreTrainedModel(PreTrainedModel):
    method _init_weights (line 694) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 743) | def _set_gradient_checkpointing(self, module, value=False):
  class ChineseCLIPTextEncoder (line 870) | class ChineseCLIPTextEncoder(nn.Module):
    method __init__ (line 871) | def __init__(self, config):
    method forward (line 877) | def forward(
  class ChineseCLIPVisionEncoder (line 968) | class ChineseCLIPVisionEncoder(nn.Module):
    method __init__ (line 977) | def __init__(self, config: ChineseCLIPConfig):
    method forward (line 983) | def forward(
  class ChineseCLIPVisionTransformer (line 1051) | class ChineseCLIPVisionTransformer(nn.Module):
    method __init__ (line 1052) | def __init__(self, config: ChineseCLIPVisionConfig):
    method forward (line 1064) | def forward(
  class ChineseCLIPTextModel (line 1112) | class ChineseCLIPTextModel(ChineseCLIPPreTrainedModel):
    method __init__ (line 1127) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 1139) | def get_input_embeddings(self):
    method set_input_embeddings (line 1142) | def set_input_embeddings(self, value):
    method _prune_heads (line 1145) | def _prune_heads(self, heads_to_prune):
    method forward (line 1159) | def forward(
  class ChineseCLIPVisionModel (line 1293) | class ChineseCLIPVisionModel(ChineseCLIPPreTrainedModel):
    method __init__ (line 1297) | def __init__(self, config: ChineseCLIPVisionConfig):
    method get_input_embeddings (line 1303) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 1308) | def forward(
  class ChineseCLIPModel (line 1348) | class ChineseCLIPModel(ChineseCLIPPreTrainedModel):
    method __init__ (line 1351) | def __init__(self, config: ChineseCLIPConfig):
    method get_text_features (line 1384) | def get_text_features(
    method get_image_features (line 1434) | def get_image_features(
    method forward (line 1485) | def forward(

FILE: transformers/models/chinese_clip/processing_chinese_clip.py
  class ChineseCLIPProcessor (line 25) | class ChineseCLIPProcessor(ProcessorMixin):
    method __init__ (line 43) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 61) | def __call__(self, text=None, images=None, return_tensors=None, **kwar...
    method batch_decode (line 114) | def batch_decode(self, *args, **kwargs):
    method decode (line 121) | def decode(self, *args, **kwargs):
    method model_input_names (line 129) | def model_input_names(self):
    method feature_extractor_class (line 135) | def feature_extractor_class(self):

FILE: transformers/models/clap/configuration_clap.py
  class ClapTextConfig (line 33) | class ClapTextConfig(PretrainedConfig):
    method __init__ (line 107) | def __init__(
    method from_pretrained (line 154) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class ClapAudioConfig (line 170) | class ClapAudioConfig(PretrainedConfig):
    method __init__ (line 260) | def __init__(
    method from_pretrained (line 322) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class ClapConfig (line 338) | class ClapConfig(PretrainedConfig):
    method __init__ (line 391) | def __init__(
    method from_text_audio_configs (line 428) | def from_text_audio_configs(cls, text_config: ClapTextConfig, audio_co...
    method to_dict (line 439) | def to_dict(self):

FILE: transformers/models/clap/convert_clap_original_pytorch_to_hf.py
  function init_clap (line 41) | def init_clap(checkpoint_path, enable_fusion=False):
  function rename_state_dict (line 54) | def rename_state_dict(state_dict):
  function convert_clap_checkpoint (line 97) | def convert_clap_checkpoint(checkpoint_path, pytorch_dump_folder_path, c...

FILE: transformers/models/clap/feature_extraction_clap.py
  class ClapFeatureExtractor (line 33) | class ClapFeatureExtractor(SequenceFeatureExtractor):
    method __init__ (line 85) | def __init__(
    method to_dict (line 138) | def to_dict(self) -> Dict[str, Any]:
    method _np_extract_fbank_features (line 154) | def _np_extract_fbank_features(self, waveform: np.array, mel_filters: ...
    method _random_mel_fusion (line 176) | def _random_mel_fusion(self, mel, total_frames, chunk_frames):
    method _get_input_mel (line 201) | def _get_input_mel(self, waveform: np.array, max_length, truncation, p...
    method __call__ (line 259) | def __call__(

FILE: transformers/models/clap/modeling_clap.py
  function interpolate (line 56) | def interpolate(hidden_states, ratio):
  function window_partition (line 73) | def window_partition(hidden_states, window_size):
  function window_reverse (line 94) | def window_reverse(windows, window_size, height, width):
  function create_position_ids_from_input_ids (line 114) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...
  function contrastive_loss (line 132) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  class ClapTextModelOutput (line 139) | class ClapTextModelOutput(ModelOutput):
  class ClapAudioModelOutput (line 168) | class ClapAudioModelOutput(ModelOutput):
  class ClapOutput (line 198) | class ClapOutput(ModelOutput):
    method to_tuple (line 227) | def to_tuple(self) -> Tuple[Any]:
  class ClapDropPath (line 235) | class ClapDropPath(nn.Module):
    method __init__ (line 241) | def __init__(self, drop_prob=None):
    method forward (line 245) | def forward(self, hidden_states):
  class ClapAudioAFFBlock (line 260) | class ClapAudioAFFBlock(nn.Module):
    method __init__ (line 266) | def __init__(self, config: ClapAudioConfig):
    method forward (line 290) | def forward(self, hidden_states, residual):
  class ClapAudioPatchEmbed (line 300) | class ClapAudioPatchEmbed(nn.Module):
    method __init__ (line 306) | def __init__(self, config: ClapAudioConfig):
    method forward (line 348) | def forward(self, hidden_states, is_longer_idx=None):
  class ClapAudioSelfAttention (line 399) | class ClapAudioSelfAttention(nn.Module):
    method __init__ (line 400) | def __init__(self, config, dim, num_heads, window_size):
    method transpose_for_scores (line 437) | def transpose_for_scores(self, x):
    method forward (line 442) | def forward(
  class ClapAudioSelfOutput (line 500) | class ClapAudioSelfOutput(nn.Module):
    method __init__ (line 501) | def __init__(self, config, dim):
    method forward (line 506) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ClapAudioAttention (line 514) | class ClapAudioAttention(nn.Module):
    method __init__ (line 515) | def __init__(self, config, dim, num_heads, window_size):
    method prune_heads (line 521) | def prune_heads(self, heads):
    method forward (line 539) | def forward(
  class ClapAudioIntermediate (line 553) | class ClapAudioIntermediate(nn.Module):
    method __init__ (line 554) | def __init__(self, config, dim):
    method forward (line 562) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ClapAudioOutput (line 569) | class ClapAudioOutput(nn.Module):
    method __init__ (line 570) | def __init__(self, config, dim):
    method forward (line 575) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ClapAudioLayer (line 582) | class ClapAudioLayer(nn.Module):
    method __init__ (line 583) | def __init__(self, config, dim, input_resolution, num_heads, shift_siz...
    method set_shift_and_window_size (line 596) | def set_shift_and_window_size(self, input_resolution):
    method get_attn_mask (line 602) | def get_attn_mask(self, height, width, dtype):
    method maybe_pad (line 630) | def maybe_pad(self, hidden_states, height, width):
    method forward (line 637) | def forward(
  class ClapAudioStage (line 706) | class ClapAudioStage(nn.Module):
    method __init__ (line 707) | def __init__(self, config, dim, input_resolution, depth, num_heads, dr...
    method forward (line 732) | def forward(
  class ClapAudioPatchMerging (line 766) | class ClapAudioPatchMerging(nn.Module):
    method __init__ (line 779) | def __init__(self, input_resolution: Tuple[int], dim: int, norm_layer:...
    method maybe_pad (line 786) | def maybe_pad(self, input_feature, height, width):
    method forward (line 794) | def forward(self, input_feature: torch.Tensor, input_dimensions: Tuple...
  class ClapAudioEncoder (line 820) | class ClapAudioEncoder(nn.Module):
    method __init__ (line 821) | def __init__(self, config):
    method reshape_mel2img (line 861) | def reshape_mel2img(self, normalized_input_features):
    method forward (line 897) | def forward(
  class ClapProjectionLayer (line 1132) | class ClapProjectionLayer(nn.Module):
    method __init__ (line 1133) | def __init__(self, config: Union[ClapAudioConfig, ClapTextConfig]):
    method forward (line 1143) | def forward(self, hidden_states):
  class ClapTextEmbeddings (line 1151) | class ClapTextEmbeddings(nn.Module):
    method __init__ (line 1157) | def __init__(self, config):
    method forward (line 1180) | def forward(
    method create_position_ids_from_inputs_embeds (line 1220) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class ClapTextSelfAttention (line 1239) | class ClapTextSelfAttention(nn.Module):
    method __init__ (line 1240) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 1266) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 1271) | def forward(
  class ClapTextSelfOutput (line 1374) | class ClapTextSelfOutput(nn.Module):
    method __init__ (line 1375) | def __init__(self, config):
    method forward (line 1381) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ClapTextAttention (line 1389) | class ClapTextAttention(nn.Module):
    method __init__ (line 1390) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 1396) | def prune_heads(self, heads):
    method forward (line 1414) | def forward(
  class ClapTextIntermediate (line 1439) | class ClapTextIntermediate(nn.Module):
    method __init__ (line 1440) | def __init__(self, config):
    method forward (line 1448) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ClapTextOutput (line 1455) | class ClapTextOutput(nn.Module):
    method __init__ (line 1456) | def __init__(self, config):
    method forward (line 1462) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ClapTextLayer (line 1470) | class ClapTextLayer(nn.Module):
    method __init__ (line 1471) | def __init__(self, config):
    method forward (line 1485) | def forward(
    method feed_forward_chunk (line 1550) | def feed_forward_chunk(self, attention_output):
  class ClapTextEncoder (line 1557) | class ClapTextEncoder(nn.Module):
    method __init__ (line 1558) | def __init__(self, config):
    method forward (line 1564) | def forward(
  class ClapTextPooler (line 1656) | class ClapTextPooler(nn.Module):
    method __init__ (line 1657) | def __init__(self, config):
    method forward (line 1662) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ClapPreTrainedModel (line 1671) | class ClapPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1682) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1704) | def _set_gradient_checkpointing(self, module, value=False):
  class ClapAudioModel (line 1709) | class ClapAudioModel(ClapPreTrainedModel):
    method __init__ (line 1713) | def __init__(self, config: ClapAudioConfig):
    method get_input_embeddings (line 1719) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 1724) | def forward(
  class ClapTextModel (line 1767) | class ClapTextModel(ClapPreTrainedModel):
    method __init__ (line 1787) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 1799) | def get_input_embeddings(self):
    method set_input_embeddings (line 1802) | def set_input_embeddings(self, value):
    method forward (line 1806) | def forward(
  class ClapModel (line 1937) | class ClapModel(ClapPreTrainedModel):
    method __init__ (line 1941) | def __init__(self, config: ClapConfig):
    method get_text_features (line 1974) | def get_text_features(
    method get_audio_features (line 2022) | def get_audio_features(
    method forward (line 2069) | def forward(
  class ClapTextModelWithProjection (line 2171) | class ClapTextModelWithProjection(ClapPreTrainedModel):
    method __init__ (line 2174) | def __init__(self, config: ClapTextConfig):
    method get_input_embeddings (line 2181) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 2184) | def set_input_embeddings(self, value):
    method forward (line 2189) | def forward(
  class ClapAudioModelWithProjection (line 2247) | class ClapAudioModelWithProjection(ClapPreTrainedModel):
    method __init__ (line 2251) | def __init__(self, config: ClapAudioConfig):
    method get_input_embeddings (line 2258) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 2263) | def forward(

FILE: transformers/models/clap/processing_clap.py
  class ClapProcessor (line 23) | class ClapProcessor(ProcessorMixin):
    method __init__ (line 39) | def __init__(self, feature_extractor, tokenizer):
    method __call__ (line 42) | def __call__(self, text=None, audios=None, return_tensors=None, **kwar...
    method batch_decode (line 98) | def batch_decode(self, *args, **kwargs):
    method decode (line 105) | def decode(self, *args, **kwargs):
    method model_input_names (line 113) | def model_input_names(self):

FILE: transformers/models/clip/configuration_clip.py
  class CLIPTextConfig (line 40) | class CLIPTextConfig(PretrainedConfig):
    method __init__ (line 94) | def __init__(
    method from_pretrained (line 129) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class CLIPVisionConfig (line 145) | class CLIPVisionConfig(PretrainedConfig):
    method __init__ (line 198) | def __init__(
    method from_pretrained (line 232) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class CLIPConfig (line 248) | class CLIPConfig(PretrainedConfig):
    method __init__ (line 297) | def __init__(
    method from_text_vision_configs (line 386) | def from_text_vision_configs(cls, text_config: CLIPTextConfig, vision_...
    method to_dict (line 397) | def to_dict(self):
  class CLIPOnnxConfig (line 411) | class CLIPOnnxConfig(OnnxConfig):
    method inputs (line 413) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 423) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 434) | def atol_for_validation(self) -> float:
    method generate_dummy_inputs (line 437) | def generate_dummy_inputs(
    method default_onnx_opset (line 453) | def default_onnx_opset(self) -> int:

FILE: transformers/models/clip/convert_clip_original_pytorch_to_hf.py
  function copy_attn_layer (line 24) | def copy_attn_layer(hf_attn_layer, pt_attn_layer):
  function copy_mlp (line 44) | def copy_mlp(hf_mlp, pt_mlp):
  function copy_linear (line 49) | def copy_linear(hf_linear, pt_linear):
  function copy_layer (line 54) | def copy_layer(hf_layer, pt_layer):
  function copy_layers (line 66) | def copy_layers(hf_layers, pt_layers):
  function copy_encoder (line 71) | def copy_encoder(hf_encoder, pt_model):
  function copy_text_model_and_projection (line 83) | def copy_text_model_and_projection(hf_model, pt_model):
  function copy_vison_model_and_projection (line 91) | def copy_vison_model_and_projection(hf_model, pt_model):
  function convert_clip_checkpoint (line 109) | def convert_clip_checkpoint(checkpoint_path, pytorch_dump_folder_path, c...

FILE: transformers/models/clip/feature_extraction_clip.py
  class CLIPFeatureExtractor (line 26) | class CLIPFeatureExtractor(CLIPImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/clip/image_processing_clip.py
  class CLIPImageProcessor (line 51) | class CLIPImageProcessor(BaseImageProcessor):
    method __init__ (line 91) | def __init__(
    method resize (line 124) | def resize(
    method center_crop (line 152) | def center_crop(
    method rescale (line 176) | def rescale(
    method normalize (line 196) | def normalize(
    method preprocess (line 219) | def preprocess(

FILE: transformers/models/clip/modeling_clip.py
  function _expand_mask (line 49) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  function contrastive_loss (line 65) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  function clip_loss (line 69) | def clip_loss(similarity: torch.Tensor) -> torch.Tensor:
  class CLIPVisionModelOutput (line 76) | class CLIPVisionModelOutput(ModelOutput):
  class CLIPTextModelOutput (line 105) | class CLIPTextModelOutput(ModelOutput):
  class CLIPOutput (line 134) | class CLIPOutput(ModelOutput):
    method to_tuple (line 163) | def to_tuple(self) -> Tuple[Any]:
  class CLIPVisionEmbeddings (line 170) | class CLIPVisionEmbeddings(nn.Module):
    method __init__ (line 171) | def __init__(self, config: CLIPVisionConfig):
    method forward (line 193) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class CLIPTextEmbeddings (line 204) | class CLIPTextEmbeddings(nn.Module):
    method __init__ (line 205) | def __init__(self, config: CLIPTextConfig):
    method forward (line 215) | def forward(
  class CLIPAttention (line 235) | class CLIPAttention(nn.Module):
    method __init__ (line 238) | def __init__(self, config):
    method _shape (line 257) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 260) | def forward(
  class CLIPMLP (line 339) | class CLIPMLP(nn.Module):
    method __init__ (line 340) | def __init__(self, config):
    method forward (line 347) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class CLIPEncoderLayer (line 354) | class CLIPEncoderLayer(nn.Module):
    method __init__ (line 355) | def __init__(self, config: CLIPConfig):
    method forward (line 363) | def forward(
  class CLIPPreTrainedModel (line 404) | class CLIPPreTrainedModel(PreTrainedModel):
    method _init_weights (line 415) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 468) | def _set_gradient_checkpointing(self, module, value=False):
  class CLIPEncoder (line 573) | class CLIPEncoder(nn.Module):
    method __init__ (line 582) | def __init__(self, config: CLIPConfig):
    method forward (line 588) | def forward(
  function _make_causal_mask (line 677) | def _make_causal_mask(
  class CLIPTextTransformer (line 694) | class CLIPTextTransformer(nn.Module):
    method __init__ (line 695) | def __init__(self, config: CLIPTextConfig):
    method forward (line 705) | def forward(
  class CLIPTextModel (line 775) | class CLIPTextModel(CLIPPreTrainedModel):
    method __init__ (line 780) | def __init__(self, config: CLIPTextConfig):
    method get_input_embeddings (line 786) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 789) | def set_input_embeddings(self, value):
    method forward (line 794) | def forward(
  class CLIPVisionTransformer (line 832) | class CLIPVisionTransformer(nn.Module):
    method __init__ (line 833) | def __init__(self, config: CLIPVisionConfig):
    method forward (line 845) | def forward(
  class CLIPVisionModel (line 894) | class CLIPVisionModel(CLIPPreTrainedModel):
    method __init__ (line 898) | def __init__(self, config: CLIPVisionConfig):
    method get_input_embeddings (line 904) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 909) | def forward(
  class CLIPModel (line 949) | class CLIPModel(CLIPPreTrainedModel):
    method __init__ (line 952) | def __init__(self, config: CLIPConfig):
    method get_text_features (line 985) | def get_text_features(
    method get_image_features (line 1032) | def get_image_features(
    method forward (line 1082) | def forward(
  class CLIPTextModelWithProjection (line 1180) | class CLIPTextModelWithProjection(CLIPPreTrainedModel):
    method __init__ (line 1185) | def __init__(self, config: CLIPTextConfig):
    method get_input_embeddings (line 1195) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 1198) | def set_input_embeddings(self, value):
    method forward (line 1203) | def forward(
  class CLIPVisionModelWithProjection (line 1261) | class CLIPVisionModelWithProjection(CLIPPreTrainedModel):
    method __init__ (line 1265) | def __init__(self, config: CLIPVisionConfig):
    method get_input_embeddings (line 1275) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 1280) | def forward(

FILE: transformers/models/clip/modeling_flax_clip.py
  class FlaxCLIPOutput (line 159) | class FlaxCLIPOutput(ModelOutput):
    method to_tuple (line 187) | def to_tuple(self) -> Tuple[Any]:
  class FlaxCLIPVisionEmbeddings (line 194) | class FlaxCLIPVisionEmbeddings(nn.Module):
    method setup (line 198) | def setup(self):
    method __call__ (line 220) | def __call__(self, pixel_values):
  class FlaxCLIPTextEmbeddings (line 232) | class FlaxCLIPTextEmbeddings(nn.Module):
    method setup (line 236) | def setup(self):
    method __call__ (line 247) | def __call__(self, input_ids, position_ids):
  class FlaxCLIPAttention (line 255) | class FlaxCLIPAttention(nn.Module):
    method setup (line 259) | def setup(self):
    method _split_heads (line 280) | def _split_heads(self, hidden_states):
    method _merge_heads (line 283) | def _merge_heads(self, hidden_states):
    method __call__ (line 286) | def __call__(
  class FlaxCLIPMLP (line 346) | class FlaxCLIPMLP(nn.Module):
    method setup (line 350) | def setup(self):
    method __call__ (line 359) | def __call__(self, hidden_states):
  class FlaxCLIPEncoderLayer (line 366) | class FlaxCLIPEncoderLayer(nn.Module):
    method setup (line 370) | def setup(self):
    method __call__ (line 376) | def __call__(
  class FlaxCLIPLayerCollection (line 408) | class FlaxCLIPLayerCollection(nn.Module):
    method setup (line 412) | def setup(self):
    method __call__ (line 418) | def __call__(
  class FlaxCLIPEncoder (line 455) | class FlaxCLIPEncoder(nn.Module):
    method setup (line 459) | def setup(self):
    method __call__ (line 462) | def __call__(
  class FlaxCLIPTextTransformer (line 481) | class FlaxCLIPTextTransformer(nn.Module):
    method setup (line 485) | def setup(self):
    method __call__ (line 490) | def __call__(
  class FlaxCLIPVisionTransformer (line 535) | class FlaxCLIPVisionTransformer(nn.Module):
    method setup (line 539) | def setup(self):
    method __call__ (line 545) | def __call__(
  class FlaxCLIPTextPreTrainedModel (line 585) | class FlaxCLIPTextPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 589) | def __init__(
    method init_weights (line 601) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 622) | def __call__(
  class FlaxCLIPVisionPreTrainedModel (line 664) | class FlaxCLIPVisionPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 669) | def __init__(
    method init_weights (line 683) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 702) | def __call__(
  class FlaxCLIPPreTrainedModel (line 736) | class FlaxCLIPPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 740) | def __init__(
    method init_weights (line 754) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 777) | def __call__(
    method get_text_features (line 822) | def get_text_features(
    method get_image_features (line 889) | def get_image_features(
  class FlaxCLIPTextModule (line 941) | class FlaxCLIPTextModule(nn.Module):
    method setup (line 945) | def setup(self):
    method __call__ (line 948) | def __call__(
  class FlaxCLIPTextModel (line 969) | class FlaxCLIPTextModel(FlaxCLIPTextPreTrainedModel):
  class FlaxCLIPVisionModule (line 998) | class FlaxCLIPVisionModule(nn.Module):
    method setup (line 1002) | def setup(self):
    method __call__ (line 1005) | def __call__(
  class FlaxCLIPVisionModel (line 1022) | class FlaxCLIPVisionModel(FlaxCLIPVisionPreTrainedModel):
  class FlaxCLIPModule (line 1056) | class FlaxCLIPModule(nn.Module):
    method setup (line 1060) | def setup(self):
    method __call__ (line 1088) | def __call__(
  class FlaxCLIPModel (line 1148) | class FlaxCLIPModel(FlaxCLIPPreTrainedModel):

FILE: transformers/models/clip/modeling_tf_clip.py
  function _expand_mask (line 63) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  function contrastive_loss (line 78) | def contrastive_loss(logits: tf.Tensor) -> tf.Tensor:
  function clip_loss (line 86) | def clip_loss(similarity: tf.Tensor) -> tf.Tensor:
  class TFCLIPOutput (line 93) | class TFCLIPOutput(ModelOutput):
    method to_tuple (line 123) | def to_tuple(self) -> Tuple[Any]:
  class TFCLIPVisionEmbeddings (line 130) | class TFCLIPVisionEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 131) | def __init__(self, config: CLIPVisionConfig, **kwargs):
    method build (line 154) | def build(self, input_shape: tf.TensorShape = None):
    method call (line 174) | def call(self, pixel_values: tf.Tensor) -> tf.Tensor:
  class TFCLIPTextEmbeddings (line 199) | class TFCLIPTextEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 200) | def __init__(self, config: CLIPTextConfig, **kwargs):
    method build (line 207) | def build(self, input_shape: tf.TensorShape = None):
    method call (line 226) | def call(
  class TFCLIPAttention (line 257) | class TFCLIPAttention(tf.keras.layers.Layer):
    method __init__ (line 260) | def __init__(self, config: CLIPConfig, **kwargs):
    method transpose_for_scores (line 295) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 302) | def call(
  class TFCLIPMLP (line 356) | class TFCLIPMLP(tf.keras.layers.Layer):
    method __init__ (line 357) | def __init__(self, config: CLIPConfig, **kwargs):
    method call (line 373) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFCLIPEncoderLayer (line 380) | class TFCLIPEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 381) | def __init__(self, config: CLIPConfig, **kwargs):
    method call (line 390) | def call(
  class TFCLIPEncoder (line 432) | class TFCLIPEncoder(tf.keras.layers.Layer):
    method __init__ (line 441) | def __init__(self, config: CLIPConfig, **kwargs):
    method call (line 446) | def call(
  class TFCLIPTextTransformer (line 487) | class TFCLIPTextTransformer(tf.keras.layers.Layer):
    method __init__ (line 488) | def __init__(self, config: CLIPTextConfig, **kwargs):
    method call (line 497) | def call(
    method _build_causal_attention_mask (line 552) | def _build_causal_attention_mask(self, batch_size, seq_length, dtype=t...
  class TFCLIPTextMainLayer (line 572) | class TFCLIPTextMainLayer(tf.keras.layers.Layer):
    method __init__ (line 575) | def __init__(self, config: CLIPTextConfig, **kwargs):
    method get_input_embeddings (line 580) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 583) | def set_input_embeddings(self, value: tf.Variable):
    method call (line 588) | def call(
  class TFCLIPVisionTransformer (line 619) | class TFCLIPVisionTransformer(tf.keras.layers.Layer):
    method __init__ (line 620) | def __init__(self, config: CLIPVisionConfig, **kwargs):
    method call (line 628) | def call(
  class TFCLIPVisionMainLayer (line 665) | class TFCLIPVisionMainLayer(tf.keras.layers.Layer):
    method __init__ (line 668) | def __init__(self, config: CLIPVisionConfig, **kwargs):
    method get_input_embeddings (line 673) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method call (line 677) | def call(
  class TFCLIPMainLayer (line 700) | class TFCLIPMainLayer(tf.keras.layers.Layer):
    method __init__ (line 703) | def __init__(self, config: CLIPConfig, **kwargs):
    method build (line 742) | def build(self, input_shape: tf.TensorShape = None):
    method get_text_features (line 753) | def get_text_features(
    method get_image_features (line 787) | def get_image_features(
    method call (line 812) | def call(
  class TFCLIPPreTrainedModel (line 887) | class TFCLIPPreTrainedModel(TFPreTrainedModel):
  class TFCLIPTextModel (line 1041) | class TFCLIPTextModel(TFCLIPPreTrainedModel):
    method __init__ (line 1044) | def __init__(self, config: CLIPTextConfig, *inputs, **kwargs):
    method call (line 1052) | def call(
  class TFCLIPVisionModel (line 1093) | class TFCLIPVisionModel(TFCLIPPreTrainedModel):
    method __init__ (line 1097) | def __init__(self, config: CLIPVisionConfig, *inputs, **kwargs):
    method call (line 1105) | def call(
  class TFCLIPModel (line 1148) | class TFCLIPModel(TFCLIPPreTrainedModel):
    method __init__ (line 1151) | def __init__(self, config: CLIPConfig, *inputs, **kwargs):
    method get_text_features (line 1158) | def get_text_features(
    method get_image_features (line 1198) | def get_image_features(
    method call (line 1241) | def call(
    method serving_output (line 1292) | def serving_output(self, output: TFCLIPOutput) -> TFCLIPOutput:

FILE: transformers/models/clip/processing_clip.py
  class CLIPProcessor (line 25) | class CLIPProcessor(ProcessorMixin):
    method __init__ (line 42) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 59) | def __call__(self, text=None, images=None, return_tensors=None, **kwar...
    method batch_decode (line 112) | def batch_decode(self, *args, **kwargs):
    method decode (line 119) | def decode(self, *args, **kwargs):
    method model_input_names (line 127) | def model_input_names(self):
    method feature_extractor_class (line 133) | def feature_extractor_class(self):
    method feature_extractor (line 141) | def feature_extractor(self):

FILE: transformers/models/clip/tokenization_clip.py
  function bytes_to_unicode (line 56) | def bytes_to_unicode():
  function get_pairs (line 80) | def get_pairs(word):
  function whitespace_clean (line 94) | def whitespace_clean(text):
  function whitespace_tokenize (line 101) | def whitespace_tokenize(text):
  class BasicTokenizer (line 111) | class BasicTokenizer(object):
    method __init__ (line 131) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 139) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 176) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 187) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 209) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 222) | def _is_chinese_char(self, cp):
    method _clean_text (line 246) | def _clean_text(self, text):
  class CLIPTokenizer (line 260) | class CLIPTokenizer(PreTrainedTokenizer):
    method __init__ (line 289) | def __init__(
    method vocab_size (line 340) | def vocab_size(self):
    method get_vocab (line 343) | def get_vocab(self):
    method build_inputs_with_special_tokens (line 346) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 373) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 401) | def create_token_type_ids_from_sequences(
    method bpe (line 424) | def bpe(self, token):
    method _tokenize (line 466) | def _tokenize(self, text):
    method _convert_token_to_id (line 481) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 485) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 489) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 496) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/clip/tokenization_clip_fast.py
  class CLIPTokenizerFast (line 50) | class CLIPTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 78) | def __init__(
    method _wrap_decode_method_backend_tokenizer (line 113) | def _wrap_decode_method_backend_tokenizer(self):
    method build_inputs_with_special_tokens (line 123) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 150) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 173) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/clipseg/configuration_clipseg.py
  class CLIPSegTextConfig (line 32) | class CLIPSegTextConfig(PretrainedConfig):
    method __init__ (line 86) | def __init__(
    method from_pretrained (line 119) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class CLIPSegVisionConfig (line 135) | class CLIPSegVisionConfig(PretrainedConfig):
    method __init__ (line 188) | def __init__(
    method from_pretrained (line 220) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class CLIPSegConfig (line 236) | class CLIPSegConfig(PretrainedConfig):
    method __init__ (line 303) | def __init__(
    method from_text_vision_configs (line 413) | def from_text_vision_configs(cls, text_config: CLIPSegTextConfig, visi...
    method to_dict (line 424) | def to_dict(self):

FILE: transformers/models/clipseg/convert_clipseg_original_pytorch_to_hf.py
  function get_clipseg_config (line 35) | def get_clipseg_config(model_name):
  function rename_key (line 51) | def rename_key(name):
  function convert_state_dict (line 114) | def convert_state_dict(orig_state_dict, config):
  function prepare_img (line 161) | def prepare_img():
  function convert_clipseg_checkpoint (line 167) | def convert_clipseg_checkpoint(model_name, checkpoint_path, pytorch_dump...

FILE: transformers/models/clipseg/modeling_clipseg.py
  function _expand_mask (line 51) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  function contrastive_loss (line 67) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  function clipseg_loss (line 72) | def clipseg_loss(similarity: torch.Tensor) -> torch.Tensor:
  class CLIPSegOutput (line 80) | class CLIPSegOutput(ModelOutput):
    method to_tuple (line 110) | def to_tuple(self) -> Tuple[Any]:
  class CLIPSegDecoderOutput (line 118) | class CLIPSegDecoderOutput(ModelOutput):
  class CLIPSegImageSegmentationOutput (line 138) | class CLIPSegImageSegmentationOutput(ModelOutput):
    method to_tuple (line 155) | def to_tuple(self) -> Tuple[Any]:
  class CLIPSegVisionEmbeddings (line 162) | class CLIPSegVisionEmbeddings(nn.Module):
    method __init__ (line 164) | def __init__(self, config: CLIPSegVisionConfig):
    method interpolate_position_embeddings (line 186) | def interpolate_position_embeddings(self, new_size):
    method forward (line 205) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class CLIPSegTextEmbeddings (line 224) | class CLIPSegTextEmbeddings(nn.Module):
    method __init__ (line 225) | def __init__(self, config: CLIPSegTextConfig):
    method forward (line 235) | def forward(
  class CLIPSegAttention (line 256) | class CLIPSegAttention(nn.Module):
    method __init__ (line 259) | def __init__(self, config):
    method _shape (line 278) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 281) | def forward(
  class CLIPSegMLP (line 361) | class CLIPSegMLP(nn.Module):
    method __init__ (line 362) | def __init__(self, config):
    method forward (line 369) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class CLIPSegEncoderLayer (line 377) | class CLIPSegEncoderLayer(nn.Module):
    method __init__ (line 378) | def __init__(self, config: CLIPSegConfig):
    method forward (line 386) | def forward(
  class CLIPSegPreTrainedModel (line 427) | class CLIPSegPreTrainedModel(PreTrainedModel):
    method _init_weights (line 438) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 481) | def _set_gradient_checkpointing(self, module, value=False):
  class CLIPSegEncoder (line 583) | class CLIPSegEncoder(nn.Module):
    method __init__ (line 592) | def __init__(self, config: CLIPSegConfig):
    method forward (line 598) | def forward(
  function _make_causal_mask (line 687) | def _make_causal_mask(
  class CLIPSegTextTransformer (line 704) | class CLIPSegTextTransformer(nn.Module):
    method __init__ (line 706) | def __init__(self, config: CLIPSegTextConfig):
    method forward (line 717) | def forward(
  class CLIPSegTextModel (line 783) | class CLIPSegTextModel(CLIPSegPreTrainedModel):
    method __init__ (line 788) | def __init__(self, config: CLIPSegTextConfig):
    method get_input_embeddings (line 794) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 797) | def set_input_embeddings(self, value):
    method forward (line 802) | def forward(
  class CLIPSegVisionTransformer (line 838) | class CLIPSegVisionTransformer(nn.Module):
    method __init__ (line 840) | def __init__(self, config: CLIPSegVisionConfig):
    method forward (line 853) | def forward(
  class CLIPSegVisionModel (line 898) | class CLIPSegVisionModel(CLIPSegPreTrainedModel):
    method __init__ (line 902) | def __init__(self, config: CLIPSegVisionConfig):
    method get_input_embeddings (line 908) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 913) | def forward(
  class CLIPSegModel (line 951) | class CLIPSegModel(CLIPSegPreTrainedModel):
    method __init__ (line 954) | def __init__(self, config: CLIPSegConfig):
    method get_text_features (line 987) | def get_text_features(
    method get_image_features (line 1034) | def get_image_features(
    method forward (line 1084) | def forward(
  class CLIPSegDecoderLayer (line 1176) | class CLIPSegDecoderLayer(nn.Module):
    method __init__ (line 1183) | def __init__(self, config: CLIPSegConfig):
    method forward (line 1191) | def forward(
  class CLIPSegDecoder (line 1233) | class CLIPSegDecoder(CLIPSegPreTrainedModel):
    method __init__ (line 1234) | def __init__(self, config: CLIPSegConfig):
    method forward (line 1276) | def forward(
  class CLIPSegForImageSegmentation (line 1339) | class CLIPSegForImageSegmentation(CLIPSegPreTrainedModel):
    method __init__ (line 1342) | def __init__(self, config: CLIPSegConfig):
    method get_conditional_embeddings (line 1355) | def get_conditional_embeddings(
    method forward (line 1386) | def forward(

FILE: transformers/models/clipseg/processing_clipseg.py
  class CLIPSegProcessor (line 25) | class CLIPSegProcessor(ProcessorMixin):
    method __init__ (line 42) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 59) | def __call__(self, text=None, images=None, visual_prompt=None, return_...
    method batch_decode (line 132) | def batch_decode(self, *args, **kwargs):
    method decode (line 139) | def decode(self, *args, **kwargs):
    method feature_extractor_class (line 147) | def feature_extractor_class(self):
    method feature_extractor (line 155) | def feature_extractor(self):

FILE: transformers/models/codegen/configuration_codegen.py
  class CodeGenConfig (line 44) | class CodeGenConfig(PretrainedConfig):
    method __init__ (line 107) | def __init__(
  class CodeGenOnnxConfig (line 154) | class CodeGenOnnxConfig(OnnxConfigWithPast):
    method __init__ (line 155) | def __init__(
    method inputs (line 168) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method num_layers (line 179) | def num_layers(self) -> int:
    method num_attention_heads (line 183) | def num_attention_heads(self) -> int:
    method generate_dummy_inputs (line 186) | def generate_dummy_inputs(
    method default_onnx_opset (line 231) | def default_onnx_opset(self) -> int:

FILE: transformers/models/codegen/modeling_codegen.py
  function create_sinusoidal_positions (line 55) | def create_sinusoidal_positions(num_pos: int, dim: int) -> torch.Tensor:
  function rotate_every_two (line 62) | def rotate_every_two(x: torch.Tensor) -> torch.Tensor:
  function apply_rotary_pos_emb (line 70) | def apply_rotary_pos_emb(tensor: torch.Tensor, sin: torch.Tensor, cos: t...
  class CodeGenAttention (line 76) | class CodeGenAttention(nn.Module):
    method __init__ (line 77) | def __init__(self, config):
    method _split_heads (line 107) | def _split_heads(self, x, n_head, dim_head, mp_num):
    method _merge_heads (line 112) | def _merge_heads(self, tensor, num_attention_heads, attn_head_size):
    method _attn (line 125) | def _attn(
    method forward (line 166) | def forward(
  class CodeGenMLP (line 245) | class CodeGenMLP(nn.Module):
    method __init__ (line 246) | def __init__(self, intermediate_size, config):  # in MLP: intermediate...
    method forward (line 256) | def forward(self, hidden_states: Optional[torch.FloatTensor]) -> torch...
  class CodeGenBlock (line 265) | class CodeGenBlock(nn.Module):
    method __init__ (line 266) | def __init__(self, config):
    method forward (line 273) | def forward(
  class CodeGenPreTrainedModel (line 308) | class CodeGenPreTrainedModel(PreTrainedModel):
    method __init__ (line 320) | def __init__(self, *inputs, **kwargs):
    method _init_weights (line 323) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 339) | def _set_gradient_checkpointing(self, module, value=False):
  class CodeGenModel (line 409) | class CodeGenModel(CodeGenPreTrainedModel):
    method __init__ (line 410) | def __init__(self, config):
    method get_input_embeddings (line 426) | def get_input_embeddings(self):
    method set_input_embeddings (line 429) | def set_input_embeddings(self, new_embeddings):
    method forward (line 438) | def forward(
  class CodeGenForCausalLM (line 602) | class CodeGenForCausalLM(CodeGenPreTrainedModel):
    method __init__ (line 605) | def __init__(self, config):
    method get_output_embeddings (line 613) | def get_output_embeddings(self):
    method set_output_embeddings (line 616) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 619) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method forward (line 652) | def forward(
    method _reorder_cache (line 721) | def _reorder_cache(

FILE: transformers/models/codegen/tokenization_codegen.py
  function bytes_to_unicode (line 60) | def bytes_to_unicode():
  function get_pairs (line 84) | def get_pairs(word):
  class CodeGenTokenizer (line 98) | class CodeGenTokenizer(PreTrainedTokenizer):
    method __init__ (line 153) | def __init__(
    method vocab_size (line 199) | def vocab_size(self):
    method get_vocab (line 202) | def get_vocab(self):
    method bpe (line 205) | def bpe(self, token):
    method build_inputs_with_special_tokens (line 247) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method _tokenize (line 260) | def _tokenize(self, text):
    method _convert_token_to_id (line 270) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 274) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 278) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 284) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method prepare_for_tokenization (line 313) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method decode (line 319) | def decode(
    method truncate (line 363) | def truncate(self, completion, truncate_before_pattern):

FILE: transformers/models/codegen/tokenization_codegen_fast.py
  class CodeGenTokenizerFast (line 63) | class CodeGenTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 122) | def __init__(
    method _batch_encode_plus (line 163) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 172) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method save_vocabulary (line 182) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method decode (line 186) | def decode(
    method truncate (line 231) | def truncate(self, completion, truncate_before_pattern):

FILE: transformers/models/conditional_detr/configuration_conditional_detr.py
  class ConditionalDetrConfig (line 37) | class ConditionalDetrConfig(PretrainedConfig):
    method __init__ (line 145) | def __init__(
    method num_attention_heads (line 234) | def num_attention_heads(self) -> int:
    method hidden_size (line 238) | def hidden_size(self) -> int:
  class ConditionalDetrOnnxConfig (line 242) | class ConditionalDetrOnnxConfig(OnnxConfig):
    method inputs (line 246) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 255) | def atol_for_validation(self) -> float:
    method default_onnx_opset (line 259) | def default_onnx_opset(self) -> int:

FILE: transformers/models/conditional_detr/convert_conditional_detr_original_pytorch_checkpoint_to_pytorch.py
  function rename_key (line 178) | def rename_key(state_dict, old, new):
  function rename_backbone_keys (line 183) | def rename_backbone_keys(state_dict):
  function read_in_q_k_v (line 195) | def read_in_q_k_v(state_dict, is_panoptic=False):
  function prepare_img (line 215) | def prepare_img():
  function convert_conditional_detr_checkpoint (line 223) | def convert_conditional_detr_checkpoint(model_name, pytorch_dump_folder_...

FILE: transformers/models/conditional_detr/feature_extraction_conditional_detr.py
  class ConditionalDetrFeatureExtractor (line 26) | class ConditionalDetrFeatureExtractor(ConditionalDetrImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/conditional_detr/image_processing_conditional_detr.py
  class AnnotionFormat (line 86) | class AnnotionFormat(ExplicitEnum):
  function get_size_with_aspect_ratio (line 95) | def get_size_with_aspect_ratio(image_size, size, max_size=None) -> Tuple...
  function get_resize_output_image_size (line 127) | def get_resize_output_image_size(
  function get_numpy_to_framework_fn (line 151) | def get_numpy_to_framework_fn(arr) -> Callable:
  function safe_squeeze (line 176) | def safe_squeeze(arr: np.ndarray, axis: Optional[int] = None) -> np.ndar...
  function normalize_annotation (line 190) | def normalize_annotation(annotation: Dict, image_size: Tuple[int, int]) ...
  function max_across_indices (line 205) | def max_across_indices(values: Iterable[Any]) -> List[Any]:
  function get_max_height_width (line 213) | def get_max_height_width(images: List[np.ndarray]) -> List[int]:
  function make_pixel_mask (line 229) | def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> ...
  function convert_coco_poly_to_mask (line 246) | def convert_coco_poly_to_mask(segmentations, height: int, width: int) ->...
  function prepare_coco_detection_annotation (line 281) | def prepare_coco_detection_annotation(image, target, return_segmentation...
  function masks_to_boxes (line 334) | def masks_to_boxes(masks: np.ndarray) -> np.ndarray:
  function prepare_coco_panoptic_annotation (line 369) | def prepare_coco_panoptic_annotation(
  function get_segmentation_image (line 407) | def get_segmentation_image(
  function get_mask_area (line 433) | def get_mask_area(seg_img: np.ndarray, target_size: Tuple[int, int], n_c...
  function score_labels_from_class_probabilities (line 443) | def score_labels_from_class_probabilities(logits: np.ndarray) -> Tuple[n...
  function post_process_panoptic_sample (line 452) | def post_process_panoptic_sample(
  function resize_annotation (line 539) | def resize_annotation(
  function binary_mask_to_rle (line 591) | def binary_mask_to_rle(mask):
  function convert_segmentation_to_rle (line 614) | def convert_segmentation_to_rle(segmentation):
  function remove_low_and_no_objects (line 636) | def remove_low_and_no_objects(masks, scores, labels, object_mask_thresho...
  function check_segment_validity (line 665) | def check_segment_validity(mask_labels, mask_probs, k, mask_threshold=0....
  function compute_segments (line 684) | def compute_segments(
  class ConditionalDetrImageProcessor (line 744) | class ConditionalDetrImageProcessor(BaseImageProcessor):
    method __init__ (line 782) | def __init__(
    method max_size (line 825) | def max_size(self):
    method from_dict (line 834) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method prepare_annotation (line 848) | def prepare_annotation(
    method prepare (line 874) | def prepare(self, image, target, return_segmentation_masks=False, mask...
    method convert_coco_poly_to_mask (line 884) | def convert_coco_poly_to_mask(self, *args, **kwargs):
    method prepare_coco_detection (line 891) | def prepare_coco_detection(self, *args, **kwargs):
    method prepare_coco_panoptic (line 898) | def prepare_coco_panoptic(self, *args, **kwargs):
    method resize (line 905) | def resize(
    method resize_annotation (line 939) | def resize_annotation(
    method rescale (line 953) | def rescale(
    method normalize (line 962) | def normalize(
    method normalize_annotation (line 975) | def normalize_annotation(self, annotation: Dict, image_size: Tuple[int...
    method pad_and_create_pixel_mask (line 983) | def pad_and_create_pixel_mask(
    method _pad_image (line 1017) | def _pad_image(
    method pad (line 1039) | def pad(
    method preprocess (line 1078) | def preprocess(
    method post_process (line 1284) | def post_process(self, outputs, target_sizes):
    method post_process_object_detection (line 1330) | def post_process_object_detection(
    method post_process_semantic_segmentation (line 1389) | def post_process_semantic_segmentation(self, outputs, target_sizes: Li...
    method post_process_instance_segmentation (line 1438) | def post_process_instance_segmentation(
    method post_process_panoptic_segmentation (line 1523) | def post_process_panoptic_segmentation(

FILE: transformers/models/conditional_detr/modeling_conditional_detr.py
  class ConditionalDetrDecoderOutput (line 65) | class ConditionalDetrDecoderOutput(BaseModelOutputWithCrossAttentions):
  class ConditionalDetrModelOutput (line 97) | class ConditionalDetrModelOutput(Seq2SeqModelOutput):
  class ConditionalDetrObjectDetectionOutput (line 140) | class ConditionalDetrObjectDetectionOutput(ModelOutput):
  class ConditionalDetrSegmentationOutput (line 204) | class ConditionalDetrSegmentationOutput(ModelOutput):
  class ConditionalDetrFrozenBatchNorm2d (line 274) | class ConditionalDetrFrozenBatchNorm2d(nn.Module):
    method __init__ (line 282) | def __init__(self, n):
    method _load_from_state_dict (line 289) | def _load_from_state_dict(
    method forward (line 300) | def forward(self, x):
  function replace_batch_norm (line 314) | def replace_batch_norm(m, name=""):
  class ConditionalDetrConvEncoder (line 330) | class ConditionalDetrConvEncoder(nn.Module):
    method __init__ (line 338) | def __init__(self, config):
    method forward (line 377) | def forward(self, pixel_values: torch.Tensor, pixel_mask: torch.Tensor):
  class ConditionalDetrConvModel (line 390) | class ConditionalDetrConvModel(nn.Module):
    method __init__ (line 395) | def __init__(self, conv_encoder, position_embedding):
    method forward (line 400) | def forward(self, pixel_values, pixel_mask):
  function _expand_mask (line 412) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, target_len: Opt...
  class ConditionalDetrSinePositionEmbedding (line 427) | class ConditionalDetrSinePositionEmbedding(nn.Module):
    method __init__ (line 433) | def __init__(self, embedding_dim=64, temperature=10000, normalize=Fals...
    method forward (line 444) | def forward(self, pixel_values, pixel_mask):
  class ConditionalDetrLearnedPositionEmbedding (line 465) | class ConditionalDetrLearnedPositionEmbedding(nn.Module):
    method __init__ (line 470) | def __init__(self, embedding_dim=256):
    method forward (line 475) | def forward(self, pixel_values, pixel_mask=None):
  function build_position_encoding (line 489) | def build_position_encoding(config):
  function gen_sine_position_embeddings (line 503) | def gen_sine_position_embeddings(pos_tensor):
  function inverse_sigmoid (line 517) | def inverse_sigmoid(x, eps=1e-5):
  class DetrAttention (line 525) | class DetrAttention(nn.Module):
    method __init__ (line 532) | def __init__(
    method _shape (line 557) | def _shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method with_pos_embed (line 560) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 563) | def forward(
  class ConditionalDetrAttention (line 656) | class ConditionalDetrAttention(nn.Module):
    method __init__ (line 664) | def __init__(
    method _qk_shape (line 693) | def _qk_shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method _v_shape (line 696) | def _v_shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method forward (line 699) | def forward(
  class ConditionalDetrEncoderLayer (line 774) | class ConditionalDetrEncoderLayer(nn.Module):
    method __init__ (line 775) | def __init__(self, config: ConditionalDetrConfig):
    method forward (line 791) | def forward(
  class ConditionalDetrDecoderLayer (line 844) | class ConditionalDetrDecoderLayer(nn.Module):
    method __init__ (line 845) | def __init__(self, config: ConditionalDetrConfig):
    method forward (line 886) | def forward(
  class ConditionalDetrClassificationHead (line 1016) | class ConditionalDetrClassificationHead(nn.Module):
    method __init__ (line 1019) | def __init__(self, input_dim: int, inner_dim: int, num_classes: int, p...
    method forward (line 1025) | def forward(self, hidden_states: torch.Tensor):
  class MLP (line 1035) | class MLP(nn.Module):
    method __init__ (line 1044) | def __init__(self, input_dim, hidden_dim, output_dim, num_layers):
    method forward (line 1050) | def forward(self, x):
  class ConditionalDetrPreTrainedModel (line 1057) | class ConditionalDetrPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1062) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1085) | def _set_gradient_checkpointing(self, module, value=False):
  class ConditionalDetrEncoder (line 1146) | class ConditionalDetrEncoder(ConditionalDetrPreTrainedModel):
    method __init__ (line 1161) | def __init__(self, config: ConditionalDetrConfig):
    method forward (line 1174) | def forward(
  class ConditionalDetrDecoder (line 1255) | class ConditionalDetrDecoder(ConditionalDetrPreTrainedModel):
    method __init__ (line 1270) | def __init__(self, config: ConditionalDetrConfig):
    method forward (line 1290) | def forward(
  class ConditionalDetrModel (line 1477) | class ConditionalDetrModel(ConditionalDetrPreTrainedModel):
    method __init__ (line 1478) | def __init__(self, config: ConditionalDetrConfig):
    method get_encoder (line 1497) | def get_encoder(self):
    method get_decoder (line 1500) | def get_decoder(self):
    method freeze_backbone (line 1503) | def freeze_backbone(self):
    method unfreeze_backbone (line 1507) | def unfreeze_backbone(self):
    method forward (line 1513) | def forward(
  class ConditionalDetrForObjectDetection (line 1646) | class ConditionalDetrForObjectDetection(ConditionalDetrPreTrainedModel):
    method __init__ (line 1647) | def __init__(self, config: ConditionalDetrConfig):
    method _set_aux_loss (line 1666) | def _set_aux_loss(self, outputs_class, outputs_coord):
    method forward (line 1674) | def forward(
  class ConditionalDetrForSegmentation (line 1834) | class ConditionalDetrForSegmentation(ConditionalDetrPreTrainedModel):
    method __init__ (line 1835) | def __init__(self, config: ConditionalDetrConfig):
    method forward (line 1858) | def forward(
  function _expand (line 2068) | def _expand(tensor, length: int):
  class ConditionalDetrMaskHeadSmallConv (line 2073) | class ConditionalDetrMaskHeadSmallConv(nn.Module):
    method __init__ (line 2078) | def __init__(self, dim, fpn_dims, context_dim):
    method forward (line 2112) | def forward(self, x: Tensor, bbox_mask: Tensor, fpns: List[Tensor]):
  class ConditionalDetrMHAttentionMap (line 2154) | class ConditionalDetrMHAttentionMap(nn.Module):
    method __init__ (line 2157) | def __init__(self, query_dim, hidden_dim, num_heads, dropout=0.0, bias...
    method forward (line 2168) | def forward(self, q, k, mask: Optional[Tensor] = None):
  function dice_loss (line 2183) | def dice_loss(inputs, targets, num_boxes):
  function sigmoid_focal_loss (line 2203) | def sigmoid_focal_loss(inputs, targets, num_boxes, alpha: float = 0.25, ...
  class ConditionalDetrLoss (line 2234) | class ConditionalDetrLoss(nn.Module):
    method __init__ (line 2252) | def __init__(self, matcher, num_classes, focal_alpha, losses):
    method loss_labels (line 2260) | def loss_labels(self, outputs, targets, indices, num_boxes):
    method loss_cardinality (line 2295) | def loss_cardinality(self, outputs, targets, indices, num_boxes):
    method loss_boxes (line 2311) | def loss_boxes(self, outputs, targets, indices, num_boxes):
    method loss_masks (line 2336) | def loss_masks(self, outputs, targets, indices, num_boxes):
    method _get_source_permutation_idx (line 2370) | def _get_source_permutation_idx(self, indices):
    method _get_target_permutation_idx (line 2377) | def _get_target_permutation_idx(self, indices):
    method get_loss (line 2384) | def get_loss(self, loss, outputs, targets, indices, num_boxes):
    method forward (line 2396) | def forward(self, outputs, targets):
  class ConditionalDetrMLPPredictionHead (line 2442) | class ConditionalDetrMLPPredictionHead(nn.Module):
    method __init__ (line 2451) | def __init__(self, input_dim, hidden_dim, output_dim, num_layers):
    method forward (line 2457) | def forward(self, x):
  class ConditionalDetrHungarianMatcher (line 2464) | class ConditionalDetrHungarianMatcher(nn.Module):
    method __init__ (line 2481) | def __init__(self, class_cost: float = 1, bbox_cost: float = 1, giou_c...
    method forward (line 2492) | def forward(self, outputs, targets):
  function _upcast (line 2545) | def _upcast(t: Tensor) -> Tensor:
  function box_area (line 2554) | def box_area(boxes: Tensor) -> Tensor:
  function box_iou (line 2571) | def box_iou(boxes1, boxes2):
  function generalized_box_iou (line 2588) | def generalized_box_iou(boxes1, boxes2):
  function _max_by_axis (line 2613) | def _max_by_axis(the_list):
  class NestedTensor (line 2623) | class NestedTensor(object):
    method __init__ (line 2624) | def __init__(self, tensors, mask: Optional[Tensor]):
    method to (line 2628) | def to(self, device):
    method decompose (line 2637) | def decompose(self):
    method __repr__ (line 2640) | def __repr__(self):
  function nested_tensor_from_tensor_list (line 2645) | def nested_tensor_from_tensor_list(tensor_list: List[Tensor]):

FILE: transformers/models/convbert/configuration_convbert.py
  class ConvBertConfig (line 37) | class ConvBertConfig(PretrainedConfig):
    method __init__ (line 101) | def __init__(
  class ConvBertOnnxConfig (line 152) | class ConvBertOnnxConfig(OnnxConfig):
    method inputs (line 154) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/convbert/convert_convbert_original_tf1_checkpoint_to_pytorch_and_tf2.py
  function convert_orig_tf1_checkpoint_to_pytorch (line 26) | def convert_orig_tf1_checkpoint_to_pytorch(tf_checkpoint_path, convbert_...

FILE: transformers/models/convbert/modeling_convbert.py
  function load_tf_weights_in_convbert (line 56) | def load_tf_weights_in_convbert(model, config, tf_checkpoint_path):
  class ConvBertEmbeddings (line 180) | class ConvBertEmbeddings(nn.Module):
    method __init__ (line 183) | def __init__(self, config):
    method forward (line 199) | def forward(
  class ConvBertPreTrainedModel (line 238) | class ConvBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 251) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 267) | def _set_gradient_checkpointing(self, module, value=False):
  class SeparableConv1D (line 272) | class SeparableConv1D(nn.Module):
    method __init__ (line 275) | def __init__(self, config, input_filters, output_filters, kernel_size,...
    method forward (line 291) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ConvBertSelfAttention (line 298) | class ConvBertSelfAttention(nn.Module):
    method __init__ (line 299) | def __init__(self, config):
    method transpose_for_scores (line 338) | def transpose_for_scores(self, x):
    method forward (line 343) | def forward(
  class ConvBertSelfOutput (line 426) | class ConvBertSelfOutput(nn.Module):
    method __init__ (line 427) | def __init__(self, config):
    method forward (line 433) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ConvBertAttention (line 440) | class ConvBertAttention(nn.Module):
    method __init__ (line 441) | def __init__(self, config):
    method prune_heads (line 447) | def prune_heads(self, heads):
    method forward (line 465) | def forward(
  class GroupedLinearLayer (line 485) | class GroupedLinearLayer(nn.Module):
    method __init__ (line 486) | def __init__(self, input_size, output_size, num_groups):
    method forward (line 496) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ConvBertIntermediate (line 507) | class ConvBertIntermediate(nn.Module):
    method __init__ (line 508) | def __init__(self, config):
    method forward (line 521) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ConvBertOutput (line 527) | class ConvBertOutput(nn.Module):
    method __init__ (line 528) | def __init__(self, config):
    method forward (line 539) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ConvBertLayer (line 546) | class ConvBertLayer(nn.Module):
    method __init__ (line 547) | def __init__(self, config):
    method forward (line 561) | def forward(
    method feed_forward_chunk (line 601) | def feed_forward_chunk(self, attention_output):
  class ConvBertEncoder (line 607) | class ConvBertEncoder(nn.Module):
    method __init__ (line 608) | def __init__(self, config):
    method forward (line 614) | def forward(
  class ConvBertPredictionHeadTransform (line 682) | class ConvBertPredictionHeadTransform(nn.Module):
    method __init__ (line 683) | def __init__(self, config):
    method forward (line 692) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ConvBertModel (line 767) | class ConvBertModel(ConvBertPreTrainedModel):
    method __init__ (line 770) | def __init__(self, config):
    method get_input_embeddings (line 782) | def get_input_embeddings(self):
    method set_input_embeddings (line 785) | def set_input_embeddings(self, value):
    method _prune_heads (line 788) | def _prune_heads(self, heads_to_prune):
    method forward (line 802) | def forward(
  class ConvBertGeneratorPredictions (line 864) | class ConvBertGeneratorPredictions(nn.Module):
    method __init__ (line 867) | def __init__(self, config):
    method forward (line 873) | def forward(self, generator_hidden_states: torch.FloatTensor) -> torch...
  class ConvBertForMaskedLM (line 882) | class ConvBertForMaskedLM(ConvBertPreTrainedModel):
    method __init__ (line 885) | def __init__(self, config):
    method get_output_embeddings (line 895) | def get_output_embeddings(self):
    method set_output_embeddings (line 898) | def set_output_embeddings(self, word_embeddings):
    method forward (line 907) | def forward(
  class ConvBertClassificationHead (line 962) | class ConvBertClassificationHead(nn.Module):
    method __init__ (line 965) | def __init__(self, config):
    method forward (line 976) | def forward(self, hidden_states: torch.Tensor, **kwargs) -> torch.Tensor:
  class ConvBertForSequenceClassification (line 993) | class ConvBertForSequenceClassification(ConvBertPreTrainedModel):
    method __init__ (line 996) | def __init__(self, config):
    method forward (line 1012) | def forward(
  class ConvBertForMultipleChoice (line 1090) | class ConvBertForMultipleChoice(ConvBertPreTrainedModel):
    method __init__ (line 1093) | def __init__(self, config):
    method forward (line 1111) | def forward(
  class ConvBertForTokenClassification (line 1185) | class ConvBertForTokenClassification(ConvBertPreTrainedModel):
    method __init__ (line 1188) | def __init__(self, config):
    method forward (line 1208) | def forward(
  class ConvBertForQuestionAnswering (line 1268) | class ConvBertForQuestionAnswering(ConvBertPreTrainedModel):
    method __init__ (line 1271) | def __init__(self, config):
    method forward (line 1287) | def forward(

FILE: transformers/models/convbert/modeling_tf_convbert.py
  class TFConvBertEmbeddings (line 71) | class TFConvBertEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 74) | def __init__(self, config: ConvBertConfig, **kwargs):
    method build (line 84) | def build(self, input_shape: tf.TensorShape):
    method call (line 109) | def call(
  class TFConvBertSelfAttention (line 150) | class TFConvBertSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 151) | def __init__(self, config, **kwargs):
    method transpose_for_scores (line 212) | def transpose_for_scores(self, x, batch_size):
    method call (line 217) | def call(self, hidden_states, attention_mask, head_mask, output_attent...
  class TFConvBertSelfOutput (line 301) | class TFConvBertSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 302) | def __init__(self, config, **kwargs):
    method call (line 311) | def call(self, hidden_states, input_tensor, training=False):
  class TFConvBertAttention (line 319) | class TFConvBertAttention(tf.keras.layers.Layer):
    method __init__ (line 320) | def __init__(self, config, **kwargs):
    method prune_heads (line 326) | def prune_heads(self, heads):
    method call (line 329) | def call(self, input_tensor, attention_mask, head_mask, output_attenti...
  class GroupedLinearLayer (line 339) | class GroupedLinearLayer(tf.keras.layers.Layer):
    method __init__ (line 340) | def __init__(self, input_size, output_size, num_groups, kernel_initial...
    method build (line 349) | def build(self, input_shape=None):
    method call (line 362) | def call(self, hidden_states):
  class TFConvBertIntermediate (line 372) | class TFConvBertIntermediate(tf.keras.layers.Layer):
    method __init__ (line 373) | def __init__(self, config, **kwargs):
    method call (line 393) | def call(self, hidden_states):
  class TFConvBertOutput (line 400) | class TFConvBertOutput(tf.keras.layers.Layer):
    method __init__ (line 401) | def __init__(self, config, **kwargs):
    method call (line 419) | def call(self, hidden_states, input_tensor, training=False):
  class TFConvBertLayer (line 427) | class TFConvBertLayer(tf.keras.layers.Layer):
    method __init__ (line 428) | def __init__(self, config, **kwargs):
    method call (line 435) | def call(self, hidden_states, attention_mask, head_mask, output_attent...
  class TFConvBertEncoder (line 447) | class TFConvBertEncoder(tf.keras.layers.Layer):
    method __init__ (line 448) | def __init__(self, config, **kwargs):
    method call (line 453) | def call(
  class TFConvBertPredictionHeadTransform (line 490) | class TFConvBertPredictionHeadTransform(tf.keras.layers.Layer):
    method __init__ (line 491) | def __init__(self, config, **kwargs):
    method call (line 505) | def call(self, hidden_states):
  class TFConvBertMainLayer (line 514) | class TFConvBertMainLayer(tf.keras.layers.Layer):
    method __init__ (line 517) | def __init__(self, config, **kwargs):
    method get_input_embeddings (line 528) | def get_input_embeddings(self):
    method set_input_embeddings (line 531) | def set_input_embeddings(self, value):
    method _prune_heads (line 535) | def _prune_heads(self, heads_to_prune):
    method get_extended_attention_mask (line 542) | def get_extended_attention_mask(self, attention_mask, input_shape, dty...
    method get_head_mask (line 563) | def get_head_mask(self, head_mask):
    method call (line 572) | def call(
  class TFConvBertPreTrainedModel (line 620) | class TFConvBertPreTrainedModel(TFPreTrainedModel):
  class TFConvBertModel (line 732) | class TFConvBertModel(TFConvBertPreTrainedModel):
    method __init__ (line 733) | def __init__(self, config, *inputs, **kwargs):
    method call (line 745) | def call(
  class TFConvBertMaskedLMHead (line 774) | class TFConvBertMaskedLMHead(tf.keras.layers.Layer):
    method __init__ (line 775) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 782) | def build(self, input_shape):
    method get_output_embeddings (line 787) | def get_output_embeddings(self):
    method set_output_embeddings (line 790) | def set_output_embeddings(self, value):
    method get_bias (line 794) | def get_bias(self):
    method set_bias (line 797) | def set_bias(self, value):
    method call (line 801) | def call(self, hidden_states):
  class TFConvBertGeneratorPredictions (line 811) | class TFConvBertGeneratorPredictions(tf.keras.layers.Layer):
    method __init__ (line 812) | def __init__(self, config, **kwargs):
    method call (line 818) | def call(self, generator_hidden_states, training=False):
  class TFConvBertForMaskedLM (line 827) | class TFConvBertForMaskedLM(TFConvBertPreTrainedModel, TFMaskedLanguageM...
    method __init__ (line 828) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 842) | def get_lm_head(self):
    method get_prefix_bias_name (line 845) | def get_prefix_bias_name(self):
    method call (line 855) | def call(
  class TFConvBertClassificationHead (line 905) | class TFConvBertClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 908) | def __init__(self, config, **kwargs):
    method call (line 924) | def call(self, hidden_states, **kwargs):
  class TFConvBertForSequenceClassification (line 941) | class TFConvBertForSequenceClassification(TFConvBertPreTrainedModel, TFS...
    method __init__ (line 942) | def __init__(self, config, *inputs, **kwargs):
    method call (line 955) | def call(
  class TFConvBertForMultipleChoice (line 1010) | class TFConvBertForMultipleChoice(TFConvBertPreTrainedModel, TFMultipleC...
    method __init__ (line 1011) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1031) | def call(
  class TFConvBertForTokenClassification (line 1103) | class TFConvBertForTokenClassification(TFConvBertPreTrainedModel, TFToke...
    method __init__ (line 1104) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1124) | def call(
  class TFConvBertForQuestionAnswering (line 1178) | class TFConvBertForQuestionAnswering(TFConvBertPreTrainedModel, TFQuesti...
    method __init__ (line 1179) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1195) | def call(

FILE: transformers/models/convbert/tokenization_convbert.py
  function load_vocab (line 54) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 66) | def whitespace_tokenize(text):
  class ConvBertTokenizer (line 76) | class ConvBertTokenizer(PreTrainedTokenizer):
    method __init__ (line 123) | def __init__(
    method do_lower_case (line 170) | def do_lower_case(self):
    method vocab_size (line 174) | def vocab_size(self):
    method get_vocab (line 177) | def get_vocab(self):
    method _tokenize (line 180) | def _tokenize(self, text):
    method _convert_token_to_id (line 193) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 197) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 201) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 206) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 231) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 259) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 288) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 310) | class BasicTokenizer(object):
    method __init__ (line 330) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 338) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 375) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 386) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 408) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 421) | def _is_chinese_char(self, cp):
    method _clean_text (line 445) | def _clean_text(self, text):
  class WordpieceTokenizer (line 460) | class WordpieceTokenizer(object):
    method __init__ (line 463) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 468) | def tokenize(self, text):

FILE: transformers/models/convbert/tokenization_convbert_fast.py
  class ConvBertTokenizerFast (line 55) | class ConvBertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 101) | def __init__(
    method build_inputs_with_special_tokens (line 143) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 167) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 196) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/convnext/configuration_convnext.py
  class ConvNextConfig (line 36) | class ConvNextConfig(BackboneConfigMixin, PretrainedConfig):
    method __init__ (line 92) | def __init__(
  class ConvNextOnnxConfig (line 128) | class ConvNextOnnxConfig(OnnxConfig):
    method inputs (line 132) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 140) | def atol_for_validation(self) -> float:

FILE: transformers/models/convnext/convert_convnext_to_pytorch.py
  function get_convnext_config (line 37) | def get_convnext_config(checkpoint_url):
  function rename_key (line 82) | def rename_key(name):
  function prepare_img (line 115) | def prepare_img():
  function convert_convnext_checkpoint (line 122) | def convert_convnext_checkpoint(checkpoint_url, pytorch_dump_folder_path):

FILE: transformers/models/convnext/feature_extraction_convnext.py
  class ConvNextFeatureExtractor (line 26) | class ConvNextFeatureExtractor(ConvNextImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/convnext/image_processing_convnext.py
  class ConvNextImageProcessor (line 50) | class ConvNextImageProcessor(BaseImageProcessor):
    method __init__ (line 88) | def __init__(
    method resize (line 116) | def resize(
    method rescale (line 161) | def rescale(
    method normalize (line 181) | def normalize(
    method preprocess (line 204) | def preprocess(

FILE: transformers/models/convnext/modeling_convnext.py
  function drop_path (line 64) | def drop_path(input, drop_prob: float = 0.0, training: bool = False):
  class ConvNextDropPath (line 85) | class ConvNextDropPath(nn.Module):
    method __init__ (line 88) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 92) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 95) | def extra_repr(self) -> str:
  class ConvNextLayerNorm (line 99) | class ConvNextLayerNorm(nn.Module):
    method __init__ (line 105) | def __init__(self, normalized_shape, eps=1e-6, data_format="channels_l...
    method forward (line 115) | def forward(self, x: torch.Tensor) -> torch.Tensor:
  class ConvNextEmbeddings (line 129) | class ConvNextEmbeddings(nn.Module):
    method __init__ (line 134) | def __init__(self, config):
    method forward (line 142) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class ConvNextLayer (line 153) | class ConvNextLayer(nn.Module):
    method __init__ (line 167) | def __init__(self, config, dim, drop_path=0):
    method forward (line 181) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class ConvNextStage (line 197) | class ConvNextStage(nn.Module):
    method __init__ (line 208) | def __init__(self, config, in_channels, out_channels, kernel_size=2, s...
    method forward (line 223) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class ConvNextEncoder (line 229) | class ConvNextEncoder(nn.Module):
    method __init__ (line 230) | def __init__(self, config):
    method forward (line 250) | def forward(
  class ConvNextPreTrainedModel (line 276) | class ConvNextPreTrainedModel(PreTrainedModel):
    method _init_weights (line 287) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 299) | def _set_gradient_checkpointing(self, module, value=False):
  class ConvNextModel (line 333) | class ConvNextModel(ConvNextPreTrainedModel):
    method __init__ (line 334) | def __init__(self, config):
    method forward (line 355) | def forward(
  class ConvNextForImageClassification (line 399) | class ConvNextForImageClassification(ConvNextPreTrainedModel):
    method __init__ (line 400) | def __init__(self, config):
    method forward (line 421) | def forward(
  class ConvNextBackbone (line 481) | class ConvNextBackbone(ConvNextPreTrainedModel, BackboneMixin):
    method __init__ (line 482) | def __init__(self, config):
    method forward (line 501) | def forward(

FILE: transformers/models/convnext/modeling_tf_convnext.py
  class TFConvNextDropPath (line 47) | class TFConvNextDropPath(tf.keras.layers.Layer):
    method __init__ (line 53) | def __init__(self, drop_path, **kwargs):
    method call (line 57) | def call(self, x, training=None):
  class TFConvNextEmbeddings (line 67) | class TFConvNextEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 72) | def __init__(self, config, **kwargs):
    method call (line 85) | def call(self, pixel_values):
  class TFConvNextLayer (line 105) | class TFConvNextLayer(tf.keras.layers.Layer):
    method __init__ (line 120) | def __init__(self, config, dim, drop_path=0.0, **kwargs):
    method build (line 158) | def build(self, input_shape: tf.TensorShape = None):
    method call (line 172) | def call(self, hidden_states, training=False):
  class TFConvNextStage (line 187) | class TFConvNextStage(tf.keras.layers.Layer):
    method __init__ (line 198) | def __init__(
    method call (line 236) | def call(self, hidden_states):
  class TFConvNextEncoder (line 244) | class TFConvNextEncoder(tf.keras.layers.Layer):
    method __init__ (line 245) | def __init__(self, config, **kwargs):
    method call (line 266) | def call(self, hidden_states, output_hidden_states=False, return_dict=...
  class TFConvNextMainLayer (line 285) | class TFConvNextMainLayer(tf.keras.layers.Layer):
    method __init__ (line 288) | def __init__(self, config: ConvNextConfig, add_pooling_layer: bool = T...
    method call (line 300) | def call(
  class TFConvNextPreTrainedModel (line 344) | class TFConvNextPreTrainedModel(TFPreTrainedModel):
  class TFConvNextModel (line 416) | class TFConvNextModel(TFConvNextPreTrainedModel):
    method __init__ (line 417) | def __init__(self, config, *inputs, add_pooling_layer=True, **kwargs):
    method call (line 424) | def call(
  class TFConvNextForImageClassification (line 483) | class TFConvNextForImageClassification(TFConvNextPreTrainedModel, TFSequ...
    method __init__ (line 484) | def __init__(self, config: ConvNextConfig, *inputs, **kwargs):
    method call (line 501) | def call(

FILE: transformers/models/convnextv2/configuration_convnextv2.py
  class ConvNextV2Config (line 30) | class ConvNextV2Config(BackboneConfigMixin, PretrainedConfig):
    method __init__ (line 84) | def __init__(

FILE: transformers/models/convnextv2/convert_convnextv2_to_pytorch.py
  function get_convnextv2_config (line 37) | def get_convnextv2_config(checkpoint_url):
  function rename_key (line 82) | def rename_key(name):
  function prepare_img (line 117) | def prepare_img():
  function convert_preprocessor (line 123) | def convert_preprocessor(checkpoint_url):
  function convert_convnextv2_checkpoint (line 144) | def convert_convnextv2_checkpoint(checkpoint_url, pytorch_dump_folder_pa...

FILE: transformers/models/convnextv2/modeling_convnextv2.py
  function drop_path (line 64) | def drop_path(input, drop_prob: float = 0.0, training: bool = False):
  class ConvNextV2DropPath (line 85) | class ConvNextV2DropPath(nn.Module):
    method __init__ (line 88) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 92) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 95) | def extra_repr(self) -> str:
  class ConvNextV2GRN (line 99) | class ConvNextV2GRN(nn.Module):
    method __init__ (line 102) | def __init__(self, dim: int):
    method forward (line 107) | def forward(self, hidden_states: torch.FloatTensor) -> torch.FloatTensor:
  class ConvNextV2LayerNorm (line 117) | class ConvNextV2LayerNorm(nn.Module):
    method __init__ (line 123) | def __init__(self, normalized_shape, eps=1e-6, data_format="channels_l...
    method forward (line 133) | def forward(self, x: torch.Tensor) -> torch.Tensor:
  class ConvNextV2Embeddings (line 148) | class ConvNextV2Embeddings(nn.Module):
    method __init__ (line 153) | def __init__(self, config):
    method forward (line 161) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class ConvNextV2Layer (line 172) | class ConvNextV2Layer(nn.Module):
    method __init__ (line 186) | def __init__(self, config, dim, drop_path=0):
    method forward (line 198) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class ConvNextV2Stage (line 216) | class ConvNextV2Stage(nn.Module):
    method __init__ (line 227) | def __init__(self, config, in_channels, out_channels, kernel_size=2, s...
    method forward (line 242) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class ConvNextV2Encoder (line 249) | class ConvNextV2Encoder(nn.Module):
    method __init__ (line 250) | def __init__(self, config):
    method forward (line 270) | def forward(
  class ConvNextV2PreTrainedModel (line 297) | class ConvNextV2PreTrainedModel(PreTrainedModel):
    method _init_weights (line 308) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 320) | def _set_gradient_checkpointing(self, module, value=False):
  class ConvNextV2Model (line 354) | class ConvNextV2Model(ConvNextV2PreTrainedModel):
    method __init__ (line 355) | def __init__(self, config):
    method forward (line 376) | def forward(
  class ConvNextV2ForImageClassification (line 421) | class ConvNextV2ForImageClassification(ConvNextV2PreTrainedModel):
    method __init__ (line 422) | def __init__(self, config):
    method forward (line 443) | def forward(
  class ConvNextV2Backbone (line 504) | class ConvNextV2Backbone(ConvNextV2PreTrainedModel, BackboneMixin):
    method __init__ (line 505) | def __init__(self, config):
    method forward (line 524) | def forward(

FILE: transformers/models/cpm/tokenization_cpm.py
  class CpmTokenizer (line 38) | class CpmTokenizer(PreTrainedTokenizer):
    method __init__ (line 41) | def __init__(
    method vocab_size (line 162) | def vocab_size(self):
    method get_vocab (line 166) | def get_vocab(self):
    method __getstate__ (line 172) | def __getstate__(self):
    method __setstate__ (line 178) | def __setstate__(self, d):
    method preprocess_text (line 189) | def preprocess_text(self, inputs):
    method _tokenize (line 205) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 226) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 231) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 236) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 242) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 268) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 297) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 328) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method _decode (line 345) | def _decode(self, *args, **kwargs):

FILE: transformers/models/cpm/tokenization_cpm_fast.py
  class CpmTokenizerFast (line 38) | class CpmTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 41) | def __init__(
    method build_inputs_with_special_tokens (line 157) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 183) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 214) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method _batch_encode_plus (line 233) | def _batch_encode_plus(self, batch_text_or_text_pairs, *args, **kwargs):
    method _decode (line 240) | def _decode(self, *args, **kwargs):

FILE: transformers/models/cpmant/configuration_cpmant.py
  class CpmAntConfig (line 29) | class CpmAntConfig(PretrainedConfig):
    method __init__ (line 89) | def __init__(

FILE: transformers/models/cpmant/modeling_cpmant.py
  class CpmAntLayerNorm (line 45) | class CpmAntLayerNorm(nn.Module):
    method __init__ (line 50) | def __init__(self, config: CpmAntConfig):
    method forward (line 57) | def forward(self, hidden_states: torch.Tensor):
  class CpmAntAttention (line 70) | class CpmAntAttention(nn.Module):
    method __init__ (line 71) | def __init__(self, config: CpmAntConfig):
    method forward (line 90) | def forward(
  class CpmAntSelfAttentionBlock (line 174) | class CpmAntSelfAttentionBlock(nn.Module):
    method __init__ (line 175) | def __init__(self, config: CpmAntConfig):
    method forward (line 184) | def forward(
  class CpmAntDenseGatedACT (line 223) | class CpmAntDenseGatedACT(nn.Module):
    method __init__ (line 224) | def __init__(self, config: CpmAntConfig):
    method forward (line 230) | def forward(self, hidden_states: torch.Tensor):
  class CpmAntFeedForward (line 243) | class CpmAntFeedForward(nn.Module):
    method __init__ (line 244) | def __init__(self, config: CpmAntConfig):
    method forward (line 254) | def forward(self, hidden_states: torch.Tensor):
  class CpmAntFFNBlock (line 269) | class CpmAntFFNBlock(nn.Module):
    method __init__ (line 270) | def __init__(self, config: CpmAntConfig):
    method forward (line 279) | def forward(
  class CpmAntTransformerBlock (line 296) | class CpmAntTransformerBlock(nn.Module):
    method __init__ (line 297) | def __init__(self, config: CpmAntConfig):
    method forward (line 302) | def forward(
  class CpmAntEncoder (line 343) | class CpmAntEncoder(nn.Module):
    method __init__ (line 344) | def __init__(self, config: CpmAntConfig):
    method forward (line 351) | def forward(
  class CpmAntIntermediate (line 409) | class CpmAntIntermediate(nn.Module):
    method __init__ (line 410) | def __init__(self, config):
    method forward (line 418) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class CpmAntSegmentPositionEmbedding (line 424) | class CpmAntSegmentPositionEmbedding(nn.Module):
    method __init__ (line 425) | def __init__(self, config: CpmAntConfig):
    method forward (line 440) | def forward(
    method _segment_relative_position_bucket (line 492) | def _segment_relative_position_bucket(self, query_segment, key_segment):
    method _position_bucket (line 495) | def _position_bucket(self, relative_position, num_buckets=32, max_dist...
  class CpmAntOutput (line 517) | class CpmAntOutput(nn.Module):
    method __init__ (line 518) | def __init__(self, config):
    method forward (line 524) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class CpmAntPreTrainedModel (line 531) | class CpmAntPreTrainedModel(PreTrainedModel):
    method _init_weights (line 542) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 560) | def _set_gradient_checkpointing(self, module, value=False):
  class CpmAntModel (line 604) | class CpmAntModel(CpmAntPreTrainedModel):
    method __init__ (line 605) | def __init__(self, config: CpmAntConfig):
    method get_input_embeddings (line 618) | def get_input_embeddings(self):
    method set_input_embeddings (line 621) | def set_input_embeddings(self, embeddings, **kwargs):
    method _prepare_attention_mask (line 624) | def _prepare_attention_mask(self, input_ids, span, context, length):
    method forward (line 648) | def forward(
  class CpmAntForCausalLM (line 751) | class CpmAntForCausalLM(CpmAntPreTrainedModel):
    method __init__ (line 754) | def __init__(self, config: CpmAntConfig):
    method forward (line 770) | def forward(
    method get_input_embeddings (line 851) | def get_input_embeddings(self):
    method set_input_embeddings (line 854) | def set_input_embeddings(self, embeddings):
    method get_output_embeddings (line 857) | def get_output_embeddings(self):
    method set_output_embeddings (line 860) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 863) | def prepare_inputs_for_generation(self, input_ids, **kwargs):
    method _reorder_cache (line 875) | def _reorder_cache(self, past_key_values, beam_idx):

FILE: transformers/models/cpmant/tokenization_cpmant.py
  function load_vocab (line 45) | def load_vocab(vocab_file):
  class WordpieceTokenizer (line 56) | class WordpieceTokenizer(object):
    method __init__ (line 57) | def __init__(self, vocab, unk_token="<unk>", max_input_chars_per_word=...
    method tokenize (line 62) | def tokenize(self, token):
  class CpmAntTokenizer (line 88) | class CpmAntTokenizer(PreTrainedTokenizer):
    method __init__ (line 119) | def __init__(
    method bod_token_id (line 161) | def bod_token_id(self):
    method eod_token_id (line 165) | def eod_token_id(self):
    method newline_id (line 169) | def newline_id(self):
    method vocab_size (line 173) | def vocab_size(self) -> int:
    method get_vocab (line 176) | def get_vocab(self):
    method _tokenize (line 179) | def _tokenize(self, text):
    method _decode (line 186) | def _decode(self, token_ids, **kwargs):
    method check (line 194) | def check(self, token):
    method convert_tokens_to_string (line 197) | def convert_tokens_to_string(self, tokens: List[str]) -> str:
    method _convert_token_to_id (line 200) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 204) | def _convert_id_to_token(self, index):
    method save_vocabulary (line 208) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 235) | def build_inputs_with_special_tokens(self, token_ids_0: List[int], tok...
    method get_special_tokens_mask (line 253) | def get_special_tokens_mask(

FILE: transformers/models/ctrl/configuration_ctrl.py
  class CTRLConfig (line 26) | class CTRLConfig(PretrainedConfig):
    method __init__ (line 87) | def __init__(

FILE: transformers/models/ctrl/modeling_ctrl.py
  function angle_defn (line 42) | def angle_defn(pos, i, d_model_size):
  function positional_encoding (line 47) | def positional_encoding(position, d_model_size, dtype):
  function scaled_dot_product_attention (line 62) | def scaled_dot_product_attention(q, k, v, mask, attention_mask=None, hea...
  class MultiHeadAttention (line 88) | class MultiHeadAttention(nn.Module):
    method __init__ (line 89) | def __init__(self, d_model_size, num_heads):
    method prune_heads (line 103) | def prune_heads(self, heads):
    method split_into_heads (line 120) | def split_into_heads(self, x, batch_size):
    method forward (line 124) | def forward(
  function point_wise_feed_forward_network (line 167) | def point_wise_feed_forward_network(d_model_size, dff):
  class EncoderLayer (line 171) | class EncoderLayer(nn.Module):
    method __init__ (line 172) | def __init__(self, d_model_size, num_heads, dff, rate=0.1):
    method forward (line 184) | def forward(
  class CTRLPreTrainedModel (line 212) | class CTRLPreTrainedModel(PreTrainedModel):
    method _init_weights (line 221) | def _init_weights(self, module):
  class CTRLModel (line 319) | class CTRLModel(CTRLPreTrainedModel):
    method __init__ (line 320) | def __init__(self, config):
    method get_input_embeddings (line 339) | def get_input_embeddings(self):
    method set_input_embeddings (line 342) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 345) | def _prune_heads(self, heads_to_prune):
    method forward (line 354) | def forward(
  class CTRLLMHeadModel (line 511) | class CTRLLMHeadModel(CTRLPreTrainedModel):
    method __init__ (line 514) | def __init__(self, config):
    method get_output_embeddings (line 522) | def get_output_embeddings(self):
    method set_output_embeddings (line 525) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 528) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method forward (line 537) | def forward(
    method _reorder_cache (line 627) | def _reorder_cache(
  class CTRLForSequenceClassification (line 653) | class CTRLForSequenceClassification(CTRLPreTrainedModel):
    method __init__ (line 654) | def __init__(self, config):
    method forward (line 665) | def forward(

FILE: transformers/models/ctrl/modeling_tf_ctrl.py
  function angle_defn (line 54) | def angle_defn(pos, i, d_model_size):
  function positional_encoding (line 59) | def positional_encoding(position, d_model_size):
  function scaled_dot_product_attention (line 70) | def scaled_dot_product_attention(q, k, v, mask, attention_mask=None, hea...
  class TFMultiHeadAttention (line 96) | class TFMultiHeadAttention(tf.keras.layers.Layer):
    method __init__ (line 97) | def __init__(self, d_model_size, num_heads, output_attentions=False, *...
    method split_into_heads (line 111) | def split_into_heads(self, x, batch_size):
    method call (line 115) | def call(self, v, k, q, mask, layer_past, attention_mask, head_mask, u...
  class TFPointWiseFeedForwardLayer (line 149) | class TFPointWiseFeedForwardLayer(tf.keras.layers.Layer):
    method __init__ (line 150) | def __init__(self, d_model_size, dff, **kwargs):
    method call (line 156) | def call(self, inputs, trainable=False):
  class TFEncoderLayer (line 163) | class TFEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 164) | def __init__(
    method call (line 182) | def call(self, x, mask, layer_past, attention_mask, head_mask, use_cac...
  class TFCTRLMainLayer (line 210) | class TFCTRLMainLayer(tf.keras.layers.Layer):
    method __init__ (line 213) | def __init__(self, config, **kwargs):
    method get_input_embeddings (line 246) | def get_input_embeddings(self):
    method set_input_embeddings (line 249) | def set_input_embeddings(self, value):
    method _prune_heads (line 253) | def _prune_heads(self, heads_to_prune):
    method call (line 260) | def call(
  class TFCTRLPreTrainedModel (line 401) | class TFCTRLPreTrainedModel(TFPreTrainedModel):
  class TFCTRLModel (line 524) | class TFCTRLModel(TFCTRLPreTrainedModel):
    method __init__ (line 525) | def __init__(self, config, *inputs, **kwargs):
    method call (line 536) | def call(
  class TFCTRLLMHead (line 568) | class TFCTRLLMHead(tf.keras.layers.Layer):
    method __init__ (line 569) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 579) | def build(self, input_shape=None):
    method get_output_embeddings (line 583) | def get_output_embeddings(self):
    method set_output_embeddings (line 586) | def set_output_embeddings(self, value):
    method get_bias (line 590) | def get_bias(self):
    method set_bias (line 593) | def set_bias(self, value):
    method call (line 597) | def call(self, hidden_states):
  class TFCTRLLMHeadModel (line 610) | class TFCTRLLMHeadModel(TFCTRLPreTrainedModel, TFCausalLanguageModelingL...
    method __init__ (line 611) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 619) | def get_lm_head(self):
    method get_prefix_bias_name (line 622) | def get_prefix_bias_name(self):
    method prepare_inputs_for_generation (line 626) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method call (line 640) | def call(
  class TFCTRLForSequenceClassification (line 715) | class TFCTRLForSequenceClassification(TFCTRLPreTrainedModel, TFSequenceC...
    method __init__ (line 716) | def __init__(self, config, *inputs, **kwargs):
    method get_output_embeddings (line 727) | def get_output_embeddings(self):
    method call (line 737) | def call(

FILE: transformers/models/ctrl/tokenization_ctrl.py
  function get_pairs (line 103) | def get_pairs(word):
  class CTRLTokenizer (line 119) | class CTRLTokenizer(PreTrainedTokenizer):
    method __init__ (line 141) | def __init__(self, vocab_file, merges_file, unk_token="<unk>", **kwargs):
    method vocab_size (line 154) | def vocab_size(self):
    method get_vocab (line 157) | def get_vocab(self):
    method bpe (line 160) | def bpe(self, token):
    method _tokenize (line 204) | def _tokenize(self, text):
    method _convert_token_to_id (line 214) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 218) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 222) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 227) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/cvt/configuration_cvt.py
  class CvtConfig (line 29) | class CvtConfig(PretrainedConfig):
    method __init__ (line 101) | def __init__(

FILE: transformers/models/cvt/convert_cvt_original_pytorch_checkpoint_to_pytorch.py
  function embeddings (line 30) | def embeddings(idx):
  function attention (line 65) | def attention(idx, cnt):
  function cls_token (line 257) | def cls_token(idx):
  function final (line 266) | def final():
  function convert_cvt_checkpoint (line 278) | def convert_cvt_checkpoint(cvt_model, image_size, cvt_file_name, pytorch...

FILE: transformers/models/cvt/modeling_cvt.py
  class BaseModelOutputWithCLSToken (line 60) | class BaseModelOutputWithCLSToken(ModelOutput):
  function drop_path (line 81) | def drop_path(input, drop_prob: float = 0.0, training: bool = False):
  class CvtDropPath (line 102) | class CvtDropPath(nn.Module):
    method __init__ (line 105) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 109) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 112) | def extra_repr(self) -> str:
  class CvtEmbeddings (line 116) | class CvtEmbeddings(nn.Module):
    method __init__ (line 121) | def __init__(self, patch_size, num_channels, embed_dim, stride, paddin...
    method forward (line 128) | def forward(self, pixel_values):
  class CvtConvEmbeddings (line 134) | class CvtConvEmbeddings(nn.Module):
    method __init__ (line 139) | def __init__(self, patch_size, num_channels, embed_dim, stride, padding):
    method forward (line 146) | def forward(self, pixel_values):
  class CvtSelfAttentionConvProjection (line 159) | class CvtSelfAttentionConvProjection(nn.Module):
    method __init__ (line 160) | def __init__(self, embed_dim, kernel_size, padding, stride):
    method forward (line 173) | def forward(self, hidden_state):
  class CvtSelfAttentionLinearProjection (line 179) | class CvtSelfAttentionLinearProjection(nn.Module):
    method forward (line 180) | def forward(self, hidden_state):
  class CvtSelfAttentionProjection (line 188) | class CvtSelfAttentionProjection(nn.Module):
    method __init__ (line 189) | def __init__(self, embed_dim, kernel_size, padding, stride, projection...
    method forward (line 195) | def forward(self, hidden_state):
  class CvtSelfAttention (line 201) | class CvtSelfAttention(nn.Module):
    method __init__ (line 202) | def __init__(
    method rearrange_for_multi_head_attention (line 243) | def rearrange_for_multi_head_attention(self, hidden_state):
    method forward (line 249) | def forward(self, hidden_state, height, width):
  class CvtSelfOutput (line 282) | class CvtSelfOutput(nn.Module):
    method __init__ (line 288) | def __init__(self, embed_dim, drop_rate):
    method forward (line 293) | def forward(self, hidden_state, input_tensor):
  class CvtAttention (line 299) | class CvtAttention(nn.Module):
    method __init__ (line 300) | def __init__(
    method prune_heads (line 332) | def prune_heads(self, heads):
    method forward (line 350) | def forward(self, hidden_state, height, width):
  class CvtIntermediate (line 356) | class CvtIntermediate(nn.Module):
    method __init__ (line 357) | def __init__(self, embed_dim, mlp_ratio):
    method forward (line 362) | def forward(self, hidden_state):
  class CvtOutput (line 368) | class CvtOutput(nn.Module):
    method __init__ (line 369) | def __init__(self, embed_dim, mlp_ratio, drop_rate):
    method forward (line 374) | def forward(self, hidden_state, input_tensor):
  class CvtLayer (line 381) | class CvtLayer(nn.Module):
    method __init__ (line 386) | def __init__(
    method forward (line 425) | def forward(self, hidden_state, height, width):
  class CvtStage (line 447) | class CvtStage(nn.Module):
    method __init__ (line 448) | def __init__(self, config, stage):
    method forward (line 488) | def forward(self, hidden_state):
  class CvtEncoder (line 508) | class CvtEncoder(nn.Module):
    method __init__ (line 509) | def __init__(self, config):
    method forward (line 516) | def forward(self, pixel_values, output_hidden_states=False, return_dic...
  class CvtPreTrainedModel (line 536) | class CvtPreTrainedModel(PreTrainedModel):
    method _init_weights (line 546) | def _init_weights(self, module):
  class CvtModel (line 590) | class CvtModel(CvtPreTrainedModel):
    method __init__ (line 591) | def __init__(self, config, add_pooling_layer=True):
    method _prune_heads (line 597) | def _prune_heads(self, heads_to_prune):
    method forward (line 613) | def forward(
  class CvtForImageClassification (line 651) | class CvtForImageClassification(CvtPreTrainedModel):
    method __init__ (line 652) | def __init__(self, config):
    method forward (line 673) | def forward(

FILE: transformers/models/cvt/modeling_tf_cvt.py
  class TFBaseModelOutputWithCLSToken (line 63) | class TFBaseModelOutputWithCLSToken(ModelOutput):
  class TFCvtDropPath (line 83) | class TFCvtDropPath(tf.keras.layers.Layer):
    method __init__ (line 89) | def __init__(self, drop_prob: float, **kwargs):
    method call (line 93) | def call(self, x: tf.Tensor, training=None):
  class TFCvtEmbeddings (line 103) | class TFCvtEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 106) | def __init__(
    method call (line 127) | def call(self, pixel_values: tf.Tensor, training: bool = False) -> tf....
  class TFCvtConvEmbeddings (line 133) | class TFCvtConvEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 136) | def __init__(self, config: CvtConfig, patch_size: int, embed_dim: int,...
    method call (line 152) | def call(self, pixel_values: tf.Tensor) -> tf.Tensor:
  class TFCvtSelfAttentionConvProjection (line 169) | class TFCvtSelfAttentionConvProjection(tf.keras.layers.Layer):
    method __init__ (line 172) | def __init__(self, config: CvtConfig, embed_dim: int, kernel_size: int...
    method call (line 188) | def call(self, hidden_state: tf.Tensor, training: bool = False) -> tf....
  class TFCvtSelfAttentionLinearProjection (line 194) | class TFCvtSelfAttentionLinearProjection(tf.keras.layers.Layer):
    method call (line 197) | def call(self, hidden_state: tf.Tensor) -> tf.Tensor:
  class TFCvtSelfAttentionProjection (line 205) | class TFCvtSelfAttentionProjection(tf.keras.layers.Layer):
    method __init__ (line 208) | def __init__(
    method call (line 225) | def call(self, hidden_state: tf.Tensor, training: bool = False) -> tf....
  class TFCvtSelfAttention (line 231) | class TFCvtSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 237) | def __init__(
    method rearrange_for_multi_head_attention (line 310) | def rearrange_for_multi_head_attention(self, hidden_state: tf.Tensor) ...
    method call (line 317) | def call(self, hidden_state: tf.Tensor, height: int, width: int, train...
  class TFCvtSelfOutput (line 352) | class TFCvtSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 355) | def __init__(self, config: CvtConfig, embed_dim: int, drop_rate: float...
    method call (line 362) | def call(self, hidden_state: tf.Tensor, training: bool = False) -> tf....
  class TFCvtAttention (line 368) | class TFCvtAttention(tf.keras.layers.Layer):
    method __init__ (line 371) | def __init__(
    method prune_heads (line 406) | def prune_heads(self, heads):
    method call (line 409) | def call(self, hidden_state: tf.Tensor, height: int, width: int, train...
  class TFCvtIntermediate (line 415) | class TFCvtIntermediate(tf.keras.layers.Layer):
    method __init__ (line 418) | def __init__(self, config: CvtConfig, embed_dim: int, mlp_ratio: int, ...
    method call (line 427) | def call(self, hidden_state: tf.Tensor) -> tf.Tensor:
  class TFCvtOutput (line 432) | class TFCvtOutput(tf.keras.layers.Layer):
    method __init__ (line 437) | def __init__(self, config: CvtConfig, embed_dim: int, drop_rate: int, ...
    method call (line 444) | def call(self, hidden_state: tf.Tensor, input_tensor: tf.Tensor, train...
  class TFCvtLayer (line 451) | class TFCvtLayer(tf.keras.layers.Layer):
    method __init__ (line 458) | def __init__(
    method call (line 506) | def call(self, hidden_state: tf.Tensor, height: int, width: int, train...
  class TFCvtStage (line 524) | class TFCvtStage(tf.keras.layers.Layer):
    method __init__ (line 536) | def __init__(self, config: CvtConfig, stage: int, **kwargs):
    method call (line 582) | def call(self, hidden_state: tf.Tensor, training: bool = False):
  class TFCvtEncoder (line 607) | class TFCvtEncoder(tf.keras.layers.Layer):
    method __init__ (line 618) | def __init__(self, config: CvtConfig, **kwargs):
    method call (line 625) | def call(
  class TFCvtMainLayer (line 660) | class TFCvtMainLayer(tf.keras.layers.Layer):
    method __init__ (line 665) | def __init__(self, config: CvtConfig, **kwargs):
    method call (line 671) | def call(
  class TFCvtPreTrainedModel (line 700) | class TFCvtPreTrainedModel(TFPreTrainedModel):
  class TFCvtModel (line 762) | class TFCvtModel(TFCvtPreTrainedModel):
    method __init__ (line 763) | def __init__(self, config: CvtConfig, *inputs, **kwargs):
    method call (line 771) | def call(
  class TFCvtForImageClassification (line 826) | class TFCvtForImageClassification(TFCvtPreTrainedModel, TFSequenceClassi...
    method __init__ (line 827) | def __init__(self, config: CvtConfig, *inputs, **kwargs):
    method call (line 847) | def call(

FILE: transformers/models/data2vec/configuration_data2vec_audio.py
  class Data2VecAudioConfig (line 31) | class Data2VecAudioConfig(PretrainedConfig):
    method __init__ (line 171) | def __init__(
    method inputs_to_logits_ratio (line 286) | def inputs_to_logits_ratio(self):

FILE: transformers/models/data2vec/configuration_data2vec_text.py
  class Data2VecTextConfig (line 31) | class Data2VecTextConfig(PretrainedConfig):
    method __init__ (line 100) | def __init__(
  class Data2VecTextOnnxConfig (line 141) | class Data2VecTextOnnxConfig(OnnxConfig):
    method inputs (line 143) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/data2vec/configuration_data2vec_vision.py
  class Data2VecVisionConfig (line 35) | class Data2VecVisionConfig(PretrainedConfig):
    method __init__ (line 116) | def __init__(
  class Data2VecVisionOnnxConfig (line 182) | class Data2VecVisionOnnxConfig(OnnxConfig):
    method inputs (line 186) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 194) | def atol_for_validation(self) -> float:

FILE: transformers/models/data2vec/convert_data2vec_audio_original_pytorch_checkpoint_to_pytorch.py
  function set_recursively (line 58) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function recursively_load_weights (line 87) | def recursively_load_weights(fairseq_model, hf_model, is_headless):
  function access_by_string (line 145) | def access_by_string(module, path):
  function set_weights (line 150) | def set_weights(full_name, module, fsq_value, hf_weight_path):
  function load_conv_layer (line 160) | def load_conv_layer(full_name, value, feature_extractor, unused_weights):
  function load_pos_conv_layer (line 178) | def load_pos_conv_layer(full_name, value, pos_conv_embeddings, unused_we...
  function convert_wav2vec2_checkpoint (line 195) | def convert_wav2vec2_checkpoint(

FILE: transformers/models/data2vec/convert_data2vec_text_original_pytorch_checkpoint_to_pytorch.py
  function convert_data2vec_checkpoint_to_pytorch (line 56) | def convert_data2vec_checkpoint_to_pytorch(

FILE: transformers/models/data2vec/convert_data2vec_vision_original_pytorch_checkpoint_to_pytorch.py
  function create_rename_keys (line 18) | def create_rename_keys(config, has_lm_head=False, is_semantic=False, hf_...
  function read_in_q_k_v (line 97) | def read_in_q_k_v(state_dict, config, has_lm_head=False, is_semantic=Fal...
  function get_args (line 139) | def get_args():
  function load_beit_model (line 150) | def load_beit_model(args, is_finetuned, is_large):
  function main (line 255) | def main():

FILE: transformers/models/data2vec/modeling_data2vec_audio.py
  function _compute_mask_indices (line 69) | def _compute_mask_indices(
  class Data2VecAudioConvLayer (line 188) | class Data2VecAudioConvLayer(nn.Module):
    method __init__ (line 189) | def __init__(self, config, layer_id=0):
    method forward (line 204) | def forward(self, hidden_states):
  class Data2VecAudioPadLayer (line 216) | class Data2VecAudioPadLayer(nn.Module):
    method __init__ (line 217) | def __init__(self, num_conv_pos_embeddings):
    method forward (line 221) | def forward(self, hidden_states):
  class Data2VecAudioPositionalConvLayer (line 227) | class Data2VecAudioPositionalConvLayer(nn.Module):
    method __init__ (line 228) | def __init__(self, config):
    method forward (line 243) | def forward(self, hidden_states):
  class Data2VecAudioPositionalConvEmbedding (line 254) | class Data2VecAudioPositionalConvEmbedding(nn.Module):
    method __init__ (line 255) | def __init__(self, config):
    method forward (line 261) | def forward(self, hidden_states):
  class Data2VecAudioFeatureEncoder (line 269) | class Data2VecAudioFeatureEncoder(nn.Module):
    method __init__ (line 272) | def __init__(self, config):
    method _freeze_parameters (line 281) | def _freeze_parameters(self):
    method forward (line 287) | def forward(self, input_values):
  class Data2VecAudioFeatureProjection (line 314) | class Data2VecAudioFeatureProjection(nn.Module):
    method __init__ (line 315) | def __init__(self, config):
    method forward (line 321) | def forward(self, hidden_states):
  class Data2VecAudioAttention (line 330) | class Data2VecAudioAttention(nn.Module):
    method __init__ (line 333) | def __init__(
    method _shape (line 360) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 363) | def forward(
  class Data2VecAudioFeedForward (line 485) | class Data2VecAudioFeedForward(nn.Module):
    method __init__ (line 486) | def __init__(self, config):
    method forward (line 499) | def forward(self, hidden_states):
  class Data2VecAudioEncoderLayer (line 510) | class Data2VecAudioEncoderLayer(nn.Module):
    method __init__ (line 511) | def __init__(self, config):
    method forward (line 524) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class Data2VecAudioEncoder (line 545) | class Data2VecAudioEncoder(nn.Module):
    method __init__ (line 546) | def __init__(self, config):
    method forward (line 555) | def forward(
  class Data2VecAudioAdapter (line 633) | class Data2VecAudioAdapter(nn.Module):
    method __init__ (line 634) | def __init__(self, config):
    method forward (line 647) | def forward(self, hidden_states):
  class Data2VecAudioAdapterLayer (line 665) | class Data2VecAudioAdapterLayer(nn.Module):
    method __init__ (line 666) | def __init__(self, config):
    method forward (line 676) | def forward(self, hidden_states):
  class Data2VecAudioPreTrainedModel (line 683) | class Data2VecAudioPreTrainedModel(PreTrainedModel):
    method _init_weights (line 695) | def _init_weights(self, module):
    method _get_feat_extract_output_lengths (line 721) | def _get_feat_extract_output_lengths(
    method _get_feature_vector_attention_mask (line 745) | def _get_feature_vector_attention_mask(
    method _set_gradient_checkpointing (line 765) | def _set_gradient_checkpointing(self, module, value=False):
  class Data2VecAudioModel (line 831) | class Data2VecAudioModel(Data2VecAudioPreTrainedModel):
    method __init__ (line 832) | def __init__(self, config: Data2VecAudioConfig):
    method freeze_feature_encoder (line 849) | def freeze_feature_encoder(self):
    method _mask_hidden_states (line 856) | def _mask_hidden_states(
    method forward (line 910) | def forward(
  class Data2VecAudioForCTC (line 967) | class Data2VecAudioForCTC(Data2VecAudioPreTrainedModel):
    method __init__ (line 968) | def __init__(self, config):
    method freeze_feature_extractor (line 989) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1001) | def freeze_feature_encoder(self):
    method forward (line 1017) | def forward(
  class Data2VecAudioForSequenceClassification (line 1096) | class Data2VecAudioForSequenceClassification(Data2VecAudioPreTrainedModel):
    method __init__ (line 1097) | def __init__(self, config):
    method freeze_feature_extractor (line 1114) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1126) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1133) | def freeze_base_model(self):
    method forward (line 1149) | def forward(
  class Data2VecAudioForAudioFrameClassification (line 1217) | class Data2VecAudioForAudioFrameClassification(Data2VecAudioPreTrainedMo...
    method __init__ (line 1218) | def __init__(self, config):
    method freeze_feature_extractor (line 1235) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1247) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1254) | def freeze_base_model(self):
    method forward (line 1270) | def forward(
  class AMSoftmaxLoss (line 1325) | class AMSoftmaxLoss(nn.Module):
    method __init__ (line 1326) | def __init__(self, input_dim, num_labels, scale=30.0, margin=0.4):
    method forward (line 1334) | def forward(self, hidden_states, labels):
  class TDNNLayer (line 1349) | class TDNNLayer(nn.Module):
    method __init__ (line 1350) | def __init__(self, config, layer_id=0):
    method forward (line 1360) | def forward(self, hidden_states):
  class Data2VecAudioForXVector (line 1381) | class Data2VecAudioForXVector(Data2VecAudioPreTrainedModel):
    method __init__ (line 1382) | def __init__(self, config):
    method freeze_feature_extractor (line 1401) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1413) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1420) | def freeze_base_model(self):
    method _get_tdnn_output_lengths (line 1428) | def _get_tdnn_output_lengths(self, input_lengths: Union[torch.LongTens...
    method forward (line 1451) | def forward(

FILE: transformers/models/data2vec/modeling_data2vec_text.py
  class Data2VecTextForTextEmbeddings (line 65) | class Data2VecTextForTextEmbeddings(nn.Module):
    method __init__ (line 71) | def __init__(self, config):
    method forward (line 94) | def forward(
    method create_position_ids_from_inputs_embeds (line 134) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class Data2VecTextSelfAttention (line 153) | class Data2VecTextSelfAttention(nn.Module):
    method __init__ (line 154) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 180) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 185) | def forward(
  class Data2VecTextSelfOutput (line 288) | class Data2VecTextSelfOutput(nn.Module):
    method __init__ (line 289) | def __init__(self, config):
    method forward (line 295) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class Data2VecTextAttention (line 303) | class Data2VecTextAttention(nn.Module):
    method __init__ (line 304) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 310) | def prune_heads(self, heads):
    method forward (line 328) | def forward(
  class Data2VecTextIntermediate (line 353) | class Data2VecTextIntermediate(nn.Module):
    method __init__ (line 354) | def __init__(self, config):
    method forward (line 362) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Data2VecTextOutput (line 369) | class Data2VecTextOutput(nn.Module):
    method __init__ (line 370) | def __init__(self, config):
    method forward (line 376) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class Data2VecTextLayer (line 384) | class Data2VecTextLayer(nn.Module):
    method __init__ (line 385) | def __init__(self, config):
    method forward (line 399) | def forward(
    method feed_forward_chunk (line 464) | def feed_forward_chunk(self, attention_output):
  class Data2VecTextEncoder (line 471) | class Data2VecTextEncoder(nn.Module):
    method __init__ (line 472) | def __init__(self, config):
    method forward (line 478) | def forward(
  class Data2VecTextPooler (line 570) | class Data2VecTextPooler(nn.Module):
    method __init__ (line 571) | def __init__(self, config):
    method forward (line 576) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Data2VecTextPreTrainedModel (line 585) | class Data2VecTextPreTrainedModel(PreTrainedModel):
    method _init_weights (line 596) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 614) | def _set_gradient_checkpointing(self, module, value=False):
    method update_keys_to_ignore (line 618) | def update_keys_to_ignore(self, config, del_keys_to_ignore):
  class Data2VecTextModel (line 701) | class Data2VecTextModel(Data2VecTextPreTrainedModel):
    method __init__ (line 719) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 731) | def get_input_embeddings(self):
    method set_input_embeddings (line 734) | def set_input_embeddings(self, value):
    method _prune_heads (line 737) | def _prune_heads(self, heads_to_prune):
    method forward (line 752) | def forward(
  class Data2VecTextForCausalLM (line 885) | class Data2VecTextForCausalLM(Data2VecTextPreTrainedModel):
    method __init__ (line 890) | def __init__(self, config):
    method get_output_embeddings (line 905) | def get_output_embeddings(self):
    method set_output_embeddings (line 908) | def set_output_embeddings(self, new_embeddings):
    method forward (line 913) | def forward(
    method prepare_inputs_for_generation (line 1019) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1031) | def _reorder_cache(self, past_key_values, beam_idx):
  class Data2VecTextForMaskedLM (line 1039) | class Data2VecTextForMaskedLM(Data2VecTextPreTrainedModel):
    method __init__ (line 1044) | def __init__(self, config):
    method get_output_embeddings (line 1062) | def get_output_embeddings(self):
    method set_output_embeddings (line 1065) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1075) | def forward(
  class Data2VecTextLMHead (line 1136) | class Data2VecTextLMHead(nn.Module):
    method __init__ (line 1139) | def __init__(self, config):
    method forward (line 1148) | def forward(self, features, **kwargs):
    method _tie_weights (line 1158) | def _tie_weights(self):
  class Data2VecTextForSequenceClassification (line 1174) | class Data2VecTextForSequenceClassification(Data2VecTextPreTrainedModel):
    method __init__ (line 1177) | def __init__(self, config):
    method forward (line 1194) | def forward(
  class Data2VecTextForMultipleChoice (line 1273) | class Data2VecTextForMultipleChoice(Data2VecTextPreTrainedModel):
    method __init__ (line 1276) | def __init__(self, config):
    method forward (line 1294) | def forward(
  class Data2VecTextForTokenClassification (line 1369) | class Data2VecTextForTokenClassification(Data2VecTextPreTrainedModel):
    method __init__ (line 1373) | def __init__(self, config):
    method forward (line 1393) | def forward(
  class Data2VecTextClassificationHead (line 1449) | class Data2VecTextClassificationHead(nn.Module):
    method __init__ (line 1452) | def __init__(self, config):
    method forward (line 1461) | def forward(self, features, **kwargs):
  class Data2VecTextForQuestionAnswering (line 1478) | class Data2VecTextForQuestionAnswering(Data2VecTextPreTrainedModel):
    method __init__ (line 1482) | def __init__(self, config):
    method forward (line 1498) | def forward(
  function create_position_ids_from_input_ids (line 1573) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/data2vec/modeling_data2vec_vision.py
  class Data2VecVisionModelOutputWithPooling (line 68) | class Data2VecVisionModelOutputWithPooling(BaseModelOutputWithPooling):
  function drop_path (line 94) | def drop_path(input: torch.Tensor, drop_prob: float = 0.0, training: boo...
  class Data2VecVisionDropPath (line 115) | class Data2VecVisionDropPath(nn.Module):
    method __init__ (line 118) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 122) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 125) | def extra_repr(self) -> str:
  class Data2VecVisionEmbeddings (line 130) | class Data2VecVisionEmbeddings(nn.Module):
    method __init__ (line 136) | def __init__(self, config: Data2VecVisionConfig) -> None:
    method forward (line 152) | def forward(self, pixel_values: torch.Tensor, bool_masked_pos: Optiona...
  class Data2VecVisionPatchEmbeddings (line 172) | class Data2VecVisionPatchEmbeddings(nn.Module):
    method __init__ (line 179) | def __init__(self, config):
    method forward (line 196) | def forward(self, pixel_values: torch.Tensor) -> torch.Tensor:
  class Data2VecVisionSelfAttention (line 212) | class Data2VecVisionSelfAttention(nn.Module):
    method __init__ (line 213) | def __init__(self, config: Data2VecVisionConfig, window_size: Optional...
    method transpose_for_scores (line 236) | def transpose_for_scores(self, x):
    method forward (line 241) | def forward(
  class Data2VecVisionSelfOutput (line 290) | class Data2VecVisionSelfOutput(nn.Module):
    method __init__ (line 296) | def __init__(self, config: Data2VecVisionConfig) -> None:
    method forward (line 301) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class Data2VecVisionAttention (line 309) | class Data2VecVisionAttention(nn.Module):
    method __init__ (line 310) | def __init__(self, config: Data2VecVisionConfig, window_size: Optional...
    method prune_heads (line 316) | def prune_heads(self, heads):
    method forward (line 334) | def forward(
  class Data2VecVisionIntermediate (line 350) | class Data2VecVisionIntermediate(nn.Module):
    method __init__ (line 351) | def __init__(self, config: Data2VecVisionConfig) -> None:
    method forward (line 359) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Data2VecVisionOutput (line 367) | class Data2VecVisionOutput(nn.Module):
    method __init__ (line 368) | def __init__(self, config: Data2VecVisionConfig) -> None:
    method forward (line 373) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Data2VecVisionLayer (line 381) | class Data2VecVisionLayer(nn.Module):
    method __init__ (line 384) | def __init__(
    method forward (line 404) | def forward(
  class Data2VecVisionRelativePositionBias (line 445) | class Data2VecVisionRelativePositionBias(nn.Module):
    method __init__ (line 446) | def __init__(self, config: Data2VecVisionConfig, window_size: tuple) -...
    method forward (line 475) | def forward(self) -> torch.Tensor:
  class Data2VecVisionEncoder (line 484) | class Data2VecVisionEncoder(nn.Module):
    method __init__ (line 485) | def __init__(self, config: Data2VecVisionConfig, window_size: Optional...
    method forward (line 507) | def forward(
  class Data2VecVisionPreTrainedModel (line 561) | class Data2VecVisionPreTrainedModel(PreTrainedModel):
    method _init_weights (line 572) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 588) | def _set_gradient_checkpointing(self, module, value=False):
  class Data2VecVisionModel (line 632) | class Data2VecVisionModel(Data2VecVisionPreTrainedModel):
    method __init__ (line 633) | def __init__(self, config: Data2VecVisionConfig, add_pooling_layer: bo...
    method get_input_embeddings (line 648) | def get_input_embeddings(self):
    method _prune_heads (line 651) | def _prune_heads(self, heads_to_prune):
    method forward (line 667) | def forward(
  class Data2VecVisionPooler (line 722) | class Data2VecVisionPooler(nn.Module):
    method __init__ (line 723) | def __init__(self, config: Data2VecVisionConfig) -> None:
    method forward (line 729) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Data2VecVisionForImageClassification (line 749) | class Data2VecVisionForImageClassification(Data2VecVisionPreTrainedModel):
    method __init__ (line 750) | def __init__(self, config: Data2VecVisionConfig) -> None:
    method forward (line 769) | def forward(
  class Data2VecVisionConvModule (line 832) | class Data2VecVisionConvModule(nn.Module):
    method __init__ (line 840) | def __init__(
    method forward (line 861) | def forward(self, input: torch.Tensor) -> torch.Tensor:
  class Data2VecVisionPyramidPoolingBlock (line 870) | class Data2VecVisionPyramidPoolingBlock(nn.Module):
    method __init__ (line 871) | def __init__(self, pool_scale: int, in_channels: int, channels: int) -...
    method forward (line 880) | def forward(self, input: torch.Tensor) -> torch.Tensor:
  class Data2VecVisionPyramidPoolingModule (line 888) | class Data2VecVisionPyramidPoolingModule(nn.Module):
    method __init__ (line 902) | def __init__(self, pool_scales: Tuple[int, ...], in_channels: int, cha...
    method forward (line 916) | def forward(self, x: torch.Tensor) -> List[torch.Tensor]:
  class Data2VecVisionUperHead (line 928) | class Data2VecVisionUperHead(nn.Module):
    method __init__ (line 936) | def __init__(self, config: Data2VecVisionConfig) -> None:
    method psp_forward (line 974) | def psp_forward(self, inputs):
    method forward (line 983) | def forward(self, encoder_hidden_states: torch.Tensor) -> torch.Tensor:
  class Data2VecVisionFCNHead (line 1014) | class Data2VecVisionFCNHead(nn.Module):
    method __init__ (line 1029) | def __init__(
    method forward (line 1067) | def forward(self, encoder_hidden_states: torch.Tensor) -> torch.Tensor:
  class Data2VecVisionForSemanticSegmentation (line 1084) | class Data2VecVisionForSemanticSegmentation(Data2VecVisionPreTrainedModel):
    method __init__ (line 1085) | def __init__(self, config: Data2VecVisionConfig) -> None:
    method compute_loss (line 1111) | def compute_loss(self, logits, auxiliary_logits, labels):
    method forward (line 1132) | def forward(

FILE: transformers/models/data2vec/modeling_tf_data2vec_vision.py
  class TFData2VecVisionModelOutputWithPooling (line 74) | class TFData2VecVisionModelOutputWithPooling(TFBaseModelOutputWithPooling):
  class TFData2VecVisionDropPath (line 104) | class TFData2VecVisionDropPath(tf.keras.layers.Layer):
    method __init__ (line 110) | def __init__(self, drop_path, **kwargs):
    method call (line 114) | def call(self, x, training=None):
  class TFData2VecVisionEmbeddings (line 124) | class TFData2VecVisionEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 130) | def __init__(self, config: Data2VecVisionConfig, **kwargs):
    method build (line 140) | def build(self, input_shape: tf.TensorShape):
    method call (line 169) | def call(self, pixel_values: tf.Tensor, bool_masked_pos: tf.Tensor | N...
  class TFData2VecVisionPatchEmbeddings (line 191) | class TFData2VecVisionPatchEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 196) | def __init__(self, config: Data2VecVisionConfig, **kwargs):
    method call (line 224) | def call(self, pixel_values: tf.Tensor, training: bool = False) -> tf....
  class TFData2VecVisionSelfAttention (line 252) | class TFData2VecVisionSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 253) | def __init__(self, config: Data2VecVisionConfig, window_size: Optional...
    method transpose_for_scores (line 288) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 295) | def call(
  class TFData2VecVisionSelfOutput (line 348) | class TFData2VecVisionSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 354) | def __init__(self, config: Data2VecVisionConfig, **kwargs):
    method call (line 362) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, gamm...
  class TFData2VecVisionAttention (line 369) | class TFData2VecVisionAttention(tf.keras.layers.Layer):
    method __init__ (line 370) | def __init__(self, config: Data2VecVisionConfig, window_size: Optional...
    method prune_heads (line 376) | def prune_heads(self, heads):
    method call (line 379) | def call(
  class TFData2VecVisionIntermediate (line 403) | class TFData2VecVisionIntermediate(tf.keras.layers.Layer):
    method __init__ (line 404) | def __init__(self, config: Data2VecVisionConfig, **kwargs):
    method call (line 416) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFData2VecVisionOutput (line 423) | class TFData2VecVisionOutput(tf.keras.layers.Layer):
    method __init__ (line 424) | def __init__(self, config: Data2VecVisionConfig, **kwargs):
    method call (line 432) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFData2VecVisionLayer (line 439) | class TFData2VecVisionLayer(tf.keras.layers.Layer):
    method __init__ (line 442) | def __init__(
    method build (line 467) | def build(self, input_shape: tf.TensorShape = None):
    method call (line 488) | def call(
  class TFData2VecVisionRelativePositionBias (line 533) | class TFData2VecVisionRelativePositionBias(tf.keras.layers.Layer):
    method __init__ (line 534) | def __init__(self, config: Data2VecVisionConfig, window_size: tuple, *...
    method build (line 545) | def build(self, input_shape):
    method get_position_index (line 556) | def get_position_index(self):
    method call (line 584) | def call(self, inputs=None) -> tf.Tensor:
  class TFData2VecVisionEncoder (line 589) | class TFData2VecVisionEncoder(tf.keras.layers.Layer):
    method __init__ (line 590) | def __init__(self, config: Data2VecVisionConfig, window_size: Optional...
    method call (line 612) | def call(
  class TFData2VecVisionMainLayer (line 655) | class TFData2VecVisionMainLayer(tf.keras.layers.Layer):
    method __init__ (line 658) | def __init__(self, config: Data2VecVisionConfig, add_pooling_layer: bo...
    method get_input_embeddings (line 678) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method _prune_heads (line 681) | def _prune_heads(self, heads_to_prune):
    method call (line 689) | def call(
  class TFData2VecVisionPooler (line 745) | class TFData2VecVisionPooler(tf.keras.layers.Layer):
    method __init__ (line 746) | def __init__(self, config: Data2VecVisionConfig, **kwargs):
    method call (line 754) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFData2VecVisionPreTrainedModel (line 766) | class TFData2VecVisionPreTrainedModel(TFPreTrainedModel):
  class TFData2VecVisionModel (line 852) | class TFData2VecVisionModel(TFData2VecVisionPreTrainedModel):
    method __init__ (line 853) | def __init__(self, config: Data2VecVisionConfig, add_pooling_layer: bo...
    method get_input_embeddings (line 861) | def get_input_embeddings(self):
    method call (line 873) | def call(
  class TFData2VecVisionForImageClassification (line 907) | class TFData2VecVisionForImageClassification(TFData2VecVisionPreTrainedM...
    method __init__ (line 908) | def __init__(self, config: Data2VecVisionConfig, *inputs, **kwargs):
    method call (line 929) | def call(
  class TFData2VecVisionConvModule (line 972) | class TFData2VecVisionConvModule(tf.keras.layers.Layer):
    method __init__ (line 980) | def __init__(
    method call (line 1001) | def call(self, input: tf.Tensor) -> tf.Tensor:
  class TFAdaptiveAvgPool1D (line 1010) | class TFAdaptiveAvgPool1D(tf.keras.layers.Layer):
    method __init__ (line 1011) | def __init__(self, output_dim, mode="dense", **kwargs):
    method build (line 1017) | def build(self, input_shape):
    method call (line 1050) | def call(self, inputs):
    method get_config (line 1059) | def get_config(self):
  class TFAdaptiveAvgPool2D (line 1065) | class TFAdaptiveAvgPool2D(tf.keras.layers.Layer):
    method __init__ (line 1066) | def __init__(self, output_shape, mode="dense", **kwargs):
    method call (line 1072) | def call(self, inputs):
    method get_config (line 1085) | def get_config(self):
  class TFData2VecVisionPyramidPoolingModule (line 1091) | class TFData2VecVisionPyramidPoolingModule(tf.keras.layers.Layer):
    method __init__ (line 1103) | def __init__(self, pool_scales: Tuple[int, ...], channels: int, **kwar...
    method call (line 1118) | def call(self, x: tf.Tensor) -> List[tf.Tensor]:
  class TFData2VecVisionUperHead (line 1132) | class TFData2VecVisionUperHead(tf.keras.layers.Layer):
    method __init__ (line 1140) | def __init__(self, config: Data2VecVisionConfig, **kwargs) -> None:
    method psp_forward (line 1166) | def psp_forward(self, inputs):
    method call (line 1175) | def call(self, encoder_hidden_states: tf.Tensor) -> tf.Tensor:
  class TFData2VecVisionFCNHead (line 1201) | class TFData2VecVisionFCNHead(tf.keras.layers.Layer):
    method __init__ (line 1215) | def __init__(
    method call (line 1261) | def call(self, encoder_hidden_states: tf.Tensor) -> tf.Tensor:
  class TFData2VecVisionForSemanticSegmentation (line 1279) | class TFData2VecVisionForSemanticSegmentation(TFData2VecVisionPreTrained...
    method __init__ (line 1280) | def __init__(self, config: Data2VecVisionConfig, *inputs, **kwargs) ->...
    method compute_loss (line 1303) | def compute_loss(self, logits, auxiliary_logits, labels):
    method call (line 1335) | def call(

FILE: transformers/models/deberta/configuration_deberta.py
  class DebertaConfig (line 40) | class DebertaConfig(PretrainedConfig):
    method __init__ (line 110) | def __init__(
  class DebertaOnnxConfig (line 164) | class DebertaOnnxConfig(OnnxConfig):
    method inputs (line 166) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method default_onnx_opset (line 179) | def default_onnx_opset(self) -> int:
    method generate_dummy_inputs (line 182) | def generate_dummy_inputs(

FILE: transformers/models/deberta/modeling_deberta.py
  class ContextPooler (line 66) | class ContextPooler(nn.Module):
    method __init__ (line 67) | def __init__(self, config):
    method forward (line 73) | def forward(self, hidden_states):
    method output_dim (line 84) | def output_dim(self):
  class XSoftmax (line 88) | class XSoftmax(torch.autograd.Function):
    method forward (line 117) | def forward(self, input, mask, dim):
    method backward (line 128) | def backward(self, grad_output):
    method symbolic (line 134) | def symbolic(g, self, mask, dim):
  class DropoutContext (line 151) | class DropoutContext(object):
    method __init__ (line 152) | def __init__(self):
  function get_mask (line 159) | def get_mask(input, local_context):
  class XDropout (line 178) | class XDropout(torch.autograd.Function):
    method forward (line 182) | def forward(ctx, input, local_ctx):
    method backward (line 192) | def backward(ctx, grad_output):
    method symbolic (line 200) | def symbolic(g: torch._C.Graph, input: torch._C.Value, local_ctx: Unio...
  class StableDropout (line 217) | class StableDropout(nn.Module):
    method __init__ (line 225) | def __init__(self, drop_prob):
    method forward (line 231) | def forward(self, x):
    method clear_context (line 242) | def clear_context(self):
    method init_context (line 246) | def init_context(self, reuse_mask=True, scale=1):
    method get_context (line 254) | def get_context(self):
  class DebertaLayerNorm (line 266) | class DebertaLayerNorm(nn.Module):
    method __init__ (line 269) | def __init__(self, size, eps=1e-12):
    method forward (line 275) | def forward(self, hidden_states):
  class DebertaSelfOutput (line 286) | class DebertaSelfOutput(nn.Module):
    method __init__ (line 287) | def __init__(self, config):
    method forward (line 293) | def forward(self, hidden_states, input_tensor):
  class DebertaAttention (line 300) | class DebertaAttention(nn.Module):
    method __init__ (line 301) | def __init__(self, config):
    method forward (line 307) | def forward(
  class DebertaIntermediate (line 337) | class DebertaIntermediate(nn.Module):
    method __init__ (line 338) | def __init__(self, config):
    method forward (line 346) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class DebertaOutput (line 352) | class DebertaOutput(nn.Module):
    method __init__ (line 353) | def __init__(self, config):
    method forward (line 360) | def forward(self, hidden_states, input_tensor):
  class DebertaLayer (line 367) | class DebertaLayer(nn.Module):
    method __init__ (line 368) | def __init__(self, config):
    method forward (line 374) | def forward(
  class DebertaEncoder (line 401) | class DebertaEncoder(nn.Module):
    method __init__ (line 404) | def __init__(self, config):
    method get_rel_embedding (line 415) | def get_rel_embedding(self):
    method get_attention_mask (line 419) | def get_attention_mask(self, attention_mask):
    method get_rel_pos (line 428) | def get_rel_pos(self, hidden_states, query_states=None, relative_pos=N...
    method forward (line 434) | def forward(
  function build_relative_position (line 508) | def build_relative_position(query_size, key_size, device):
  function c2p_dynamic_expand (line 534) | def c2p_dynamic_expand(c2p_pos, query_layer, relative_pos):
  function p2c_dynamic_expand (line 539) | def p2c_dynamic_expand(c2p_pos, query_layer, key_layer):
  function pos_dynamic_expand (line 544) | def pos_dynamic_expand(pos_index, p2c_att, key_layer):
  class DisentangledSelfAttention (line 548) | class DisentangledSelfAttention(nn.Module):
    method __init__ (line 559) | def __init__(self, config):
    method transpose_for_scores (line 594) | def transpose_for_scores(self, x):
    method forward (line 599) | def forward(
    method disentangled_att_bias (line 690) | def disentangled_att_bias(self, query_layer, key_layer, relative_pos, ...
  class DebertaEmbeddings (line 742) | class DebertaEmbeddings(nn.Module):
    method __init__ (line 745) | def __init__(self, config):
    method forward (line 769) | def forward(self, input_ids=None, token_type_ids=None, position_ids=No...
  class DebertaPreTrainedModel (line 816) | class DebertaPreTrainedModel(PreTrainedModel):
    method _init_weights (line 828) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 841) | def _set_gradient_checkpointing(self, module, value=False):
  class DebertaModel (line 911) | class DebertaModel(DebertaPreTrainedModel):
    method __init__ (line 912) | def __init__(self, config):
    method get_input_embeddings (line 922) | def get_input_embeddings(self):
    method set_input_embeddings (line 925) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 928) | def _prune_heads(self, heads_to_prune):
    method forward (line 941) | def forward(
  class DebertaForMaskedLM (line 1022) | class DebertaForMaskedLM(DebertaPreTrainedModel):
    method __init__ (line 1026) | def __init__(self, config):
    method get_output_embeddings (line 1035) | def get_output_embeddings(self):
    method set_output_embeddings (line 1038) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1050) | def forward(
  class DebertaPredictionHeadTransform (line 1103) | class DebertaPredictionHeadTransform(nn.Module):
    method __init__ (line 1104) | def __init__(self, config):
    method forward (line 1113) | def forward(self, hidden_states):
  class DebertaLMPredictionHead (line 1121) | class DebertaLMPredictionHead(nn.Module):
    method __init__ (line 1122) | def __init__(self, config):
    method forward (line 1135) | def forward(self, hidden_states):
  class DebertaOnlyMLMHead (line 1142) | class DebertaOnlyMLMHead(nn.Module):
    method __init__ (line 1143) | def __init__(self, config):
    method forward (line 1147) | def forward(self, sequence_output):
  class DebertaForSequenceClassification (line 1159) | class DebertaForSequenceClassification(DebertaPreTrainedModel):
    method __init__ (line 1160) | def __init__(self, config):
    method get_input_embeddings (line 1178) | def get_input_embeddings(self):
    method set_input_embeddings (line 1181) | def set_input_embeddings(self, new_embeddings):
    method forward (line 1190) | def forward(
  class DebertaForTokenClassification (line 1277) | class DebertaForTokenClassification(DebertaPreTrainedModel):
    method __init__ (line 1280) | def __init__(self, config):
    method forward (line 1297) | def forward(
  class DebertaForQuestionAnswering (line 1352) | class DebertaForQuestionAnswering(DebertaPreTrainedModel):
    method __init__ (line 1355) | def __init__(self, config):
    method forward (line 1375) | def forward(

FILE: transformers/models/deberta/modeling_tf_deberta.py
  class TFDebertaContextPooler (line 61) | class TFDebertaContextPooler(tf.keras.layers.Layer):
    method __init__ (line 62) | def __init__(self, config: DebertaConfig, **kwargs):
    method call (line 68) | def call(self, hidden_states, training: bool = False):
    method output_dim (line 78) | def output_dim(self) -> int:
  class TFDebertaXSoftmax (line 82) | class TFDebertaXSoftmax(tf.keras.layers.Layer):
    method __init__ (line 92) | def __init__(self, axis=-1, **kwargs):
    method call (line 96) | def call(self, inputs: tf.Tensor, mask: tf.Tensor):
  class TFDebertaStableDropout (line 104) | class TFDebertaStableDropout(tf.keras.layers.Layer):
    method __init__ (line 112) | def __init__(self, drop_prob, **kwargs):
    method xdropout (line 117) | def xdropout(self, inputs):
    method call (line 138) | def call(self, inputs: tf.Tensor, training: tf.Tensor = False):
  class TFDebertaLayerNorm (line 144) | class TFDebertaLayerNorm(tf.keras.layers.Layer):
    method __init__ (line 147) | def __init__(self, size, eps=1e-12, **kwargs):
    method build (line 152) | def build(self, input_shape):
    method call (line 157) | def call(self, x: tf.Tensor) -> tf.Tensor:
  class TFDebertaSelfOutput (line 164) | class TFDebertaSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 165) | def __init__(self, config: DebertaConfig, **kwargs):
    method call (line 171) | def call(self, hidden_states, input_tensor, training: bool = False):
  class TFDebertaAttention (line 178) | class TFDebertaAttention(tf.keras.layers.Layer):
    method __init__ (line 179) | def __init__(self, config: DebertaConfig, **kwargs):
    method call (line 185) | def call(
  class TFDebertaIntermediate (line 215) | class TFDebertaIntermediate(tf.keras.layers.Layer):
    method __init__ (line 216) | def __init__(self, config: DebertaConfig, **kwargs):
    method call (line 228) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFDebertaOutput (line 235) | class TFDebertaOutput(tf.keras.layers.Layer):
    method __init__ (line 236) | def __init__(self, config: DebertaConfig, **kwargs):
    method call (line 245) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFDebertaLayer (line 253) | class TFDebertaLayer(tf.keras.layers.Layer):
    method __init__ (line 254) | def __init__(self, config: DebertaConfig, **kwargs):
    method call (line 261) | def call(
  class TFDebertaEncoder (line 290) | class TFDebertaEncoder(tf.keras.layers.Layer):
    method __init__ (line 291) | def __init__(self, config: DebertaConfig, **kwargs):
    method build (line 302) | def build(self, input_shape):
    method get_rel_embedding (line 311) | def get_rel_embedding(self):
    method get_attention_mask (line 315) | def get_attention_mask(self, attention_mask):
    method get_rel_pos (line 325) | def get_rel_pos(self, hidden_states, query_states=None, relative_pos=N...
    method call (line 331) | def call(
  function build_relative_position (line 392) | def build_relative_position(query_size, key_size):
  function c2p_dynamic_expand (line 416) | def c2p_dynamic_expand(c2p_pos, query_layer, relative_pos):
  function p2c_dynamic_expand (line 426) | def p2c_dynamic_expand(c2p_pos, query_layer, key_layer):
  function pos_dynamic_expand (line 436) | def pos_dynamic_expand(pos_index, p2c_att, key_layer):
  function torch_gather (line 441) | def torch_gather(x, indices, gather_axis):
  class TFDebertaDisentangledSelfAttention (line 465) | class TFDebertaDisentangledSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 476) | def __init__(self, config: DebertaConfig, **kwargs):
    method build (line 532) | def build(self, input_shape):
    method transpose_for_scores (line 541) | def transpose_for_scores(self, tensor: tf.Tensor) -> tf.Tensor:
    method call (line 549) | def call(
    method disentangled_att_bias (line 660) | def disentangled_att_bias(self, query_layer, key_layer, relative_pos, ...
  class TFDebertaEmbeddings (line 716) | class TFDebertaEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 719) | def __init__(self, config, **kwargs):
    method build (line 733) | def build(self, input_shape: tf.TensorShape):
    method call (line 763) | def call(
  class TFDebertaPredictionHeadTransform (line 819) | class TFDebertaPredictionHeadTransform(tf.keras.layers.Layer):
    method __init__ (line 820) | def __init__(self, config: DebertaConfig, **kwargs):
    method call (line 835) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFDebertaLMPredictionHead (line 843) | class TFDebertaLMPredictionHead(tf.keras.layers.Layer):
    method __init__ (line 844) | def __init__(self, config: DebertaConfig, input_embeddings: tf.keras.l...
    method build (line 856) | def build(self, input_shape: tf.TensorShape):
    method get_output_embeddings (line 861) | def get_output_embeddings(self) -> tf.keras.layers.Layer:
    method set_output_embeddings (line 864) | def set_output_embeddings(self, value: tf.Variable):
    method get_bias (line 868) | def get_bias(self) -> Dict[str, tf.Variable]:
    method set_bias (line 871) | def set_bias(self, value: tf.Variable):
    method call (line 875) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFDebertaOnlyMLMHead (line 886) | class TFDebertaOnlyMLMHead(tf.keras.layers.Layer):
    method __init__ (line 887) | def __init__(self, config: DebertaConfig, input_embeddings: tf.keras.l...
    method call (line 891) | def call(self, sequence_output: tf.Tensor) -> tf.Tensor:
  class TFDebertaMainLayer (line 898) | class TFDebertaMainLayer(tf.keras.layers.Layer):
    method __init__ (line 901) | def __init__(self, config: DebertaConfig, **kwargs):
    method get_input_embeddings (line 909) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 912) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 916) | def _prune_heads(self, heads_to_prune):
    method call (line 924) | def call(
  class TFDebertaPreTrainedModel (line 981) | class TFDebertaPreTrainedModel(TFPreTrainedModel):
  class TFDebertaModel (line 1081) | class TFDebertaModel(TFDebertaPreTrainedModel):
    method __init__ (line 1082) | def __init__(self, config: DebertaConfig, *inputs, **kwargs):
    method call (line 1094) | def call(
  class TFDebertaForMaskedLM (line 1122) | class TFDebertaForMaskedLM(TFDebertaPreTrainedModel, TFMaskedLanguageMod...
    method __init__ (line 1123) | def __init__(self, config: DebertaConfig, *inputs, **kwargs):
    method get_lm_head (line 1135) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method call (line 1145) | def call(
  class TFDebertaForSequenceClassification (line 1198) | class TFDebertaForSequenceClassification(TFDebertaPreTrainedModel, TFSeq...
    method __init__ (line 1199) | def __init__(self, config: DebertaConfig, *inputs, **kwargs):
    method call (line 1223) | def call(
  class TFDebertaForTokenClassification (line 1279) | class TFDebertaForTokenClassification(TFDebertaPreTrainedModel, TFTokenC...
    method __init__ (line 1280) | def __init__(self, config: DebertaConfig, *inputs, **kwargs):
    method call (line 1298) | def call(
  class TFDebertaForQuestionAnswering (line 1350) | class TFDebertaForQuestionAnswering(TFDebertaPreTrainedModel, TFQuestion...
    method __init__ (line 1351) | def __init__(self, config: DebertaConfig, *inputs, **kwargs):
    method call (line 1368) | def call(

FILE: transformers/models/deberta/tokenization_deberta.py
  function bytes_to_unicode (line 73) | def bytes_to_unicode():
  function get_pairs (line 98) | def get_pairs(word):
  class DebertaTokenizer (line 112) | class DebertaTokenizer(PreTrainedTokenizer):
    method __init__ (line 182) | def __init__(
    method vocab_size (line 241) | def vocab_size(self):
    method get_vocab (line 245) | def get_vocab(self):
    method bpe (line 249) | def bpe(self, token):
    method build_inputs_with_special_tokens (line 291) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 316) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 343) | def create_token_type_ids_from_sequences(
    method _tokenize (line 374) | def _tokenize(self, text):
    method _convert_token_to_id (line 385) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 390) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 395) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 402) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method prepare_for_tokenization (line 431) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method _build_conversation_input_ids (line 438) | def _build_conversation_input_ids(self, conversation: "Conversation") ...

FILE: transformers/models/deberta/tokenization_deberta_fast.py
  class DebertaTokenizerFast (line 74) | class DebertaTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 145) | def __init__(
    method mask_token (line 187) | def mask_token(self) -> str:
    method mask_token (line 202) | def mask_token(self, value):
    method build_inputs_with_special_tokens (line 211) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 236) | def create_token_type_ids_from_sequences(
    method _batch_encode_plus (line 267) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 277) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method save_vocabulary (line 288) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method _build_conversation_input_ids (line 293) | def _build_conversation_input_ids(self, conversation: "Conversation") ...

FILE: transformers/models/deberta_v2/configuration_deberta_v2.py
  class DebertaV2Config (line 42) | class DebertaV2Config(PretrainedConfig):
    method __init__ (line 112) | def __init__(
  class DebertaV2OnnxConfig (line 165) | class DebertaV2OnnxConfig(OnnxConfig):
    method inputs (line 167) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method default_onnx_opset (line 180) | def default_onnx_opset(self) -> int:
    method generate_dummy_inputs (line 183) | def generate_dummy_inputs(

FILE: transformers/models/deberta_v2/modeling_deberta_v2.py
  class ContextPooler (line 56) | class ContextPooler(nn.Module):
    method __init__ (line 57) | def __init__(self, config):
    method forward (line 63) | def forward(self, hidden_states):
    method output_dim (line 74) | def output_dim(self):
  class XSoftmax (line 79) | class XSoftmax(torch.autograd.Function):
    method forward (line 108) | def forward(self, input, mask, dim):
    method backward (line 119) | def backward(self, grad_output):
    method symbolic (line 125) | def symbolic(g, self, mask, dim):
  class DropoutContext (line 143) | class DropoutContext(object):
    method __init__ (line 144) | def __init__(self):
  function get_mask (line 152) | def get_mask(input, local_context):
  class XDropout (line 172) | class XDropout(torch.autograd.Function):
    method forward (line 176) | def forward(ctx, input, local_ctx):
    method backward (line 186) | def backward(ctx, grad_output):
    method symbolic (line 194) | def symbolic(g: torch._C.Graph, input: torch._C.Value, local_ctx: Unio...
  class StableDropout (line 212) | class StableDropout(nn.Module):
    method __init__ (line 220) | def __init__(self, drop_prob):
    method forward (line 226) | def forward(self, x):
    method clear_context (line 237) | def clear_context(self):
    method init_context (line 241) | def init_context(self, reuse_mask=True, scale=1):
    method get_context (line 249) | def get_context(self):
  class DebertaV2SelfOutput (line 262) | class DebertaV2SelfOutput(nn.Module):
    method __init__ (line 263) | def __init__(self, config):
    method forward (line 269) | def forward(self, hidden_states, input_tensor):
  class DebertaV2Attention (line 277) | class DebertaV2Attention(nn.Module):
    method __init__ (line 278) | def __init__(self, config):
    method forward (line 284) | def forward(
  class DebertaV2Intermediate (line 314) | class DebertaV2Intermediate(nn.Module):
    method __init__ (line 315) | def __init__(self, config):
    method forward (line 323) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class DebertaV2Output (line 330) | class DebertaV2Output(nn.Module):
    method __init__ (line 331) | def __init__(self, config):
    method forward (line 338) | def forward(self, hidden_states, input_tensor):
  class DebertaV2Layer (line 346) | class DebertaV2Layer(nn.Module):
    method __init__ (line 347) | def __init__(self, config):
    method forward (line 353) | def forward(
  class ConvLayer (line 380) | class ConvLayer(nn.Module):
    method __init__ (line 381) | def __init__(self, config):
    method forward (line 393) | def forward(self, hidden_states, residual_states, input_mask):
  class DebertaV2Encoder (line 416) | class DebertaV2Encoder(nn.Module):
    method __init__ (line 419) | def __init__(self, config):
    method get_rel_embedding (line 446) | def get_rel_embedding(self):
    method get_attention_mask (line 452) | def get_attention_mask(self, attention_mask):
    method get_rel_pos (line 461) | def get_rel_pos(self, hidden_states, query_states=None, relative_pos=N...
    method forward (line 473) | def forward(
  function make_log_bucket_position (line 555) | def make_log_bucket_position(relative_pos, bucket_size, max_position):
  function build_relative_position (line 570) | def build_relative_position(query_size, key_size, bucket_size=-1, max_po...
  function c2p_dynamic_expand (line 602) | def c2p_dynamic_expand(c2p_pos, query_layer, relative_pos):
  function p2c_dynamic_expand (line 608) | def p2c_dynamic_expand(c2p_pos, query_layer, key_layer):
  function pos_dynamic_expand (line 614) | def pos_dynamic_expand(pos_index, p2c_att, key_layer):
  class DisentangledSelfAttention (line 618) | class DisentangledSelfAttention(nn.Module):
    method __init__ (line 629) | def __init__(self, config):
    method transpose_for_scores (line 667) | def transpose_for_scores(self, x, attention_heads):
    method forward (line 672) | def forward(
    method disentangled_attention_bias (line 756) | def disentangled_attention_bias(self, query_layer, key_layer, relative...
  class DebertaV2Embeddings (line 840) | class DebertaV2Embeddings(nn.Module):
    method __init__ (line 843) | def __init__(self, config):
    method forward (line 867) | def forward(self, input_ids=None, token_type_ids=None, position_ids=No...
  class DebertaV2PreTrainedModel (line 915) | class DebertaV2PreTrainedModel(PreTrainedModel):
    method _init_weights (line 927) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 940) | def _set_gradient_checkpointing(self, module, value=False):
  class DebertaV2Model (line 1011) | class DebertaV2Model(DebertaV2PreTrainedModel):
    method __init__ (line 1012) | def __init__(self, config):
    method get_input_embeddings (line 1022) | def get_input_embeddings(self):
    method set_input_embeddings (line 1025) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 1028) | def _prune_heads(self, heads_to_prune):
    method forward (line 1041) | def forward(
  class DebertaV2ForMaskedLM (line 1122) | class DebertaV2ForMaskedLM(DebertaV2PreTrainedModel):
    method __init__ (line 1126) | def __init__(self, config):
    method get_output_embeddings (line 1135) | def get_output_embeddings(self):
    method set_output_embeddings (line 1138) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1149) | def forward(
  class DebertaV2PredictionHeadTransform (line 1202) | class DebertaV2PredictionHeadTransform(nn.Module):
    method __init__ (line 1203) | def __init__(self, config):
    method forward (line 1212) | def forward(self, hidden_states):
  class DebertaV2LMPredictionHead (line 1220) | class DebertaV2LMPredictionHead(nn.Module):
    method __init__ (line 1221) | def __init__(self, config):
    method forward (line 1234) | def forward(self, hidden_states):
  class DebertaV2OnlyMLMHead (line 1241) | class DebertaV2OnlyMLMHead(nn.Module):
    method __init__ (line 1242) | def __init__(self, config):
    method forward (line 1246) | def forward(self, sequence_output):
  class DebertaV2ForSequenceClassification (line 1258) | class DebertaV2ForSequenceClassification(DebertaV2PreTrainedModel):
    method __init__ (line 1259) | def __init__(self, config):
    method get_input_embeddings (line 1277) | def get_input_embeddings(self):
    method set_input_embeddings (line 1280) | def set_input_embeddings(self, new_embeddings):
    method forward (line 1290) | def forward(
  class DebertaV2ForTokenClassification (line 1378) | class DebertaV2ForTokenClassification(DebertaV2PreTrainedModel):
    method __init__ (line 1381) | def __init__(self, config):
    method forward (line 1398) | def forward(
  class DebertaV2ForQuestionAnswering (line 1453) | class DebertaV2ForQuestionAnswering(DebertaV2PreTrainedModel):
    method __init__ (line 1456) | def __init__(self, config):
    method forward (line 1475) | def forward(
  class DebertaV2ForMultipleChoice (line 1555) | class DebertaV2ForMultipleChoice(DebertaV2PreTrainedModel):
    method __init__ (line 1556) | def __init__(self, config):
    method get_input_embeddings (line 1573) | def get_input_embeddings(self):
    method set_input_embeddings (line 1576) | def set_input_embeddings(self, new_embeddings):
    method forward (line 1585) | def forward(

FILE: transformers/models/deberta_v2/modeling_tf_deberta_v2.py
  class TFDebertaV2ContextPooler (line 61) | class TFDebertaV2ContextPooler(tf.keras.layers.Layer):
    method __init__ (line 62) | def __init__(self, config: DebertaV2Config, **kwargs):
    method call (line 68) | def call(self, hidden_states, training: bool = False):
    method output_dim (line 78) | def output_dim(self) -> int:
  class TFDebertaV2XSoftmax (line 83) | class TFDebertaV2XSoftmax(tf.keras.layers.Layer):
    method __init__ (line 93) | def __init__(self, axis=-1, **kwargs):
    method call (line 97) | def call(self, inputs: tf.Tensor, mask: tf.Tensor):
  class TFDebertaV2StableDropout (line 106) | class TFDebertaV2StableDropout(tf.keras.layers.Layer):
    method __init__ (line 114) | def __init__(self, drop_prob, **kwargs):
    method xdropout (line 119) | def xdropout(self, inputs):
    method call (line 140) | def call(self, inputs: tf.Tensor, training: tf.Tensor = False):
  class TFDebertaV2SelfOutput (line 147) | class TFDebertaV2SelfOutput(tf.keras.layers.Layer):
    method __init__ (line 148) | def __init__(self, config: DebertaV2Config, **kwargs):
    method call (line 154) | def call(self, hidden_states, input_tensor, training: bool = False):
  class TFDebertaV2Attention (line 162) | class TFDebertaV2Attention(tf.keras.layers.Layer):
    method __init__ (line 163) | def __init__(self, config: DebertaV2Config, **kwargs):
    method call (line 169) | def call(
  class TFDebertaV2Intermediate (line 200) | class TFDebertaV2Intermediate(tf.keras.layers.Layer):
    method __init__ (line 201) | def __init__(self, config: DebertaV2Config, **kwargs):
    method call (line 213) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFDebertaV2Output (line 221) | class TFDebertaV2Output(tf.keras.layers.Layer):
    method __init__ (line 222) | def __init__(self, config: DebertaV2Config, **kwargs):
    method call (line 231) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFDebertaV2Layer (line 240) | class TFDebertaV2Layer(tf.keras.layers.Layer):
    method __init__ (line 241) | def __init__(self, config: DebertaV2Config, **kwargs):
    method call (line 248) | def call(
  class TFDebertaV2ConvLayer (line 277) | class TFDebertaV2ConvLayer(tf.keras.layers.Layer):
    method __init__ (line 278) | def __init__(self, config: DebertaV2Config, **kwargs):
    method build (line 289) | def build(self, input_shape):
    method call (line 301) | def call(
  class TFDebertaV2Encoder (line 332) | class TFDebertaV2Encoder(tf.keras.layers.Layer):
    method __init__ (line 333) | def __init__(self, config: DebertaV2Config, **kwargs):
    method build (line 357) | def build(self, input_shape):
    method get_rel_embedding (line 366) | def get_rel_embedding(self):
    method get_attention_mask (line 372) | def get_attention_mask(self, attention_mask):
    method get_rel_pos (line 382) | def get_rel_pos(self, hidden_states, query_states=None, relative_pos=N...
    method call (line 393) | def call(
  function make_log_bucket_position (line 454) | def make_log_bucket_position(relative_pos, bucket_size, max_position):
  function build_relative_position (line 470) | def build_relative_position(query_size, key_size, bucket_size=-1, max_po...
  function c2p_dynamic_expand (line 498) | def c2p_dynamic_expand(c2p_pos, query_layer, relative_pos):
  function p2c_dynamic_expand (line 508) | def p2c_dynamic_expand(c2p_pos, query_layer, key_layer):
  function pos_dynamic_expand (line 518) | def pos_dynamic_expand(pos_index, p2c_att, key_layer):
  function take_along_axis (line 523) | def take_along_axis(x, indices):
  class TFDebertaV2DisentangledSelfAttention (line 542) | class TFDebertaV2DisentangledSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 553) | def __init__(self, config: DebertaV2Config, **kwargs):
    method transpose_for_scores (line 615) | def transpose_for_scores(self, tensor: tf.Tensor, attention_heads: int...
    method call (line 626) | def call(
    method disentangled_att_bias (line 715) | def disentangled_att_bias(self, query_layer, key_layer, relative_pos, ...
  class TFDebertaV2Embeddings (line 806) | class TFDebertaV2Embeddings(tf.keras.layers.Layer):
    method __init__ (line 809) | def __init__(self, config, **kwargs):
    method build (line 823) | def build(self, input_shape: tf.TensorShape):
    method call (line 853) | def call(
  class TFDebertaV2PredictionHeadTransform (line 910) | class TFDebertaV2PredictionHeadTransform(tf.keras.layers.Layer):
    method __init__ (line 911) | def __init__(self, config: DebertaV2Config, **kwargs):
    method call (line 926) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFDebertaV2LMPredictionHead (line 935) | class TFDebertaV2LMPredictionHead(tf.keras.layers.Layer):
    method __init__ (line 936) | def __init__(self, config: DebertaV2Config, input_embeddings: tf.keras...
    method build (line 948) | def build(self, input_shape: tf.TensorShape):
    method get_output_embeddings (line 953) | def get_output_embeddings(self) -> tf.keras.layers.Layer:
    method set_output_embeddings (line 956) | def set_output_embeddings(self, value: tf.Variable):
    method get_bias (line 960) | def get_bias(self) -> Dict[str, tf.Variable]:
    method set_bias (line 963) | def set_bias(self, value: tf.Variable):
    method call (line 967) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFDebertaV2OnlyMLMHead (line 979) | class TFDebertaV2OnlyMLMHead(tf.keras.layers.Layer):
    method __init__ (line 980) | def __init__(self, config: DebertaV2Config, input_embeddings: tf.keras...
    method call (line 984) | def call(self, sequence_output: tf.Tensor) -> tf.Tensor:
  class TFDebertaV2MainLayer (line 991) | class TFDebertaV2MainLayer(tf.keras.layers.Layer):
    method __init__ (line 994) | def __init__(self, config: DebertaV2Config, **kwargs):
    method get_input_embeddings (line 1002) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 1005) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 1009) | def _prune_heads(self, heads_to_prune):
    method call (line 1017) | def call(
  class TFDebertaV2PreTrainedModel (line 1075) | class TFDebertaV2PreTrainedModel(TFPreTrainedModel):
  class TFDebertaV2Model (line 1176) | class TFDebertaV2Model(TFDebertaV2PreTrainedModel):
    method __init__ (line 1177) | def __init__(self, config: DebertaV2Config, *inputs, **kwargs):
    method call (line 1189) | def call(
  class TFDebertaV2ForMaskedLM (line 1218) | class TFDebertaV2ForMaskedLM(TFDebertaV2PreTrainedModel, TFMaskedLanguag...
    method __init__ (line 1219) | def __init__(self, config: DebertaV2Config, *inputs, **kwargs):
    method get_lm_head (line 1231) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method call (line 1241) | def call(
  class TFDebertaV2ForSequenceClassification (line 1295) | class TFDebertaV2ForSequenceClassification(TFDebertaV2PreTrainedModel, T...
    method __init__ (line 1296) | def __init__(self, config: DebertaV2Config, *inputs, **kwargs):
    method call (line 1320) | def call(
  class TFDebertaV2ForTokenClassification (line 1377) | class TFDebertaV2ForTokenClassification(TFDebertaV2PreTrainedModel, TFTo...
    method __init__ (line 1378) | def __init__(self, config: DebertaV2Config, *inputs, **kwargs):
    method call (line 1396) | def call(
  class TFDebertaV2ForQuestionAnswering (line 1449) | class TFDebertaV2ForQuestionAnswering(TFDebertaV2PreTrainedModel, TFQues...
    method __init__ (line 1450) | def __init__(self, config: DebertaV2Config, *inputs, **kwargs):
    method call (line 1467) | def call(

FILE: transformers/models/deberta_v2/tokenization_deberta_v2.py
  class DebertaV2Tokenizer (line 56) | class DebertaV2Tokenizer(PreTrainedTokenizer):
    method __init__ (line 110) | def __init__(
    method vocab_size (line 154) | def vocab_size(self):
    method vocab (line 158) | def vocab(self):
    method get_vocab (line 161) | def get_vocab(self):
    method _tokenize (line 166) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 172) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 176) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 180) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 184) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method get_special_tokens_mask (line 208) | def get_special_tokens_mask(self, token_ids_0, token_ids_1=None, alrea...
    method create_token_type_ids_from_sequences (line 234) | def create_token_type_ids_from_sequences(self, token_ids_0, token_ids_...
    method prepare_for_tokenization (line 261) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method save_vocabulary (line 267) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class SPMTokenizer (line 271) | class SPMTokenizer:
    method __init__ (line 296) | def __init__(
    method __getstate__ (line 321) | def __getstate__(self):
    method __setstate__ (line 326) | def __setstate__(self, d):
    method tokenize (line 336) | def tokenize(self, text):
    method convert_ids_to_tokens (line 339) | def convert_ids_to_tokens(self, ids):
    method decode (line 345) | def decode(self, tokens, start=-1, end=-1, raw_text=None):
    method add_special_token (line 377) | def add_special_token(self, token):
    method part_of_whole_word (line 385) | def part_of_whole_word(self, token, is_bos=False):
    method pad (line 397) | def pad(self):
    method bos (line 400) | def bos(self):
    method eos (line 403) | def eos(self):
    method unk (line 406) | def unk(self):
    method mask (line 409) | def mask(self):
    method sym (line 412) | def sym(self, id):
    method id (line 415) | def id(self, sym):
    method _encode_as_pieces (line 418) | def _encode_as_pieces(self, text):
    method split_to_words (line 427) | def split_to_words(self, text):
    method _run_strip_accents (line 463) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 474) | def _run_split_on_punc(self, text):
    method save_pretrained (line 494) | def save_pretrained(self, path: str, filename_prefix: str = None):
  function _is_whitespace (line 504) | def _is_whitespace(char):
  function _is_control (line 516) | def _is_control(char):
  function _is_punctuation (line 528) | def _is_punctuation(char):
  function convert_to_unicode (line 543) | def convert_to_unicode(text):

FILE: transformers/models/deberta_v2/tokenization_deberta_v2_fast.py
  class DebertaV2TokenizerFast (line 63) | class DebertaV2TokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 118) | def __init__(
    method build_inputs_with_special_tokens (line 153) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method get_special_tokens_mask (line 177) | def get_special_tokens_mask(self, token_ids_0, token_ids_1=None, alrea...
    method create_token_type_ids_from_sequences (line 203) | def create_token_type_ids_from_sequences(self, token_ids_0, token_ids_...
    method save_vocabulary (line 230) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/decision_transformer/configuration_decision_transformer.py
  class DecisionTransformerConfig (line 31) | class DecisionTransformerConfig(PretrainedConfig):
    method __init__ (line 111) | def __init__(

FILE: transformers/models/decision_transformer/modeling_decision_transformer.py
  function load_tf_weights_in_gpt2 (line 53) | def load_tf_weights_in_gpt2(model, config, gpt2_checkpoint_path):
  class DecisionTransformerGPT2Attention (line 111) | class DecisionTransformerGPT2Attention(nn.Module):
    method __init__ (line 112) | def __init__(self, config, is_cross_attention=False, layer_idx=None):
    method prune_heads (line 155) | def prune_heads(self, heads):
    method _attn (line 170) | def _attn(self, query, key, value, attention_mask=None, head_mask=None):
    method _upcast_and_reordered_attn (line 210) | def _upcast_and_reordered_attn(self, query, key, value, attention_mask...
    method _split_heads (line 262) | def _split_heads(self, tensor, num_heads, attn_head_size):
    method _merge_heads (line 270) | def _merge_heads(self, tensor, num_heads, attn_head_size):
    method forward (line 278) | def forward(
  class DecisionTransformerGPT2MLP (line 333) | class DecisionTransformerGPT2MLP(nn.Module):
    method __init__ (line 334) | def __init__(self, intermediate_size, config):
    method forward (line 342) | def forward(self, hidden_states: Optional[Tuple[torch.FloatTensor]]) -...
  class DecisionTransformerGPT2Block (line 351) | class DecisionTransformerGPT2Block(nn.Module):
    method __init__ (line 352) | def __init__(self, config, layer_idx=None):
    method forward (line 369) | def forward(
  class DecisionTransformerGPT2PreTrainedModel (line 431) | class DecisionTransformerGPT2PreTrainedModel(PreTrainedModel):
    method __init__ (line 443) | def __init__(self, *inputs, **kwargs):
    method _init_weights (line 446) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 473) | def _set_gradient_checkpointing(self, module, value=False):
  class DecisionTransformerGPT2Model (line 478) | class DecisionTransformerGPT2Model(DecisionTransformerGPT2PreTrainedModel):
    method __init__ (line 481) | def __init__(self, config):
    method get_input_embeddings (line 503) | def get_input_embeddings(self):
    method set_input_embeddings (line 506) | def set_input_embeddings(self, new_embeddings):
    method forward (line 510) | def forward(
  class DecisionTransformerOutput (line 706) | class DecisionTransformerOutput(ModelOutput):
  class DecisionTransformerPreTrainedModel (line 740) | class DecisionTransformerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 753) | def _init_weights(self, module):
  class DecisionTransformerModel (line 800) | class DecisionTransformerModel(DecisionTransformerPreTrainedModel):
    method __init__ (line 808) | def __init__(self, config):
    method forward (line 835) | def forward(

FILE: transformers/models/deformable_detr/configuration_deformable_detr.py
  class DeformableDetrConfig (line 30) | class DeformableDetrConfig(PretrainedConfig):
    method __init__ (line 152) | def __init__(
    method num_attention_heads (line 257) | def num_attention_heads(self) -> int:
    method hidden_size (line 261) | def hidden_size(self) -> int:

FILE: transformers/models/deformable_detr/convert_deformable_detr_to_pytorch.py
  function rename_key (line 35) | def rename_key(orig_key):
  function read_in_q_k_v (line 64) | def read_in_q_k_v(state_dict):
  function prepare_img (line 80) | def prepare_img():
  function convert_deformable_detr_checkpoint (line 88) | def convert_deformable_detr_checkpoint(

FILE: transformers/models/deformable_detr/feature_extraction_deformable_detr.py
  class DeformableDetrFeatureExtractor (line 26) | class DeformableDetrFeatureExtractor(DeformableDetrImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/deformable_detr/image_processing_deformable_detr.py
  class AnnotionFormat (line 85) | class AnnotionFormat(ExplicitEnum):
  function get_size_with_aspect_ratio (line 94) | def get_size_with_aspect_ratio(image_size, size, max_size=None) -> Tuple...
  function get_resize_output_image_size (line 126) | def get_resize_output_image_size(
  function get_numpy_to_framework_fn (line 150) | def get_numpy_to_framework_fn(arr) -> Callable:
  function safe_squeeze (line 175) | def safe_squeeze(arr: np.ndarray, axis: Optional[int] = None) -> np.ndar...
  function normalize_annotation (line 189) | def normalize_annotation(annotation: Dict, image_size: Tuple[int, int]) ...
  function max_across_indices (line 204) | def max_across_indices(values: Iterable[Any]) -> List[Any]:
  function get_max_height_width (line 212) | def get_max_height_width(images: List[np.ndarray]) -> List[int]:
  function make_pixel_mask (line 228) | def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> ...
  function convert_coco_poly_to_mask (line 245) | def convert_coco_poly_to_mask(segmentations, height: int, width: int) ->...
  function prepare_coco_detection_annotation (line 280) | def prepare_coco_detection_annotation(image, target, return_segmentation...
  function masks_to_boxes (line 333) | def masks_to_boxes(masks: np.ndarray) -> np.ndarray:
  function prepare_coco_panoptic_annotation (line 368) | def prepare_coco_panoptic_annotation(
  function get_segmentation_image (line 406) | def get_segmentation_image(
  function get_mask_area (line 432) | def get_mask_area(seg_img: np.ndarray, target_size: Tuple[int, int], n_c...
  function score_labels_from_class_probabilities (line 442) | def score_labels_from_class_probabilities(logits: np.ndarray) -> Tuple[n...
  function post_process_panoptic_sample (line 451) | def post_process_panoptic_sample(
  function resize_annotation (line 537) | def resize_annotation(
  function binary_mask_to_rle (line 589) | def binary_mask_to_rle(mask):
  function convert_segmentation_to_rle (line 612) | def convert_segmentation_to_rle(segmentation):
  function remove_low_and_no_objects (line 634) | def remove_low_and_no_objects(masks, scores, labels, object_mask_thresho...
  function check_segment_validity (line 663) | def check_segment_validity(mask_labels, mask_probs, k, mask_threshold=0....
  function compute_segments (line 682) | def compute_segments(
  class DeformableDetrImageProcessor (line 742) | class DeformableDetrImageProcessor(BaseImageProcessor):
    method __init__ (line 780) | def __init__(
    method max_size (line 823) | def max_size(self):
    method from_dict (line 832) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method prepare_annotation (line 846) | def prepare_annotation(
    method prepare (line 872) | def prepare(self, image, target, return_segmentation_masks=None, masks...
    method convert_coco_poly_to_mask (line 882) | def convert_coco_poly_to_mask(self, *args, **kwargs):
    method prepare_coco_detection (line 889) | def prepare_coco_detection(self, *args, **kwargs):
    method prepare_coco_panoptic (line 896) | def prepare_coco_panoptic(self, *args, **kwargs):
    method resize (line 903) | def resize(
    method resize_annotation (line 937) | def resize_annotation(
    method rescale (line 951) | def rescale(
    method normalize (line 960) | def normalize(
    method normalize_annotation (line 973) | def normalize_annotation(self, annotation: Dict, image_size: Tuple[int...
    method pad_and_create_pixel_mask (line 981) | def pad_and_create_pixel_mask(
    method _pad_image (line 1015) | def _pad_image(
    method pad (line 1037) | def pad(
    method preprocess (line 1076) | def preprocess(
    method post_process (line 1282) | def post_process(self, outputs, target_sizes):
    method post_process_object_detection (line 1327) | def post_process_object_detection(

FILE: transformers/models/deformable_detr/load_custom.py
  function load_cuda_kernels (line 20) | def load_cuda_kernels():

FILE: transformers/models/deformable_detr/modeling_deformable_detr.py
  class MultiScaleDeformableAttentionFunction (line 68) | class MultiScaleDeformableAttentionFunction(Function):
    method forward (line 70) | def forward(
    method backward (line 95) | def backward(context, grad_output):
  class DeformableDetrDecoderOutput (line 134) | class DeformableDetrDecoderOutput(ModelOutput):
  class DeformableDetrModelOutput (line 171) | class DeformableDetrModelOutput(ModelOutput):
  class DeformableDetrObjectDetectionOutput (line 229) | class DeformableDetrObjectDetectionOutput(ModelOutput):
  function _get_clones (line 308) | def _get_clones(module, N):
  function inverse_sigmoid (line 312) | def inverse_sigmoid(x, eps=1e-5):
  class DeformableDetrFrozenBatchNorm2d (line 320) | class DeformableDetrFrozenBatchNorm2d(nn.Module):
    method __init__ (line 328) | def __init__(self, n):
    method _load_from_state_dict (line 335) | def _load_from_state_dict(
    method forward (line 346) | def forward(self, x):
  function replace_batch_norm (line 360) | def replace_batch_norm(m, name=""):
  class DeformableDetrConvEncoder (line 375) | class DeformableDetrConvEncoder(nn.Module):
    method __init__ (line 383) | def __init__(self, config):
    method forward (line 423) | def forward(self, pixel_values: torch.Tensor, pixel_mask: torch.Tensor):
  class DeformableDetrConvModel (line 436) | class DeformableDetrConvModel(nn.Module):
    method __init__ (line 441) | def __init__(self, conv_encoder, position_embedding):
    method forward (line 446) | def forward(self, pixel_values, pixel_mask):
  function _expand_mask (line 458) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, target_len: Opt...
  class DeformableDetrSinePositionEmbedding (line 472) | class DeformableDetrSinePositionEmbedding(nn.Module):
    method __init__ (line 478) | def __init__(self, embedding_dim=64, temperature=10000, normalize=Fals...
    method forward (line 489) | def forward(self, pixel_values, pixel_mask):
  class DeformableDetrLearnedPositionEmbedding (line 511) | class DeformableDetrLearnedPositionEmbedding(nn.Module):
    method __init__ (line 516) | def __init__(self, embedding_dim=256):
    method forward (line 521) | def forward(self, pixel_values, pixel_mask=None):
  function build_position_encoding (line 535) | def build_position_encoding(config):
  function multi_scale_deformable_attention (line 548) | def multi_scale_deformable_attention(
  class DeformableDetrMultiscaleDeformableAttention (line 587) | class DeformableDetrMultiscaleDeformableAttention(nn.Module):
    method __init__ (line 592) | def __init__(self, config: DeformableDetrConfig, num_heads: int, n_poi...
    method _reset_parameters (line 623) | def _reset_parameters(self):
    method with_pos_embed (line 643) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 646) | def forward(
  class DeformableDetrMultiheadAttention (line 720) | class DeformableDetrMultiheadAttention(nn.Module):
    method __init__ (line 727) | def __init__(
    method _shape (line 751) | def _shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method with_pos_embed (line 754) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 757) | def forward(
  class DeformableDetrEncoderLayer (line 837) | class DeformableDetrEncoderLayer(nn.Module):
    method __init__ (line 838) | def __init__(self, config: DeformableDetrConfig):
    method forward (line 852) | def forward(
  class DeformableDetrDecoderLayer (line 922) | class DeformableDetrDecoderLayer(nn.Module):
    method __init__ (line 923) | def __init__(self, config: DeformableDetrConfig):
    method forward (line 950) | def forward(
  class DeformableDetrClassificationHead (line 1034) | class DeformableDetrClassificationHead(nn.Module):
    method __init__ (line 1037) | def __init__(self, input_dim: int, inner_dim: int, num_classes: int, p...
    method forward (line 1043) | def forward(self, hidden_states: torch.Tensor):
  class DeformableDetrPreTrainedModel (line 1052) | class DeformableDetrPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1057) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1081) | def _set_gradient_checkpointing(self, module, value=False):
  class DeformableDetrEncoder (line 1141) | class DeformableDetrEncoder(DeformableDetrPreTrainedModel):
    method __init__ (line 1152) | def __init__(self, config: DeformableDetrConfig):
    method get_reference_points (line 1162) | def get_reference_points(spatial_shapes, valid_ratios, device):
    method forward (line 1192) | def forward(
  class DeformableDetrDecoder (line 1271) | class DeformableDetrDecoder(DeformableDetrPreTrainedModel):
    method __init__ (line 1286) | def __init__(self, config: DeformableDetrConfig):
    method forward (line 1300) | def forward(
  class DeformableDetrModel (line 1467) | class DeformableDetrModel(DeformableDetrPreTrainedModel):
    method __init__ (line 1468) | def __init__(self, config: DeformableDetrConfig):
    method get_encoder (line 1525) | def get_encoder(self):
    method get_decoder (line 1528) | def get_decoder(self):
    method freeze_backbone (line 1531) | def freeze_backbone(self):
    method unfreeze_backbone (line 1535) | def unfreeze_backbone(self):
    method get_valid_ratio (line 1539) | def get_valid_ratio(self, mask):
    method get_proposal_pos_embed (line 1550) | def get_proposal_pos_embed(self, proposals):
    method gen_encoder_output_proposals (line 1567) | def gen_encoder_output_proposals(self, enc_output, padding_mask, spati...
    method forward (line 1618) | def forward(
  class DeformableDetrForObjectDetection (line 1824) | class DeformableDetrForObjectDetection(DeformableDetrPreTrainedModel):
    method __init__ (line 1828) | def __init__(self, config: DeformableDetrConfig):
    method _set_aux_loss (line 1870) | def _set_aux_loss(self, outputs_class, outputs_coord):
    method forward (line 1878) | def forward(
  function dice_loss (line 2049) | def dice_loss(inputs, targets, num_boxes):
  function sigmoid_focal_loss (line 2069) | def sigmoid_focal_loss(inputs, targets, num_boxes, alpha: float = 0.25, ...
  class DeformableDetrLoss (line 2100) | class DeformableDetrLoss(nn.Module):
    method __init__ (line 2117) | def __init__(self, matcher, num_classes, focal_alpha, losses):
    method loss_labels (line 2125) | def loss_labels(self, outputs, targets, indices, num_boxes):
    method loss_cardinality (line 2160) | def loss_cardinality(self, outputs, targets, indices, num_boxes):
    method loss_boxes (line 2176) | def loss_boxes(self, outputs, targets, indices, num_boxes):
    method _get_source_permutation_idx (line 2201) | def _get_source_permutation_idx(self, indices):
    method _get_target_permutation_idx (line 2208) | def _get_target_permutation_idx(self, indices):
    method get_loss (line 2214) | def get_loss(self, loss, outputs, targets, indices, num_boxes):
    method forward (line 2224) | def forward(self, outputs, targets):
  class DeformableDetrMLPPredictionHead (line 2282) | class DeformableDetrMLPPredictionHead(nn.Module):
    method __init__ (line 2291) | def __init__(self, input_dim, hidden_dim, output_dim, num_layers):
    method forward (line 2297) | def forward(self, x):
  class DeformableDetrHungarianMatcher (line 2303) | class DeformableDetrHungarianMatcher(nn.Module):
    method __init__ (line 2320) | def __init__(self, class_cost: float = 1, bbox_cost: float = 1, giou_c...
    method forward (line 2331) | def forward(self, outputs, targets):
  function _upcast (line 2384) | def _upcast(t: Tensor) -> Tensor:
  function box_area (line 2393) | def box_area(boxes: Tensor) -> Tensor:
  function box_iou (line 2410) | def box_iou(boxes1, boxes2):
  function generalized_box_iou (line 2427) | def generalized_box_iou(boxes1, boxes2):
  function _max_by_axis (line 2452) | def _max_by_axis(the_list):
  class NestedTensor (line 2462) | class NestedTensor(object):
    method __init__ (line 2463) | def __init__(self, tensors, mask: Optional[Tensor]):
    method to (line 2467) | def to(self, device):
    method decompose (line 2476) | def decompose(self):
    method __repr__ (line 2479) | def __repr__(self):
  function nested_tensor_from_tensor_list (line 2484) | def nested_tensor_from_tensor_list(tensor_list: List[Tensor]):

FILE: transformers/models/deit/configuration_deit.py
  class DeiTConfig (line 37) | class DeiTConfig(PretrainedConfig):
    method __init__ (line 96) | def __init__(
  class DeiTOnnxConfig (line 132) | class DeiTOnnxConfig(OnnxConfig):
    method inputs (line 136) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 144) | def atol_for_validation(self) -> float:

FILE: transformers/models/deit/convert_deit_timm_to_pytorch.py
  function create_rename_keys (line 37) | def create_rename_keys(config, base_model=False):
  function read_in_q_k_v (line 93) | def read_in_q_k_v(state_dict, config, base_model=False):
  function rename_key (line 119) | def rename_key(dct, old, new):
  function prepare_img (line 125) | def prepare_img():
  function convert_deit_checkpoint (line 132) | def convert_deit_checkpoint(deit_name, pytorch_dump_folder_path):

FILE: transformers/models/deit/feature_extraction_deit.py
  class DeiTFeatureExtractor (line 26) | class DeiTFeatureExtractor(DeiTImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/deit/image_processing_deit.py
  class DeiTImageProcessor (line 43) | class DeiTImageProcessor(BaseImageProcessor):
    method __init__ (line 79) | def __init__(
    method resize (line 110) | def resize(
    method center_crop (line 138) | def center_crop(
    method rescale (line 162) | def rescale(
    method normalize (line 182) | def normalize(
    method preprocess (line 205) | def preprocess(

FILE: transformers/models/deit/modeling_deit.py
  class DeiTEmbeddings (line 68) | class DeiTEmbeddings(nn.Module):
    method __init__ (line 73) | def __init__(self, config: DeiTConfig, use_mask_token: bool = False) -...
    method forward (line 84) | def forward(self, pixel_values: torch.Tensor, bool_masked_pos: Optiona...
  class DeiTPatchEmbeddings (line 102) | class DeiTPatchEmbeddings(nn.Module):
    method __init__ (line 109) | def __init__(self, config):
    method forward (line 124) | def forward(self, pixel_values: torch.Tensor) -> torch.Tensor:
  class DeiTSelfAttention (line 139) | class DeiTSelfAttention(nn.Module):
    method __init__ (line 140) | def __init__(self, config: DeiTConfig) -> None:
    method transpose_for_scores (line 158) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 163) | def forward(
  class DeiTSelfOutput (line 200) | class DeiTSelfOutput(nn.Module):
    method __init__ (line 206) | def __init__(self, config: DeiTConfig) -> None:
    method forward (line 211) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class DeiTAttention (line 219) | class DeiTAttention(nn.Module):
    method __init__ (line 220) | def __init__(self, config: DeiTConfig) -> None:
    method prune_heads (line 226) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 244) | def forward(
  class DeiTIntermediate (line 259) | class DeiTIntermediate(nn.Module):
    method __init__ (line 260) | def __init__(self, config: DeiTConfig) -> None:
    method forward (line 268) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class DeiTOutput (line 276) | class DeiTOutput(nn.Module):
    method __init__ (line 277) | def __init__(self, config: DeiTConfig) -> None:
    method forward (line 282) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class DeiTLayer (line 292) | class DeiTLayer(nn.Module):
    method __init__ (line 295) | def __init__(self, config: DeiTConfig) -> None:
    method forward (line 305) | def forward(
  class DeiTEncoder (line 335) | class DeiTEncoder(nn.Module):
    method __init__ (line 336) | def __init__(self, config: DeiTConfig) -> None:
    method forward (line 342) | def forward(
  class DeiTPreTrainedModel (line 392) | class DeiTPreTrainedModel(PreTrainedModel):
    method _init_weights (line 404) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 418) | def _set_gradient_checkpointing(self, module: DeiTEncoder, value: bool...
  class DeiTModel (line 461) | class DeiTModel(DeiTPreTrainedModel):
    method __init__ (line 462) | def __init__(self, config: DeiTConfig, add_pooling_layer: bool = True,...
    method get_input_embeddings (line 475) | def get_input_embeddings(self) -> DeiTPatchEmbeddings:
    method _prune_heads (line 478) | def _prune_heads(self, heads_to_prune):
    method forward (line 494) | def forward(
  class DeiTPooler (line 554) | class DeiTPooler(nn.Module):
    method __init__ (line 555) | def __init__(self, config: DeiTConfig):
    method forward (line 560) | def forward(self, hidden_states):
  class DeiTForMaskedImageModeling (line 581) | class DeiTForMaskedImageModeling(DeiTPreTrainedModel):
    method __init__ (line 582) | def __init__(self, config: DeiTConfig) -> None:
    method forward (line 601) | def forward(
  class DeiTForImageClassification (line 693) | class DeiTForImageClassification(DeiTPreTrainedModel):
    method __init__ (line 694) | def __init__(self, config: DeiTConfig) -> None:
    method forward (line 708) | def forward(
  class DeiTForImageClassificationWithTeacherOutput (line 801) | class DeiTForImageClassificationWithTeacherOutput(ModelOutput):
  class DeiTForImageClassificationWithTeacher (line 843) | class DeiTForImageClassificationWithTeacher(DeiTPreTrainedModel):
    method __init__ (line 844) | def __init__(self, config: DeiTConfig) -> None:
    method forward (line 868) | def forward(

FILE: transformers/models/deit/modeling_tf_deit.py
  class TFDeiTForImageClassificationWithTeacherOutput (line 74) | class TFDeiTForImageClassificationWithTeacherOutput(ModelOutput):
  class TFDeiTEmbeddings (line 104) | class TFDeiTEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 109) | def __init__(self, config: DeiTConfig, use_mask_token: bool = False, *...
    method build (line 116) | def build(self, input_shape: tf.TensorShape):
    method call (line 146) | def call(
  class TFDeiTPatchEmbeddings (line 167) | class TFDeiTPatchEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 174) | def __init__(self, config: DeiTConfig, **kwargs) -> None:
    method call (line 191) | def call(self, pixel_values: tf.Tensor) -> tf.Tensor:
  class TFDeiTSelfAttention (line 208) | class TFDeiTSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 209) | def __init__(self, config: DeiTConfig, **kwargs):
    method transpose_for_scores (line 234) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 241) | def call(
  class TFDeiTSelfOutput (line 284) | class TFDeiTSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 290) | def __init__(self, config: DeiTConfig, **kwargs):
    method call (line 298) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFDeiTAttention (line 306) | class TFDeiTAttention(tf.keras.layers.Layer):
    method __init__ (line 307) | def __init__(self, config: DeiTConfig, **kwargs):
    method prune_heads (line 313) | def prune_heads(self, heads):
    method call (line 316) | def call(
  class TFDeiTIntermediate (line 335) | class TFDeiTIntermediate(tf.keras.layers.Layer):
    method __init__ (line 336) | def __init__(self, config: DeiTConfig, **kwargs):
    method call (line 348) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFDeiTOutput (line 356) | class TFDeiTOutput(tf.keras.layers.Layer):
    method __init__ (line 357) | def __init__(self, config: DeiTConfig, **kwargs):
    method call (line 365) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFDeiTLayer (line 373) | class TFDeiTLayer(tf.keras.layers.Layer):
    method __init__ (line 376) | def __init__(self, config: DeiTConfig, **kwargs):
    method call (line 390) | def call(
  class TFDeiTEncoder (line 424) | class TFDeiTEncoder(tf.keras.layers.Layer):
    method __init__ (line 425) | def __init__(self, config: DeiTConfig, **kwargs):
    method call (line 430) | def call(
  class TFDeiTMainLayer (line 470) | class TFDeiTMainLayer(tf.keras.layers.Layer):
    method __init__ (line 473) | def __init__(
    method get_input_embeddings (line 485) | def get_input_embeddings(self) -> TFDeiTPatchEmbeddings:
    method _prune_heads (line 488) | def _prune_heads(self, heads_to_prune):
    method get_head_mask (line 495) | def get_head_mask(self, head_mask):
    method call (line 504) | def call(
  class TFDeiTPreTrainedModel (line 561) | class TFDeiTPreTrainedModel(TFPreTrainedModel):
  class TFDeiTModel (line 610) | class TFDeiTModel(TFDeiTPreTrainedModel):
    method __init__ (line 611) | def __init__(
    method call (line 629) | def call(
  class TFDeiTPooler (line 652) | class TFDeiTPooler(tf.keras.layers.Layer):
    method __init__ (line 653) | def __init__(self, config: DeiTConfig, **kwargs):
    method call (line 663) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFDeitPixelShuffle (line 672) | class TFDeitPixelShuffle(tf.keras.layers.Layer):
    method __init__ (line 675) | def __init__(self, upscale_factor: int, **kwargs) -> None:
    method call (line 681) | def call(self, x: tf.Tensor) -> tf.Tensor:
  class TFDeitDecoder (line 698) | class TFDeitDecoder(tf.keras.layers.Layer):
    method __init__ (line 699) | def __init__(self, config: DeiTConfig, **kwargs) -> None:
    method call (line 706) | def call(self, inputs: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFDeiTForMaskedImageModeling (line 718) | class TFDeiTForMaskedImageModeling(TFDeiTPreTrainedModel):
    method __init__ (line 719) | def __init__(self, config: DeiTConfig) -> None:
    method call (line 728) | def call(
  class TFDeiTForImageClassification (line 833) | class TFDeiTForImageClassification(TFDeiTPreTrainedModel, TFSequenceClas...
    method __init__ (line 834) | def __init__(self, config: DeiTConfig):
    method call (line 850) | def call(
  class TFDeiTForImageClassificationWithTeacher (line 935) | class TFDeiTForImageClassificationWithTeacher(TFDeiTPreTrainedModel):
    method __init__ (line 936) | def __init__(self, config: DeiTConfig) -> None:
    method call (line 962) | def call(

FILE: transformers/models/deta/configuration_deta.py
  class DetaConfig (line 31) | class DetaConfig(PretrainedConfig):
    method __init__ (line 134) | def __init__(
    method num_attention_heads (line 227) | def num_attention_heads(self) -> int:
    method hidden_size (line 231) | def hidden_size(self) -> int:
    method to_dict (line 234) | def to_dict(self):

FILE: transformers/models/deta/convert_deta_resnet_to_pytorch.py
  function get_deta_config (line 37) | def get_deta_config():
  function create_rename_keys (line 61) | def create_rename_keys(config):
  function rename_key (line 185) | def rename_key(dct, old, new):
  function read_in_decoder_q_k_v (line 190) | def read_in_decoder_q_k_v(state_dict, config):
  function prepare_img (line 209) | def prepare_img():
  function convert_deta_checkpoint (line 217) | def convert_deta_checkpoint(model_name, pytorch_dump_folder_path, push_t...

FILE: transformers/models/deta/convert_deta_swin_to_pytorch.py
  function get_deta_config (line 37) | def get_deta_config(model_name):
  function create_rename_keys (line 76) | def create_rename_keys(config):
  function rename_key (line 160) | def rename_key(dct, old, new):
  function read_in_swin_q_k_v (line 166) | def read_in_swin_q_k_v(state_dict, backbone_config):
  function read_in_decoder_q_k_v (line 191) | def read_in_decoder_q_k_v(state_dict, config):
  function prepare_img (line 210) | def prepare_img():
  function convert_deta_checkpoint (line 218) | def convert_deta_checkpoint(model_name, pytorch_dump_folder_path, push_t...

FILE: transformers/models/deta/image_processing_deta.py
  class AnnotionFormat (line 77) | class AnnotionFormat(ExplicitEnum):
  function get_size_with_aspect_ratio (line 86) | def get_size_with_aspect_ratio(image_size, size, max_size=None) -> Tuple...
  function get_resize_output_image_size (line 118) | def get_resize_output_image_size(
  function get_numpy_to_framework_fn (line 142) | def get_numpy_to_framework_fn(arr) -> Callable:
  function safe_squeeze (line 167) | def safe_squeeze(arr: np.ndarray, axis: Optional[int] = None) -> np.ndar...
  function normalize_annotation (line 181) | def normalize_annotation(annotation: Dict, image_size: Tuple[int, int]) ...
  function max_across_indices (line 196) | def max_across_indices(values: Iterable[Any]) -> List[Any]:
  function get_max_height_width (line 204) | def get_max_height_width(images: List[np.ndarray]) -> List[int]:
  function make_pixel_mask (line 220) | def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> ...
  function convert_coco_poly_to_mask (line 237) | def convert_coco_poly_to_mask(segmentations, height: int, width: int) ->...
  function prepare_coco_detection_annotation (line 272) | def prepare_coco_detection_annotation(image, target, return_segmentation...
  function masks_to_boxes (line 325) | def masks_to_boxes(masks: np.ndarray) -> np.ndarray:
  function prepare_coco_panoptic_annotation (line 360) | def prepare_coco_panoptic_annotation(
  function resize_annotation (line 398) | def resize_annotation(
  class DetaImageProcessor (line 449) | class DetaImageProcessor(BaseImageProcessor):
    method __init__ (line 486) | def __init__(
    method prepare_annotation (line 519) | def prepare_annotation(
    method prepare (line 545) | def prepare(self, image, target, return_segmentation_masks=None, masks...
    method convert_coco_poly_to_mask (line 555) | def convert_coco_poly_to_mask(self, *args, **kwargs):
    method prepare_coco_detection (line 562) | def prepare_coco_detection(self, *args, **kwargs):
    method prepare_coco_panoptic (line 569) | def prepare_coco_panoptic(self, *args, **kwargs):
    method resize (line 575) | def resize(
    method resize_annotation (line 601) | def resize_annotation(
    method rescale (line 615) | def rescale(
    method normalize (line 624) | def normalize(
    method normalize_annotation (line 637) | def normalize_annotation(self, annotation: Dict, image_size: Tuple[int...
    method pad_and_create_pixel_mask (line 645) | def pad_and_create_pixel_mask(
    method _pad_image (line 679) | def _pad_image(
    method pad (line 701) | def pad(
    method preprocess (line 739) | def preprocess(
    method post_process_object_detection (line 936) | def post_process_object_detection(

FILE: transformers/models/deta/modeling_deta.py
  class DetaDecoderOutput (line 70) | class DetaDecoderOutput(ModelOutput):
  class DetaModelOutput (line 108) | class DetaModelOutput(ModelOutput):
  class DetaObjectDetectionOutput (line 167) | class DetaObjectDetectionOutput(ModelOutput):
  function _get_clones (line 246) | def _get_clones(module, N):
  function inverse_sigmoid (line 250) | def inverse_sigmoid(x, eps=1e-5):
  class DetaFrozenBatchNorm2d (line 258) | class DetaFrozenBatchNorm2d(nn.Module):
    method __init__ (line 266) | def __init__(self, n):
    method _load_from_state_dict (line 273) | def _load_from_state_dict(
    method forward (line 284) | def forward(self, x):
  function replace_batch_norm (line 298) | def replace_batch_norm(m, name=""):
  class DetaBackboneWithPositionalEncodings (line 313) | class DetaBackboneWithPositionalEncodings(nn.Module):
    method __init__ (line 320) | def __init__(self, config):
    method forward (line 337) | def forward(self, pixel_values: torch.Tensor, pixel_mask: torch.Tensor):
  function _expand_mask (line 359) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, target_len: Opt...
  class DetaSinePositionEmbedding (line 374) | class DetaSinePositionEmbedding(nn.Module):
    method __init__ (line 380) | def __init__(self, embedding_dim=64, temperature=10000, normalize=Fals...
    method forward (line 391) | def forward(self, pixel_values, pixel_mask):
  class DetaLearnedPositionEmbedding (line 413) | class DetaLearnedPositionEmbedding(nn.Module):
    method __init__ (line 418) | def __init__(self, embedding_dim=256):
    method forward (line 423) | def forward(self, pixel_values, pixel_mask=None):
  function build_position_encoding (line 437) | def build_position_encoding(config):
  function multi_scale_deformable_attention (line 451) | def multi_scale_deformable_attention(
  class DetaMultiscaleDeformableAttention (line 490) | class DetaMultiscaleDeformableAttention(nn.Module):
    method __init__ (line 495) | def __init__(self, embed_dim: int, num_heads: int, n_levels: int, n_po...
    method _reset_parameters (line 524) | def _reset_parameters(self):
    method with_pos_embed (line 544) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 547) | def forward(
  class DetaMultiheadAttention (line 606) | class DetaMultiheadAttention(nn.Module):
    method __init__ (line 613) | def __init__(
    method _shape (line 637) | def _shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method with_pos_embed (line 640) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 643) | def forward(
  class DetaEncoderLayer (line 723) | class DetaEncoderLayer(nn.Module):
    method __init__ (line 724) | def __init__(self, config: DetaConfig):
    method forward (line 741) | def forward(
  class DetaDecoderLayer (line 811) | class DetaDecoderLayer(nn.Module):
    method __init__ (line 812) | def __init__(self, config: DetaConfig):
    method forward (line 840) | def forward(
  class DetaClassificationHead (line 924) | class DetaClassificationHead(nn.Module):
    method __init__ (line 927) | def __init__(self, input_dim: int, inner_dim: int, num_classes: int, p...
    method forward (line 933) | def forward(self, hidden_states: torch.Tensor):
  class DetaPreTrainedModel (line 943) | class DetaPreTrainedModel(PreTrainedModel):
    method _init_weights (line 948) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 972) | def _set_gradient_checkpointing(self, module, value=False):
  class DetaEncoder (line 1032) | class DetaEncoder(DetaPreTrainedModel):
    method __init__ (line 1043) | def __init__(self, config: DetaConfig):
    method get_reference_points (line 1053) | def get_reference_points(spatial_shapes, valid_ratios, device):
    method forward (line 1083) | def forward(
  class DetaDecoder (line 1163) | class DetaDecoder(DetaPreTrainedModel):
    method __init__ (line 1178) | def __init__(self, config: DetaConfig):
    method forward (line 1192) | def forward(
  class DetaModel (line 1359) | class DetaModel(DetaPreTrainedModel):
    method __init__ (line 1360) | def __init__(self, config: DetaConfig):
    method get_encoder (line 1425) | def get_encoder(self):
    method get_decoder (line 1429) | def get_decoder(self):
    method freeze_backbone (line 1433) | def freeze_backbone(self):
    method unfreeze_backbone (line 1438) | def unfreeze_backbone(self):
    method get_valid_ratio (line 1443) | def get_valid_ratio(self, mask):
    method get_proposal_pos_embed (line 1455) | def get_proposal_pos_embed(self, proposals):
    method gen_encoder_output_proposals (line 1472) | def gen_encoder_output_proposals(self, enc_output, padding_mask, spati...
    method forward (line 1526) | def forward(
  class DetaForObjectDetection (line 1776) | class DetaForObjectDetection(DetaPreTrainedModel):
    method __init__ (line 1781) | def __init__(self, config: DetaConfig):
    method _set_aux_loss (line 1823) | def _set_aux_loss(self, outputs_class, outputs_coord):
    method forward (line 1831) | def forward(
  function dice_loss (line 2007) | def dice_loss(inputs, targets, num_boxes):
  function sigmoid_focal_loss (line 2027) | def sigmoid_focal_loss(inputs, targets, num_boxes, alpha: float = 0.25, ...
  class DetaLoss (line 2058) | class DetaLoss(nn.Module):
    method __init__ (line 2075) | def __init__(
    method loss_labels (line 2099) | def loss_labels(self, outputs, targets, indices, num_boxes):
    method loss_cardinality (line 2134) | def loss_cardinality(self, outputs, targets, indices, num_boxes):
    method loss_boxes (line 2150) | def loss_boxes(self, outputs, targets, indices, num_boxes):
    method _get_source_permutation_idx (line 2175) | def _get_source_permutation_idx(self, indices):
    method _get_target_permutation_idx (line 2182) | def _get_target_permutation_idx(self, indices):
    method get_loss (line 2189) | def get_loss(self, loss, outputs, targets, indices, num_boxes):
    method forward (line 2199) | def forward(self, outputs, targets):
  class DetaMLPPredictionHead (line 2264) | class DetaMLPPredictionHead(nn.Module):
    method __init__ (line 2273) | def __init__(self, input_dim, hidden_dim, output_dim, num_layers):
    method forward (line 2279) | def forward(self, x):
  class DetaHungarianMatcher (line 2286) | class DetaHungarianMatcher(nn.Module):
    method __init__ (line 2303) | def __init__(self, class_cost: float = 1, bbox_cost: float = 1, giou_c...
    method forward (line 2314) | def forward(self, outputs, targets):
  function _upcast (line 2367) | def _upcast(t: Tensor) -> Tensor:
  function box_area (line 2376) | def box_area(boxes: Tensor) -> Tensor:
  function box_iou (line 2393) | def box_iou(boxes1, boxes2):
  function generalized_box_iou (line 2410) | def generalized_box_iou(boxes1, boxes2):
  function nonzero_tuple (line 2435) | def nonzero_tuple(x):
  class DetaMatcher (line 2449) | class DetaMatcher(object):
    method __init__ (line 2462) | def __init__(self, thresholds: List[float], labels: List[int], allow_l...
    method __call__ (line 2497) | def __call__(self, match_quality_matrix):
    method set_low_quality_matches_ (line 2538) | def set_low_quality_matches_(self, match_labels, match_quality_matrix):
  function subsample_labels (line 2559) | def subsample_labels(labels: torch.Tensor, num_samples: int, positive_fr...
  function sample_topk_per_gt (line 2602) | def sample_topk_per_gt(pr_inds, gt_inds, iou, k):
  class DetaStage2Assigner (line 2617) | class DetaStage2Assigner(nn.Module):
    method __init__ (line 2618) | def __init__(self, num_queries, max_k=4):
    method _sample_proposals (line 2626) | def _sample_proposals(self, matched_idxs: torch.Tensor, matched_labels...
    method forward (line 2662) | def forward(self, outputs, targets, return_cost_matrix=False):
    method postprocess_indices (line 2691) | def postprocess_indices(self, pr_inds, gt_inds, iou):
  class DetaStage1Assigner (line 2696) | class DetaStage1Assigner(nn.Module):
    method __init__ (line 2697) | def __init__(self, t_low=0.3, t_high=0.7, max_k=4):
    method _subsample_labels (line 2708) | def _subsample_labels(self, label):
    method forward (line 2723) | def forward(self, outputs, targets):
    method postprocess_indices (line 2753) | def postprocess_indices(self, pr_inds, gt_inds, iou):

FILE: transformers/models/detr/configuration_detr.py
  class DetrConfig (line 37) | class DetrConfig(PretrainedConfig):
    method __init__ (line 143) | def __init__(
    method num_attention_heads (line 232) | def num_attention_heads(self) -> int:
    method hidden_size (line 236) | def hidden_size(self) -> int:
    method from_backbone_config (line 240) | def from_backbone_config(cls, backbone_config: PretrainedConfig, **kwa...
    method to_dict (line 251) | def to_dict(self) -> Dict[str, any]:
  class DetrOnnxConfig (line 263) | class DetrOnnxConfig(OnnxConfig):
    method inputs (line 267) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 276) | def atol_for_validation(self) -> float:
    method default_onnx_opset (line 280) | def default_onnx_opset(self) -> int:

FILE: transformers/models/detr/convert_detr_original_pytorch_checkpoint_to_pytorch.py
  function rename_key (line 111) | def rename_key(state_dict, old, new):
  function rename_backbone_keys (line 116) | def rename_backbone_keys(state_dict):
  function read_in_q_k_v (line 128) | def read_in_q_k_v(state_dict, is_panoptic=False):
  function prepare_img (line 172) | def prepare_img():
  function convert_detr_checkpoint (line 180) | def convert_detr_checkpoint(model_name, pytorch_dump_folder_path):

FILE: transformers/models/detr/convert_detr_to_pytorch.py
  function get_detr_config (line 35) | def get_detr_config(model_name):
  function create_rename_keys (line 62) | def create_rename_keys(config):
  function rename_key (line 232) | def rename_key(state_dict, old, new):
  function read_in_q_k_v (line 237) | def read_in_q_k_v(state_dict, is_panoptic=False):
  function prepare_img (line 281) | def prepare_img():
  function convert_detr_checkpoint (line 289) | def convert_detr_checkpoint(model_name, pytorch_dump_folder_path=None, p...

FILE: transformers/models/detr/feature_extraction_detr.py
  class DetrFeatureExtractor (line 26) | class DetrFeatureExtractor(DetrImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/detr/image_processing_detr.py
  class AnnotionFormat (line 85) | class AnnotionFormat(ExplicitEnum):
  function get_size_with_aspect_ratio (line 93) | def get_size_with_aspect_ratio(image_size, size, max_size=None) -> Tuple...
  function get_resize_output_image_size (line 124) | def get_resize_output_image_size(
  function get_numpy_to_framework_fn (line 147) | def get_numpy_to_framework_fn(arr) -> Callable:
  function safe_squeeze (line 171) | def safe_squeeze(arr: np.ndarray, axis: Optional[int] = None) -> np.ndar...
  function normalize_annotation (line 184) | def normalize_annotation(annotation: Dict, image_size: Tuple[int, int]) ...
  function max_across_indices (line 199) | def max_across_indices(values: Iterable[Any]) -> List[Any]:
  function get_max_height_width (line 207) | def get_max_height_width(images: List[np.ndarray]) -> List[int]:
  function make_pixel_mask (line 223) | def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> ...
  function convert_coco_poly_to_mask (line 240) | def convert_coco_poly_to_mask(segmentations, height: int, width: int) ->...
  function prepare_coco_detection_annotation (line 275) | def prepare_coco_detection_annotation(image, target, return_segmentation...
  function masks_to_boxes (line 327) | def masks_to_boxes(masks: np.ndarray) -> np.ndarray:
  function prepare_coco_panoptic_annotation (line 361) | def prepare_coco_panoptic_annotation(
  function get_segmentation_image (line 398) | def get_segmentation_image(
  function get_mask_area (line 423) | def get_mask_area(seg_img: np.ndarray, target_size: Tuple[int, int], n_c...
  function score_labels_from_class_probabilities (line 432) | def score_labels_from_class_probabilities(logits: np.ndarray) -> Tuple[n...
  function post_process_panoptic_sample (line 440) | def post_process_panoptic_sample(
  function resize_annotation (line 525) | def resize_annotation(
  function binary_mask_to_rle (line 577) | def binary_mask_to_rle(mask):
  function convert_segmentation_to_rle (line 600) | def convert_segmentation_to_rle(segmentation):
  function remove_low_and_no_objects (line 621) | def remove_low_and_no_objects(masks, scores, labels, object_mask_thresho...
  function check_segment_validity (line 649) | def check_segment_validity(mask_labels, mask_probs, k, mask_threshold=0....
  function compute_segments (line 667) | def compute_segments(
  class DetrImageProcessor (line 727) | class DetrImageProcessor(BaseImageProcessor):
    method __init__ (line 764) | def __init__(
    method max_size (line 806) | def max_size(self):
    method from_dict (line 814) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method prepare_annotation (line 827) | def prepare_annotation(
    method prepare (line 852) | def prepare(self, image, target, return_segmentation_masks=None, masks...
    method convert_coco_poly_to_mask (line 861) | def convert_coco_poly_to_mask(self, *args, **kwargs):
    method prepare_coco_detection (line 867) | def prepare_coco_detection(self, *args, **kwargs):
    method prepare_coco_panoptic (line 873) | def prepare_coco_panoptic(self, *args, **kwargs):
    method resize (line 879) | def resize(
    method resize_annotation (line 912) | def resize_annotation(
    method rescale (line 925) | def rescale(
    method normalize (line 933) | def normalize(
    method normalize_annotation (line 945) | def normalize_annotation(self, annotation: Dict, image_size: Tuple[int...
    method pad_and_create_pixel_mask (line 952) | def pad_and_create_pixel_mask(
    method _pad_image (line 985) | def _pad_image(
    method pad (line 1006) | def pad(
    method preprocess (line 1044) | def preprocess(
    method post_process (line 1251) | def post_process(self, outputs, target_sizes):
    method post_process_segmentation (line 1292) | def post_process_segmentation(self, outputs, target_sizes, threshold=0...
    method post_process_instance (line 1337) | def post_process_instance(self, results, outputs, orig_target_sizes, m...
    method post_process_panoptic (line 1384) | def post_process_panoptic(self, outputs, processed_sizes, target_sizes...
    method post_process_object_detection (line 1522) | def post_process_object_detection(
    method post_process_semantic_segmentation (line 1575) | def post_process_semantic_segmentation(self, outputs, target_sizes: Li...
    method post_process_instance_segmentation (line 1623) | def post_process_instance_segmentation(
    method post_process_panoptic_segmentation (line 1707) | def post_process_panoptic_segmentation(

FILE: transformers/models/detr/modeling_detr.py
  class DetrDecoderOutput (line 65) | class DetrDecoderOutput(BaseModelOutputWithCrossAttentions):
  class DetrModelOutput (line 95) | class DetrModelOutput(Seq2SeqModelOutput):
  class DetrObjectDetectionOutput (line 135) | class DetrObjectDetectionOutput(ModelOutput):
  class DetrSegmentationOutput (line 198) | class DetrSegmentationOutput(ModelOutput):
  class DetrFrozenBatchNorm2d (line 269) | class DetrFrozenBatchNorm2d(nn.Module):
    method __init__ (line 277) | def __init__(self, n):
    method _load_from_state_dict (line 284) | def _load_from_state_dict(
    method forward (line 295) | def forward(self, x):
  function replace_batch_norm (line 308) | def replace_batch_norm(m, name=""):
  class DetrConvEncoder (line 323) | class DetrConvEncoder(nn.Module):
    method __init__ (line 331) | def __init__(self, config):
    method forward (line 370) | def forward(self, pixel_values: torch.Tensor, pixel_mask: torch.Tensor):
  class DetrConvModel (line 382) | class DetrConvModel(nn.Module):
    method __init__ (line 387) | def __init__(self, conv_encoder, position_embedding):
    method forward (line 392) | def forward(self, pixel_values, pixel_mask):
  function _expand_mask (line 403) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, target_len: Opt...
  class DetrSinePositionEmbedding (line 417) | class DetrSinePositionEmbedding(nn.Module):
    method __init__ (line 423) | def __init__(self, embedding_dim=64, temperature=10000, normalize=Fals...
    method forward (line 434) | def forward(self, pixel_values, pixel_mask):
  class DetrLearnedPositionEmbedding (line 454) | class DetrLearnedPositionEmbedding(nn.Module):
    method __init__ (line 459) | def __init__(self, embedding_dim=256):
    method forward (line 464) | def forward(self, pixel_values, pixel_mask=None):
  function build_position_encoding (line 477) | def build_position_encoding(config):
  class DetrAttention (line 490) | class DetrAttention(nn.Module):
    method __init__ (line 497) | def __init__(
    method _shape (line 522) | def _shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method with_pos_embed (line 525) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 528) | def forward(
  class DetrEncoderLayer (line 621) | class DetrEncoderLayer(nn.Module):
    method __init__ (line 622) | def __init__(self, config: DetrConfig):
    method forward (line 638) | def forward(
  class DetrDecoderLayer (line 691) | class DetrDecoderLayer(nn.Module):
    method __init__ (line 692) | def __init__(self, config: DetrConfig):
    method forward (line 718) | def forward(
  class DetrClassificationHead (line 798) | class DetrClassificationHead(nn.Module):
    method __init__ (line 801) | def __init__(self, input_dim: int, inner_dim: int, num_classes: int, p...
    method forward (line 807) | def forward(self, hidden_states: torch.Tensor):
  class DetrPreTrainedModel (line 816) | class DetrPreTrainedModel(PreTrainedModel):
    method _init_weights (line 821) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 844) | def _set_gradient_checkpointing(self, module, value=False):
  class DetrEncoder (line 903) | class DetrEncoder(DetrPreTrainedModel):
    method __init__ (line 918) | def __init__(self, config: DetrConfig):
    method forward (line 931) | def forward(
  class DetrDecoder (line 1012) | class DetrDecoder(DetrPreTrainedModel):
    method __init__ (line 1027) | def __init__(self, config: DetrConfig):
    method forward (line 1040) | def forward(
  class DetrModel (line 1200) | class DetrModel(DetrPreTrainedModel):
    method __init__ (line 1201) | def __init__(self, config: DetrConfig):
    method get_encoder (line 1220) | def get_encoder(self):
    method get_decoder (line 1223) | def get_decoder(self):
    method freeze_backbone (line 1226) | def freeze_backbone(self):
    method unfreeze_backbone (line 1230) | def unfreeze_backbone(self):
    method forward (line 1236) | def forward(
  class DetrForObjectDetection (line 1368) | class DetrForObjectDetection(DetrPreTrainedModel):
    method __init__ (line 1369) | def __init__(self, config: DetrConfig):
    method _set_aux_loss (line 1388) | def _set_aux_loss(self, outputs_class, outputs_coord):
    method forward (line 1396) | def forward(
  class DetrForSegmentation (line 1542) | class DetrForSegmentation(DetrPreTrainedModel):
    method __init__ (line 1543) | def __init__(self, config: DetrConfig):
    method forward (line 1566) | def forward(
  function _expand (line 1768) | def _expand(tensor, length: int):
  class DetrMaskHeadSmallConv (line 1773) | class DetrMaskHeadSmallConv(nn.Module):
    method __init__ (line 1778) | def __init__(self, dim, fpn_dims, context_dim):
    method forward (line 1812) | def forward(self, x: Tensor, bbox_mask: Tensor, fpns: List[Tensor]):
  class DetrMHAttentionMap (line 1853) | class DetrMHAttentionMap(nn.Module):
    method __init__ (line 1856) | def __init__(self, query_dim, hidden_dim, num_heads, dropout=0.0, bias...
    method forward (line 1867) | def forward(self, q, k, mask: Optional[Tensor] = None):
  function dice_loss (line 1881) | def dice_loss(inputs, targets, num_boxes):
  function sigmoid_focal_loss (line 1900) | def sigmoid_focal_loss(inputs, targets, num_boxes, alpha: float = 0.25, ...
  class DetrLoss (line 1932) | class DetrLoss(nn.Module):
    method __init__ (line 1957) | def __init__(self, matcher, num_classes, eos_coef, losses):
    method loss_labels (line 1968) | def loss_labels(self, outputs, targets, indices, num_boxes):
    method loss_cardinality (line 1990) | def loss_cardinality(self, outputs, targets, indices, num_boxes):
    method loss_boxes (line 2005) | def loss_boxes(self, outputs, targets, indices, num_boxes):
    method loss_masks (line 2029) | def loss_masks(self, outputs, targets, indices, num_boxes):
    method _get_source_permutation_idx (line 2062) | def _get_source_permutation_idx(self, indices):
    method _get_target_permutation_idx (line 2068) | def _get_target_permutation_idx(self, indices):
    method get_loss (line 2074) | def get_loss(self, loss, outputs, targets, indices, num_boxes):
    method forward (line 2085) | def forward(self, outputs, targets):
  class DetrMLPPredictionHead (line 2131) | class DetrMLPPredictionHead(nn.Module):
    method __init__ (line 2140) | def __init__(self, input_dim, hidden_dim, output_dim, num_layers):
    method forward (line 2146) | def forward(self, x):
  class DetrHungarianMatcher (line 2153) | class DetrHungarianMatcher(nn.Module):
    method __init__ (line 2170) | def __init__(self, class_cost: float = 1, bbox_cost: float = 1, giou_c...
    method forward (line 2181) | def forward(self, outputs, targets):
  function _upcast (line 2234) | def _upcast(t: Tensor) -> Tensor:
  function box_area (line 2242) | def box_area(boxes: Tensor) -> Tensor:
  function box_iou (line 2259) | def box_iou(boxes1, boxes2):
  function generalized_box_iou (line 2275) | def generalized_box_iou(boxes1, boxes2):
  function _max_by_axis (line 2300) | def _max_by_axis(the_list):
  class NestedTensor (line 2309) | class NestedTensor(object):
    method __init__ (line 2310) | def __init__(self, tensors, mask: Optional[Tensor]):
    method to (line 2314) | def to(self, device):
    method decompose (line 2323) | def decompose(self):
    method __repr__ (line 2326) | def __repr__(self):
  function nested_tensor_from_tensor_list (line 2330) | def nested_tensor_from_tensor_list(tensor_list: List[Tensor]):

FILE: transformers/models/dialogpt/convert_dialogpt_original_pytorch_checkpoint_to_pytorch.py
  function convert_dialogpt_checkpoint (line 29) | def convert_dialogpt_checkpoint(checkpoint_path: str, pytorch_dump_folde...

FILE: transformers/models/dinat/configuration_dinat.py
  class DinatConfig (line 30) | class DinatConfig(BackboneConfigMixin, PretrainedConfig):
    method __init__ (line 104) | def __init__(

FILE: transformers/models/dinat/modeling_dinat.py
  function natten2dqkrpb (line 50) | def natten2dqkrpb(*args, **kwargs):
  function natten2dav (line 53) | def natten2dav(*args, **kwargs):
  class DinatEncoderOutput (line 81) | class DinatEncoderOutput(ModelOutput):
  class DinatModelOutput (line 115) | class DinatModelOutput(ModelOutput):
  class DinatImageClassifierOutput (line 152) | class DinatImageClassifierOutput(ModelOutput):
  class DinatEmbeddings (line 188) | class DinatEmbeddings(nn.Module):
    method __init__ (line 193) | def __init__(self, config):
    method forward (line 201) | def forward(self, pixel_values: Optional[torch.FloatTensor]) -> Tuple[...
  class DinatPatchEmbeddings (line 211) | class DinatPatchEmbeddings(nn.Module):
    method __init__ (line 218) | def __init__(self, config):
    method forward (line 235) | def forward(self, pixel_values: Optional[torch.FloatTensor]) -> torch....
  class DinatDownsampler (line 248) | class DinatDownsampler(nn.Module):
    method __init__ (line 259) | def __init__(self, dim: int, norm_layer: nn.Module = nn.LayerNorm) -> ...
    method forward (line 265) | def forward(self, input_feature: torch.Tensor) -> torch.Tensor:
  function drop_path (line 272) | def drop_path(input, drop_prob=0.0, training=False, scale_by_keep=True):
  class DinatDropPath (line 293) | class DinatDropPath(nn.Module):
    method __init__ (line 296) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 300) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 303) | def extra_repr(self) -> str:
  class NeighborhoodAttention (line 307) | class NeighborhoodAttention(nn.Module):
    method __init__ (line 308) | def __init__(self, config, dim, num_heads, kernel_size, dilation):
    method transpose_for_scores (line 331) | def transpose_for_scores(self, x):
    method forward (line 336) | def forward(
  class NeighborhoodAttentionOutput (line 371) | class NeighborhoodAttentionOutput(nn.Module):
    method __init__ (line 372) | def __init__(self, config, dim):
    method forward (line 377) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class NeighborhoodAttentionModule (line 384) | class NeighborhoodAttentionModule(nn.Module):
    method __init__ (line 385) | def __init__(self, config, dim, num_heads, kernel_size, dilation):
    method prune_heads (line 392) | def prune_heads(self, heads):
    method forward (line 411) | def forward(
  class DinatIntermediate (line 423) | class DinatIntermediate(nn.Module):
    method __init__ (line 424) | def __init__(self, config, dim):
    method forward (line 432) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class DinatOutput (line 439) | class DinatOutput(nn.Module):
    method __init__ (line 440) | def __init__(self, config, dim):
    method forward (line 445) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class DinatLayer (line 451) | class DinatLayer(nn.Module):
    method __init__ (line 452) | def __init__(self, config, dim, num_heads, dilation, drop_path_rate=0.0):
    method maybe_pad (line 472) | def maybe_pad(self, hidden_states, height, width):
    method forward (line 483) | def forward(
  class DinatStage (line 522) | class DinatStage(nn.Module):
    method __init__ (line 523) | def __init__(self, config, dim, depth, num_heads, dilations, drop_path...
    method forward (line 549) | def forward(
  class DinatEncoder (line 570) | class DinatEncoder(nn.Module):
    method __init__ (line 571) | def __init__(self, config):
    method forward (line 592) | def forward(
  class DinatPreTrainedModel (line 641) | class DinatPreTrainedModel(PreTrainedModel):
    method _init_weights (line 651) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 663) | def _set_gradient_checkpointing(self, module: DinatEncoder, value: boo...
  class DinatModel (line 700) | class DinatModel(DinatPreTrainedModel):
    method __init__ (line 701) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 719) | def get_input_embeddings(self):
    method _prune_heads (line 722) | def _prune_heads(self, heads_to_prune):
    method forward (line 738) | def forward(
  class DinatForImageClassification (line 792) | class DinatForImageClassification(DinatPreTrainedModel):
    method __init__ (line 793) | def __init__(self, config):
    method forward (line 816) | def forward(
  class DinatBackbone (line 883) | class DinatBackbone(DinatPreTrainedModel, BackboneMixin):
    method __init__ (line 884) | def __init__(self, config):
    method get_input_embeddings (line 903) | def get_input_embeddings(self):
    method forward (line 908) | def forward(

FILE: transformers/models/distilbert/configuration_distilbert.py
  class DistilBertConfig (line 45) | class DistilBertConfig(PretrainedConfig):
    method __init__ (line 108) | def __init__(
  class DistilBertOnnxConfig (line 142) | class DistilBertOnnxConfig(OnnxConfig):
    method inputs (line 144) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/distilbert/modeling_distilbert.py
  function create_sinusoidal_embeddings (line 72) | def create_sinusoidal_embeddings(n_pos: int, dim: int, out: torch.Tensor):
  function _create_sinusoidal_embeddings (line 83) | def _create_sinusoidal_embeddings(n_pos: int, dim: int, out: torch.Tensor):
  class Embeddings (line 91) | class Embeddings(nn.Module):
    method __init__ (line 92) | def __init__(self, config: PretrainedConfig):
    method forward (line 107) | def forward(self, input_ids: torch.Tensor, input_embeds: Optional[torc...
  class MultiHeadSelfAttention (line 141) | class MultiHeadSelfAttention(nn.Module):
    method __init__ (line 142) | def __init__(self, config: PretrainedConfig):
    method prune_heads (line 162) | def prune_heads(self, heads: List[int]):
    method forward (line 178) | def forward(
  class FFN (line 243) | class FFN(nn.Module):
    method __init__ (line 244) | def __init__(self, config: PretrainedConfig):
    method forward (line 253) | def forward(self, input: torch.Tensor) -> torch.Tensor:
    method ff_chunk (line 256) | def ff_chunk(self, input: torch.Tensor) -> torch.Tensor:
  class TransformerBlock (line 264) | class TransformerBlock(nn.Module):
    method __init__ (line 265) | def __init__(self, config: PretrainedConfig):
    method forward (line 278) | def forward(
  class Transformer (line 322) | class Transformer(nn.Module):
    method __init__ (line 323) | def __init__(self, config: PretrainedConfig):
    method forward (line 328) | def forward(
  class DistilBertPreTrainedModel (line 386) | class DistilBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 396) | def _init_weights(self, module: nn.Module):
  class DistilBertModel (line 470) | class DistilBertModel(DistilBertPreTrainedModel):
    method __init__ (line 471) | def __init__(self, config: PretrainedConfig):
    method get_position_embeddings (line 480) | def get_position_embeddings(self) -> nn.Embedding:
    method resize_position_embeddings (line 486) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_input_embeddings (line 528) | def get_input_embeddings(self) -> nn.Embedding:
    method set_input_embeddings (line 531) | def set_input_embeddings(self, new_embeddings: nn.Embedding):
    method _prune_heads (line 534) | def _prune_heads(self, heads_to_prune: Dict[int, List[List[int]]]):
    method forward (line 548) | def forward(
  class DistilBertForMaskedLM (line 597) | class DistilBertForMaskedLM(DistilBertPreTrainedModel):
    method __init__ (line 600) | def __init__(self, config: PretrainedConfig):
    method get_position_embeddings (line 615) | def get_position_embeddings(self) -> nn.Embedding:
    method resize_position_embeddings (line 621) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_output_embeddings (line 635) | def get_output_embeddings(self) -> nn.Module:
    method set_output_embeddings (line 638) | def set_output_embeddings(self, new_embeddings: nn.Module):
    method forward (line 647) | def forward(
  class DistilBertForSequenceClassification (line 704) | class DistilBertForSequenceClassification(DistilBertPreTrainedModel):
    method __init__ (line 705) | def __init__(self, config: PretrainedConfig):
    method get_position_embeddings (line 718) | def get_position_embeddings(self) -> nn.Embedding:
    method resize_position_embeddings (line 724) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method forward (line 744) | def forward(
  class DistilBertForQuestionAnswering (line 821) | class DistilBertForQuestionAnswering(DistilBertPreTrainedModel):
    method __init__ (line 822) | def __init__(self, config: PretrainedConfig):
    method get_position_embeddings (line 835) | def get_position_embeddings(self) -> nn.Embedding:
    method resize_position_embeddings (line 841) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method forward (line 861) | def forward(
  class DistilBertForTokenClassification (line 939) | class DistilBertForTokenClassification(DistilBertPreTrainedModel):
    method __init__ (line 940) | def __init__(self, config: PretrainedConfig):
    method get_position_embeddings (line 951) | def get_position_embeddings(self) -> nn.Embedding:
    method resize_position_embeddings (line 957) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method forward (line 977) | def forward(
  class DistilBertForMultipleChoice (line 1033) | class DistilBertForMultipleChoice(DistilBertPreTrainedModel):
    method __init__ (line 1034) | def __init__(self, config: PretrainedConfig):
    method get_position_embeddings (line 1045) | def get_position_embeddings(self) -> nn.Embedding:
    method resize_position_embeddings (line 1051) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method forward (line 1069) | def forward(

FILE: transformers/models/distilbert/modeling_flax_distilbert.py
  function get_angles (line 95) | def get_angles(pos, i, d_model):
  function positional_encoding (line 100) | def positional_encoding(position, d_model):
  class FlaxEmbeddings (line 115) | class FlaxEmbeddings(nn.Module):
    method setup (line 121) | def setup(self):
    method __call__ (line 138) | def __call__(self, input_ids, deterministic: bool = True):
  class FlaxMultiHeadSelfAttention (line 160) | class FlaxMultiHeadSelfAttention(nn.Module):
    method setup (line 164) | def setup(self):
    method __call__ (line 193) | def __call__(
  class FlaxFFN (line 243) | class FlaxFFN(nn.Module):
    method setup (line 247) | def setup(self):
    method __call__ (line 264) | def __call__(self, hidden_states, deterministic: bool = True):
  class FlaxTransformerBlock (line 272) | class FlaxTransformerBlock(nn.Module):
    method setup (line 276) | def setup(self):
    method __call__ (line 287) | def __call__(
  class FlaxTransformer (line 319) | class FlaxTransformer(nn.Module):
    method setup (line 323) | def setup(self):
    method __call__ (line 328) | def __call__(
  class FlaxTransformerEncoder (line 370) | class FlaxTransformerEncoder(nn.Module):
    method setup (line 374) | def setup(self):
    method __call__ (line 377) | def __call__(
  class FlaxDistilBertLMDecoder (line 396) | class FlaxDistilBertLMDecoder(nn.Module):
    method setup (line 401) | def setup(self):
    method __call__ (line 404) | def __call__(self, inputs, kernel):
  class FlaxDistilBertPreTrainedModel (line 413) | class FlaxDistilBertPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 423) | def __init__(
    method init_weights (line 435) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 456) | def __call__(
  class FlaxDistilBertModule (line 494) | class FlaxDistilBertModule(nn.Module):
    method setup (line 498) | def setup(self):
    method __call__ (line 502) | def __call__(
  class FlaxDistilBertModel (line 532) | class FlaxDistilBertModel(FlaxDistilBertPreTrainedModel):
  class FlaxDistilBertForMaskedLMModule (line 539) | class FlaxDistilBertForMaskedLMModule(nn.Module):
    method setup (line 543) | def setup(self):
    method __call__ (line 563) | def __call__(
  class FlaxDistilBertForMaskedLM (line 605) | class FlaxDistilBertForMaskedLM(FlaxDistilBertPreTrainedModel):
  class FlaxDistilBertForSequenceClassificationModule (line 612) | class FlaxDistilBertForSequenceClassificationModule(nn.Module):
    method setup (line 616) | def setup(self):
    method __call__ (line 629) | def __call__(
  class FlaxDistilBertForSequenceClassification (line 672) | class FlaxDistilBertForSequenceClassification(FlaxDistilBertPreTrainedMo...
  class FlaxDistilBertForMultipleChoiceModule (line 684) | class FlaxDistilBertForMultipleChoiceModule(nn.Module):
    method setup (line 688) | def setup(self):
    method __call__ (line 701) | def __call__(
  class FlaxDistilBertForMultipleChoice (line 751) | class FlaxDistilBertForMultipleChoice(FlaxDistilBertPreTrainedModel):
  class FlaxDistilBertForTokenClassificationModule (line 766) | class FlaxDistilBertForTokenClassificationModule(nn.Module):
    method setup (line 770) | def setup(self):
    method __call__ (line 775) | def __call__(
  class FlaxDistilBertForTokenClassification (line 816) | class FlaxDistilBertForTokenClassification(FlaxDistilBertPreTrainedModel):
  class FlaxDistilBertForQuestionAnsweringModule (line 828) | class FlaxDistilBertForQuestionAnsweringModule(nn.Module):
    method setup (line 832) | def setup(self):
    method __call__ (line 838) | def __call__(
  class FlaxDistilBertForQuestionAnswering (line 885) | class FlaxDistilBertForQuestionAnswering(FlaxDistilBertPreTrainedModel):

FILE: transformers/models/distilbert/modeling_tf_distilbert.py
  class TFEmbeddings (line 75) | class TFEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 78) | def __init__(self, config, **kwargs):
    method build (line 87) | def build(self, input_shape: tf.TensorShape):
    method call (line 104) | def call(self, input_ids=None, position_ids=None, inputs_embeds=None, ...
  class TFMultiHeadSelfAttention (line 130) | class TFMultiHeadSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 131) | def __init__(self, config, **kwargs):
    method prune_heads (line 156) | def prune_heads(self, heads):
    method call (line 159) | def call(self, query, key, value, mask, head_mask, output_attentions, ...
  class TFFFN (line 216) | class TFFFN(tf.keras.layers.Layer):
    method __init__ (line 217) | def __init__(self, config, **kwargs):
    method call (line 228) | def call(self, input, training=False):
  class TFTransformerBlock (line 236) | class TFTransformerBlock(tf.keras.layers.Layer):
    method __init__ (line 237) | def __init__(self, config, **kwargs):
    method call (line 257) | def call(self, x, attn_mask, head_mask, output_attentions, training=Fa...
  class TFTransformer (line 285) | class TFTransformer(tf.keras.layers.Layer):
    method __init__ (line 286) | def __init__(self, config, **kwargs):
    method call (line 294) | def call(self, x, attn_mask, head_mask, output_attentions, output_hidd...
  class TFDistilBertMainLayer (line 341) | class TFDistilBertMainLayer(tf.keras.layers.Layer):
    method __init__ (line 344) | def __init__(self, config, **kwargs):
    method get_input_embeddings (line 356) | def get_input_embeddings(self):
    method set_input_embeddings (line 359) | def set_input_embeddings(self, value):
    method _prune_heads (line 363) | def _prune_heads(self, heads_to_prune):
    method call (line 367) | def call(
  class TFDistilBertPreTrainedModel (line 417) | class TFDistilBertPreTrainedModel(TFPreTrainedModel):
  class TFDistilBertModel (line 516) | class TFDistilBertModel(TFDistilBertPreTrainedModel):
    method __init__ (line 517) | def __init__(self, config, *inputs, **kwargs):
    method call (line 528) | def call(
  class TFDistilBertLMHead (line 552) | class TFDistilBertLMHead(tf.keras.layers.Layer):
    method __init__ (line 553) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 563) | def build(self, input_shape):
    method get_output_embeddings (line 568) | def get_output_embeddings(self):
    method set_output_embeddings (line 571) | def set_output_embeddings(self, value):
    method get_bias (line 575) | def get_bias(self):
    method set_bias (line 578) | def set_bias(self, value):
    method call (line 582) | def call(self, hidden_states):
  class TFDistilBertForMaskedLM (line 596) | class TFDistilBertForMaskedLM(TFDistilBertPreTrainedModel, TFMaskedLangu...
    method __init__ (line 597) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 609) | def get_lm_head(self):
    method get_prefix_bias_name (line 612) | def get_prefix_bias_name(self):
    method call (line 623) | def call(
  class TFDistilBertForSequenceClassification (line 678) | class TFDistilBertForSequenceClassification(TFDistilBertPreTrainedModel,...
    method __init__ (line 679) | def __init__(self, config, *inputs, **kwargs):
    method call (line 702) | def call(
  class TFDistilBertForTokenClassification (line 757) | class TFDistilBertForTokenClassification(TFDistilBertPreTrainedModel, TF...
    method __init__ (line 758) | def __init__(self, config, *inputs, **kwargs):
    method call (line 775) | def call(
  class TFDistilBertForMultipleChoice (line 825) | class TFDistilBertForMultipleChoice(TFDistilBertPreTrainedModel, TFMulti...
    method __init__ (line 826) | def __init__(self, config, *inputs, **kwargs):
    method call (line 850) | def call(
  class TFDistilBertForQuestionAnswering (line 919) | class TFDistilBertForQuestionAnswering(TFDistilBertPreTrainedModel, TFQu...
    method __init__ (line 920) | def __init__(self, config, *inputs, **kwargs):
    method call (line 937) | def call(

FILE: transformers/models/distilbert/tokenization_distilbert.py
  function load_vocab (line 68) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 80) | def whitespace_tokenize(text):
  class DistilBertTokenizer (line 89) | class DistilBertTokenizer(PreTrainedTokenizer):
    method __init__ (line 137) | def __init__(
    method do_lower_case (line 185) | def do_lower_case(self):
    method vocab_size (line 190) | def vocab_size(self):
    method get_vocab (line 194) | def get_vocab(self):
    method _tokenize (line 198) | def _tokenize(self, text):
    method _convert_token_to_id (line 212) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 217) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 222) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 228) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 254) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 283) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 313) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 335) | class BasicTokenizer(object):
    method __init__ (line 355) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 363) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 400) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 411) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 433) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 446) | def _is_chinese_char(self, cp):
    method _clean_text (line 470) | def _clean_text(self, text):
  class WordpieceTokenizer (line 485) | class WordpieceTokenizer(object):
    method __init__ (line 488) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 493) | def tokenize(self, text):

FILE: transformers/models/distilbert/tokenization_distilbert_fast.py
  class DistilBertTokenizerFast (line 84) | class DistilBertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 131) | def __init__(
    method build_inputs_with_special_tokens (line 174) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 199) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 229) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/dit/convert_dit_unilm_to_pytorch.py
  function create_rename_keys (line 37) | def create_rename_keys(config, has_lm_head=False, is_semantic=False):
  function read_in_q_k_v (line 92) | def read_in_q_k_v(state_dict, config, has_lm_head=False, is_semantic=Fal...
  function rename_key (line 121) | def rename_key(dct, old, new):
  function prepare_img (line 127) | def prepare_img():
  function convert_dit_checkpoint (line 134) | def convert_dit_checkpoint(checkpoint_url, pytorch_dump_folder_path, pus...

FILE: transformers/models/donut/configuration_donut_swin.py
  class DonutSwinConfig (line 29) | class DonutSwinConfig(PretrainedConfig):
    method __init__ (line 95) | def __init__(

FILE: transformers/models/donut/convert_donut_to_pytorch.py
  function get_configs (line 35) | def get_configs(model):
  function rename_key (line 62) | def rename_key(name):
  function convert_state_dict (line 95) | def convert_state_dict(orig_state_dict, model):
  function convert_donut_checkpoint (line 135) | def convert_donut_checkpoint(model_name, pytorch_dump_folder_path=None, ...

FILE: transformers/models/donut/feature_extraction_donut.py
  class DonutFeatureExtractor (line 26) | class DonutFeatureExtractor(DonutImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/donut/image_processing_donut.py
  class DonutImageProcessor (line 52) | class DonutImageProcessor(BaseImageProcessor):
    method __init__ (line 91) | def __init__(
    method align_long_axis (line 126) | def align_long_axis(
    method rotate_image (line 154) | def rotate_image(self, *args, **kwargs):
    method pad_image (line 160) | def pad_image(
    method pad (line 199) | def pad(self, *args, **kwargs):
    method thumbnail (line 203) | def thumbnail(
    method resize (line 244) | def resize(
    method rescale (line 272) | def rescale(
    method normalize (line 292) | def normalize(
    method preprocess (line 315) | def preprocess(

FILE: transformers/models/donut/modeling_donut_swin.py
  class DonutSwinEncoderOutput (line 59) | class DonutSwinEncoderOutput(ModelOutput):
  class DonutSwinModelOutput (line 93) | class DonutSwinModelOutput(ModelOutput):
  function window_partition (line 129) | def window_partition(input_feature, window_size):
  function window_reverse (line 142) | def window_reverse(windows, window_size, height, width):
  class DonutSwinEmbeddings (line 153) | class DonutSwinEmbeddings(nn.Module):
    method __init__ (line 158) | def __init__(self, config, use_mask_token=False):
    method forward (line 174) | def forward(
  class DonutSwinPatchEmbeddings (line 196) | class DonutSwinPatchEmbeddings(nn.Module):
    method __init__ (line 203) | def __init__(self, config):
    method maybe_pad (line 218) | def maybe_pad(self, pixel_values, height, width):
    method forward (line 227) | def forward(self, pixel_values: Optional[torch.FloatTensor]) -> Tuple[...
  class DonutSwinPatchMerging (line 244) | class DonutSwinPatchMerging(nn.Module):
    method __init__ (line 257) | def __init__(self, input_resolution: Tuple[int], dim: int, norm_layer:...
    method maybe_pad (line 264) | def maybe_pad(self, input_feature, height, width):
    method forward (line 272) | def forward(self, input_feature: torch.Tensor, input_dimensions: Tuple...
  function drop_path (line 299) | def drop_path(input, drop_prob=0.0, training=False, scale_by_keep=True):
  class DonutSwinDropPath (line 320) | class DonutSwinDropPath(nn.Module):
    method __init__ (line 323) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 327) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 330) | def extra_repr(self) -> str:
  class DonutSwinSelfAttention (line 335) | class DonutSwinSelfAttention(nn.Module):
    method __init__ (line 336) | def __init__(self, config, dim, num_heads, window_size):
    method transpose_for_scores (line 373) | def transpose_for_scores(self, x):
    method forward (line 378) | def forward(
  class DonutSwinSelfOutput (line 436) | class DonutSwinSelfOutput(nn.Module):
    method __init__ (line 437) | def __init__(self, config, dim):
    method forward (line 442) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class DonutSwinAttention (line 450) | class DonutSwinAttention(nn.Module):
    method __init__ (line 451) | def __init__(self, config, dim, num_heads, window_size):
    method prune_heads (line 457) | def prune_heads(self, heads):
    method forward (line 475) | def forward(
  class DonutSwinIntermediate (line 489) | class DonutSwinIntermediate(nn.Module):
    method __init__ (line 490) | def __init__(self, config, dim):
    method forward (line 498) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class DonutSwinOutput (line 505) | class DonutSwinOutput(nn.Module):
    method __init__ (line 506) | def __init__(self, config, dim):
    method forward (line 511) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class DonutSwinLayer (line 518) | class DonutSwinLayer(nn.Module):
    method __init__ (line 519) | def __init__(self, config, dim, input_resolution, num_heads, shift_siz...
    method set_shift_and_window_size (line 532) | def set_shift_and_window_size(self, input_resolution):
    method get_attn_mask (line 538) | def get_attn_mask(self, height, width, dtype):
    method maybe_pad (line 566) | def maybe_pad(self, hidden_states, height, width):
    method forward (line 573) | def forward(
  class DonutSwinStage (line 642) | class DonutSwinStage(nn.Module):
    method __init__ (line 643) | def __init__(self, config, dim, input_resolution, depth, num_heads, dr...
    method forward (line 668) | def forward(
  class DonutSwinEncoder (line 702) | class DonutSwinEncoder(nn.Module):
    method __init__ (line 703) | def __init__(self, config, grid_size):
    method forward (line 725) | def forward(
  class DonutSwinPreTrainedModel (line 806) | class DonutSwinPreTrainedModel(PreTrainedModel):
    method _init_weights (line 817) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 829) | def _set_gradient_checkpointing(self, module, value=False):
  class DonutSwinModel (line 871) | class DonutSwinModel(DonutSwinPreTrainedModel):
    method __init__ (line 872) | def __init__(self, config, add_pooling_layer=True, use_mask_token=False):
    method get_input_embeddings (line 886) | def get_input_embeddings(self):
    method _prune_heads (line 889) | def _prune_heads(self, heads_to_prune):
    method forward (line 905) | def forward(

FILE: transformers/models/donut/processing_donut.py
  class DonutProcessor (line 25) | class DonutProcessor(ProcessorMixin):
    method __init__ (line 44) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 63) | def __call__(self, *args, **kwargs):
    method batch_decode (line 96) | def batch_decode(self, *args, **kwargs):
    method decode (line 103) | def decode(self, *args, **kwargs):
    method as_target_processor (line 111) | def as_target_processor(self):
    method token2json (line 126) | def token2json(self, tokens, is_inner_value=False, added_vocab=None):
    method feature_extractor_class (line 177) | def feature_extractor_class(self):
    method feature_extractor (line 185) | def feature_extractor(self):

FILE: transformers/models/dpr/configuration_dpr.py
  class DPRConfig (line 45) | class DPRConfig(PretrainedConfig):
    method __init__ (line 112) | def __init__(

FILE: transformers/models/dpr/convert_dpr_original_checkpoint_to_pytorch.py
  function load_states_from_checkpoint (line 30) | def load_states_from_checkpoint(model_file: str) -> CheckpointState:
  class DPRState (line 36) | class DPRState:
    method __init__ (line 37) | def __init__(self, src_file: Path):
    method load_dpr_model (line 40) | def load_dpr_model(self):
    method from_type (line 44) | def from_type(comp_type: str, *args, **kwargs) -> "DPRState":
  class DPRContextEncoderState (line 55) | class DPRContextEncoderState(DPRState):
    method load_dpr_model (line 56) | def load_dpr_model(self):
  class DPRQuestionEncoderState (line 73) | class DPRQuestionEncoderState(DPRState):
    method load_dpr_model (line 74) | def load_dpr_model(self):
  class DPRReaderState (line 91) | class DPRReaderState(DPRState):
    method load_dpr_model (line 92) | def load_dpr_model(self):
  function convert (line 108) | def convert(comp_type: str, src_file: Path, dest_dir: Path):

FILE: transformers/models/dpr/modeling_dpr.py
  class DPRContextEncoderOutput (line 62) | class DPRContextEncoderOutput(ModelOutput):
  class DPRQuestionEncoderOutput (line 90) | class DPRQuestionEncoderOutput(ModelOutput):
  class DPRReaderOutput (line 118) | class DPRReaderOutput(ModelOutput):
  class DPRPreTrainedModel (line 150) | class DPRPreTrainedModel(PreTrainedModel):
    method _init_weights (line 151) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 167) | def _set_gradient_checkpointing(self, module, value=False):
  class DPREncoder (line 172) | class DPREncoder(DPRPreTrainedModel):
    method __init__ (line 175) | def __init__(self, config: DPRConfig):
    method forward (line 186) | def forward(
    method embeddings_size (line 222) | def embeddings_size(self) -> int:
  class DPRSpanPredictor (line 228) | class DPRSpanPredictor(DPRPreTrainedModel):
    method __init__ (line 231) | def __init__(self, config: DPRConfig):
    method forward (line 239) | def forward(
  class DPRPretrainedContextEncoder (line 290) | class DPRPretrainedContextEncoder(DPRPreTrainedModel):
  class DPRPretrainedQuestionEncoder (line 303) | class DPRPretrainedQuestionEncoder(DPRPreTrainedModel):
  class DPRPretrainedReader (line 316) | class DPRPretrainedReader(DPRPreTrainedModel):
  class DPRContextEncoder (line 446) | class DPRContextEncoder(DPRPretrainedContextEncoder):
    method __init__ (line 447) | def __init__(self, config: DPRConfig):
    method forward (line 456) | def forward(
  class DPRQuestionEncoder (line 527) | class DPRQuestionEncoder(DPRPretrainedQuestionEncoder):
    method __init__ (line 528) | def __init__(self, config: DPRConfig):
    method forward (line 537) | def forward(
  class DPRReader (line 608) | class DPRReader(DPRPretrainedReader):
    method __init__ (line 609) | def __init__(self, config: DPRConfig):
    method forward (line 618) | def forward(

FILE: transformers/models/dpr/modeling_tf_dpr.py
  class TFDPRContextEncoderOutput (line 62) | class TFDPRContextEncoderOutput(ModelOutput):
  class TFDPRQuestionEncoderOutput (line 90) | class TFDPRQuestionEncoderOutput(ModelOutput):
  class TFDPRReaderOutput (line 118) | class TFDPRReaderOutput(ModelOutput):
  class TFDPREncoderLayer (line 150) | class TFDPREncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 153) | def __init__(self, config: DPRConfig, **kwargs):
    method call (line 169) | def call(
    method embeddings_size (line 207) | def embeddings_size(self) -> int:
  class TFDPRSpanPredictorLayer (line 213) | class TFDPRSpanPredictorLayer(tf.keras.layers.Layer):
    method __init__ (line 216) | def __init__(self, config: DPRConfig, **kwargs):
    method call (line 229) | def call(
  class TFDPRSpanPredictor (line 277) | class TFDPRSpanPredictor(TFPreTrainedModel):
    method __init__ (line 280) | def __init__(self, config: DPRConfig, **kwargs):
    method call (line 285) | def call(
  class TFDPREncoder (line 309) | class TFDPREncoder(TFPreTrainedModel):
    method __init__ (line 312) | def __init__(self, config: DPRConfig, **kwargs):
    method call (line 318) | def call(
  class TFDPRPretrainedContextEncoder (line 346) | class TFDPRPretrainedContextEncoder(TFPreTrainedModel):
  class TFDPRPretrainedQuestionEncoder (line 356) | class TFDPRPretrainedQuestionEncoder(TFPreTrainedModel):
  class TFDPRPretrainedReader (line 366) | class TFDPRPretrainedReader(TFPreTrainedModel):
  class TFDPRContextEncoder (line 526) | class TFDPRContextEncoder(TFDPRPretrainedContextEncoder):
    method __init__ (line 527) | def __init__(self, config: DPRConfig, *args, **kwargs):
    method get_input_embeddings (line 531) | def get_input_embeddings(self):
    method call (line 541) | def call(
  class TFDPRQuestionEncoder (line 607) | class TFDPRQuestionEncoder(TFDPRPretrainedQuestionEncoder):
    method __init__ (line 608) | def __init__(self, config: DPRConfig, *args, **kwargs):
    method get_input_embeddings (line 612) | def get_input_embeddings(self):
    method call (line 622) | def call(
  class TFDPRReader (line 687) | class TFDPRReader(TFDPRPretrainedReader):
    method __init__ (line 688) | def __init__(self, config: DPRConfig, *args, **kwargs):
    method get_input_embeddings (line 692) | def get_input_embeddings(self):
    method call (line 702) | def call(

FILE: transformers/models/dpr/tokenization_dpr.py
  class DPRContextEncoderTokenizer (line 113) | class DPRContextEncoderTokenizer(BertTokenizer):
  class DPRQuestionEncoderTokenizer (line 129) | class DPRQuestionEncoderTokenizer(BertTokenizer):
  class CustomDPRReaderTokenizerMixin (line 222) | class CustomDPRReaderTokenizerMixin:
    method __call__ (line 223) | def __call__(
    method decode_best_spans (line 282) | def decode_best_spans(
    method _get_best_spans (line 358) | def _get_best_spans(
  class DPRReaderTokenizer (line 397) | class DPRReaderTokenizer(CustomDPRReaderTokenizerMixin, BertTokenizer):

FILE: transformers/models/dpr/tokenization_dpr_fast.py
  class DPRContextEncoderTokenizerFast (line 114) | class DPRContextEncoderTokenizerFast(BertTokenizerFast):
  class DPRQuestionEncoderTokenizerFast (line 131) | class DPRQuestionEncoderTokenizerFast(BertTokenizerFast):
  class CustomDPRReaderTokenizerMixin (line 223) | class CustomDPRReaderTokenizerMixin:
    method __call__ (line 224) | def __call__(
    method decode_best_spans (line 282) | def decode_best_spans(
    method _get_best_spans (line 358) | def _get_best_spans(
  class DPRReaderTokenizerFast (line 395) | class DPRReaderTokenizerFast(CustomDPRReaderTokenizerMixin, BertTokenize...

FILE: transformers/models/dpt/configuration_dpt.py
  class DPTConfig (line 32) | class DPTConfig(PretrainedConfig):
    method __init__ (line 125) | def __init__(
    method to_dict (line 220) | def to_dict(self):

FILE: transformers/models/dpt/convert_dpt_hybrid_to_pytorch.py
  function get_dpt_config (line 35) | def get_dpt_config(checkpoint_url):
  function remove_ignore_keys_ (line 74) | def remove_ignore_keys_(state_dict):
  function rename_key (line 80) | def rename_key(name):
  function read_in_q_k_v (line 193) | def read_in_q_k_v(state_dict, config):
  function prepare_img (line 214) | def prepare_img():
  function convert_dpt_checkpoint (line 221) | def convert_dpt_checkpoint(checkpoint_url, pytorch_dump_folder_path, pus...

FILE: transformers/models/dpt/convert_dpt_to_pytorch.py
  function get_dpt_config (line 35) | def get_dpt_config(checkpoint_url):
  function remove_ignore_keys_ (line 62) | def remove_ignore_keys_(state_dict):
  function rename_key (line 68) | def rename_key(name):
  function read_in_q_k_v (line 161) | def read_in_q_k_v(state_dict, config):
  function prepare_img (line 182) | def prepare_img():
  function convert_dpt_checkpoint (line 189) | def convert_dpt_checkpoint(checkpoint_url, pytorch_dump_folder_path, pus...

FILE: transformers/models/dpt/feature_extraction_dpt.py
  class DPTFeatureExtractor (line 26) | class DPTFeatureExtractor(DPTImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/dpt/image_processing_dpt.py
  function get_resize_output_image_size (line 50) | def get_resize_output_image_size(
  class DPTImageProcessor (line 88) | class DPTImageProcessor(BaseImageProcessor):
    method __init__ (line 123) | def __init__(
    method resize (line 151) | def resize(
    method rescale (line 194) | def rescale(
    method normalize (line 214) | def normalize(
    method preprocess (line 237) | def preprocess(
    method post_process_semantic_segmentation (line 342) | def post_process_semantic_segmentation(self, outputs, target_sizes: Li...

FILE: transformers/models/dpt/modeling_dpt.py
  class BaseModelOutputWithIntermediateActivations (line 66) | class BaseModelOutputWithIntermediateActivations(ModelOutput):
  class BaseModelOutputWithPoolingAndIntermediateActivations (line 83) | class BaseModelOutputWithPoolingAndIntermediateActivations(ModelOutput):
  class DPTViTHybridEmbeddings (line 118) | class DPTViTHybridEmbeddings(nn.Module):
    method __init__ (line 125) | def __init__(self, config, feature_size=None):
    method _resize_pos_embed (line 161) | def _resize_pos_embed(self, posemb, grid_size_height, grid_size_width,...
    method forward (line 175) | def forward(
  class DPTViTEmbeddings (line 219) | class DPTViTEmbeddings(nn.Module):
    method __init__ (line 225) | def __init__(self, config):
    method _resize_pos_embed (line 235) | def _resize_pos_embed(self, posemb, grid_size_height, grid_size_width,...
    method forward (line 249) | def forward(self, pixel_values, return_dict=False):
  class DPTViTPatchEmbeddings (line 277) | class DPTViTPatchEmbeddings(nn.Module):
    method __init__ (line 283) | def __init__(self, config):
    method forward (line 298) | def forward(self, pixel_values):
  class DPTViTSelfAttention (line 309) | class DPTViTSelfAttention(nn.Module):
    method __init__ (line 310) | def __init__(self, config: DPTConfig) -> None:
    method transpose_for_scores (line 328) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 333) | def forward(
  class DPTViTSelfOutput (line 370) | class DPTViTSelfOutput(nn.Module):
    method __init__ (line 376) | def __init__(self, config: DPTConfig) -> None:
    method forward (line 381) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class DPTViTAttention (line 388) | class DPTViTAttention(nn.Module):
    method __init__ (line 389) | def __init__(self, config: DPTConfig) -> None:
    method prune_heads (line 396) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 415) | def forward(
  class DPTViTIntermediate (line 430) | class DPTViTIntermediate(nn.Module):
    method __init__ (line 431) | def __init__(self, config: DPTConfig) -> None:
    method forward (line 439) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class DPTViTOutput (line 447) | class DPTViTOutput(nn.Module):
    method __init__ (line 448) | def __init__(self, config: DPTConfig) -> None:
    method forward (line 453) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class DPTViTLayer (line 463) | class DPTViTLayer(nn.Module):
    method __init__ (line 466) | def __init__(self, config: DPTConfig) -> None:
    method forward (line 476) | def forward(
  class DPTViTEncoder (line 506) | class DPTViTEncoder(nn.Module):
    method __init__ (line 507) | def __init__(self, config: DPTConfig) -> None:
    method forward (line 513) | def forward(
  class DPTReassembleStage (line 563) | class DPTReassembleStage(nn.Module):
    method __init__ (line 579) | def __init__(self, config):
    method _init_reassemble_dpt_hybrid (line 591) | def _init_reassemble_dpt_hybrid(self, config):
    method _init_reassemble_dpt (line 616) | def _init_reassemble_dpt(self, config):
    method forward (line 627) | def forward(self, hidden_states: List[torch.Tensor]) -> List[torch.Ten...
  class DPTReassembleLayer (line 662) | class DPTReassembleLayer(nn.Module):
    method __init__ (line 663) | def __init__(self, config, channels, factor):
    method forward (line 677) | def forward(self, hidden_state):
  class DPTFeatureFusionStage (line 683) | class DPTFeatureFusionStage(nn.Module):
    method __init__ (line 684) | def __init__(self, config):
    method forward (line 690) | def forward(self, hidden_states):
  class DPTPreActResidualLayer (line 706) | class DPTPreActResidualLayer(nn.Module):
    method __init__ (line 715) | def __init__(self, config):
    method forward (line 743) | def forward(self, hidden_state: torch.Tensor) -> torch.Tensor:
  class DPTFeatureFusionLayer (line 761) | class DPTFeatureFusionLayer(nn.Module):
    method __init__ (line 771) | def __init__(self, config, align_corners=True):
    method forward (line 781) | def forward(self, hidden_state, residual=None):
  class DPTPreTrainedModel (line 798) | class DPTPreTrainedModel(PreTrainedModel):
    method _init_weights (line 809) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 821) | def _set_gradient_checkpointing(self, module, value=False):
  class DPTModel (line 864) | class DPTModel(DPTPreTrainedModel):
    method __init__ (line 865) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 882) | def get_input_embeddings(self):
    method _prune_heads (line 888) | def _prune_heads(self, heads_to_prune):
    method forward (line 904) | def forward(
  class DPTViTPooler (line 955) | class DPTViTPooler(nn.Module):
    method __init__ (line 956) | def __init__(self, config: DPTConfig):
    method forward (line 961) | def forward(self, hidden_states):
  class DPTNeck (line 970) | class DPTNeck(nn.Module):
    method __init__ (line 982) | def __init__(self, config):
    method forward (line 995) | def forward(self, hidden_states: List[torch.Tensor]) -> List[torch.Ten...
  class DPTDepthEstimationHead (line 1013) | class DPTDepthEstimationHead(nn.Module):
    method __init__ (line 1020) | def __init__(self, config):
    method forward (line 1035) | def forward(self, hidden_states: List[torch.Tensor]) -> torch.Tensor:
  class DPTForDepthEstimation (line 1052) | class DPTForDepthEstimation(DPTPreTrainedModel):
    method __init__ (line 1053) | def __init__(self, config):
    method forward (line 1069) | def forward(
  class DPTSemanticSegmentationHead (line 1170) | class DPTSemanticSegmentationHead(nn.Module):
    method __init__ (line 1171) | def __init__(self, config):
    method forward (line 1186) | def forward(self, hidden_states: List[torch.Tensor]) -> torch.Tensor:
  class DPTAuxiliaryHead (line 1195) | class DPTAuxiliaryHead(nn.Module):
    method __init__ (line 1196) | def __init__(self, config):
    method forward (line 1208) | def forward(self, hidden_states):
  class DPTForSemanticSegmentation (line 1220) | class DPTForSemanticSegmentation(DPTPreTrainedModel):
    method __init__ (line 1221) | def __init__(self, config):
    method forward (line 1238) | def forward(

FILE: transformers/models/efficientformer/configuration_efficientformer.py
  class EfficientFormerConfig (line 32) | class EfficientFormerConfig(PretrainedConfig):
    method __init__ (line 114) | def __init__(

FILE: transformers/models/efficientformer/convert_efficientformer_original_pytorch_checkpoint_to_pytorch.py
  function rename_key (line 38) | def rename_key(old_name, num_meta4D_last_stage):
  function convert_torch_checkpoint (line 107) | def convert_torch_checkpoint(checkpoint, num_meta4D_last_stage):
  function prepare_img (line 116) | def prepare_img():
  function convert_efficientformer_checkpoint (line 123) | def convert_efficientformer_checkpoint(

FILE: transformers/models/efficientformer/image_processing_efficientformer.py
  class EfficientFormerImageProcessor (line 46) | class EfficientFormerImageProcessor(BaseImageProcessor):
    method __init__ (line 85) | def __init__(
    method resize (line 116) | def resize(
    method center_crop (line 154) | def center_crop(
    method rescale (line 178) | def rescale(
    method normalize (line 200) | def normalize(
    method preprocess (line 229) | def preprocess(

FILE: transformers/models/efficientformer/modeling_efficientformer.py
  class EfficientFormerPatchEmbeddings (line 59) | class EfficientFormerPatchEmbeddings(nn.Module):
    method __init__ (line 65) | def __init__(self, config: EfficientFormerConfig, num_channels: int, e...
    method forward (line 78) | def forward(self, pixel_values: torch.Tensor) -> torch.Tensor:
  class EfficientFormerSelfAttention (line 91) | class EfficientFormerSelfAttention(nn.Module):
    method __init__ (line 92) | def __init__(self, dim: int, key_dim: int, num_heads: int, attention_r...
    method train (line 119) | def train(self, mode=True):
    method forward (line 126) | def forward(self, hidden_states: torch.Tensor, output_attentions: bool...
  class EfficientFormerConvStem (line 155) | class EfficientFormerConvStem(nn.Module):
    method __init__ (line 156) | def __init__(self, config: EfficientFormerConfig, out_channels: int):
    method forward (line 167) | def forward(self, pixel_values: torch.Tensor) -> torch.Tensor:
  class EfficientFormerPooling (line 176) | class EfficientFormerPooling(nn.Module):
    method __init__ (line 177) | def __init__(self, pool_size: int):
    method forward (line 181) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class EfficientFormerDenseMlp (line 186) | class EfficientFormerDenseMlp(nn.Module):
    method __init__ (line 187) | def __init__(
    method forward (line 203) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class EfficientFormerConvMlp (line 213) | class EfficientFormerConvMlp(nn.Module):
    method __init__ (line 214) | def __init__(
    method forward (line 234) | def forward(self, hidden_state: torch.Tensor) -> torch.Tensor:
  function drop_path (line 249) | def drop_path(input, drop_prob: float = 0.0, training: bool = False):
  class EfficientFormerDropPath (line 270) | class EfficientFormerDropPath(nn.Module):
    method __init__ (line 273) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 277) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 280) | def extra_repr(self) -> str:
  class EfficientFormerFlat (line 284) | class EfficientFormerFlat(nn.Module):
    method __init__ (line 285) | def __init__(self):
    method forward (line 288) | def forward(self, hidden_states: torch.Tensor) -> Tuple[torch.Tensor]:
  class EfficientFormerMeta3D (line 293) | class EfficientFormerMeta3D(nn.Module):
    method __init__ (line 294) | def __init__(self, config: EfficientFormerConfig, dim: int, drop_path:...
    method forward (line 317) | def forward(self, hidden_states: torch.Tensor, output_attentions: bool...
  class EfficientFormerMeta3DLayers (line 338) | class EfficientFormerMeta3DLayers(nn.Module):
    method __init__ (line 339) | def __init__(self, config: EfficientFormerConfig):
    method forward (line 349) | def forward(self, hidden_states: torch.Tensor, output_attentions: bool...
  class EfficientFormerMeta4D (line 368) | class EfficientFormerMeta4D(nn.Module):
    method __init__ (line 369) | def __init__(self, config: EfficientFormerConfig, dim: int, drop_path:...
    method forward (line 384) | def forward(self, hidden_states: torch.Tensor) -> Tuple[torch.Tensor]:
  class EfficientFormerMeta4DLayers (line 400) | class EfficientFormerMeta4DLayers(nn.Module):
    method __init__ (line 401) | def __init__(self, config: EfficientFormerConfig, stage_idx: int):
    method forward (line 417) | def forward(self, hidden_states: torch.Tensor) -> Tuple[torch.Tensor]:
  class EfficientFormerIntermediateStage (line 423) | class EfficientFormerIntermediateStage(nn.Module):
    method __init__ (line 424) | def __init__(self, config: EfficientFormerConfig, index: int):
    method forward (line 428) | def forward(self, hidden_states: torch.Tensor) -> Tuple[torch.Tensor]:
  class EfficientFormerLastStage (line 433) | class EfficientFormerLastStage(nn.Module):
    method __init__ (line 434) | def __init__(self, config: EfficientFormerConfig):
    method forward (line 440) | def forward(self, hidden_states: torch.Tensor, output_attentions: bool...
  class EfficientFormerEncoder (line 448) | class EfficientFormerEncoder(nn.Module):
    method __init__ (line 449) | def __init__(self, config: EfficientFormerConfig):
    method forward (line 469) | def forward(
  class EfficientFormerPreTrainedModel (line 505) | class EfficientFormerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 516) | def _init_weights(self, module: nn.Module):
  class EfficientFormerModel (line 557) | class EfficientFormerModel(EfficientFormerPreTrainedModel):
    method __init__ (line 558) | def __init__(self, config: EfficientFormerConfig):
    method forward (line 577) | def forward(
  class EfficientFormerForImageClassification (line 619) | class EfficientFormerForImageClassification(EfficientFormerPreTrainedMod...
    method __init__ (line 620) | def __init__(self, config: EfficientFormerConfig):
    method forward (line 641) | def forward(
  class EfficientFormerForImageClassificationWithTeacherOutput (line 704) | class EfficientFormerForImageClassificationWithTeacherOutput(ModelOutput):
  class EfficientFormerForImageClassificationWithTeacher (line 749) | class EfficientFormerForImageClassificationWithTeacher(EfficientFormerPr...
    method __init__ (line 750) | def __init__(self, config: EfficientFormerConfig):
    method forward (line 773) | def forward(

FILE: transformers/models/efficientformer/modeling_tf_efficientformer.py
  class TFEfficientFormerPatchEmbeddings (line 67) | class TFEfficientFormerPatchEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 73) | def __init__(
    method call (line 94) | def call(self, pixel_values: tf.Tensor, training: bool = False) -> tf....
  class TFEfficientFormerSelfAttention (line 104) | class TFEfficientFormerSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 105) | def __init__(
    method build (line 134) | def build(self, input_shape: tf.TensorShape) -> None:
    method call (line 165) | def call(
  class TFEfficientFormerConvStem (line 202) | class TFEfficientFormerConvStem(tf.keras.layers.Layer):
    method __init__ (line 203) | def __init__(self, config: EfficientFormerConfig, out_channels: int, *...
    method call (line 229) | def call(self, pixel_values: tf.Tensor, training: bool = False) -> tf....
  class TFEfficientFormerPooling (line 237) | class TFEfficientFormerPooling(tf.keras.layers.Layer):
    method __init__ (line 238) | def __init__(self, pool_size: int, **kwargs):
    method call (line 242) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFEfficientFormerDenseMlp (line 248) | class TFEfficientFormerDenseMlp(tf.keras.layers.Layer):
    method __init__ (line 249) | def __init__(
    method call (line 271) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFEfficientFormerConvMlp (line 281) | class TFEfficientFormerConvMlp(tf.keras.layers.Layer):
    method __init__ (line 282) | def __init__(
    method call (line 322) | def call(self, hidden_state: tf.Tensor, training: bool = False) -> tf....
  class TFEfficientFormerDropPath (line 334) | class TFEfficientFormerDropPath(tf.keras.layers.Layer):
    method __init__ (line 340) | def __init__(self, drop_path, **kwargs):
    method call (line 344) | def call(self, x, training=None):
  class TFEfficientFormerFlat (line 354) | class TFEfficientFormerFlat(tf.keras.layers.Layer):
    method __init__ (line 355) | def __init__(self, **kwargs):
    method call (line 358) | def call(self, hidden_states: tf.Tensor) -> Tuple[tf.Tensor]:
  class TFEfficientFormerMeta3D (line 364) | class TFEfficientFormerMeta3D(tf.keras.layers.Layer):
    method __init__ (line 365) | def __init__(self, config: EfficientFormerConfig, dim: int, drop_path:...
    method build (line 393) | def build(self, input_shape: tf.TensorShape):
    method call (line 412) | def call(
  class TFEfficientFormerMeta3DLayers (line 446) | class TFEfficientFormerMeta3DLayers(tf.keras.layers.Layer):
    method __init__ (line 447) | def __init__(self, config: EfficientFormerConfig, **kwargs):
    method call (line 458) | def call(
  class TFEfficientFormerMeta4D (line 480) | class TFEfficientFormerMeta4D(tf.keras.layers.Layer):
    method __init__ (line 481) | def __init__(self, config: EfficientFormerConfig, dim: int, drop_path:...
    method build (line 498) | def build(self, input_shape: tf.TensorShape):
    method call (line 517) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> Tu...
  class TFEfficientFormerMeta4DLayers (line 541) | class TFEfficientFormerMeta4DLayers(tf.keras.layers.Layer):
    method __init__ (line 542) | def __init__(self, config: EfficientFormerConfig, stage_idx: int, **kw...
    method call (line 558) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> Tu...
  class TFEfficientFormerIntermediateStage (line 564) | class TFEfficientFormerIntermediateStage(tf.keras.layers.Layer):
    method __init__ (line 565) | def __init__(self, config: EfficientFormerConfig, index: int, **kwargs):
    method call (line 569) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> Tu...
  class TFEfficientFormerLastStage (line 574) | class TFEfficientFormerLastStage(tf.keras.layers.Layer):
    method __init__ (line 575) | def __init__(self, config: EfficientFormerConfig, **kwargs):
    method call (line 581) | def call(
  class TFEfficientFormerEncoder (line 593) | class TFEfficientFormerEncoder(tf.keras.layers.Layer):
    method __init__ (line 594) | def __init__(self, config: EfficientFormerConfig, **kwargs):
    method call (line 624) | def call(
  class TFEfficientFormerMainLayer (line 663) | class TFEfficientFormerMainLayer(tf.keras.layers.Layer):
    method __init__ (line 666) | def __init__(self, config: EfficientFormerConfig, **kwargs) -> None:
    method call (line 675) | def call(
  class TFEfficientFormerPreTrainedModel (line 732) | class TFEfficientFormerPreTrainedModel(TFPreTrainedModel):
  class TFEfficientFormerModel (line 775) | class TFEfficientFormerModel(TFEfficientFormerPreTrainedModel):
    method __init__ (line 776) | def __init__(self, config: EfficientFormerConfig, **kwargs) -> None:
    method call (line 790) | def call(
  class TFEfficientFormerForImageClassification (line 815) | class TFEfficientFormerForImageClassification(TFEfficientFormerPreTraine...
    method __init__ (line 816) | def __init__(self, config: EfficientFormerConfig):
    method call (line 837) | def call(
  class TFEfficientFormerForImageClassificationWithTeacherOutput (line 878) | class TFEfficientFormerForImageClassificationWithTeacherOutput(ModelOutp...
  class TFEfficientFormerForImageClassificationWithTeacher (line 920) | class TFEfficientFormerForImageClassificationWithTeacher(TFEfficientForm...
    method __init__ (line 921) | def __init__(self, config: EfficientFormerConfig) -> None:
    method call (line 947) | def call(

FILE: transformers/models/efficientnet/configuration_efficientnet.py
  class EfficientNetConfig (line 34) | class EfficientNetConfig(PretrainedConfig):
    method __init__ (line 105) | def __init__(
  class EfficientNetOnnxConfig (line 156) | class EfficientNetOnnxConfig(OnnxConfig):
    method inputs (line 160) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 168) | def atol_for_validation(self) -> float:

FILE: transformers/models/efficientnet/convert_efficientnet_to_pytorch.py
  function get_efficientnet_config (line 122) | def get_efficientnet_config(model_name):
  function prepare_img (line 143) | def prepare_img():
  function convert_image_processor (line 149) | def convert_image_processor(model_name):
  function rename_keys (line 161) | def rename_keys(original_param_names):
  function replace_params (line 229) | def replace_params(hf_params, tf_params, key_mapping):
  function convert_efficientnet_checkpoint (line 250) | def convert_efficientnet_checkpoint(model_name, pytorch_dump_folder_path...

FILE: transformers/models/efficientnet/image_processing_efficientnet.py
  class EfficientNetImageProcessor (line 43) | class EfficientNetImageProcessor(BaseImageProcessor):
    method __init__ (line 84) | def __init__(
    method resize (line 119) | def resize(
    method center_crop (line 147) | def center_crop(
    method rescale (line 171) | def rescale(
    method normalize (line 201) | def normalize(
    method preprocess (line 224) | def preprocess(

FILE: transformers/models/efficientnet/modeling_efficientnet.py
  function round_filters (line 86) | def round_filters(config: EfficientNetConfig, num_channels: int):
  function correct_pad (line 101) | def correct_pad(kernel_size: Union[int, Tuple], adjust: bool = True):
  class EfficientNetEmbeddings (line 121) | class EfficientNetEmbeddings(nn.Module):
    method __init__ (line 126) | def __init__(self, config: EfficientNetConfig):
    method forward (line 137) | def forward(self, pixel_values: torch.Tensor) -> torch.Tensor:
  class EfficientNetDepthwiseConv2d (line 146) | class EfficientNetDepthwiseConv2d(nn.Conv2d):
    method __init__ (line 147) | def __init__(
  class EfficientNetExpansionLayer (line 172) | class EfficientNetExpansionLayer(nn.Module):
    method __init__ (line 177) | def __init__(self, config: EfficientNetConfig, in_dim: int, out_dim: i...
    method forward (line 189) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class EfficientNetDepthwiseLayer (line 198) | class EfficientNetDepthwiseLayer(nn.Module):
    method __init__ (line 203) | def __init__(
    method forward (line 225) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class EfficientNetSqueezeExciteLayer (line 237) | class EfficientNetSqueezeExciteLayer(nn.Module):
    method __init__ (line 242) | def __init__(self, config: EfficientNetConfig, in_dim: int, expand_dim...
    method forward (line 263) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class EfficientNetFinalBlockLayer (line 276) | class EfficientNetFinalBlockLayer(nn.Module):
    method __init__ (line 281) | def __init__(
    method forward (line 298) | def forward(self, embeddings: torch.FloatTensor, hidden_states: torch....
  class EfficientNetBlock (line 309) | class EfficientNetBlock(nn.Module):
    method __init__ (line 336) | def __init__(
    method forward (line 377) | def forward(self, hidden_states: torch.FloatTensor) -> torch.Tensor:
  class EfficientNetEncoder (line 390) | class EfficientNetEncoder(nn.Module):
    method __init__ (line 399) | def __init__(self, config: EfficientNetConfig):
    method forward (line 454) | def forward(
  class EfficientNetPreTrainedModel (line 480) | class EfficientNetPreTrainedModel(PreTrainedModel):
    method _init_weights (line 491) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 503) | def _set_gradient_checkpointing(self, module, value=False):
  class EfficientNetModel (line 512) | class EfficientNetModel(EfficientNetPreTrainedModel):
    method __init__ (line 513) | def __init__(self, config: EfficientNetConfig):
    method forward (line 538) | def forward(
  class EfficientNetForImageClassification (line 582) | class EfficientNetForImageClassification(EfficientNetPreTrainedModel):
    method __init__ (line 583) | def __init__(self, config):
    method forward (line 603) | def forward(

FILE: transformers/models/electra/configuration_electra.py
  class ElectraConfig (line 44) | class ElectraConfig(PretrainedConfig):
    method __init__ (line 135) | def __init__(
  class ElectraOnnxConfig (line 185) | class ElectraOnnxConfig(OnnxConfig):
    method inputs (line 187) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/electra/convert_electra_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 29) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, config_file, py...

FILE: transformers/models/electra/modeling_electra.py
  function load_tf_weights_in_electra (line 67) | def load_tf_weights_in_electra(model, config, tf_checkpoint_path, discri...
  class ElectraEmbeddings (line 149) | class ElectraEmbeddings(nn.Module):
    method __init__ (line 152) | def __init__(self, config):
    method forward (line 171) | def forward(
  class ElectraSelfAttention (line 214) | class ElectraSelfAttention(nn.Module):
    method __init__ (line 215) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 241) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 246) | def forward(
  class ElectraSelfOutput (line 349) | class ElectraSelfOutput(nn.Module):
    method __init__ (line 350) | def __init__(self, config):
    method forward (line 356) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ElectraAttention (line 364) | class ElectraAttention(nn.Module):
    method __init__ (line 365) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 371) | def prune_heads(self, heads):
    method forward (line 389) | def forward(
  class ElectraIntermediate (line 414) | class ElectraIntermediate(nn.Module):
    method __init__ (line 415) | def __init__(self, config):
    method forward (line 423) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ElectraOutput (line 430) | class ElectraOutput(nn.Module):
    method __init__ (line 431) | def __init__(self, config):
    method forward (line 437) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ElectraLayer (line 445) | class ElectraLayer(nn.Module):
    method __init__ (line 446) | def __init__(self, config):
    method forward (line 460) | def forward(
    method feed_forward_chunk (line 525) | def feed_forward_chunk(self, attention_output):
  class ElectraEncoder (line 532) | class ElectraEncoder(nn.Module):
    method __init__ (line 533) | def __init__(self, config):
    method forward (line 539) | def forward(
  class ElectraDiscriminatorPredictions (line 630) | class ElectraDiscriminatorPredictions(nn.Module):
    method __init__ (line 633) | def __init__(self, config):
    method forward (line 640) | def forward(self, discriminator_hidden_states):
  class ElectraGeneratorPredictions (line 648) | class ElectraGeneratorPredictions(nn.Module):
    method __init__ (line 651) | def __init__(self, config):
    method forward (line 657) | def forward(self, generator_hidden_states):
  class ElectraPreTrainedModel (line 665) | class ElectraPreTrainedModel(PreTrainedModel):
    method _init_weights (line 679) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 695) | def _set_gradient_checkpointing(self, module, value=False):
  class ElectraForPreTrainingOutput (line 701) | class ElectraForPreTrainingOutput(ModelOutput):
  class ElectraModel (line 813) | class ElectraModel(ElectraPreTrainedModel):
    method __init__ (line 814) | def __init__(self, config):
    method get_input_embeddings (line 826) | def get_input_embeddings(self):
    method set_input_embeddings (line 829) | def set_input_embeddings(self, value):
    method _prune_heads (line 832) | def _prune_heads(self, heads_to_prune):
    method forward (line 846) | def forward(
  class ElectraClassificationHead (line 935) | class ElectraClassificationHead(nn.Module):
    method __init__ (line 938) | def __init__(self, config):
    method forward (line 947) | def forward(self, features, **kwargs):
  class ElectraForSequenceClassification (line 964) | class ElectraForSequenceClassification(ElectraPreTrainedModel):
    method __init__ (line 965) | def __init__(self, config):
    method forward (line 983) | def forward(
  class ElectraForPreTraining (line 1062) | class ElectraForPreTraining(ElectraPreTrainedModel):
    method __init__ (line 1063) | def __init__(self, config):
    method forward (line 1073) | def forward(
  class ElectraForMaskedLM (line 1168) | class ElectraForMaskedLM(ElectraPreTrainedModel):
    method __init__ (line 1171) | def __init__(self, config):
    method get_output_embeddings (line 1181) | def get_output_embeddings(self):
    method set_output_embeddings (line 1184) | def set_output_embeddings(self, word_embeddings):
    method forward (line 1196) | def forward(
  class ElectraForTokenClassification (line 1259) | class ElectraForTokenClassification(ElectraPreTrainedModel):
    method __init__ (line 1260) | def __init__(self, config):
    method forward (line 1281) | def forward(
  class ElectraForQuestionAnswering (line 1340) | class ElectraForQuestionAnswering(ElectraPreTrainedModel):
    method __init__ (line 1344) | def __init__(self, config):
    method forward (line 1364) | def forward(
  class ElectraForMultipleChoice (line 1448) | class ElectraForMultipleChoice(ElectraPreTrainedModel):
    method __init__ (line 1449) | def __init__(self, config):
    method forward (line 1465) | def forward(
  class ElectraForCausalLM (line 1535) | class ElectraForCausalLM(ElectraPreTrainedModel):
    method __init__ (line 1538) | def __init__(self, config):
    method get_output_embeddings (line 1550) | def get_output_embeddings(self):
    method set_output_embeddings (line 1553) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1558) | def forward(
    method prepare_inputs_for_generation (line 1663) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1676) | def _reorder_cache(self, past_key_values, beam_idx):

FILE: transformers/models/electra/modeling_flax_electra.py
  class FlaxElectraForPreTrainingOutput (line 60) | class FlaxElectraForPreTrainingOutput(ModelOutput):
  class FlaxElectraEmbeddings (line 146) | class FlaxElectraEmbeddings(nn.Module):
    method setup (line 152) | def setup(self):
    method __call__ (line 172) | def __call__(self, input_ids, token_type_ids, position_ids, attention_...
  class FlaxElectraSelfAttention (line 188) | class FlaxElectraSelfAttention(nn.Module):
    method setup (line 193) | def setup(self):
    method _split_heads (line 222) | def _split_heads(self, hidden_states):
    method _merge_heads (line 225) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 230) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 261) | def __call__(
  class FlaxElectraSelfOutput (line 360) | class FlaxElectraSelfOutput(nn.Module):
    method setup (line 364) | def setup(self):
    method __call__ (line 373) | def __call__(self, hidden_states, input_tensor, deterministic: bool = ...
  class FlaxElectraAttention (line 381) | class FlaxElectraAttention(nn.Module):
    method setup (line 386) | def setup(self):
    method __call__ (line 390) | def __call__(
  class FlaxElectraIntermediate (line 424) | class FlaxElectraIntermediate(nn.Module):
    method setup (line 428) | def setup(self):
    method __call__ (line 436) | def __call__(self, hidden_states):
  class FlaxElectraOutput (line 443) | class FlaxElectraOutput(nn.Module):
    method setup (line 447) | def setup(self):
    method __call__ (line 456) | def __call__(self, hidden_states, attention_output, deterministic: boo...
  class FlaxElectraLayer (line 464) | class FlaxElectraLayer(nn.Module):
    method setup (line 468) | def setup(self):
    method __call__ (line 475) | def __call__(
  class FlaxElectraLayerCollection (line 522) | class FlaxElectraLayerCollection(nn.Module):
    method setup (line 527) | def setup(self):
    method __call__ (line 540) | def __call__(
  class FlaxElectraEncoder (line 605) | class FlaxElectraEncoder(nn.Module):
    method setup (line 610) | def setup(self):
    method __call__ (line 617) | def __call__(
  class FlaxElectraGeneratorPredictions (line 644) | class FlaxElectraGeneratorPredictions(nn.Module):
    method setup (line 648) | def setup(self):
    method __call__ (line 652) | def __call__(self, hidden_states):
  class FlaxElectraDiscriminatorPredictions (line 659) | class FlaxElectraDiscriminatorPredictions(nn.Module):
    method setup (line 665) | def setup(self):
    method __call__ (line 669) | def __call__(self, hidden_states):
  class FlaxElectraPreTrainedModel (line 676) | class FlaxElectraPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 686) | def __init__(
    method enable_gradient_checkpointing (line 700) | def enable_gradient_checkpointing(self):
    method init_weights (line 708) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 751) | def init_cache(self, batch_size, max_length):
    method __call__ (line 771) | def __call__(
  class FlaxElectraModule (line 868) | class FlaxElectraModule(nn.Module):
    method setup (line 873) | def setup(self):
    method __call__ (line 881) | def __call__(
  class FlaxElectraModel (line 920) | class FlaxElectraModel(FlaxElectraPreTrainedModel):
  class FlaxElectraTiedDense (line 927) | class FlaxElectraTiedDense(nn.Module):
    method setup (line 933) | def setup(self):
    method __call__ (line 936) | def __call__(self, x, kernel):
  class FlaxElectraForMaskedLMModule (line 949) | class FlaxElectraForMaskedLMModule(nn.Module):
    method setup (line 954) | def setup(self):
    method __call__ (line 964) | def __call__(
  class FlaxElectraForMaskedLM (line 1007) | class FlaxElectraForMaskedLM(FlaxElectraPreTrainedModel):
  class FlaxElectraForPreTrainingModule (line 1014) | class FlaxElectraForPreTrainingModule(nn.Module):
    method setup (line 1019) | def setup(self):
    method __call__ (line 1025) | def __call__(
  class FlaxElectraForPreTraining (line 1071) | class FlaxElectraForPreTraining(FlaxElectraPreTrainedModel):
  class FlaxElectraForTokenClassificationModule (line 1102) | class FlaxElectraForTokenClassificationModule(nn.Module):
    method setup (line 1107) | def setup(self):
    method __call__ (line 1119) | def __call__(
  class FlaxElectraForTokenClassification (line 1166) | class FlaxElectraForTokenClassification(FlaxElectraPreTrainedModel):
  function identity (line 1178) | def identity(x, **kwargs):
  class FlaxElectraSequenceSummary (line 1182) | class FlaxElectraSequenceSummary(nn.Module):
    method setup (line 1202) | def setup(self):
    method __call__ (line 1226) | def __call__(self, hidden_states, cls_index=None, deterministic: bool ...
  class FlaxElectraForMultipleChoiceModule (line 1248) | class FlaxElectraForMultipleChoiceModule(nn.Module):
    method setup (line 1253) | def setup(self):
    method __call__ (line 1260) | def __call__(
  class FlaxElectraForMultipleChoice (line 1313) | class FlaxElectraForMultipleChoice(FlaxElectraPreTrainedModel):
  class FlaxElectraForQuestionAnsweringModule (line 1329) | class FlaxElectraForQuestionAnsweringModule(nn.Module):
    method setup (line 1334) | def setup(self):
    method __call__ (line 1340) | def __call__(
  class FlaxElectraForQuestionAnswering (line 1388) | class FlaxElectraForQuestionAnswering(FlaxElectraPreTrainedModel):
  class FlaxElectraClassificationHead (line 1400) | class FlaxElectraClassificationHead(nn.Module):
    method setup (line 1406) | def setup(self):
    method __call__ (line 1416) | def __call__(self, hidden_states, deterministic: bool = True):
  class FlaxElectraForSequenceClassificationModule (line 1426) | class FlaxElectraForSequenceClassificationModule(nn.Module):
    method setup (line 1431) | def setup(self):
    method __call__ (line 1437) | def __call__(
  class FlaxElectraForSequenceClassification (line 1481) | class FlaxElectraForSequenceClassification(FlaxElectraPreTrainedModel):
  class FlaxElectraForCausalLMModule (line 1493) | class FlaxElectraForCausalLMModule(nn.Module):
    method setup (line 1498) | def setup(self):
    method __call__ (line 1508) | def __call__(
  class FlaxElectraForCausalLM (line 1565) | class FlaxElectraForCausalLM(FlaxElectraPreTrainedModel):
    method prepare_inputs_for_generation (line 1568) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 1589) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/electra/modeling_tf_electra.py
  class TFElectraSelfAttention (line 79) | class TFElectraSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 80) | def __init__(self, config: ElectraConfig, **kwargs):
    method transpose_for_scores (line 107) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 114) | def call(
  class TFElectraSelfOutput (line 197) | class TFElectraSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 198) | def __init__(self, config: ElectraConfig, **kwargs):
    method call (line 207) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFElectraAttention (line 216) | class TFElectraAttention(tf.keras.layers.Layer):
    method __init__ (line 217) | def __init__(self, config: ElectraConfig, **kwargs):
    method prune_heads (line 223) | def prune_heads(self, heads):
    method call (line 226) | def call(
  class TFElectraIntermediate (line 257) | class TFElectraIntermediate(tf.keras.layers.Layer):
    method __init__ (line 258) | def __init__(self, config: ElectraConfig, **kwargs):
    method call (line 270) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFElectraOutput (line 278) | class TFElectraOutput(tf.keras.layers.Layer):
    method __init__ (line 279) | def __init__(self, config: ElectraConfig, **kwargs):
    method call (line 288) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFElectraLayer (line 297) | class TFElectraLayer(tf.keras.layers.Layer):
    method __init__ (line 298) | def __init__(self, config: ElectraConfig, **kwargs):
    method call (line 311) | def call(
  class TFElectraEncoder (line 384) | class TFElectraEncoder(tf.keras.layers.Layer):
    method __init__ (line 385) | def __init__(self, config: ElectraConfig, **kwargs):
    method call (line 390) | def call(
  class TFElectraPooler (line 454) | class TFElectraPooler(tf.keras.layers.Layer):
    method __init__ (line 455) | def __init__(self, config: ElectraConfig, **kwargs):
    method call (line 465) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFElectraEmbeddings (line 475) | class TFElectraEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 478) | def __init__(self, config: ElectraConfig, **kwargs):
    method build (line 488) | def build(self, input_shape: tf.TensorShape):
    method call (line 513) | def call(
  class TFElectraDiscriminatorPredictions (line 554) | class TFElectraDiscriminatorPredictions(tf.keras.layers.Layer):
    method __init__ (line 555) | def __init__(self, config, **kwargs):
    method call (line 562) | def call(self, discriminator_hidden_states, training=False):
  class TFElectraGeneratorPredictions (line 570) | class TFElectraGeneratorPredictions(tf.keras.layers.Layer):
    method __init__ (line 571) | def __init__(self, config, **kwargs):
    method call (line 577) | def call(self, generator_hidden_states, training=False):
  class TFElectraPreTrainedModel (line 585) | class TFElectraPreTrainedModel(TFPreTrainedModel):
  class TFElectraMainLayer (line 599) | class TFElectraMainLayer(tf.keras.layers.Layer):
    method __init__ (line 602) | def __init__(self, config, **kwargs):
    method get_input_embeddings (line 615) | def get_input_embeddings(self):
    method set_input_embeddings (line 618) | def set_input_embeddings(self, value):
    method _prune_heads (line 622) | def _prune_heads(self, heads_to_prune):
    method get_extended_attention_mask (line 629) | def get_extended_attention_mask(self, attention_mask, input_shape, dty...
    method get_head_mask (line 678) | def get_head_mask(self, head_mask):
    method call (line 687) | def call(
  class TFElectraForPreTrainingOutput (line 786) | class TFElectraForPreTrainingOutput(ModelOutput):
  class TFElectraModel (line 911) | class TFElectraModel(TFElectraPreTrainedModel):
    method __init__ (line 912) | def __init__(self, config, *inputs, **kwargs):
    method call (line 924) | def call(
  class TFElectraForPreTraining (line 990) | class TFElectraForPreTraining(TFElectraPreTrainedModel):
    method __init__ (line 991) | def __init__(self, config, **kwargs):
    method call (line 1000) | def call(
  class TFElectraMaskedLMHead (line 1053) | class TFElectraMaskedLMHead(tf.keras.layers.Layer):
    method __init__ (line 1054) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 1061) | def build(self, input_shape):
    method get_output_embeddings (line 1066) | def get_output_embeddings(self):
    method set_output_embeddings (line 1069) | def set_output_embeddings(self, value):
    method get_bias (line 1073) | def get_bias(self):
    method set_bias (line 1076) | def set_bias(self, value):
    method call (line 1080) | def call(self, hidden_states):
  class TFElectraForMaskedLM (line 1099) | class TFElectraForMaskedLM(TFElectraPreTrainedModel, TFMaskedLanguageMod...
    method __init__ (line 1100) | def __init__(self, config, **kwargs):
    method get_lm_head (line 1114) | def get_lm_head(self):
    method get_prefix_bias_name (line 1117) | def get_prefix_bias_name(self):
    method call (line 1131) | def call(
  class TFElectraClassificationHead (line 1181) | class TFElectraClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 1184) | def __init__(self, config, **kwargs):
    method call (line 1200) | def call(self, inputs, **kwargs):
  class TFElectraForSequenceClassification (line 1218) | class TFElectraForSequenceClassification(TFElectraPreTrainedModel, TFSeq...
    method __init__ (line 1219) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1234) | def call(
  class TFElectraForMultipleChoice (line 1289) | class TFElectraForMultipleChoice(TFElectraPreTrainedModel, TFMultipleCho...
    method __init__ (line 1290) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1308) | def call(
  class TFElectraForTokenClassification (line 1382) | class TFElectraForTokenClassification(TFElectraPreTrainedModel, TFTokenC...
    method __init__ (line 1383) | def __init__(self, config, **kwargs):
    method call (line 1404) | def call(
  class TFElectraForQuestionAnswering (line 1459) | class TFElectraForQuestionAnswering(TFElectraPreTrainedModel, TFQuestion...
    method __init__ (line 1460) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1480) | def call(

FILE: transformers/models/electra/tokenization_electra.py
  function load_vocab (line 71) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 83) | def whitespace_tokenize(text):
  class ElectraTokenizer (line 93) | class ElectraTokenizer(PreTrainedTokenizer):
    method __init__ (line 140) | def __init__(
    method do_lower_case (line 187) | def do_lower_case(self):
    method vocab_size (line 191) | def vocab_size(self):
    method get_vocab (line 194) | def get_vocab(self):
    method _tokenize (line 197) | def _tokenize(self, text):
    method _convert_token_to_id (line 210) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 214) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 218) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 223) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 248) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 276) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 305) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 327) | class BasicTokenizer(object):
    method __init__ (line 347) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 355) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 392) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 403) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 425) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 438) | def _is_chinese_char(self, cp):
    method _clean_text (line 462) | def _clean_text(self, text):
  class WordpieceTokenizer (line 477) | class WordpieceTokenizer(object):
    method __init__ (line 480) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 485) | def tokenize(self, text):

FILE: transformers/models/electra/tokenization_electra_fast.py
  class ElectraTokenizerFast (line 88) | class ElectraTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 134) | def __init__(
    method build_inputs_with_special_tokens (line 176) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 200) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 229) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/encoder_decoder/configuration_encoder_decoder.py
  class EncoderDecoderConfig (line 26) | class EncoderDecoderConfig(PretrainedConfig):
    method __init__ (line 75) | def __init__(self, **kwargs):
    method from_encoder_decoder_configs (line 92) | def from_encoder_decoder_configs(
    method to_dict (line 108) | def to_dict(self):

FILE: transformers/models/encoder_decoder/modeling_encoder_decoder.py
  function shift_tokens_right (line 151) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  class EncoderDecoderModel (line 170) | class EncoderDecoderModel(PreTrainedModel):
    method __init__ (line 182) | def __init__(
    method tie_weights (line 259) | def tie_weights(self):
    method _set_gradient_checkpointing (line 268) | def _set_gradient_checkpointing(self, module, value=False):
    method get_encoder (line 273) | def get_encoder(self):
    method get_decoder (line 276) | def get_decoder(self):
    method get_input_embeddings (line 279) | def get_input_embeddings(self):
    method get_output_embeddings (line 282) | def get_output_embeddings(self):
    method set_output_embeddings (line 285) | def set_output_embeddings(self, new_embeddings):
    method from_pretrained (line 289) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
    method from_encoder_decoder_pretrained (line 390) | def from_encoder_decoder_pretrained(
    method forward (line 541) | def forward(
    method prepare_decoder_input_ids_from_labels (line 665) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method prepare_inputs_for_generation (line 668) | def prepare_inputs_for_generation(
    method resize_token_embeddings (line 683) | def resize_token_embeddings(self, *args, **kwargs):
    method _reorder_cache (line 690) | def _reorder_cache(self, past_key_values, beam_idx):

FILE: transformers/models/encoder_decoder/modeling_flax_encoder_decoder.py
  class FlaxEncoderDecoderModule (line 207) | class FlaxEncoderDecoderModule(nn.Module):
    method setup (line 211) | def setup(self):
    method _get_encoder_module (line 237) | def _get_encoder_module(self):
    method _get_projection_module (line 240) | def _get_projection_module(self):
    method _get_decoder_module (line 243) | def _get_decoder_module(self):
    method __call__ (line 246) | def __call__(
  class FlaxEncoderDecoderModel (line 302) | class FlaxEncoderDecoderModel(FlaxPreTrainedModel):
    method __init__ (line 313) | def __init__(
    method init_weights (line 342) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 387) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 430) | def encode(
    method decode (line 504) | def decode(
    method __call__ (line 629) | def __call__(
    method prepare_inputs_for_generation (line 721) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 754) | def update_inputs_for_generation(self, model_outputs, model_kwargs):
    method from_encoder_decoder_pretrained (line 760) | def from_encoder_decoder_pretrained(

FILE: transformers/models/encoder_decoder/modeling_tf_encoder_decoder.py
  function shift_tokens_right (line 166) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int, decoder_...
  class TFEncoderDecoderModel (line 193) | class TFEncoderDecoderModel(TFPreTrainedModel, TFCausalLanguageModelingL...
    method __init__ (line 204) | def __init__(
    method get_encoder (line 278) | def get_encoder(self):
    method get_decoder (line 281) | def get_decoder(self):
    method get_input_embeddings (line 284) | def get_input_embeddings(self):
    method get_output_embeddings (line 287) | def get_output_embeddings(self):
    method set_output_embeddings (line 290) | def set_output_embeddings(self, new_embeddings):
    method from_pretrained (line 294) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
    method from_encoder_decoder_pretrained (line 323) | def from_encoder_decoder_pretrained(
    method call (line 472) | def call(
    method prepare_inputs_for_generation (line 631) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 651) | def prepare_decoder_input_ids_from_labels(self, labels: tf.Tensor):
    method resize_token_embeddings (line 654) | def resize_token_embeddings(self, *args, **kwargs):
    method _reorder_cache (line 661) | def _reorder_cache(self, past, beam_idx):

FILE: transformers/models/ernie/configuration_ernie.py
  class ErnieConfig (line 41) | class ErnieConfig(PretrainedConfig):
    method __init__ (line 114) | def __init__(
  class ErnieOnnxConfig (line 157) | class ErnieOnnxConfig(OnnxConfig):
    method inputs (line 159) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/ernie/modeling_ernie.py
  class ErnieEmbeddings (line 74) | class ErnieEmbeddings(nn.Module):
    method __init__ (line 77) | def __init__(self, config):
    method forward (line 97) | def forward(
  class ErnieSelfAttention (line 149) | class ErnieSelfAttention(nn.Module):
    method __init__ (line 150) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 176) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 181) | def forward(
  class ErnieSelfOutput (line 284) | class ErnieSelfOutput(nn.Module):
    method __init__ (line 285) | def __init__(self, config):
    method forward (line 291) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ErnieAttention (line 299) | class ErnieAttention(nn.Module):
    method __init__ (line 300) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 306) | def prune_heads(self, heads):
    method forward (line 324) | def forward(
  class ErnieIntermediate (line 349) | class ErnieIntermediate(nn.Module):
    method __init__ (line 350) | def __init__(self, config):
    method forward (line 358) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ErnieOutput (line 365) | class ErnieOutput(nn.Module):
    method __init__ (line 366) | def __init__(self, config):
    method forward (line 372) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ErnieLayer (line 380) | class ErnieLayer(nn.Module):
    method __init__ (line 381) | def __init__(self, config):
    method forward (line 395) | def forward(
    method feed_forward_chunk (line 460) | def feed_forward_chunk(self, attention_output):
  class ErnieEncoder (line 467) | class ErnieEncoder(nn.Module):
    method __init__ (line 468) | def __init__(self, config):
    method forward (line 474) | def forward(
  class ErniePooler (line 566) | class ErniePooler(nn.Module):
    method __init__ (line 567) | def __init__(self, config):
    method forward (line 572) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ErniePredictionHeadTransform (line 582) | class ErniePredictionHeadTransform(nn.Module):
    method __init__ (line 583) | def __init__(self, config):
    method forward (line 592) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ErnieLMPredictionHead (line 600) | class ErnieLMPredictionHead(nn.Module):
    method __init__ (line 601) | def __init__(self, config):
    method forward (line 614) | def forward(self, hidden_states):
  class ErnieOnlyMLMHead (line 621) | class ErnieOnlyMLMHead(nn.Module):
    method __init__ (line 622) | def __init__(self, config):
    method forward (line 626) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class ErnieOnlyNSPHead (line 632) | class ErnieOnlyNSPHead(nn.Module):
    method __init__ (line 633) | def __init__(self, config):
    method forward (line 637) | def forward(self, pooled_output):
  class ErniePreTrainingHeads (line 643) | class ErniePreTrainingHeads(nn.Module):
    method __init__ (line 644) | def __init__(self, config):
    method forward (line 649) | def forward(self, sequence_output, pooled_output):
  class ErniePreTrainedModel (line 655) | class ErniePreTrainedModel(PreTrainedModel):
    method _init_weights (line 666) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 682) | def _set_gradient_checkpointing(self, module, value=False):
  class ErnieForPreTrainingOutput (line 689) | class ErnieForPreTrainingOutput(ModelOutput):
  class ErnieModel (line 797) | class ErnieModel(ErniePreTrainedModel):
    method __init__ (line 811) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 824) | def get_input_embeddings(self):
    method set_input_embeddings (line 828) | def set_input_embeddings(self, value):
    method _prune_heads (line 832) | def _prune_heads(self, heads_to_prune):
    method forward (line 846) | def forward(
  class ErnieForPreTraining (line 985) | class ErnieForPreTraining(ErniePreTrainedModel):
    method __init__ (line 989) | def __init__(self, config):
    method get_output_embeddings (line 999) | def get_output_embeddings(self):
    method set_output_embeddings (line 1003) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1008) | def forward(
  class ErnieForCausalLM (line 1096) | class ErnieForCausalLM(ErniePreTrainedModel):
    method __init__ (line 1101) | def __init__(self, config):
    method get_output_embeddings (line 1114) | def get_output_embeddings(self):
    method set_output_embeddings (line 1118) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1127) | def forward(
    method prepare_inputs_for_generation (line 1215) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1235) | def _reorder_cache(self, past_key_values, beam_idx):
  class ErnieForMaskedLM (line 1243) | class ErnieForMaskedLM(ErniePreTrainedModel):
    method __init__ (line 1248) | def __init__(self, config):
    method get_output_embeddings (line 1264) | def get_output_embeddings(self):
    method set_output_embeddings (line 1268) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1279) | def forward(
    method prepare_inputs_for_generation (line 1339) | def prepare_inputs_for_generation(self, input_ids, attention_mask=None...
  class ErnieForNextSentencePrediction (line 1360) | class ErnieForNextSentencePrediction(ErniePreTrainedModel):
    method __init__ (line 1362) | def __init__(self, config):
    method forward (line 1373) | def forward(
  class ErnieForSequenceClassification (line 1468) | class ErnieForSequenceClassification(ErniePreTrainedModel):
    method __init__ (line 1470) | def __init__(self, config):
    method forward (line 1486) | def forward(
  class ErnieForMultipleChoice (line 1567) | class ErnieForMultipleChoice(ErniePreTrainedModel):
    method __init__ (line 1569) | def __init__(self, config):
    method forward (line 1588) | def forward(
  class ErnieForTokenClassification (line 1664) | class ErnieForTokenClassification(ErniePreTrainedModel):
    method __init__ (line 1668) | def __init__(self, config):
    method forward (line 1683) | def forward(
  class ErnieForQuestionAnswering (line 1745) | class ErnieForQuestionAnswering(ErniePreTrainedModel):
    method __init__ (line 1749) | def __init__(self, config):
    method forward (line 1760) | def forward(

FILE: transformers/models/ernie_m/configuration_ernie_m.py
  class ErnieMConfig (line 31) | class ErnieMConfig(PretrainedConfig):
    method __init__ (line 84) | def __init__(

FILE: transformers/models/ernie_m/modeling_ernie_m.py
  class ErnieMEmbeddings (line 55) | class ErnieMEmbeddings(nn.Module):
    method __init__ (line 58) | def __init__(self, config):
    method forward (line 69) | def forward(
  class ErnieMSelfAttention (line 97) | class ErnieMSelfAttention(nn.Module):
    method __init__ (line 98) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 124) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 129) | def forward(
  class ErnieMAttention (line 231) | class ErnieMAttention(nn.Module):
    method __init__ (line 232) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 238) | def prune_heads(self, heads):
    method forward (line 256) | def forward(
  class ErnieMEncoderLayer (line 280) | class ErnieMEncoderLayer(nn.Module):
    method __init__ (line 281) | def __init__(self, config):
    method forward (line 300) | def forward(
  class ErnieMEncoder (line 343) | class ErnieMEncoder(nn.Module):
    method __init__ (line 344) | def __init__(self, config):
    method forward (line 349) | def forward(
  class ErnieMPooler (line 391) | class ErnieMPooler(nn.Module):
    method __init__ (line 392) | def __init__(self, config):
    method forward (line 397) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ErnieMPreTrainedModel (line 406) | class ErnieMPreTrainedModel(PreTrainedModel):
    method _init_weights (line 417) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 433) | def _set_gradient_checkpointing(self, module, value=False):
  class ErnieMModel (line 500) | class ErnieMModel(ErnieMPreTrainedModel):
    method __init__ (line 501) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 509) | def get_input_embeddings(self):
    method set_input_embeddings (line 512) | def set_input_embeddings(self, value):
    method _prune_heads (line 515) | def _prune_heads(self, heads_to_prune):
    method forward (line 530) | def forward(
  class ErnieMForSequenceClassification (line 614) | class ErnieMForSequenceClassification(ErnieMPreTrainedModel):
    method __init__ (line 616) | def __init__(self, config):
    method forward (line 638) | def forward(
  class ErnieMForMultipleChoice (line 716) | class ErnieMForMultipleChoice(ErnieMPreTrainedModel):
    method __init__ (line 718) | def __init__(self, config):
    method forward (line 737) | def forward(
  class ErnieMForTokenClassification (line 806) | class ErnieMForTokenClassification(ErnieMPreTrainedModel):
    method __init__ (line 808) | def __init__(self, config):
    method forward (line 829) | def forward(
  class ErnieMForQuestionAnswering (line 887) | class ErnieMForQuestionAnswering(ErnieMPreTrainedModel):
    method __init__ (line 889) | def __init__(self, config):
    method forward (line 906) | def forward(
  class ErnieMForInformationExtraction (line 985) | class ErnieMForInformationExtraction(ErnieMPreTrainedModel):
    method __init__ (line 986) | def __init__(self, config):
    method forward (line 995) | def forward(

FILE: transformers/models/ernie_m/tokenization_ernie_m.py
  class ErnieMTokenizer (line 62) | class ErnieMTokenizer(PreTrainedTokenizer):
    method __init__ (line 97) | def __init__(
    method get_offset_mapping (line 139) | def get_offset_mapping(self, text):
    method vocab_size (line 172) | def vocab_size(self):
    method get_vocab (line 175) | def get_vocab(self):
    method __getstate__ (line 178) | def __getstate__(self):
    method __setstate__ (line 183) | def __setstate__(self, d):
    method clean_text (line 193) | def clean_text(self, text):
    method _tokenize (line 197) | def _tokenize(self, text, enable_sampling=False, nbest_size=64, alpha=...
    method convert_tokens_to_string (line 240) | def convert_tokens_to_string(self, tokens):
    method convert_ids_to_string (line 245) | def convert_ids_to_string(self, ids):
    method _convert_token_to_id (line 254) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 258) | def _convert_id_to_token(self, index):
    method build_inputs_with_special_tokens (line 262) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method build_offset_mapping_with_special_tokens (line 284) | def build_offset_mapping_with_special_tokens(self, offset_mapping_0, o...
    method get_special_tokens_mask (line 305) | def get_special_tokens_mask(self, token_ids_0, token_ids_1=None, alrea...
    method create_token_type_ids_from_sequences (line 334) | def create_token_type_ids_from_sequences(
    method is_ch_char (line 358) | def is_ch_char(self, char):
    method is_alpha (line 366) | def is_alpha(self, char):
    method is_punct (line 374) | def is_punct(self, char):
    method is_whitespace (line 382) | def is_whitespace(self, char):
    method load_vocab (line 394) | def load_vocab(self, filepath):
    method save_vocabulary (line 403) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/esm/configuration_esm.py
  class EsmConfig (line 33) | class EsmConfig(PretrainedConfig):
    method __init__ (line 102) | def __init__(
    method to_dict (line 160) | def to_dict(self):
  class EsmFoldConfig (line 174) | class EsmFoldConfig:
    method __post_init__ (line 188) | def __post_init__(self):
    method to_dict (line 194) | def to_dict(self):
  class TrunkConfig (line 207) | class TrunkConfig:
    method __post_init__ (line 221) | def __post_init__(self):
    method to_dict (line 259) | def to_dict(self):
  class StructureModuleConfig (line 272) | class StructureModuleConfig:
    method to_dict (line 323) | def to_dict(self):
  function get_default_vocab_list (line 327) | def get_default_vocab_list():

FILE: transformers/models/esm/convert_esm.py
  function get_esmfold_tokenizer (line 78) | def get_esmfold_tokenizer():
  function transfer_and_check_weights (line 88) | def transfer_and_check_weights(original_module, our_module):
  function convert_esm_checkpoint_to_pytorch (line 96) | def convert_esm_checkpoint_to_pytorch(

FILE: transformers/models/esm/modeling_esm.py
  function rotate_half (line 51) | def rotate_half(x):
  function apply_rotary_pos_emb (line 56) | def apply_rotary_pos_emb(x, cos, sin):
  function gelu (line 63) | def gelu(x):
  function symmetrize (line 70) | def symmetrize(x):
  function average_product_correct (line 75) | def average_product_correct(x):
  class RotaryEmbedding (line 87) | class RotaryEmbedding(torch.nn.Module):
    method __init__ (line 94) | def __init__(self, dim: int):
    method _update_cos_sin_tables (line 105) | def _update_cos_sin_tables(self, x, seq_dimension=2):
    method forward (line 121) | def forward(self, q: torch.Tensor, k: torch.Tensor) -> Tuple[torch.Ten...
  class EsmContactPredictionHead (line 130) | class EsmContactPredictionHead(nn.Module):
    method __init__ (line 133) | def __init__(
    method forward (line 145) | def forward(self, tokens, attentions):
  class EsmEmbeddings (line 165) | class EsmEmbeddings(nn.Module):
    method __init__ (line 170) | def __init__(self, config):
    method forward (line 190) | def forward(
    method create_position_ids_from_inputs_embeds (line 235) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class EsmSelfAttention (line 253) | class EsmSelfAttention(nn.Module):
    method __init__ (line 254) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 283) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 288) | def forward(
  class EsmSelfOutput (line 391) | class EsmSelfOutput(nn.Module):
    method __init__ (line 392) | def __init__(self, config):
    method forward (line 397) | def forward(self, hidden_states, input_tensor):
  class EsmAttention (line 404) | class EsmAttention(nn.Module):
    method __init__ (line 405) | def __init__(self, config):
    method prune_heads (line 412) | def prune_heads(self, heads):
    method forward (line 430) | def forward(
  class EsmIntermediate (line 455) | class EsmIntermediate(nn.Module):
    method __init__ (line 456) | def __init__(self, config):
    method forward (line 460) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class EsmOutput (line 466) | class EsmOutput(nn.Module):
    method __init__ (line 467) | def __init__(self, config):
    method forward (line 472) | def forward(self, hidden_states, input_tensor):
  class EsmLayer (line 479) | class EsmLayer(nn.Module):
    method __init__ (line 480) | def __init__(self, config):
    method forward (line 495) | def forward(
    method feed_forward_chunk (line 558) | def feed_forward_chunk(self, attention_output):
  class EsmEncoder (line 565) | class EsmEncoder(nn.Module):
    method __init__ (line 566) | def __init__(self, config):
    method forward (line 573) | def forward(
  class EsmPooler (line 668) | class EsmPooler(nn.Module):
    method __init__ (line 669) | def __init__(self, config):
    method forward (line 674) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class EsmPreTrainedModel (line 683) | class EsmPreTrainedModel(PreTrainedModel):
    method _init_weights (line 694) | def _init_weights(self, module):
  class EsmModel (line 773) | class EsmModel(EsmPreTrainedModel):
    method __init__ (line 789) | def __init__(self, config, add_pooling_layer=True):
    method _set_gradient_checkpointing (line 805) | def _set_gradient_checkpointing(self, module, value=False):
    method get_input_embeddings (line 809) | def get_input_embeddings(self):
    method set_input_embeddings (line 812) | def set_input_embeddings(self, value):
    method _prune_heads (line 815) | def _prune_heads(self, heads_to_prune):
    method forward (line 829) | def forward(
    method predict_contacts (line 949) | def predict_contacts(self, tokens, attention_mask):
  class EsmForMaskedLM (line 962) | class EsmForMaskedLM(EsmPreTrainedModel):
    method __init__ (line 966) | def __init__(self, config):
    method get_output_embeddings (line 980) | def get_output_embeddings(self):
    method set_output_embeddings (line 983) | def set_output_embeddings(self, new_embeddings):
    method forward (line 993) | def forward(
    method predict_contacts (line 1050) | def predict_contacts(self, tokens, attention_mask):
  class EsmLMHead (line 1054) | class EsmLMHead(nn.Module):
    method __init__ (line 1057) | def __init__(self, config):
    method forward (line 1065) | def forward(self, features, **kwargs):
  class EsmForSequenceClassification (line 1082) | class EsmForSequenceClassification(EsmPreTrainedModel):
    method __init__ (line 1085) | def __init__(self, config):
    method forward (line 1101) | def forward(
  class EsmForTokenClassification (line 1178) | class EsmForTokenClassification(EsmPreTrainedModel):
    method __init__ (line 1182) | def __init__(self, config):
    method forward (line 1198) | def forward(
  class EsmClassificationHead (line 1251) | class EsmClassificationHead(nn.Module):
    method __init__ (line 1254) | def __init__(self, config):
    method forward (line 1260) | def forward(self, features, **kwargs):
  function create_position_ids_from_input_ids (line 1270) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/esm/modeling_esmfold.py
  class EsmForProteinFoldingOutput (line 60) | class EsmForProteinFoldingOutput(ModelOutput):
  function is_fp16_enabled (line 173) | def is_fp16_enabled():
  function is_deepspeed_initialized (line 181) | def is_deepspeed_initialized():
  function collate_dense_tensors (line 194) | def collate_dense_tensors(samples: List[torch.Tensor], pad_v: float = 0)...
  function flatten_final_dims (line 217) | def flatten_final_dims(t: torch.Tensor, no_dims: int):
  function permute_final_dims (line 221) | def permute_final_dims(tensor: torch.Tensor, inds: List[int]):
  function dict_multimap (line 227) | def dict_multimap(fn, dicts):
  function trunc_normal_init_ (line 240) | def trunc_normal_init_(weights, scale=1.0, fan="fan_in"):
  function ipa_point_weights_init_ (line 261) | def ipa_point_weights_init_(weights):
  class EsmFoldLinear (line 267) | class EsmFoldLinear(nn.Linear):
    method __init__ (line 274) | def __init__(
  class EsmFoldLayerNorm (line 313) | class EsmFoldLayerNorm(nn.Module):
    method __init__ (line 314) | def __init__(self, c_in, eps=1e-5):
    method forward (line 323) | def forward(self, x):
  function softmax_no_cast (line 335) | def softmax_no_cast(t: torch.Tensor, dim: int = -1) -> torch.Tensor:
  class EsmFoldAttention (line 349) | class EsmFoldAttention(nn.Module):
    method __init__ (line 354) | def __init__(
    method _prep_qkv (line 401) | def _prep_qkv(self, q_x: torch.Tensor, kv_x: torch.Tensor) -> Tuple[to...
    method _wrap_up (line 421) | def _wrap_up(self, o: torch.Tensor, q_x: torch.Tensor) -> torch.Tensor:
    method forward (line 437) | def forward(
  class EsmFoldTriangleAttention (line 501) | class EsmFoldTriangleAttention(nn.Module):
    method __init__ (line 502) | def __init__(self, c_in, c_hidden, no_heads, starting=True, inf=1e9):
    method _chunk (line 527) | def _chunk(
    method forward (line 551) | def forward(
  class EsmFoldTriangleMultiplicativeUpdate (line 611) | class EsmFoldTriangleMultiplicativeUpdate(nn.Module):
    method __init__ (line 616) | def __init__(self, config, _outgoing=True):
    method _combine_projections (line 633) | def _combine_projections(
    method _inference_forward (line 659) | def _inference_forward(
    method forward (line 886) | def forward(
  class EsmFoldPreTrainedModel (line 940) | class EsmFoldPreTrainedModel(EsmPreTrainedModel):
    method _init_weights (line 947) | def _init_weights(self, module):
  class EsmFoldSelfAttention (line 992) | class EsmFoldSelfAttention(nn.Module):
    method __init__ (line 993) | def __init__(self, embed_dim, num_heads, head_width, gated=False):
    method forward (line 1013) | def forward(self, x, mask=None, bias=None, indices=None):
  class EsmFoldDropout (line 1054) | class EsmFoldDropout(nn.Module):
    method __init__ (line 1059) | def __init__(self, r: float, batch_dim: Union[int, List[int]]):
    method forward (line 1068) | def forward(self, x: torch.Tensor) -> torch.Tensor:
  class EsmFoldSequenceToPair (line 1076) | class EsmFoldSequenceToPair(nn.Module):
    method __init__ (line 1077) | def __init__(self, sequence_state_dim, inner_dim, pairwise_state_dim):
    method forward (line 1087) | def forward(self, sequence_state):
  class EsmFoldPairToSequence (line 1114) | class EsmFoldPairToSequence(nn.Module):
    method __init__ (line 1115) | def __init__(self, pairwise_state_dim, num_heads):
    method forward (line 1121) | def forward(self, pairwise_state):
  class EsmFoldResidueMLP (line 1135) | class EsmFoldResidueMLP(nn.Module):
    method __init__ (line 1136) | def __init__(self, embed_dim, inner_dim, dropout=0):
    method forward (line 1147) | def forward(self, x):
  class EsmFoldTriangularSelfAttentionBlock (line 1151) | class EsmFoldTriangularSelfAttentionBlock(nn.Module):
    method __init__ (line 1152) | def __init__(self, config):
    method forward (line 1186) | def forward(self, sequence_state, pairwise_state, mask=None, chunk_siz...
  class EsmCategoricalMixture (line 1255) | class EsmCategoricalMixture:
    method __init__ (line 1256) | def __init__(self, param, bins=50, start=0, end=1):
    method log_prob (line 1262) | def log_prob(self, true):
    method mean (line 1270) | def mean(self):
  function categorical_lddt (line 1274) | def categorical_lddt(logits, bins=50):
  function get_axial_mask (line 1279) | def get_axial_mask(mask):
  class EsmFoldRelativePosition (line 1301) | class EsmFoldRelativePosition(nn.Module):
    method __init__ (line 1302) | def __init__(self, config):
    method forward (line 1310) | def forward(self, residue_index, mask=None):
  class EsmFoldAngleResnetBlock (line 1337) | class EsmFoldAngleResnetBlock(nn.Module):
    method __init__ (line 1338) | def __init__(self, config):
    method forward (line 1346) | def forward(self, a: torch.Tensor) -> torch.Tensor:
  class EsmFoldAngleResnet (line 1357) | class EsmFoldAngleResnet(nn.Module):
    method __init__ (line 1362) | def __init__(self, config):
    method forward (line 1378) | def forward(self, s: torch.Tensor, s_initial: torch.Tensor) -> Tuple[t...
  class EsmFoldInvariantPointAttention (line 1422) | class EsmFoldInvariantPointAttention(nn.Module):
    method __init__ (line 1427) | def __init__(self, config):
    method forward (line 1462) | def forward(
  class EsmFoldBackboneUpdate (line 1619) | class EsmFoldBackboneUpdate(nn.Module):
    method __init__ (line 1624) | def __init__(self, config):
    method forward (line 1629) | def forward(self, s: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]:
  class EsmFoldStructureModuleTransitionLayer (line 1642) | class EsmFoldStructureModuleTransitionLayer(nn.Module):
    method __init__ (line 1643) | def __init__(self, config):
    method forward (line 1652) | def forward(self, s):
  class EsmFoldStructureModuleTransition (line 1665) | class EsmFoldStructureModuleTransition(nn.Module):
    method __init__ (line 1666) | def __init__(self, config):
    method forward (line 1678) | def forward(self, s):
  class EsmFoldStructureModule (line 1688) | class EsmFoldStructureModule(nn.Module):
    method __init__ (line 1689) | def __init__(self, config):
    method forward (line 1713) | def forward(
    method _init_residue_constants (line 1826) | def _init_residue_constants(self, float_dtype, device):
    method torsion_angles_to_frames (line 1871) | def torsion_angles_to_frames(self, r, alpha, f):
    method frames_and_literature_positions_to_atom14_pos (line 1877) | def frames_and_literature_positions_to_atom14_pos(self, r, f):  # [*, ...
  class EsmFoldingTrunk (line 1890) | class EsmFoldingTrunk(nn.Module):
    method __init__ (line 1891) | def __init__(self, config):
    method set_chunk_size (line 1914) | def set_chunk_size(self, chunk_size):
    method forward (line 1921) | def forward(self, seq_feats, pair_feats, true_aa, residx, mask, no_rec...
    method distogram (line 1987) | def distogram(coords, min_bin, max_bin, num_bins):
  class EsmForProteinFolding (line 2020) | class EsmForProteinFolding(EsmPreTrainedModel):
    method __init__ (line 2021) | def __init__(self, config):
    method _af2_to_esm_from_vocab_list (line 2077) | def _af2_to_esm_from_vocab_list(vocab_list: List[str]) -> torch.Tensor:
    method forward (line 2084) | def forward(
    method af2_idx_to_esm_idx (line 2208) | def af2_idx_to_esm_idx(self, aa, mask):
    method compute_language_model_representations (line 2215) | def compute_language_model_representations(self, esmaa: torch.Tensor) ...
    method bert_mask (line 2240) | def bert_mask(self, aa, esmaa, mask, pattern):
    method infer (line 2250) | def infer(
    method output_to_pdb (line 2290) | def output_to_pdb(output: Dict) -> List[str]:
    method infer_pdb (line 2311) | def infer_pdb(self, seqs, *args, **kwargs) -> str:
    method infer_pdbs (line 2317) | def infer_pdbs(self, seqs: List[str], *args, **kwargs) -> List[str]:

FILE: transformers/models/esm/modeling_tf_esm.py
  function rotate_half (line 64) | def rotate_half(x):
  function apply_rotary_pos_emb (line 69) | def apply_rotary_pos_emb(x, cos, sin):
  function symmetrize (line 76) | def symmetrize(x):
  function average_product_correct (line 81) | def average_product_correct(x):
  class TFRotaryEmbedding (line 93) | class TFRotaryEmbedding(Layer):
    method __init__ (line 100) | def __init__(self, dim: int, name=None):
    method build (line 110) | def build(self, input_shape):
    method _compute_cos_sin (line 119) | def _compute_cos_sin(self, x, seq_dimension=2):
    method call (line 128) | def call(self, q: tf.Tensor, k: tf.Tensor) -> Tuple[tf.Tensor, tf.Tens...
  class TFEsmContactPredictionHead (line 137) | class TFEsmContactPredictionHead(Layer):
    method __init__ (line 140) | def __init__(
    method build (line 152) | def build(self, input_shape):
    method call (line 157) | def call(self, tokens, attentions):
  class TFEsmEmbeddings (line 174) | class TFEsmEmbeddings(Layer):
    method __init__ (line 179) | def __init__(self, config, name=None):
    method call (line 209) | def call(
    method create_position_ids_from_inputs_embeds (line 254) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class TFEsmSelfAttention (line 272) | class TFEsmSelfAttention(Layer):
    method __init__ (line 273) | def __init__(self, config, position_embedding_type=None, name=None):
    method transpose_for_scores (line 310) | def transpose_for_scores(self, x: tf.Tensor) -> tf.Tensor:
    method call (line 315) | def call(
  class TFEsmSelfOutput (line 419) | class TFEsmSelfOutput(Layer):
    method __init__ (line 420) | def __init__(self, config, name=None):
    method call (line 427) | def call(self, hidden_states, input_tensor, training=False):
  class TFEsmAttention (line 434) | class TFEsmAttention(Layer):
    method __init__ (line 435) | def __init__(self, config, name=None):
    method prune_heads (line 442) | def prune_heads(self, heads):
    method call (line 445) | def call(
  class TFEsmIntermediate (line 472) | class TFEsmIntermediate(tf.keras.layers.Layer):
    method __init__ (line 473) | def __init__(self, config: EsmConfig, **kwargs):
    method call (line 482) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFEsmOutput (line 488) | class TFEsmOutput(Layer):
    method __init__ (line 489) | def __init__(self, config, name=None):
    method call (line 496) | def call(self, hidden_states, input_tensor, training=False):
  class TFEsmLayer (line 503) | class TFEsmLayer(Layer):
    method __init__ (line 504) | def __init__(self, config, name=None):
    method call (line 519) | def call(
  class TFEsmEncoder (line 590) | class TFEsmEncoder(Layer):
    method __init__ (line 591) | def __init__(self, config, name=None):
    method call (line 597) | def call(
  class TFEsmPooler (line 670) | class TFEsmPooler(Layer):
    method __init__ (line 671) | def __init__(self, config: EsmConfig, **kwargs):
    method call (line 681) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFEsmPreTrainedModel (line 690) | class TFEsmPreTrainedModel(TFPreTrainedModel):
  class TFEsmMainLayer (line 761) | class TFEsmMainLayer(Layer):
    method __init__ (line 776) | def __init__(self, config, add_pooling_layer=True, name=None, **kwargs):
    method build (line 790) | def build(self, input_shape):
    method get_input_embeddings (line 795) | def get_input_embeddings(self):
    method set_input_embeddings (line 798) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 802) | def _prune_heads(self, heads_to_prune):
    method call (line 805) | def call(
    method predict_contacts (line 958) | def predict_contacts(self, tokens, attention_mask):
  class TFEsmModel (line 975) | class TFEsmModel(TFEsmPreTrainedModel):
    method __init__ (line 976) | def __init__(self, config: EsmConfig, add_pooling_layer=True, *inputs,...
    method call (line 988) | def call(
    method predict_contacts (line 1041) | def predict_contacts(self, tokens, attention_mask):
  class TFEsmForMaskedLM (line 1046) | class TFEsmForMaskedLM(TFEsmPreTrainedModel, TFMaskedLanguageModelingLoss):
    method __init__ (line 1050) | def __init__(self, config):
    method get_output_embeddings (line 1067) | def get_output_embeddings(self):
    method set_output_embeddings (line 1070) | def set_output_embeddings(self, new_embeddings):
    method get_lm_head (line 1073) | def get_lm_head(self):
    method call (line 1084) | def call(
    method predict_contacts (line 1140) | def predict_contacts(self, tokens, attention_mask):
  class TFEsmLMHead (line 1144) | class TFEsmLMHead(Layer):
    method __init__ (line 1147) | def __init__(self, config, name=None):
    method build (line 1165) | def build(self, input_shape):
    method get_bias (line 1171) | def get_bias(self):
    method call (line 1174) | def call(self, features):
  class TFEsmForSequenceClassification (line 1194) | class TFEsmForSequenceClassification(TFEsmPreTrainedModel, TFSequenceCla...
    method __init__ (line 1197) | def __init__(self, config):
    method call (line 1212) | def call(
  class TFEsmForTokenClassification (line 1268) | class TFEsmForTokenClassification(TFEsmPreTrainedModel, TFTokenClassific...
    method __init__ (line 1272) | def __init__(self, config):
    method call (line 1287) | def call(
  class TFEsmClassificationHead (line 1337) | class TFEsmClassificationHead(Layer):
    method __init__ (line 1340) | def __init__(self, config, name=None):
    method call (line 1356) | def call(self, features, training=False):
  function create_position_ids_from_input_ids (line 1365) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/esm/openfold_utils/chunk_utils.py
  function _fetch_dims (line 24) | def _fetch_dims(tree: Union[dict, list, tuple, torch.Tensor]) -> List[Tu...
  function _flat_idx_to_idx (line 41) | def _flat_idx_to_idx(flat_idx: int, dims: Tuple[int, ...]) -> Tuple[int,...
  function _get_minimal_slice_set (line 51) | def _get_minimal_slice_set(
  function _chunk_slice (line 166) | def _chunk_slice(t: torch.Tensor, flat_start: int, flat_end: int, no_bat...
  function chunk_layer (line 194) | def chunk_layer(
  class ChunkSizeTuner (line 314) | class ChunkSizeTuner:
    method __init__ (line 315) | def __init__(
    method _determine_favorable_chunk_size (line 325) | def _determine_favorable_chunk_size(self, fn: Callable, args: tuple, m...
    method _compare_arg_caches (line 356) | def _compare_arg_caches(self, ac1: Iterable, ac2: Iterable) -> bool:
    method tune_chunk_size (line 371) | def tune_chunk_size(

FILE: transformers/models/esm/openfold_utils/data_transforms.py
  function make_atom14_masks (line 25) | def make_atom14_masks(protein: Dict[str, torch.Tensor]) -> Dict[str, tor...
  function make_atom14_masks_np (line 90) | def make_atom14_masks_np(batch: Dict[str, torch.Tensor]) -> Dict[str, np...

FILE: transformers/models/esm/openfold_utils/feats.py
  function pseudo_beta_fn (line 28) | def pseudo_beta_fn(aatype: torch.Tensor, all_atom_positions: torch.Tenso...
  function pseudo_beta_fn (line 33) | def pseudo_beta_fn(
  function pseudo_beta_fn (line 39) | def pseudo_beta_fn(aatype, all_atom_positions, all_atom_masks):
  function atom14_to_atom37 (line 60) | def atom14_to_atom37(atom14: torch.Tensor, batch: Dict[str, torch.Tensor...
  function build_template_angle_feat (line 73) | def build_template_angle_feat(template_feats: Dict[str, torch.Tensor]) -...
  function build_template_pair_feat (line 91) | def build_template_pair_feat(
  function build_extra_msa_feat (line 152) | def build_extra_msa_feat(batch: Dict[str, torch.Tensor]) -> torch.Tensor:
  function torsion_angles_to_frames (line 162) | def torsion_angles_to_frames(
  function frames_and_literature_positions_to_atom14_pos (line 224) | def frames_and_literature_positions_to_atom14_pos(

FILE: transformers/models/esm/openfold_utils/loss.py
  function _calculate_bin_centers (line 21) | def _calculate_bin_centers(boundaries: torch.Tensor) -> torch.Tensor:
  function _calculate_expected_aligned_error (line 28) | def _calculate_expected_aligned_error(
  function compute_predicted_aligned_error (line 39) | def compute_predicted_aligned_error(
  function compute_tm (line 74) | def compute_tm(

FILE: transformers/models/esm/openfold_utils/protein.py
  class Protein (line 33) | class Protein:
  function from_proteinnet_string (line 70) | def from_proteinnet_string(proteinnet_str: str) -> Protein:
  function get_pdb_headers (line 120) | def get_pdb_headers(prot: Protein, chain_id: int = 0) -> List[str]:
  function add_pdb_headers (line 140) | def add_pdb_headers(prot: Protein, pdb_str: str) -> str:
  function to_pdb (line 190) | def to_pdb(prot: Protein) -> str:
  function ideal_atom_mask (line 283) | def ideal_atom_mask(prot: Protein) -> np.ndarray:
  function from_prediction (line 298) | def from_prediction(

FILE: transformers/models/esm/openfold_utils/residue_constants.py
  function map_structure_with_atom_order (line 387) | def map_structure_with_atom_order(in_list: list, first_call: bool = True...
  function load_stereo_chemical_props (line 402) | def load_stereo_chemical_props() -> (
  function sequence_to_onehot (line 606) | def sequence_to_onehot(sequence: str, mapping: Mapping[str, int], map_un...
  function _make_standard_atom_mask (line 752) | def _make_standard_atom_mask() -> np.ndarray:
  function chi_angle_atom (line 770) | def chi_angle_atom(atom_index: int) -> np.ndarray:
  function _make_rigid_transformation_4x4 (line 812) | def _make_rigid_transformation_4x4(ex: np.ndarray, ey: np.ndarray, trans...
  function _make_rigid_group_constants (line 841) | def _make_rigid_group_constants() -> None:
  function make_atom14_dists_bounds (line 915) | def make_atom14_dists_bounds(
  function _make_atom14_ambiguity_feats (line 967) | def _make_atom14_ambiguity_feats() -> None:
  function aatype_to_str_sequence (line 982) | def aatype_to_str_sequence(aatype: Sequence[int]) -> str:

FILE: transformers/models/esm/openfold_utils/rigid_utils.py
  function rot_matmul (line 25) | def rot_matmul(a: torch.Tensor, b: torch.Tensor) -> torch.Tensor:
  function rot_vec_mul (line 56) | def rot_vec_mul(r: torch.Tensor, t: torch.Tensor) -> torch.Tensor:
  function identity_rot_mats (line 78) | def identity_rot_mats(
  function identity_trans (line 93) | def identity_trans(
  function identity_quats (line 104) | def identity_quats(
  function _to_mat (line 123) | def _to_mat(pairs: List[Tuple[str, int]]) -> np.ndarray:
  function quat_to_rot (line 144) | def quat_to_rot(quat: torch.Tensor) -> torch.Tensor:
  function rot_to_quat (line 167) | def rot_to_quat(rot: torch.Tensor) -> torch.Tensor:
  function _get_quat (line 223) | def _get_quat(quat_key: str, dtype: torch.dtype, device: torch.device) -...
  function quat_multiply (line 227) | def quat_multiply(quat1: torch.Tensor, quat2: torch.Tensor) -> torch.Ten...
  function quat_multiply_by_vec (line 234) | def quat_multiply_by_vec(quat: torch.Tensor, vec: torch.Tensor) -> torch...
  function invert_rot_mat (line 241) | def invert_rot_mat(rot_mat: torch.Tensor) -> torch.Tensor:
  function invert_quat (line 245) | def invert_quat(quat: torch.Tensor) -> torch.Tensor:
  class Rotation (line 252) | class Rotation:
    method __init__ (line 260) | def __init__(
    method identity (line 295) | def identity(
    method __getitem__ (line 335) | def __getitem__(self, index: Any) -> Rotation:
    method __mul__ (line 358) | def __mul__(self, right: torch.Tensor) -> Rotation:
    method __rmul__ (line 380) | def __rmul__(self, left: torch.Tensor) -> Rotation:
    method shape (line 395) | def shape(self) -> torch.Size:
    method dtype (line 412) | def dtype(self) -> torch.dtype:
    method device (line 427) | def device(self) -> torch.device:
    method requires_grad (line 442) | def requires_grad(self) -> bool:
    method get_rot_mats (line 456) | def get_rot_mats(self) -> torch.Tensor:
    method get_quats (line 470) | def get_quats(self) -> torch.Tensor:
    method get_cur_rot (line 486) | def get_cur_rot(self) -> torch.Tensor:
    method compose_q_update_vec (line 502) | def compose_q_update_vec(self, q_update_vec: torch.Tensor, normalize_q...
    method compose_r (line 524) | def compose_r(self, r: Rotation) -> Rotation:
    method compose_q (line 539) | def compose_q(self, r: Rotation, normalize_quats: bool = True) -> Rota...
    method apply (line 557) | def apply(self, pts: torch.Tensor) -> torch.Tensor:
    method invert_apply (line 570) | def invert_apply(self, pts: torch.Tensor) -> torch.Tensor:
    method invert (line 584) | def invert(self) -> Rotation:
    method unsqueeze (line 604) | def unsqueeze(self, dim: int) -> Rotation:
    method cat (line 626) | def cat(rs: Sequence[Rotation], dim: int) -> Rotation:
    method map_tensor_fn (line 648) | def map_tensor_fn(self, fn: Callable[[torch.Tensor], torch.Tensor]) ->...
    method cuda (line 670) | def cuda(self) -> Rotation:
    method to (line 684) | def to(self, device: Optional[torch.device], dtype: Optional[torch.dty...
    method detach (line 710) | def detach(self) -> Rotation:
  class Rigid (line 729) | class Rigid:
    method __init__ (line 736) | def __init__(self, rots: Optional[Rotation], trans: Optional[torch.Ten...
    method identity (line 786) | def identity(
    method __getitem__ (line 813) | def __getitem__(self, index: Any) -> Rigid:
    method __mul__ (line 838) | def __mul__(self, right: torch.Tensor) -> Rigid:
    method __rmul__ (line 856) | def __rmul__(self, left: torch.Tensor) -> Rigid:
    method shape (line 869) | def shape(self) -> torch.Size:
    method device (line 879) | def device(self) -> torch.device:
    method get_rots (line 888) | def get_rots(self) -> Rotation:
    method get_trans (line 897) | def get_trans(self) -> torch.Tensor:
    method compose_q_update_vec (line 906) | def compose_q_update_vec(self, q_update_vec: torch.Tensor) -> Rigid:
    method compose (line 924) | def compose(self, r: Rigid) -> Rigid:
    method apply (line 938) | def apply(self, pts: torch.Tensor) -> torch.Tensor:
    method invert_apply (line 950) | def invert_apply(self, pts: torch.Tensor) -> torch.Tensor:
    method invert (line 962) | def invert(self) -> Rigid:
    method map_tensor_fn (line 974) | def map_tensor_fn(self, fn: Callable[[torch.Tensor], torch.Tensor]) ->...
    method to_tensor_4x4 (line 990) | def to_tensor_4x4(self) -> torch.Tensor:
    method from_tensor_4x4 (line 1004) | def from_tensor_4x4(t: torch.Tensor) -> Rigid:
    method to_tensor_7 (line 1021) | def to_tensor_7(self) -> torch.Tensor:
    method from_tensor_7 (line 1036) | def from_tensor_7(t: torch.Tensor, normalize_quats: bool = False) -> R...
    method from_3_points (line 1047) | def from_3_points(
    method unsqueeze (line 1087) | def unsqueeze(self, dim: int) -> Rigid:
    method cat (line 1104) | def cat(ts: Sequence[Rigid], dim: int) -> Rigid:
    method apply_rot_fn (line 1121) | def apply_rot_fn(self, fn: Callable[[Rotation], Rotation]) -> Rigid:
    method apply_trans_fn (line 1132) | def apply_trans_fn(self, fn: Callable[[torch.Tensor], torch.Tensor]) -...
    method scale_translation (line 1144) | def scale_translation(self, trans_scale_factor: float) -> Rigid:
    method stop_rot_gradient (line 1156) | def stop_rot_gradient(self) -> Rigid:
    method make_transform_from_reference (line 1166) | def make_transform_from_reference(
    method cuda (line 1235) | def cuda(self) -> Rigid:

FILE: transformers/models/esm/openfold_utils/tensor_utils.py
  function add (line 24) | def add(m1: torch.Tensor, m2: torch.Tensor, inplace: bool) -> torch.Tensor:
  function permute_final_dims (line 35) | def permute_final_dims(tensor: torch.Tensor, inds: List[int]) -> torch.T...
  function flatten_final_dims (line 41) | def flatten_final_dims(t: torch.Tensor, no_dims: int) -> torch.Tensor:
  function masked_mean (line 45) | def masked_mean(mask: torch.Tensor, value: torch.Tensor, dim: int, eps: ...
  function pts_to_distogram (line 50) | def pts_to_distogram(
  function dict_multimap (line 58) | def dict_multimap(fn: Callable[[list], Any], dicts: List[dict]) -> dict:
  function one_hot (line 71) | def one_hot(x: torch.Tensor, v_bins: torch.Tensor) -> torch.Tensor:
  function batched_gather (line 78) | def batched_gather(data: torch.Tensor, inds: torch.Tensor, dim: int = 0,...
  function dict_map (line 97) | def dict_map(
  function tree_map (line 111) | def tree_map(fn: Callable[[T], Any], tree: T, leaf_type: Type[T]) -> Any:
  function tree_map (line 116) | def tree_map(fn: Callable[[T], Any], tree: dict, leaf_type: Type[T]) -> ...
  function tree_map (line 121) | def tree_map(fn: Callable[[T], Any], tree: list, leaf_type: Type[T]) -> ...
  function tree_map (line 126) | def tree_map(fn: Callable[[T], Any], tree: tuple, leaf_type: Type[T]) ->...
  function tree_map (line 130) | def tree_map(fn, tree, leaf_type):

FILE: transformers/models/esm/tokenization_esm.py
  function load_vocab_file (line 41) | def load_vocab_file(vocab_file):
  class EsmTokenizer (line 47) | class EsmTokenizer(PreTrainedTokenizer):
    method __init__ (line 57) | def __init__(
    method _convert_id_to_token (line 79) | def _convert_id_to_token(self, index: int) -> str:
    method _convert_token_to_id (line 82) | def _convert_token_to_id(self, token: str) -> int:
    method _tokenize (line 85) | def _tokenize(self, text, **kwargs):
    method get_vocab_size (line 88) | def get_vocab_size(self, with_added_tokens=False):
    method get_vocab (line 91) | def get_vocab(self):
    method token_to_id (line 94) | def token_to_id(self, token: str) -> int:
    method id_to_token (line 97) | def id_to_token(self, index: int) -> str:
    method build_inputs_with_special_tokens (line 100) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 114) | def get_special_tokens_mask(
    method save_vocabulary (line 145) | def save_vocabulary(self, save_directory, filename_prefix):
    method vocab_size (line 152) | def vocab_size(self) -> int:
    method _add_tokens (line 155) | def _add_tokens(self, new_tokens: Union[List[str], List[AddedToken]], ...

FILE: transformers/models/flaubert/configuration_flaubert.py
  class FlaubertConfig (line 34) | class FlaubertConfig(PretrainedConfig):
    method __init__ (line 146) | def __init__(
  class FlaubertOnnxConfig (line 226) | class FlaubertOnnxConfig(OnnxConfig):
    method inputs (line 228) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/flaubert/modeling_flaubert.py
  function create_sinusoidal_embeddings (line 65) | def create_sinusoidal_embeddings(n_pos, dim, out):
  function get_masks (line 74) | def get_masks(slen, lengths, causal, padding_mask=None):
  class MultiHeadAttention (line 100) | class MultiHeadAttention(nn.Module):
    method __init__ (line 103) | def __init__(self, n_heads, dim, config):
    method prune_heads (line 117) | def prune_heads(self, heads):
    method forward (line 132) | def forward(self, input, mask, kv=None, cache=None, head_mask=None, ou...
  class TransformerFFN (line 197) | class TransformerFFN(nn.Module):
    method __init__ (line 198) | def __init__(self, in_dim, dim_hidden, out_dim, config):
    method forward (line 207) | def forward(self, input):
    method ff_chunk (line 210) | def ff_chunk(self, input):
  class FlaubertPredLayer (line 298) | class FlaubertPredLayer(nn.Module):
    method __init__ (line 303) | def __init__(self, config):
    method forward (line 321) | def forward(self, x, y=None):
  class FlaubertPreTrainedModel (line 341) | class FlaubertPreTrainedModel(PreTrainedModel):
    method __init__ (line 351) | def __init__(self, *inputs, **kwargs):
    method dummy_inputs (line 355) | def dummy_inputs(self):
    method _init_weights (line 364) | def _init_weights(self, module):
  class FlaubertModel (line 381) | class FlaubertModel(FlaubertPreTrainedModel):
    method __init__ (line 384) | def __init__(self, config):  # , dico, is_encoder, with_output):
    method get_input_embeddings (line 461) | def get_input_embeddings(self):
    method set_input_embeddings (line 465) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 469) | def _prune_heads(self, heads_to_prune):
    method forward (line 483) | def forward(
  class FlaubertWithLMHeadModel (line 656) | class FlaubertWithLMHeadModel(FlaubertPreTrainedModel):
    method __init__ (line 659) | def __init__(self, config):
    method get_output_embeddings (line 667) | def get_output_embeddings(self):
    method set_output_embeddings (line 670) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 673) | def prepare_inputs_for_generation(self, input_ids, **kwargs):
    method forward (line 693) | def forward(
  class FlaubertForSequenceClassification (line 754) | class FlaubertForSequenceClassification(FlaubertPreTrainedModel):
    method __init__ (line 755) | def __init__(self, config):
    method forward (line 772) | def forward(
  class FlaubertForTokenClassification (line 857) | class FlaubertForTokenClassification(FlaubertPreTrainedModel):
    method __init__ (line 858) | def __init__(self, config):
    method forward (line 875) | def forward(
  class FlaubertForQuestionAnsweringSimple (line 942) | class FlaubertForQuestionAnsweringSimple(FlaubertPreTrainedModel):
    method __init__ (line 943) | def __init__(self, config):
    method forward (line 958) | def forward(
  class FlaubertForQuestionAnsweringOutput (line 1048) | class FlaubertForQuestionAnsweringOutput(ModelOutput):
  class FlaubertForQuestionAnswering (line 1091) | class FlaubertForQuestionAnswering(FlaubertPreTrainedModel):
    method __init__ (line 1092) | def __init__(self, config):
    method forward (line 1103) | def forward(
  class FlaubertForMultipleChoice (line 1213) | class FlaubertForMultipleChoice(FlaubertPreTrainedModel):
    method __init__ (line 1214) | def __init__(self, config, *inputs, **kwargs):
    method forward (line 1232) | def forward(

FILE: transformers/models/flaubert/modeling_tf_flaubert.py
  function get_masks (line 190) | def get_masks(slen, lengths, causal, padding_mask=None):
  class TFFlaubertPreTrainedModel (line 219) | class TFFlaubertPreTrainedModel(TFPreTrainedModel):
    method dummy_inputs (line 229) | def dummy_inputs(self):
  class TFFlaubertModel (line 247) | class TFFlaubertModel(TFFlaubertPreTrainedModel):
    method __init__ (line 248) | def __init__(self, config, *inputs, **kwargs):
    method call (line 259) | def call(
  class TFFlaubertMultiHeadAttention (line 295) | class TFFlaubertMultiHeadAttention(tf.keras.layers.Layer):
    method __init__ (line 298) | def __init__(self, n_heads, dim, config, **kwargs):
    method prune_heads (line 313) | def prune_heads(self, heads):
    method call (line 316) | def call(self, input, mask, kv, cache, head_mask, output_attentions, t...
  class TFFlaubertTransformerFFN (line 388) | class TFFlaubertTransformerFFN(tf.keras.layers.Layer):
    method __init__ (line 389) | def __init__(self, in_dim, dim_hidden, out_dim, config, **kwargs):
    method call (line 397) | def call(self, input, training=False):
  class TFFlaubertMainLayer (line 407) | class TFFlaubertMainLayer(tf.keras.layers.Layer):
    method __init__ (line 410) | def __init__(self, config, **kwargs):
    method build (line 457) | def build(self, input_shape):
    method get_input_embeddings (line 475) | def get_input_embeddings(self):
    method set_input_embeddings (line 478) | def set_input_embeddings(self, value):
    method call (line 483) | def call(
  class TFFlaubertPredLayer (line 680) | class TFFlaubertPredLayer(tf.keras.layers.Layer):
    method __init__ (line 685) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 704) | def build(self, input_shape):
    method get_output_embeddings (line 710) | def get_output_embeddings(self):
    method set_output_embeddings (line 713) | def set_output_embeddings(self, value):
    method get_bias (line 717) | def get_bias(self):
    method set_bias (line 720) | def set_bias(self, value):
    method call (line 724) | def call(self, hidden_states):
  class TFFlaubertWithLMHeadModelOutput (line 732) | class TFFlaubertWithLMHeadModelOutput(ModelOutput):
  class TFFlaubertWithLMHeadModel (line 764) | class TFFlaubertWithLMHeadModel(TFFlaubertPreTrainedModel):
    method __init__ (line 765) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 772) | def get_lm_head(self):
    method get_prefix_bias_name (line 775) | def get_prefix_bias_name(self):
    method prepare_inputs_for_generation (line 779) | def prepare_inputs_for_generation(self, inputs, **kwargs):
    method call (line 800) | def call(
  class TFFlaubertForSequenceClassification (line 850) | class TFFlaubertForSequenceClassification(TFFlaubertPreTrainedModel, TFS...
    method __init__ (line 851) | def __init__(self, config, *inputs, **kwargs):
    method call (line 865) | def call(
  class TFFlaubertForQuestionAnsweringSimple (line 929) | class TFFlaubertForQuestionAnsweringSimple(TFFlaubertPreTrainedModel, TF...
    method __init__ (line 930) | def __init__(self, config, *inputs, **kwargs):
    method call (line 944) | def call(
  class TFFlaubertForTokenClassification (line 1021) | class TFFlaubertForTokenClassification(TFFlaubertPreTrainedModel, TFToke...
    method __init__ (line 1022) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1039) | def call(
  class TFFlaubertForMultipleChoice (line 1102) | class TFFlaubertForMultipleChoice(TFFlaubertPreTrainedModel, TFMultipleC...
    method __init__ (line 1103) | def __init__(self, config, *inputs, **kwargs):
    method dummy_inputs (line 1113) | def dummy_inputs(self):
    method call (line 1140) | def call(

FILE: transformers/models/flaubert/tokenization_flaubert.py
  function convert_to_unicode (line 77) | def convert_to_unicode(text):
  function get_pairs (line 94) | def get_pairs(word):
  function replace_unicode_punct (line 108) | def replace_unicode_punct(text):
  function remove_non_printing_char (line 152) | def remove_non_printing_char(text):
  class FlaubertTokenizer (line 165) | class FlaubertTokenizer(PreTrainedTokenizer):
    method __init__ (line 223) | def __init__(
    method do_lower_case (line 308) | def do_lower_case(self):
    method moses_punct_norm (line 312) | def moses_punct_norm(self, text, lang):
    method moses_tokenize (line 321) | def moses_tokenize(self, text, lang):
    method moses_pipeline (line 330) | def moses_pipeline(self, text, lang):
    method ja_tokenize (line 337) | def ja_tokenize(self, text):
    method vocab_size (line 360) | def vocab_size(self):
    method get_vocab (line 364) | def get_vocab(self):
    method bpe (line 368) | def bpe(self, token):
    method preprocess_text (line 412) | def preprocess_text(self, text):
    method _tokenize (line 422) | def _tokenize(self, text, bypass_tokenizer=False):
    method _convert_token_to_id (line 460) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 465) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 470) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 476) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 504) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 533) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 563) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method __getstate__ (line 592) | def __getstate__(self):
    method __setstate__ (line 598) | def __setstate__(self, d):

FILE: transformers/models/flava/configuration_flava.py
  class FlavaImageConfig (line 32) | class FlavaImageConfig(PretrainedConfig):
    method __init__ (line 95) | def __init__(
    method from_pretrained (line 133) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class FlavaTextConfig (line 149) | class FlavaTextConfig(PretrainedConfig):
    method __init__ (line 222) | def __init__(
    method from_pretrained (line 260) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class FlavaMultimodalConfig (line 276) | class FlavaMultimodalConfig(PretrainedConfig):
    method __init__ (line 331) | def __init__(
    method from_pretrained (line 361) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class FlavaImageCodebookConfig (line 377) | class FlavaImageCodebookConfig(PretrainedConfig):
    method __init__ (line 423) | def __init__(
    method from_pretrained (line 444) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class FlavaConfig (line 460) | class FlavaConfig(PretrainedConfig):
    method __init__ (line 533) | def __init__(
    method from_configs (line 736) | def from_configs(
    method to_dict (line 760) | def to_dict(self):

FILE: transformers/models/flava/convert_dalle_to_flava_codebook.py
  function rreplace (line 24) | def rreplace(s, old, new, occurrence):
  function count_parameters (line 29) | def count_parameters(state_dict):
  function upgrade_state_dict (line 34) | def upgrade_state_dict(state_dict):
  function convert_dalle_checkpoint (line 57) | def convert_dalle_checkpoint(checkpoint_path, pytorch_dump_folder_path, ...

FILE: transformers/models/flava/convert_flava_original_pytorch_to_hf.py
  function count_parameters (line 25) | def count_parameters(state_dict):
  function upgrade_state_dict (line 30) | def upgrade_state_dict(state_dict, codebook_state_dict):
  function convert_flava_checkpoint (line 62) | def convert_flava_checkpoint(checkpoint_path, codebook_path, pytorch_dum...

FILE: transformers/models/flava/feature_extraction_flava.py
  class FlavaFeatureExtractor (line 26) | class FlavaFeatureExtractor(FlavaImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/flava/image_processing_flava.py
  class FlavaMaskingGenerator (line 55) | class FlavaMaskingGenerator:
    method __init__ (line 56) | def __init__(
    method __repr__ (line 78) | def __repr__(self):
    method get_shape (line 90) | def get_shape(self):
    method _mask (line 93) | def _mask(self, mask, max_mask_patches):
    method __call__ (line 117) | def __call__(self):
  class FlavaImageProcessor (line 133) | class FlavaImageProcessor(BaseImageProcessor):
    method __init__ (line 223) | def __init__(
    method from_dict (line 303) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method masking_generator (line 316) | def masking_generator(
    method resize (line 334) | def resize(
    method center_crop (line 362) | def center_crop(
    method rescale (line 386) | def rescale(
    method normalize (line 406) | def normalize(
    method map_pixels (line 429) | def map_pixels(self, image: np.ndarray) -> np.ndarray:
    method _preprocess_image (line 432) | def _preprocess_image(
    method preprocess (line 480) | def preprocess(

FILE: transformers/models/flava/modeling_flava.py
  class FlavaModelOutput (line 70) | class FlavaModelOutput(ModelOutput):
    method to_tuple (line 100) | def to_tuple(self) -> Tuple[Any]:
  class FlavaLosses (line 108) | class FlavaLosses(ModelOutput):
    method all_none (line 135) | def all_none(self) -> bool:
  class FlavaForPreTrainingOutput (line 145) | class FlavaForPreTrainingOutput(ModelOutput):
    method to_tuple (line 232) | def to_tuple(self) -> Tuple[Any]:
  class FlavaImageEmbeddings (line 246) | class FlavaImageEmbeddings(nn.Module):
    method __init__ (line 251) | def __init__(self, config: FlavaImageConfig, use_mask_token: bool = Fa...
    method interpolate_pos_encoding (line 268) | def interpolate_pos_encoding(self, embeddings: torch.Tensor, height: i...
    method forward (line 303) | def forward(
  class PatchEmbeddings (line 339) | class PatchEmbeddings(nn.Module):
    method __init__ (line 344) | def __init__(
    method forward (line 363) | def forward(self, pixel_values: torch.Tensor, interpolate_pos_encoding...
  class FlavaTextEmbeddings (line 375) | class FlavaTextEmbeddings(nn.Module):
    method __init__ (line 378) | def __init__(self, config):
    method forward (line 395) | def forward(
  class FlavaSelfAttention (line 430) | class FlavaSelfAttention(nn.Module):
    method __init__ (line 431) | def __init__(self, config: FlavaPossibleConfigs) -> None:
    method transpose_for_scores (line 449) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 454) | def forward(
  class FlavaSelfOutput (line 499) | class FlavaSelfOutput(nn.Module):
    method __init__ (line 505) | def __init__(self, config: FlavaPossibleConfigs) -> None:
    method forward (line 510) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class FlavaAttention (line 517) | class FlavaAttention(nn.Module):
    method __init__ (line 518) | def __init__(self, config: FlavaPossibleConfigs) -> None:
    method prune_heads (line 524) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 542) | def forward(
  class FlavaIntermediate (line 559) | class FlavaIntermediate(nn.Module):
    method __init__ (line 560) | def __init__(self, config: FlavaPossibleConfigs) -> None:
    method forward (line 569) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class FlavaOutput (line 576) | class FlavaOutput(nn.Module):
    method __init__ (line 577) | def __init__(self, config: FlavaPossibleConfigs) -> None:
    method forward (line 583) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class FlavaLayer (line 592) | class FlavaLayer(nn.Module):
    method __init__ (line 595) | def __init__(self, config: FlavaPossibleConfigs) -> None:
    method forward (line 607) | def forward(
  class FlavaEncoder (line 638) | class FlavaEncoder(nn.Module):
    method __init__ (line 639) | def __init__(self, config: FlavaConfig) -> None:
    method forward (line 645) | def forward(
  class FlavaPooler (line 695) | class FlavaPooler(nn.Module):
    method __init__ (line 696) | def __init__(self, config: FlavaPossibleConfigs):
    method forward (line 701) | def forward(self, hidden_states: torch.Tensor):
  class FlavaPreTrainedModel (line 854) | class FlavaPreTrainedModel(PreTrainedModel):
    method _init_weights (line 864) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 880) | def _set_gradient_checkpointing(self, module: FlavaEncoder, value: boo...
  class FlavaImageModel (line 889) | class FlavaImageModel(FlavaPreTrainedModel):
    method __init__ (line 895) | def __init__(self, config: FlavaImageConfig, add_pooling_layer: bool =...
    method get_input_embeddings (line 908) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 911) | def set_input_embeddings(self, value: nn.Module):
    method _prune_heads (line 914) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
    method forward (line 930) | def forward(
  class FlavaTextModel (line 988) | class FlavaTextModel(FlavaPreTrainedModel):
    method __init__ (line 993) | def __init__(self, config: FlavaTextConfig, add_pooling_layer: bool = ...
    method get_input_embeddings (line 1005) | def get_input_embeddings(self) -> PatchEmbeddings:
    method set_input_embeddings (line 1008) | def set_input_embeddings(self, value: nn.Module):
    method _prune_heads (line 1011) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
    method forward (line 1025) | def forward(
  class FlavaMultimodalModel (line 1093) | class FlavaMultimodalModel(FlavaPreTrainedModel):
    method __init__ (line 1099) | def __init__(self, config: FlavaMultimodalConfig, add_pooling_layer=Tr...
    method _prune_heads (line 1113) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
    method forward (line 1129) | def forward(
  class FlavaModel (line 1191) | class FlavaModel(FlavaPreTrainedModel):
    method __init__ (line 1194) | def __init__(self, config: FlavaConfig):
    method get_text_features (line 1238) | def get_text_features(
    method get_image_features (line 1284) | def get_image_features(
    method forward (line 1339) | def forward(
  class FlavaImageCodebookResPath (line 1443) | class FlavaImageCodebookResPath(nn.Module):
    method __init__ (line 1444) | def __init__(self, in_size: int, out_size: int, **kwargs):
    method forward (line 1460) | def forward(self, x: torch.Tensor) -> torch.Tensor:
  class FlavaImageCodebookBlock (line 1464) | class FlavaImageCodebookBlock(nn.Module):
    method __init__ (line 1465) | def __init__(self, in_size: int, out_size: int, num_layers: int, **kwa...
    method forward (line 1477) | def forward(self, x: torch.Tensor) -> torch.Tensor:
  class FlavaImageCodebookLayerGroup (line 1481) | class FlavaImageCodebookLayerGroup(nn.Module):
    method __init__ (line 1482) | def __init__(self, num_blocks: int, num_layers: int, in_size: int, out...
    method forward (line 1496) | def forward(self, x: torch.Tensor) -> torch.Tensor:
  class FlavaImageCodebook (line 1509) | class FlavaImageCodebook(FlavaPreTrainedModel):
    method __init__ (line 1515) | def __init__(
    method get_codebook_indices (line 1559) | def get_codebook_indices(self, pixel_values: torch.Tensor) -> torch.Te...
    method get_codebook_probs (line 1589) | def get_codebook_probs(self, pixel_values: torch.Tensor) -> torch.Tensor:
    method forward (line 1593) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class FlavaPredictionHeadTransform (line 1630) | class FlavaPredictionHeadTransform(nn.Module):
    method __init__ (line 1631) | def __init__(self, config):
    method forward (line 1640) | def forward(self, hidden_states):
  class FlavaMaskedPredictionHead (line 1647) | class FlavaMaskedPredictionHead(nn.Module):
    method __init__ (line 1648) | def __init__(self, config, weight=None):
    method forward (line 1660) | def forward(self, x):
  class FlavaITMHead (line 1666) | class FlavaITMHead(nn.Module):
    method __init__ (line 1667) | def __init__(self, config):
    method forward (line 1673) | def forward(self, x):
  class FlavaGlobalContrastiveHead (line 1679) | class FlavaGlobalContrastiveHead(nn.Module):
    method __init__ (line 1680) | def __init__(self, config):
    method forward (line 1685) | def forward(self, image_embeddings, text_embeddings, logit_scale):
  class FlavaForPreTraining (line 1725) | class FlavaForPreTraining(FlavaPreTrainedModel):
    method __init__ (line 1734) | def __init__(self, config: FlavaConfig, image_codebook: Optional[nn.Mo...
    method _resize_to_2d (line 1764) | def _resize_to_2d(self, x: torch.Tensor):
    method forward (line 1773) | def forward(

FILE: transformers/models/flava/processing_flava.py
  class FlavaProcessor (line 28) | class FlavaProcessor(ProcessorMixin):
    method __init__ (line 43) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 61) | def __call__(
    method batch_decode (line 129) | def batch_decode(self, *args, **kwargs):
    method decode (line 136) | def decode(self, *args, **kwargs):
    method model_input_names (line 144) | def model_input_names(self):
    method feature_extractor_class (line 150) | def feature_extractor_class(self):
    method feature_extractor (line 158) | def feature_extractor(self):

FILE: transformers/models/fnet/configuration_fnet.py
  class FNetConfig (line 30) | class FNetConfig(PretrainedConfig):
    method __init__ (line 89) | def __init__(

FILE: transformers/models/fnet/convert_fnet_original_flax_checkpoint_to_pytorch.py
  function convert_flax_checkpoint_to_pytorch (line 30) | def convert_flax_checkpoint_to_pytorch(flax_checkpoint_path, fnet_config...

FILE: transformers/models/fnet/modeling_fnet.py
  function _two_dim_matmul (line 70) | def _two_dim_matmul(x, matrix_dim_one, matrix_dim_two):
  function two_dim_matmul (line 79) | def two_dim_matmul(x, matrix_dim_one, matrix_dim_two):
  function fftn (line 84) | def fftn(x):
  class FNetEmbeddings (line 100) | class FNetEmbeddings(nn.Module):
    method __init__ (line 103) | def __init__(self, config):
    method forward (line 123) | def forward(self, input_ids=None, token_type_ids=None, position_ids=No...
  class FNetBasicFourierTransform (line 159) | class FNetBasicFourierTransform(nn.Module):
    method __init__ (line 160) | def __init__(self, config):
    method _init_fourier_transform (line 164) | def _init_fourier_transform(self, config):
    method forward (line 187) | def forward(self, hidden_states):
  class FNetBasicOutput (line 197) | class FNetBasicOutput(nn.Module):
    method __init__ (line 198) | def __init__(self, config):
    method forward (line 202) | def forward(self, hidden_states, input_tensor):
  class FNetFourierTransform (line 207) | class FNetFourierTransform(nn.Module):
    method __init__ (line 208) | def __init__(self, config):
    method forward (line 213) | def forward(self, hidden_states):
  class FNetIntermediate (line 221) | class FNetIntermediate(nn.Module):
    method __init__ (line 222) | def __init__(self, config):
    method forward (line 230) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class FNetOutput (line 237) | class FNetOutput(nn.Module):
    method __init__ (line 238) | def __init__(self, config):
    method forward (line 244) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class FNetLayer (line 251) | class FNetLayer(nn.Module):
    method __init__ (line 252) | def __init__(self, config):
    method forward (line 260) | def forward(self, hidden_states):
    method feed_forward_chunk (line 272) | def feed_forward_chunk(self, fourier_output):
  class FNetEncoder (line 278) | class FNetEncoder(nn.Module):
    method __init__ (line 279) | def __init__(self, config):
    method forward (line 285) | def forward(self, hidden_states, output_hidden_states=False, return_di...
  class FNetPooler (line 316) | class FNetPooler(nn.Module):
    method __init__ (line 317) | def __init__(self, config):
    method forward (line 322) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class FNetPredictionHeadTransform (line 332) | class FNetPredictionHeadTransform(nn.Module):
    method __init__ (line 333) | def __init__(self, config):
    method forward (line 342) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class FNetLMPredictionHead (line 349) | class FNetLMPredictionHead(nn.Module):
    method __init__ (line 350) | def __init__(self, config):
    method forward (line 361) | def forward(self, hidden_states):
    method _tie_weights (line 366) | def _tie_weights(self):
  class FNetOnlyMLMHead (line 371) | class FNetOnlyMLMHead(nn.Module):
    method __init__ (line 372) | def __init__(self, config):
    method forward (line 376) | def forward(self, sequence_output):
  class FNetOnlyNSPHead (line 382) | class FNetOnlyNSPHead(nn.Module):
    method __init__ (line 383) | def __init__(self, config):
    method forward (line 387) | def forward(self, pooled_output):
  class FNetPreTrainingHeads (line 393) | class FNetPreTrainingHeads(nn.Module):
    method __init__ (line 394) | def __init__(self, config):
    method forward (line 399) | def forward(self, sequence_output, pooled_output):
  class FNetPreTrainedModel (line 405) | class FNetPreTrainedModel(PreTrainedModel):
    method _init_weights (line 416) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 433) | def _set_gradient_checkpointing(self, module, value=False):
  class FNetForPreTrainingOutput (line 439) | class FNetForPreTrainingOutput(ModelOutput):
  class FNetModel (line 514) | class FNetModel(FNetPreTrainedModel):
    method __init__ (line 522) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 534) | def get_input_embeddings(self):
    method set_input_embeddings (line 537) | def set_input_embeddings(self, value):
    method forward (line 546) | def forward(
  class FNetForPreTraining (line 623) | class FNetForPreTraining(FNetPreTrainedModel):
    method __init__ (line 626) | def __init__(self, config):
    method get_output_embeddings (line 635) | def get_output_embeddings(self):
    method set_output_embeddings (line 638) | def set_output_embeddings(self, new_embeddings):
    method forward (line 643) | def forward(
  class FNetForMaskedLM (line 717) | class FNetForMaskedLM(FNetPreTrainedModel):
    method __init__ (line 720) | def __init__(self, config):
    method get_output_embeddings (line 729) | def get_output_embeddings(self):
    method set_output_embeddings (line 732) | def set_output_embeddings(self, new_embeddings):
    method forward (line 741) | def forward(
  class FNetForNextSentencePrediction (line 787) | class FNetForNextSentencePrediction(FNetPreTrainedModel):
    method __init__ (line 788) | def __init__(self, config):
    method forward (line 799) | def forward(
  class FNetForSequenceClassification (line 882) | class FNetForSequenceClassification(FNetPreTrainedModel):
    method __init__ (line 883) | def __init__(self, config):
    method forward (line 900) | def forward(
  class FNetForMultipleChoice (line 967) | class FNetForMultipleChoice(FNetPreTrainedModel):
    method __init__ (line 968) | def __init__(self, config):
    method forward (line 984) | def forward(
  class FNetForTokenClassification (line 1046) | class FNetForTokenClassification(FNetPreTrainedModel):
    method __init__ (line 1047) | def __init__(self, config):
    method forward (line 1065) | def forward(
  class FNetForQuestionAnswering (line 1115) | class FNetForQuestionAnswering(FNetPreTrainedModel):
    method __init__ (line 1116) | def __init__(self, config):
    method forward (line 1133) | def forward(

FILE: transformers/models/fnet/tokenization_fnet.py
  class FNetTokenizer (line 47) | class FNetTokenizer(PreTrainedTokenizer):
    method __init__ (line 104) | def __init__(
    method vocab_size (line 150) | def vocab_size(self):
    method get_vocab (line 153) | def get_vocab(self):
    method __getstate__ (line 158) | def __getstate__(self):
    method __setstate__ (line 163) | def __setstate__(self, d):
    method preprocess_text (line 173) | def preprocess_text(self, inputs):
    method _tokenize (line 188) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 208) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 212) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 216) | def convert_tokens_to_string(self, tokens):
    method _decode (line 235) | def _decode(
    method build_inputs_with_special_tokens (line 283) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 308) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 336) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 365) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/fnet/tokenization_fnet_fast.py
  class FNetTokenizerFast (line 54) | class FNetTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 95) | def __init__(
    method build_inputs_with_special_tokens (line 137) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 162) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 192) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/focalnet/configuration_focalnet.py
  class FocalNetConfig (line 29) | class FocalNetConfig(BackboneConfigMixin, PretrainedConfig):
    method __init__ (line 109) | def __init__(

FILE: transformers/models/focalnet/convert_focalnet_to_hf_format.py
  function get_focalnet_config (line 30) | def get_focalnet_config(model_name):
  function rename_key (line 90) | def rename_key(name):
  function convert_focalnet_checkpoint (line 123) | def convert_focalnet_checkpoint(model_name, pytorch_dump_folder_path, pu...

FILE: transformers/models/focalnet/modeling_focalnet.py
  class FocalNetEncoderOutput (line 64) | class FocalNetEncoderOutput(ModelOutput):
  class FocalNetModelOutput (line 91) | class FocalNetModelOutput(ModelOutput):
  class FocalNetMaskedImageModelingOutput (line 120) | class FocalNetMaskedImageModelingOutput(ModelOutput):
  class FocalNetImageClassifierOutput (line 149) | class FocalNetImageClassifierOutput(ModelOutput):
  class FocalNetEmbeddings (line 177) | class FocalNetEmbeddings(nn.Module):
    method __init__ (line 182) | def __init__(self, config, use_mask_token=False):
    method forward (line 200) | def forward(
  class FocalNetPatchEmbeddings (line 217) | class FocalNetPatchEmbeddings(nn.Module):
    method __init__ (line 218) | def __init__(
    method maybe_pad (line 260) | def maybe_pad(self, pixel_values, height, width):
    method forward (line 269) | def forward(self, pixel_values: Optional[torch.FloatTensor]) -> Tuple[...
  function drop_path (line 289) | def drop_path(input, drop_prob=0.0, training=False, scale_by_keep=True):
  class FocalNetDropPath (line 310) | class FocalNetDropPath(nn.Module):
    method __init__ (line 313) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 317) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 320) | def extra_repr(self) -> str:
  class FocalNetModulation (line 324) | class FocalNetModulation(nn.Module):
    method __init__ (line 325) | def __init__(self, config, index, dim, focal_factor=2, bias=True, proj...
    method forward (line 358) | def forward(self, hidden_state):
  class FocalNetMlp (line 395) | class FocalNetMlp(nn.Module):
    method __init__ (line 396) | def __init__(self, config, in_features, hidden_features=None, out_feat...
    method forward (line 405) | def forward(self, hidden_state):
  class FocalNetLayer (line 414) | class FocalNetLayer(nn.Module):
    method __init__ (line 430) | def __init__(self, config, index, dim, input_resolution, drop_path=0.0):
    method forward (line 462) | def forward(self, hidden_state, input_dimensions):
  class FocalNetStage (line 483) | class FocalNetStage(nn.Module):
    method __init__ (line 484) | def __init__(self, config, index, input_resolution):
    method forward (line 528) | def forward(self, hidden_states: torch.Tensor, input_dimensions: Tuple...
  class FocalNetEncoder (line 549) | class FocalNetEncoder(nn.Module):
    method __init__ (line 550) | def __init__(self, config, grid_size):
    method forward (line 568) | def forward(
  class FocalNetPreTrainedModel (line 639) | class FocalNetPreTrainedModel(PreTrainedModel):
    method _init_weights (line 650) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 662) | def _set_gradient_checkpointing(self, module, value=False):
  class FocalNetModel (line 696) | class FocalNetModel(FocalNetPreTrainedModel):
    method __init__ (line 697) | def __init__(self, config, add_pooling_layer=True, use_mask_token=False):
    method get_input_embeddings (line 712) | def get_input_embeddings(self):
    method forward (line 723) | def forward(
  class FocalNetForMaskedImageModeling (line 786) | class FocalNetForMaskedImageModeling(FocalNetPreTrainedModel):
    method __init__ (line 787) | def __init__(self, config):
    method forward (line 806) | def forward(
  class FocalNetForImageClassification (line 894) | class FocalNetForImageClassification(FocalNetPreTrainedModel):
    method __init__ (line 896) | def __init__(self, config):
    method forward (line 917) | def forward(
  class FocalNetBackbone (line 983) | class FocalNetBackbone(FocalNetPreTrainedModel, BackboneMixin):
    method __init__ (line 984) | def __init__(self, config: FocalNetConfig):
    method forward (line 996) | def forward(

FILE: transformers/models/fsmt/configuration_fsmt.py
  class DecoderConfig (line 29) | class DecoderConfig(PretrainedConfig):
    method __init__ (line 35) | def __init__(self, vocab_size=0, bos_token_id=0):
  class FSMTConfig (line 41) | class FSMTConfig(PretrainedConfig):
    method __init__ (line 141) | def __init__(
    method to_dict (line 220) | def to_dict(self):

FILE: transformers/models/fsmt/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py
  function rewrite_dict_keys (line 78) | def rewrite_dict_keys(d):
  function convert_fsmt_checkpoint_to_pytorch (line 90) | def convert_fsmt_checkpoint_to_pytorch(fsmt_checkpoint_path, pytorch_dum...

FILE: transformers/models/fsmt/modeling_fsmt.py
  function invert_mask (line 300) | def invert_mask(attention_mask):
  function triu_onnx (line 306) | def triu_onnx(x, diagonal=0):
  function _prepare_fsmt_decoder_inputs (line 317) | def _prepare_fsmt_decoder_inputs(
  class PretrainedFSMTModel (line 343) | class PretrainedFSMTModel(PreTrainedModel):
    method _init_weights (line 347) | def _init_weights(self, module):
    method dummy_inputs (line 361) | def dummy_inputs(self):
  function _make_linear_from_emb (line 371) | def _make_linear_from_emb(emb):
  function _check_shapes (line 379) | def _check_shapes(shape_1, shape2):
  function shift_tokens_right (line 384) | def shift_tokens_right(input_ids, pad_token_id):
  function make_padding_mask (line 397) | def make_padding_mask(input_ids, padding_idx=1):
  class EncoderLayer (line 408) | class EncoderLayer(nn.Module):
    method __init__ (line 409) | def __init__(self, config: FSMTConfig):
    method forward (line 421) | def forward(self, x, encoder_padding_mask, layer_head_mask, output_att...
  class FSMTEncoder (line 457) | class FSMTEncoder(nn.Module):
    method __init__ (line 465) | def __init__(self, config: FSMTConfig, embed_tokens):
    method forward (line 480) | def forward(
  class DecoderLayer (line 578) | class DecoderLayer(nn.Module):
    method __init__ (line 579) | def __init__(self, config: FSMTConfig):
    method forward (line 604) | def forward(
  class FSMTDecoder (line 666) | class FSMTDecoder(nn.Module):
    method __init__ (line 675) | def __init__(self, config: FSMTConfig, embed_tokens: nn.Embedding):
    method forward (line 700) | def forward(
  function _reorder_buffer (line 849) | def _reorder_buffer(attn_cache, new_order):
  class Attention (line 856) | class Attention(nn.Module):
    method __init__ (line 859) | def __init__(
    method _shape (line 882) | def _shape(self, tensor, seq_len, bsz):
    method forward (line 885) | def forward(
    method _use_saved_state (line 990) | def _use_saved_state(self, k, v, saved_state, key_padding_mask, static...
  function fill_with_neg_inf (line 1022) | def fill_with_neg_inf(t):
  function _get_shape (line 1028) | def _get_shape(t):
  class FSMTModel (line 1036) | class FSMTModel(PretrainedFSMTModel):
    method __init__ (line 1039) | def __init__(self, config: FSMTConfig):
    method get_encoder (line 1052) | def get_encoder(self):
    method get_decoder (line 1055) | def get_decoder(self):
    method forward (line 1064) | def forward(
    method get_input_embeddings (line 1156) | def get_input_embeddings(self):
    method set_input_embeddings (line 1159) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1162) | def get_output_embeddings(self):
    method set_output_embeddings (line 1165) | def set_output_embeddings(self, value):
  class FSMTForConditionalGeneration (line 1172) | class FSMTForConditionalGeneration(PretrainedFSMTModel):
    method __init__ (line 1184) | def __init__(self, config: FSMTConfig):
    method forward (line 1195) | def forward(
    method prepare_inputs_for_generation (line 1269) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1293) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 1297) | def _reorder_cache(past_key_values, beam_idx):
    method get_encoder (line 1307) | def get_encoder(self):
    method get_decoder (line 1310) | def get_decoder(self):
    method get_output_embeddings (line 1313) | def get_output_embeddings(self):
    method set_output_embeddings (line 1316) | def set_output_embeddings(self, value):
  class SinusoidalPositionalEmbedding (line 1320) | class SinusoidalPositionalEmbedding(nn.Embedding):
    method __init__ (line 1331) | def __init__(self, num_positions, embedding_dim, padding_idx):
    method make_weight (line 1334) | def make_weight(self, num_positions, embedding_dim, padding_idx):
    method get_embedding (line 1347) | def get_embedding(num_embeddings, embedding_dim, padding_idx):
    method make_positions (line 1367) | def make_positions(tensor, padding_idx: int):
    method forward (line 1380) | def forward(

FILE: transformers/models/fsmt/tokenization_fsmt.py
  function get_pairs (line 57) | def get_pairs(word):
  function replace_unicode_punct (line 70) | def replace_unicode_punct(text):
  function remove_non_printing_char (line 113) | def remove_non_printing_char(text):
  class FSMTTokenizer (line 135) | class FSMTTokenizer(PreTrainedTokenizer):
    method __init__ (line 187) | def __init__(
    method get_vocab (line 255) | def get_vocab(self) -> Dict[str, int]:
    method vocab_size (line 260) | def vocab_size(self) -> int:
    method moses_punct_norm (line 263) | def moses_punct_norm(self, text, lang):
    method moses_tokenize (line 269) | def moses_tokenize(self, text, lang):
    method moses_detokenize (line 277) | def moses_detokenize(self, tokens, lang):
    method moses_pipeline (line 283) | def moses_pipeline(self, text, lang):
    method src_vocab_size (line 290) | def src_vocab_size(self):
    method tgt_vocab_size (line 294) | def tgt_vocab_size(self):
    method get_src_vocab (line 297) | def get_src_vocab(self):
    method get_tgt_vocab (line 300) | def get_tgt_vocab(self):
    method bpe (line 303) | def bpe(self, token):
    method _tokenize (line 347) | def _tokenize(self, text, lang="en", bypass_tokenizer=False):
    method _convert_token_to_id (line 386) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 390) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 394) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 404) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 430) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 458) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 491) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method __getstate__ (line 527) | def __getstate__(self):
    method __setstate__ (line 532) | def __setstate__(self, d):

FILE: transformers/models/funnel/configuration_funnel.py
  class FunnelConfig (line 41) | class FunnelConfig(PretrainedConfig):
    method __init__ (line 105) | def __init__(
    method num_hidden_layers (line 164) | def num_hidden_layers(self):
    method num_hidden_layers (line 168) | def num_hidden_layers(self, value):
    method num_blocks (line 174) | def num_blocks(self):
    method num_blocks (line 178) | def num_blocks(self, value):

FILE: transformers/models/funnel/convert_funnel_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 29) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, config_file, py...

FILE: transformers/models/funnel/modeling_funnel.py
  function load_tf_weights_in_funnel (line 68) | def load_tf_weights_in_funnel(model, config, tf_checkpoint_path):
  class FunnelEmbeddings (line 158) | class FunnelEmbeddings(nn.Module):
    method __init__ (line 159) | def __init__(self, config: FunnelConfig) -> None:
    method forward (line 165) | def forward(
  class FunnelAttentionStructure (line 175) | class FunnelAttentionStructure(nn.Module):
    method __init__ (line 182) | def __init__(self, config: FunnelConfig) -> None:
    method init_attention_inputs (line 191) | def init_attention_inputs(
    method token_type_ids_to_mat (line 211) | def token_type_ids_to_mat(self, token_type_ids: torch.Tensor) -> torch...
    method get_position_embeds (line 219) | def get_position_embeds(
    method stride_pool_pos (line 299) | def stride_pool_pos(self, pos_id: torch.Tensor, block_index: int):
    method relative_pos (line 314) | def relative_pos(self, pos: torch.Tensor, stride: int, pooled_pos=None...
    method stride_pool (line 328) | def stride_pool(
    method pool_tensor (line 361) | def pool_tensor(
    method pre_attention_pooling (line 399) | def pre_attention_pooling(
    method post_attention_pooling (line 421) | def post_attention_pooling(self, attention_inputs: Tuple[torch.Tensor]...
  function _relative_shift_gather (line 435) | def _relative_shift_gather(positional_attn: torch.Tensor, context_len: i...
  class FunnelRelMultiheadAttention (line 451) | class FunnelRelMultiheadAttention(nn.Module):
    method __init__ (line 452) | def __init__(self, config: FunnelConfig, block_index: int) -> None:
    method relative_positional_attention (line 475) | def relative_positional_attention(self, position_embeds, q_head, conte...
    method relative_token_type_attention (line 517) | def relative_token_type_attention(self, token_type_mat, q_head, cls_ma...
    method forward (line 541) | def forward(
  class FunnelPositionwiseFFN (line 595) | class FunnelPositionwiseFFN(nn.Module):
    method __init__ (line 596) | def __init__(self, config: FunnelConfig) -> None:
    method forward (line 605) | def forward(self, hidden: torch.Tensor) -> torch.Tensor:
  class FunnelLayer (line 614) | class FunnelLayer(nn.Module):
    method __init__ (line 615) | def __init__(self, config: FunnelConfig, block_index: int) -> None:
    method forward (line 620) | def forward(
  class FunnelEncoder (line 633) | class FunnelEncoder(nn.Module):
    method __init__ (line 634) | def __init__(self, config: FunnelConfig) -> None:
    method forward (line 645) | def forward(
  function upsample (line 696) | def upsample(
  class FunnelDecoder (line 718) | class FunnelDecoder(nn.Module):
    method __init__ (line 719) | def __init__(self, config: FunnelConfig) -> None:
    method forward (line 725) | def forward(
  class FunnelDiscriminatorPredictions (line 767) | class FunnelDiscriminatorPredictions(nn.Module):
    method __init__ (line 770) | def __init__(self, config: FunnelConfig) -> None:
    method forward (line 776) | def forward(self, discriminator_hidden_states: torch.Tensor) -> torch....
  class FunnelPreTrainedModel (line 783) | class FunnelPreTrainedModel(PreTrainedModel):
    method _init_weights (line 793) | def _init_weights(self, module):
  class FunnelClassificationHead (line 818) | class FunnelClassificationHead(nn.Module):
    method __init__ (line 819) | def __init__(self, config: FunnelConfig, n_labels: int) -> None:
    method forward (line 825) | def forward(self, hidden: torch.Tensor) -> torch.Tensor:
  class FunnelForPreTrainingOutput (line 833) | class FunnelForPreTrainingOutput(ModelOutput):
  class FunnelBaseModel (line 926) | class FunnelBaseModel(FunnelPreTrainedModel):
    method __init__ (line 927) | def __init__(self, config: FunnelConfig) -> None:
    method get_input_embeddings (line 936) | def get_input_embeddings(self) -> nn.Embedding:
    method set_input_embeddings (line 939) | def set_input_embeddings(self, new_embeddings: nn.Embedding) -> None:
    method forward (line 948) | def forward(
  class FunnelModel (line 1002) | class FunnelModel(FunnelPreTrainedModel):
    method __init__ (line 1003) | def __init__(self, config: FunnelConfig) -> None:
    method get_input_embeddings (line 1013) | def get_input_embeddings(self) -> nn.Embedding:
    method set_input_embeddings (line 1016) | def set_input_embeddings(self, new_embeddings: nn.Embedding) -> None:
    method forward (line 1025) | def forward(
  class FunnelForPreTraining (line 1109) | class FunnelForPreTraining(FunnelPreTrainedModel):
    method __init__ (line 1110) | def __init__(self, config: FunnelConfig) -> None:
    method forward (line 1120) | def forward(
  class FunnelForMaskedLM (line 1192) | class FunnelForMaskedLM(FunnelPreTrainedModel):
    method __init__ (line 1195) | def __init__(self, config: FunnelConfig) -> None:
    method get_output_embeddings (line 1204) | def get_output_embeddings(self) -> nn.Linear:
    method set_output_embeddings (line 1207) | def set_output_embeddings(self, new_embeddings: nn.Embedding) -> None:
    method forward (line 1217) | def forward(
  class FunnelForSequenceClassification (line 1273) | class FunnelForSequenceClassification(FunnelPreTrainedModel):
    method __init__ (line 1274) | def __init__(self, config: FunnelConfig) -> None:
    method forward (line 1290) | def forward(
  class FunnelForMultipleChoice (line 1365) | class FunnelForMultipleChoice(FunnelPreTrainedModel):
    method __init__ (line 1366) | def __init__(self, config: FunnelConfig) -> None:
    method forward (line 1380) | def forward(
  class FunnelForTokenClassification (line 1448) | class FunnelForTokenClassification(FunnelPreTrainedModel):
    method __init__ (line 1449) | def __init__(self, config: FunnelConfig) -> None:
    method forward (line 1466) | def forward(
  class FunnelForQuestionAnswering (line 1521) | class FunnelForQuestionAnswering(FunnelPreTrainedModel):
    method __init__ (line 1522) | def __init__(self, config: FunnelConfig) -> None:
    method forward (line 1538) | def forward(

FILE: transformers/models/funnel/modeling_tf_funnel.py
  class TFFunnelEmbeddings (line 80) | class TFFunnelEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 83) | def __init__(self, config, **kwargs):
    method build (line 93) | def build(self, input_shape):
    method call (line 103) | def call(self, input_ids=None, inputs_embeds=None, training=False):
  class TFFunnelAttentionStructure (line 123) | class TFFunnelAttentionStructure:
    method __init__ (line 130) | def __init__(self, config):
    method init_attention_inputs (line 145) | def init_attention_inputs(self, inputs_embeds, attention_mask=None, to...
    method token_type_ids_to_mat (line 160) | def token_type_ids_to_mat(self, token_type_ids):
    method get_position_embeds (line 168) | def get_position_embeds(self, seq_len, training=False):
    method stride_pool_pos (line 251) | def stride_pool_pos(self, pos_id, block_index):
    method relative_pos (line 266) | def relative_pos(self, pos, stride, pooled_pos=None, shift=1):
    method stride_pool (line 280) | def stride_pool(self, tensor, axis):
    method pool_tensor (line 307) | def pool_tensor(self, tensor, mode="mean", stride=2):
    method pre_attention_pooling (line 335) | def pre_attention_pooling(self, output, attention_inputs):
    method post_attention_pooling (line 355) | def post_attention_pooling(self, attention_inputs):
  function _relative_shift_gather (line 369) | def _relative_shift_gather(positional_attn, context_len, shift):
  class TFFunnelRelMultiheadAttention (line 385) | class TFFunnelRelMultiheadAttention(tf.keras.layers.Layer):
    method __init__ (line 386) | def __init__(self, config, block_index, **kwargs):
    method build (line 410) | def build(self, input_shape):
    method relative_positional_attention (line 431) | def relative_positional_attention(self, position_embeds, q_head, conte...
    method relative_token_type_attention (line 477) | def relative_token_type_attention(self, token_type_mat, q_head, cls_ma...
    method call (line 504) | def call(self, query, key, value, attention_inputs, output_attentions=...
  class TFFunnelPositionwiseFFN (line 550) | class TFFunnelPositionwiseFFN(tf.keras.layers.Layer):
    method __init__ (line 551) | def __init__(self, config, **kwargs):
    method call (line 561) | def call(self, hidden, training=False):
  class TFFunnelLayer (line 570) | class TFFunnelLayer(tf.keras.layers.Layer):
    method __init__ (line 571) | def __init__(self, config, block_index, **kwargs):
    method call (line 576) | def call(self, query, key, value, attention_inputs, output_attentions=...
  class TFFunnelEncoder (line 584) | class TFFunnelEncoder(tf.keras.layers.Layer):
    method __init__ (line 585) | def __init__(self, config, **kwargs):
    method call (line 596) | def call(
  function upsample (line 654) | def upsample(x, stride, target_len, separate_cls=True, truncate_seq=False):
  class TFFunnelDecoder (line 674) | class TFFunnelDecoder(tf.keras.layers.Layer):
    method __init__ (line 675) | def __init__(self, config, **kwargs):
    method call (line 683) | def call(
  class TFFunnelBaseLayer (line 730) | class TFFunnelBaseLayer(tf.keras.layers.Layer):
    method __init__ (line 735) | def __init__(self, config, **kwargs):
    method get_input_embeddings (line 746) | def get_input_embeddings(self):
    method set_input_embeddings (line 749) | def set_input_embeddings(self, value):
    method _prune_heads (line 753) | def _prune_heads(self, heads_to_prune):
    method call (line 757) | def call(
  class TFFunnelMainLayer (line 800) | class TFFunnelMainLayer(tf.keras.layers.Layer):
    method __init__ (line 805) | def __init__(self, config, **kwargs):
    method get_input_embeddings (line 818) | def get_input_embeddings(self):
    method set_input_embeddings (line 821) | def set_input_embeddings(self, value):
    method _prune_heads (line 825) | def _prune_heads(self, heads_to_prune):
    method call (line 829) | def call(
  class TFFunnelDiscriminatorPredictions (line 899) | class TFFunnelDiscriminatorPredictions(tf.keras.layers.Layer):
    method __init__ (line 902) | def __init__(self, config, **kwargs):
    method call (line 909) | def call(self, discriminator_hidden_states):
  class TFFunnelMaskedLMHead (line 916) | class TFFunnelMaskedLMHead(tf.keras.layers.Layer):
    method __init__ (line 917) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 923) | def build(self, input_shape):
    method get_output_embeddings (line 928) | def get_output_embeddings(self):
    method set_output_embeddings (line 931) | def set_output_embeddings(self, value):
    method get_bias (line 935) | def get_bias(self):
    method set_bias (line 938) | def set_bias(self, value):
    method call (line 942) | def call(self, hidden_states, training=False):
  class TFFunnelClassificationHead (line 952) | class TFFunnelClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 953) | def __init__(self, config, n_labels, **kwargs):
    method call (line 962) | def call(self, hidden, training=False):
  class TFFunnelPreTrainedModel (line 969) | class TFFunnelPreTrainedModel(TFPreTrainedModel):
    method dummy_inputs (line 979) | def dummy_inputs(self):
  class TFFunnelForPreTrainingOutput (line 985) | class TFFunnelForPreTrainingOutput(ModelOutput):
  class TFFunnelBaseModel (line 1107) | class TFFunnelBaseModel(TFFunnelPreTrainedModel):
    method __init__ (line 1108) | def __init__(self, config: FunnelConfig, *inputs, **kwargs) -> None:
    method call (line 1119) | def call(
    method serving_output (line 1141) | def serving_output(self, output):
  class TFFunnelModel (line 1155) | class TFFunnelModel(TFFunnelPreTrainedModel):
    method __init__ (line 1156) | def __init__(self, config: FunnelConfig, *inputs, **kwargs) -> None:
    method call (line 1167) | def call(
    method serving_output (line 1189) | def serving_output(self, output):
  class TFFunnelForPreTraining (line 1205) | class TFFunnelForPreTraining(TFFunnelPreTrainedModel):
    method __init__ (line 1206) | def __init__(self, config: FunnelConfig, **kwargs) -> None:
    method call (line 1215) | def call(
    method serving_output (line 1264) | def serving_output(self, output):
  class TFFunnelForMaskedLM (line 1273) | class TFFunnelForMaskedLM(TFFunnelPreTrainedModel, TFMaskedLanguageModel...
    method __init__ (line 1274) | def __init__(self, config: FunnelConfig, *inputs, **kwargs) -> None:
    method get_lm_head (line 1280) | def get_lm_head(self) -> TFFunnelMaskedLMHead:
    method get_prefix_bias_name (line 1283) | def get_prefix_bias_name(self) -> str:
    method call (line 1294) | def call(
    method serving_output (line 1338) | def serving_output(self, output: TFMaskedLMOutput) -> TFMaskedLMOutput:
  class TFFunnelForSequenceClassification (line 1351) | class TFFunnelForSequenceClassification(TFFunnelPreTrainedModel, TFSeque...
    method __init__ (line 1352) | def __init__(self, config: FunnelConfig, *inputs, **kwargs) -> None:
    method call (line 1366) | def call(
    method serving_output (line 1411) | def serving_output(self, output: TFSequenceClassifierOutput) -> TFSequ...
  class TFFunnelForMultipleChoice (line 1426) | class TFFunnelForMultipleChoice(TFFunnelPreTrainedModel, TFMultipleChoic...
    method __init__ (line 1427) | def __init__(self, config: FunnelConfig, *inputs, **kwargs) -> None:
    method dummy_inputs (line 1434) | def dummy_inputs(self):
    method call (line 1444) | def call(
    method serving_output (line 1506) | def serving_output(self, output: TFMultipleChoiceModelOutput) -> TFMul...
  class TFFunnelForTokenClassification (line 1521) | class TFFunnelForTokenClassification(TFFunnelPreTrainedModel, TFTokenCla...
    method __init__ (line 1522) | def __init__(self, config: FunnelConfig, *inputs, **kwargs) -> None:
    method call (line 1539) | def call(
    method serving_output (line 1583) | def serving_output(self, output: TFTokenClassifierOutput) -> TFTokenCl...
  class TFFunnelForQuestionAnswering (line 1598) | class TFFunnelForQuestionAnswering(TFFunnelPreTrainedModel, TFQuestionAn...
    method __init__ (line 1599) | def __init__(self, config: FunnelConfig, *inputs, **kwargs) -> None:
    method call (line 1615) | def call(
    method serving_output (line 1673) | def serving_output(self, output: TFQuestionAnsweringModelOutput) -> TF...

FILE: transformers/models/funnel/tokenization_funnel.py
  function load_vocab (line 70) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 82) | def whitespace_tokenize(text):
  class FunnelTokenizer (line 91) | class FunnelTokenizer(PreTrainedTokenizer):
    method __init__ (line 143) | def __init__(
    method do_lower_case (line 195) | def do_lower_case(self):
    method vocab_size (line 200) | def vocab_size(self):
    method get_vocab (line 204) | def get_vocab(self):
    method _tokenize (line 208) | def _tokenize(self, text):
    method _convert_token_to_id (line 222) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 227) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 232) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 238) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 264) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 292) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 322) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 344) | class BasicTokenizer(object):
    method __init__ (line 364) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 372) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 409) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 420) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 442) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 455) | def _is_chinese_char(self, cp):
    method _clean_text (line 479) | def _clean_text(self, text):
  class WordpieceTokenizer (line 494) | class WordpieceTokenizer(object):
    method __init__ (line 497) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 502) | def tokenize(self, text):

FILE: transformers/models/funnel/tokenization_funnel_fast.py
  class FunnelTokenizerFast (line 94) | class FunnelTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 145) | def __init__(
    method build_inputs_with_special_tokens (line 196) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 220) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 250) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/git/configuration_git.py
  class GitVisionConfig (line 31) | class GitVisionConfig(PretrainedConfig):
    method __init__ (line 81) | def __init__(
    method from_pretrained (line 111) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class GitConfig (line 127) | class GitConfig(PretrainedConfig):
    method __init__ (line 192) | def __init__(
    method to_dict (line 241) | def to_dict(self):

FILE: transformers/models/git/convert_git_to_pytorch.py
  function get_git_config (line 46) | def get_git_config(model_name):
  function create_rename_keys (line 71) | def create_rename_keys(config, prefix=""):
  function rename_key (line 155) | def rename_key(dct, old, new):
  function read_in_q_k_v (line 161) | def read_in_q_k_v(state_dict, config, prefix=""):
  function prepare_img (line 185) | def prepare_img(model_name):
  function prepare_video (line 196) | def prepare_video():
  function convert_git_checkpoint (line 223) | def convert_git_checkpoint(model_name, pytorch_dump_folder_path, push_to...

FILE: transformers/models/git/modeling_git.py
  class GitVisionModelOutput (line 55) | class GitVisionModelOutput(ModelOutput):
  function _expand_mask (line 84) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class GitEmbeddings (line 98) | class GitEmbeddings(nn.Module):
    method __init__ (line 101) | def __init__(self, config):
    method forward (line 114) | def forward(
  class GitSelfAttention (line 144) | class GitSelfAttention(nn.Module):
    method __init__ (line 145) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 172) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 177) | def forward(
  class GitSelfOutput (line 270) | class GitSelfOutput(nn.Module):
    method __init__ (line 271) | def __init__(self, config):
    method forward (line 277) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class GitAttention (line 284) | class GitAttention(nn.Module):
    method __init__ (line 286) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 293) | def prune_heads(self, heads):
    method forward (line 311) | def forward(
  class GitIntermediate (line 334) | class GitIntermediate(nn.Module):
    method __init__ (line 335) | def __init__(self, config):
    method forward (line 343) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class GitOutput (line 350) | class GitOutput(nn.Module):
    method __init__ (line 351) | def __init__(self, config):
    method forward (line 357) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class GitLayer (line 364) | class GitLayer(nn.Module):
    method __init__ (line 365) | def __init__(self, config):
    method forward (line 373) | def forward(
    method feed_forward_chunk (line 408) | def feed_forward_chunk(self, attention_output):
  class GitEncoder (line 414) | class GitEncoder(nn.Module):
    method __init__ (line 416) | def __init__(self, config):
    method forward (line 422) | def forward(
  class GitPreTrainedModel (line 504) | class GitPreTrainedModel(PreTrainedModel):
    method _init_weights (line 515) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 535) | def _set_gradient_checkpointing(self, module, value=False):
  class GitVisionEmbeddings (line 605) | class GitVisionEmbeddings(nn.Module):
    method __init__ (line 606) | def __init__(self, config: GitVisionConfig):
    method forward (line 628) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class GitVisionMLP (line 640) | class GitVisionMLP(nn.Module):
    method __init__ (line 641) | def __init__(self, config):
    method forward (line 648) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class GitVisionAttention (line 656) | class GitVisionAttention(nn.Module):
    method __init__ (line 659) | def __init__(self, config):
    method _shape (line 678) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 681) | def forward(
  class GitVisionEncoderLayer (line 761) | class GitVisionEncoderLayer(nn.Module):
    method __init__ (line 762) | def __init__(self, config: GitVisionConfig):
    method forward (line 770) | def forward(
  class GitVisionEncoder (line 812) | class GitVisionEncoder(nn.Module):
    method __init__ (line 821) | def __init__(self, config: GitVisionConfig):
    method forward (line 827) | def forward(
  class GitVisionTransformer (line 931) | class GitVisionTransformer(nn.Module):
    method __init__ (line 933) | def __init__(self, config: GitVisionConfig):
    method forward (line 945) | def forward(
  class GitVisionModel (line 993) | class GitVisionModel(GitPreTrainedModel):
    method __init__ (line 998) | def __init__(self, config: GitVisionConfig):
    method get_input_embeddings (line 1004) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 1009) | def forward(
  class GitProjection (line 1047) | class GitProjection(nn.Module):
    method __init__ (line 1048) | def __init__(self, config: GitConfig):
    method forward (line 1056) | def forward(self, embeddings: torch.Tensor) -> torch.Tensor:
  class GitModel (line 1065) | class GitModel(GitPreTrainedModel):
    method __init__ (line 1066) | def __init__(self, config):
    method get_input_embeddings (line 1085) | def get_input_embeddings(self):
    method set_input_embeddings (line 1088) | def set_input_embeddings(self, value):
    method _prune_heads (line 1091) | def _prune_heads(self, heads_to_prune):
    method _generate_future_mask (line 1099) | def _generate_future_mask(self, size: int, dtype: torch.dtype, device:...
    method create_attention_mask (line 1105) | def create_attention_mask(self, tgt, memory, tgt_mask, past_key_values...
    method forward (line 1157) | def forward(
  class GitForCausalLM (line 1326) | class GitForCausalLM(GitPreTrainedModel):
    method __init__ (line 1327) | def __init__(self, config):
    method get_output_embeddings (line 1336) | def get_output_embeddings(self):
    method set_output_embeddings (line 1339) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1344) | def forward(
    method prepare_inputs_for_generation (line 1534) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1554) | def _reorder_cache(self, past_key_values, beam_idx):

FILE: transformers/models/git/processing_git.py
  class GitProcessor (line 23) | class GitProcessor(ProcessorMixin):
    method __init__ (line 40) | def __init__(self, image_processor, tokenizer):
    method __call__ (line 44) | def __call__(self, text=None, images=None, return_tensors=None, **kwar...
    method batch_decode (line 97) | def batch_decode(self, *args, **kwargs):
    method decode (line 104) | def decode(self, *args, **kwargs):
    method model_input_names (line 112) | def model_input_names(self):

FILE: transformers/models/glpn/configuration_glpn.py
  class GLPNConfig (line 29) | class GLPNConfig(PretrainedConfig):
    method __init__ (line 95) | def __init__(

FILE: transformers/models/glpn/convert_glpn_to_pytorch.py
  function rename_keys (line 34) | def rename_keys(state_dict):
  function read_in_k_v (line 99) | def read_in_k_v(state_dict, config):
  function prepare_img (line 118) | def prepare_img():
  function convert_glpn_checkpoint (line 126) | def convert_glpn_checkpoint(checkpoint_path, pytorch_dump_folder_path, p...

FILE: transformers/models/glpn/feature_extraction_glpn.py
  class GLPNFeatureExtractor (line 26) | class GLPNFeatureExtractor(GLPNImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/glpn/image_processing_glpn.py
  class GLPNImageProcessor (line 38) | class GLPNImageProcessor(BaseImageProcessor):
    method __init__ (line 58) | def __init__(
    method resize (line 72) | def resize(
    method rescale (line 104) | def rescale(
    method preprocess (line 126) | def preprocess(

FILE: transformers/models/glpn/modeling_glpn.py
  function drop_path (line 56) | def drop_path(input, drop_prob: float = 0.0, training: bool = False):
  class GLPNDropPath (line 77) | class GLPNDropPath(nn.Module):
    method __init__ (line 80) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 84) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 87) | def extra_repr(self) -> str:
  class GLPNOverlapPatchEmbeddings (line 92) | class GLPNOverlapPatchEmbeddings(nn.Module):
    method __init__ (line 95) | def __init__(self, patch_size, stride, num_channels, hidden_size):
    method forward (line 107) | def forward(self, pixel_values):
  class GLPNEfficientSelfAttention (line 118) | class GLPNEfficientSelfAttention(nn.Module):
    method __init__ (line 122) | def __init__(self, config, hidden_size, num_attention_heads, sequence_...
    method transpose_for_scores (line 149) | def transpose_for_scores(self, hidden_states):
    method forward (line 154) | def forward(
  class GLPNSelfOutput (line 200) | class GLPNSelfOutput(nn.Module):
    method __init__ (line 201) | def __init__(self, config, hidden_size):
    method forward (line 206) | def forward(self, hidden_states, input_tensor):
  class GLPNAttention (line 213) | class GLPNAttention(nn.Module):
    method __init__ (line 214) | def __init__(self, config, hidden_size, num_attention_heads, sequence_...
    method prune_heads (line 225) | def prune_heads(self, heads):
    method forward (line 243) | def forward(self, hidden_states, height, width, output_attentions=False):
  class GLPNDWConv (line 252) | class GLPNDWConv(nn.Module):
    method __init__ (line 253) | def __init__(self, dim=768):
    method forward (line 257) | def forward(self, hidden_states, height, width):
  class GLPNMixFFN (line 267) | class GLPNMixFFN(nn.Module):
    method __init__ (line 268) | def __init__(self, config, in_features, hidden_features=None, out_feat...
    method forward (line 280) | def forward(self, hidden_states, height, width):
  class GLPNLayer (line 291) | class GLPNLayer(nn.Module):
    method __init__ (line 294) | def __init__(self, config, hidden_size, num_attention_heads, drop_path...
    method forward (line 308) | def forward(self, hidden_states, height, width, output_attentions=False):
  class GLPNEncoder (line 334) | class GLPNEncoder(nn.Module):
    method __init__ (line 335) | def __init__(self, config):
    method forward (line 383) | def forward(
  class GLPNPreTrainedModel (line 422) | class GLPNPreTrainedModel(PreTrainedModel):
    method _init_weights (line 433) | def _init_weights(self, module):
  class GLPNModel (line 483) | class GLPNModel(GLPNPreTrainedModel):
    method __init__ (line 485) | def __init__(self, config):
    method _prune_heads (line 495) | def _prune_heads(self, heads_to_prune):
    method forward (line 512) | def forward(
  class GLPNSelectiveFeatureFusion (line 543) | class GLPNSelectiveFeatureFusion(nn.Module):
    method __init__ (line 549) | def __init__(self, in_channel=64):
    method forward (line 570) | def forward(self, local_features, global_features):
  class GLPNDecoderStage (line 587) | class GLPNDecoderStage(nn.Module):
    method __init__ (line 588) | def __init__(self, in_channels, out_channels):
    method forward (line 595) | def forward(self, hidden_state, residual=None):
  class GLPNDecoder (line 607) | class GLPNDecoder(nn.Module):
    method __init__ (line 608) | def __init__(self, config):
    method forward (line 622) | def forward(self, hidden_states: List[torch.Tensor]) -> List[torch.Ten...
  class SiLogLoss (line 634) | class SiLogLoss(nn.Module):
    method __init__ (line 643) | def __init__(self, lambd=0.5):
    method forward (line 647) | def forward(self, pred, target):
  class GLPNDepthEstimationHead (line 655) | class GLPNDepthEstimationHead(nn.Module):
    method __init__ (line 656) | def __init__(self, config):
    method forward (line 668) | def forward(self, hidden_states: List[torch.Tensor]) -> torch.Tensor:
  class GLPNForDepthEstimation (line 684) | class GLPNForDepthEstimation(GLPNPreTrainedModel):
    method __init__ (line 685) | def __init__(self, config):
    method forward (line 697) | def forward(

FILE: transformers/models/gpt2/configuration_gpt2.py
  class GPT2Config (line 37) | class GPT2Config(PretrainedConfig):
    method __init__ (line 140) | def __init__(
  class GPT2OnnxConfig (line 195) | class GPT2OnnxConfig(OnnxConfigWithPast):
    method __init__ (line 196) | def __init__(
    method inputs (line 209) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method num_layers (line 220) | def num_layers(self) -> int:
    method num_attention_heads (line 224) | def num_attention_heads(self) -> int:
    method generate_dummy_inputs (line 227) | def generate_dummy_inputs(
    method default_onnx_opset (line 272) | def default_onnx_opset(self) -> int:

FILE: transformers/models/gpt2/convert_gpt2_original_tf_checkpoint_to_pytorch.py
  function convert_gpt2_checkpoint_to_pytorch (line 29) | def convert_gpt2_checkpoint_to_pytorch(gpt2_checkpoint_path, gpt2_config...

FILE: transformers/models/gpt2/modeling_flax_gpt2.py
  class FlaxConv1D (line 110) | class FlaxConv1D(nn.Module):
    method __call__ (line 117) | def __call__(self, inputs):
  class FlaxGPT2Attention (line 129) | class FlaxGPT2Attention(nn.Module):
    method setup (line 135) | def setup(self):
    method _split_heads (line 155) | def _split_heads(self, hidden_states):
    method _merge_heads (line 158) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 162) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 193) | def __call__(
  class FlaxGPT2MLP (line 282) | class FlaxGPT2MLP(nn.Module):
    method setup (line 287) | def setup(self):
    method __call__ (line 294) | def __call__(self, hidden_states, deterministic: bool = True):
  class FlaxGPT2Block (line 302) | class FlaxGPT2Block(nn.Module):
    method setup (line 306) | def setup(self):
    method __call__ (line 322) | def __call__(
  class FlaxGPT2PreTrainedModel (line 380) | class FlaxGPT2PreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 390) | def __init__(
    method init_weights (line 402) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 437) | def init_cache(self, batch_size, max_length):
    method __call__ (line 457) | def __call__(
  class FlaxGPT2BlockCollection (line 535) | class FlaxGPT2BlockCollection(nn.Module):
    method setup (line 539) | def setup(self):
    method __call__ (line 544) | def __call__(
  class FlaxGPT2Module (line 587) | class FlaxGPT2Module(nn.Module):
    method setup (line 591) | def setup(self):
    method __call__ (line 610) | def __call__(
  class FlaxGPT2Model (line 665) | class FlaxGPT2Model(FlaxGPT2PreTrainedModel):
  class FlaxGPT2LMHeadModule (line 677) | class FlaxGPT2LMHeadModule(nn.Module):
    method setup (line 681) | def setup(self):
    method __call__ (line 690) | def __call__(
  class FlaxGPT2LMHeadModel (line 742) | class FlaxGPT2LMHeadModel(FlaxGPT2PreTrainedModel):
    method prepare_inputs_for_generation (line 745) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 766) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/gpt2/modeling_gpt2.py
  function load_tf_weights_in_gpt2 (line 67) | def load_tf_weights_in_gpt2(model, config, gpt2_checkpoint_path):
  class GPT2Attention (line 124) | class GPT2Attention(nn.Module):
    method __init__ (line 125) | def __init__(self, config, is_cross_attention=False, layer_idx=None):
    method prune_heads (line 168) | def prune_heads(self, heads):
    method _attn (line 183) | def _attn(self, query, key, value, attention_mask=None, head_mask=None):
    method _upcast_and_reordered_attn (line 223) | def _upcast_and_reordered_attn(self, query, key, value, attention_mask...
    method _split_heads (line 275) | def _split_heads(self, tensor, num_heads, attn_head_size):
    method _merge_heads (line 283) | def _merge_heads(self, tensor, num_heads, attn_head_size):
    method forward (line 291) | def forward(
  class GPT2MLP (line 345) | class GPT2MLP(nn.Module):
    method __init__ (line 346) | def __init__(self, intermediate_size, config):
    method forward (line 354) | def forward(self, hidden_states: Optional[Tuple[torch.FloatTensor]]) -...
  class GPT2Block (line 362) | class GPT2Block(nn.Module):
    method __init__ (line 363) | def __init__(self, config, layer_idx=None):
    method forward (line 378) | def forward(
  class GPT2PreTrainedModel (line 440) | class GPT2PreTrainedModel(PreTrainedModel):
    method __init__ (line 454) | def __init__(self, *inputs, **kwargs):
    method _init_weights (line 457) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 484) | def _set_gradient_checkpointing(self, module, value=False):
  class GPT2DoubleHeadsModelOutput (line 490) | class GPT2DoubleHeadsModelOutput(ModelOutput):
  class GPT2Model (line 670) | class GPT2Model(GPT2PreTrainedModel):
    method __init__ (line 674) | def __init__(self, config):
    method parallelize (line 695) | def parallelize(self, device_map=None):
    method deparallelize (line 722) | def deparallelize(self):
    method get_input_embeddings (line 738) | def get_input_embeddings(self):
    method set_input_embeddings (line 741) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 744) | def _prune_heads(self, heads_to_prune):
    method forward (line 757) | def forward(
  class GPT2LMHeadModel (line 959) | class GPT2LMHeadModel(GPT2PreTrainedModel):
    method __init__ (line 963) | def __init__(self, config):
    method parallelize (line 976) | def parallelize(self, device_map=None):
    method deparallelize (line 995) | def deparallelize(self):
    method get_output_embeddings (line 1006) | def get_output_embeddings(self):
    method set_output_embeddings (line 1009) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 1012) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method forward (line 1055) | def forward(
    method _reorder_cache (line 1129) | def _reorder_cache(
  class GPT2DoubleHeadsModel (line 1152) | class GPT2DoubleHeadsModel(GPT2PreTrainedModel):
    method __init__ (line 1156) | def __init__(self, config):
    method parallelize (line 1171) | def parallelize(self, device_map=None):
    method deparallelize (line 1191) | def deparallelize(self):
    method get_output_embeddings (line 1203) | def get_output_embeddings(self):
    method set_output_embeddings (line 1206) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 1209) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method forward (line 1240) | def forward(
    method _reorder_cache (line 1352) | def _reorder_cache(
  class GPT2ForSequenceClassification (line 1381) | class GPT2ForSequenceClassification(GPT2PreTrainedModel):
    method __init__ (line 1385) | def __init__(self, config):
    method forward (line 1404) | def forward(
  class GPT2ForTokenClassification (line 1507) | class GPT2ForTokenClassification(GPT2PreTrainedModel):
    method __init__ (line 1508) | def __init__(self, config):
    method forward (line 1539) | def forward(
  class GPT2ForQuestionAnswering (line 1605) | class GPT2ForQuestionAnswering(GPT2PreTrainedModel):
    method __init__ (line 1609) | def __init__(self, config):
    method forward (line 1630) | def forward(

FILE: transformers/models/gpt2/modeling_tf_gpt2.py
  class TFAttention (line 70) | class TFAttention(tf.keras.layers.Layer):
    method __init__ (line 71) | def __init__(self, nx, config, scale=False, is_cross_attention=False, ...
    method prune_heads (line 95) | def prune_heads(self, heads):
    method causal_attention_mask (line 99) | def causal_attention_mask(nd, ns, dtype):
    method _attn (line 109) | def _attn(self, q, k, v, attention_mask, head_mask, output_attentions,...
    method merge_heads (line 142) | def merge_heads(self, x):
    method split_heads (line 148) | def split_heads(self, x):
    method call (line 154) | def call(
  class TFMLP (line 206) | class TFMLP(tf.keras.layers.Layer):
    method __init__ (line 207) | def __init__(self, n_state, config, **kwargs):
    method call (line 215) | def call(self, x, training=False):
  class TFBlock (line 222) | class TFBlock(tf.keras.layers.Layer):
    method __init__ (line 223) | def __init__(self, config, scale=False, **kwargs):
    method call (line 239) | def call(
  class TFGPT2MainLayer (line 301) | class TFGPT2MainLayer(tf.keras.layers.Layer):
    method __init__ (line 304) | def __init__(self, config, *inputs, **kwargs):
    method get_input_embeddings (line 334) | def get_input_embeddings(self):
    method set_input_embeddings (line 337) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 340) | def _prune_heads(self, heads_to_prune):
    method call (line 347) | def call(
  class TFGPT2PreTrainedModel (line 513) | class TFGPT2PreTrainedModel(TFPreTrainedModel):
  class TFGPT2DoubleHeadsModelOutput (line 526) | class TFGPT2DoubleHeadsModelOutput(ModelOutput):
  class TFGPT2Model (line 675) | class TFGPT2Model(TFGPT2PreTrainedModel):
    method __init__ (line 676) | def __init__(self, config, *inputs, **kwargs):
    method call (line 687) | def call(
  class TFGPT2LMHeadModel (line 752) | class TFGPT2LMHeadModel(TFGPT2PreTrainedModel, TFCausalLanguageModelingL...
    method __init__ (line 753) | def __init__(self, config, *inputs, **kwargs):
    method get_output_embeddings (line 757) | def get_output_embeddings(self):
    method set_output_embeddings (line 760) | def set_output_embeddings(self, value):
    method prepare_inputs_for_generation (line 763) | def prepare_inputs_for_generation(self, inputs, past_key_values=None, ...
    method call (line 795) | def call(
  class TFGPT2DoubleHeadsModel (line 886) | class TFGPT2DoubleHeadsModel(TFGPT2PreTrainedModel):
    method __init__ (line 887) | def __init__(self, config, *inputs, **kwargs):
    method call (line 898) | def call(
    method input_signature (line 998) | def input_signature(self):
  class TFGPT2ForSequenceClassification (line 1021) | class TFGPT2ForSequenceClassification(TFGPT2PreTrainedModel, TFSequenceC...
    method __init__ (line 1022) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1040) | def call(

FILE: transformers/models/gpt2/tokenization_gpt2.py
  function bytes_to_unicode (line 66) | def bytes_to_unicode():
  function get_pairs (line 90) | def get_pairs(word):
  class GPT2Tokenizer (line 104) | class GPT2Tokenizer(PreTrainedTokenizer):
    method __init__ (line 159) | def __init__(
    method vocab_size (line 205) | def vocab_size(self):
    method get_vocab (line 208) | def get_vocab(self):
    method bpe (line 211) | def bpe(self, token):
    method build_inputs_with_special_tokens (line 253) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method get_special_tokens_mask (line 266) | def get_special_tokens_mask(
    method _tokenize (line 298) | def _tokenize(self, text):
    method _convert_token_to_id (line 308) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 312) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 316) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 322) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method prepare_for_tokenization (line 351) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method _build_conversation_input_ids (line 357) | def _build_conversation_input_ids(self, conversation: "Conversation") ...

FILE: transformers/models/gpt2/tokenization_gpt2_fast.py
  class GPT2TokenizerFast (line 70) | class GPT2TokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 129) | def __init__(
    method _batch_encode_plus (line 161) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 170) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method save_vocabulary (line 180) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method _build_conversation_input_ids (line 184) | def _build_conversation_input_ids(self, conversation: "Conversation") ...

FILE: transformers/models/gpt2/tokenization_gpt2_tf.py
  class TFGPT2Tokenizer (line 11) | class TFGPT2Tokenizer(tf.keras.layers.Layer):
    method __init__ (line 27) | def __init__(self, vocab: Dict[str, int], merges: List[str], max_lengt...
    method from_tokenizer (line 36) | def from_tokenizer(cls, tokenizer: GPT2Tokenizer, *args, **kwargs):
    method from_pretrained (line 56) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
    method from_config (line 74) | def from_config(cls, config):
    method get_config (line 82) | def get_config(self):
    method call (line 90) | def call(self, x, max_length: int = None):

FILE: transformers/models/gpt_bigcode/configuration_gpt_bigcode.py
  class GPTBigCodeConfig (line 28) | class GPTBigCodeConfig(PretrainedConfig):
    method __init__ (line 101) | def __init__(

FILE: transformers/models/gpt_bigcode/modeling_gpt_bigcode.py
  function upcast_masked_softmax (line 56) | def upcast_masked_softmax(
  function upcast_softmax (line 67) | def upcast_softmax(x: torch.Tensor, scale: float, softmax_dtype: torch.d...
  function masked_softmax (line 75) | def masked_softmax(x: torch.Tensor, mask: torch.Tensor, mask_value: torc...
  class GPTBigCodeAttention (line 81) | class GPTBigCodeAttention(nn.Module):
    method __init__ (line 82) | def __init__(self, config, is_cross_attention=False, layer_idx=None):
    method _get_mask_value (line 122) | def _get_mask_value(self, device, dtype):
    method _attn (line 128) | def _attn(self, query, key, value, attention_mask=None, head_mask=None):
    method forward (line 205) | def forward(
  class GPTBigCodeMLP (line 265) | class GPTBigCodeMLP(nn.Module):
    method __init__ (line 266) | def __init__(self, intermediate_size, config):
    method forward (line 275) | def forward(self, hidden_states: Optional[Tuple[torch.Tensor]]) -> tor...
  class GPTBigCodeBlock (line 283) | class GPTBigCodeBlock(nn.Module):
    method __init__ (line 284) | def __init__(self, config, layer_idx=None):
    method forward (line 301) | def forward(
  class GPTBigCodePreTrainedModel (line 365) | class GPTBigCodePreTrainedModel(PreTrainedModel):
    method __init__ (line 377) | def __init__(self, *inputs, **kwargs):
    method _init_weights (line 380) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 408) | def _set_gradient_checkpointing(self, module, value=False):
  class GPTBigCodeModel (line 502) | class GPTBigCodeModel(GPTBigCodePreTrainedModel):
    method __init__ (line 505) | def __init__(self, config):
    method get_input_embeddings (line 527) | def get_input_embeddings(self):
    method set_input_embeddings (line 530) | def set_input_embeddings(self, new_embeddings):
    method forward (line 539) | def forward(
  class GPTBigCodeForCausalLM (line 724) | class GPTBigCodeForCausalLM(GPTBigCodePreTrainedModel):
    method __init__ (line 727) | def __init__(self, config):
    method get_output_embeddings (line 735) | def get_output_embeddings(self):
    method set_output_embeddings (line 738) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 741) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method forward (line 784) | def forward(
    method _reorder_cache (line 851) | def _reorder_cache(
  class GPTBigCodeForSequenceClassification (line 877) | class GPTBigCodeForSequenceClassification(GPTBigCodePreTrainedModel):
    method __init__ (line 880) | def __init__(self, config):
    method forward (line 890) | def forward(
  class GPTBigCodeForTokenClassification (line 995) | class GPTBigCodeForTokenClassification(GPTBigCodePreTrainedModel):
    method __init__ (line 996) | def __init__(self, config):
    method forward (line 1014) | def forward(

FILE: transformers/models/gpt_neo/configuration_gpt_neo.py
  class GPTNeoConfig (line 34) | class GPTNeoConfig(PretrainedConfig):
    method __init__ (line 104) | def __init__(
    method expand_attention_types_params (line 161) | def expand_attention_types_params(attention_types):
  function custom_unfold (line 169) | def custom_unfold(input, dimension, size, step):
  function custom_get_block_length_and_num_blocks (line 191) | def custom_get_block_length_and_num_blocks(seq_length, window_size):
  class GPTNeoOnnxConfig (line 206) | class GPTNeoOnnxConfig(OnnxConfigWithPast):
    method inputs (line 208) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method num_attention_heads (line 219) | def num_attention_heads(self) -> int:
    method generate_dummy_inputs (line 222) | def generate_dummy_inputs(
    method default_onnx_opset (line 267) | def default_onnx_opset(self) -> int:

FILE: transformers/models/gpt_neo/convert_gpt_neo_mesh_tf_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 28) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, config_file, py...

FILE: transformers/models/gpt_neo/modeling_flax_gpt_neo.py
  class FlaxGPTNeoSelfAttention (line 108) | class FlaxGPTNeoSelfAttention(nn.Module):
    method setup (line 113) | def setup(self):
    method _split_heads (line 141) | def _split_heads(self, hidden_states):
    method _merge_heads (line 144) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 148) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 179) | def __call__(
  class FlaxGPTNeoAttention (line 249) | class FlaxGPTNeoAttention(nn.Module):
    method setup (line 254) | def setup(self):
    method __call__ (line 258) | def __call__(
  class FlaxGPTNeoMLP (line 275) | class FlaxGPTNeoMLP(nn.Module):
    method setup (line 280) | def setup(self):
    method __call__ (line 288) | def __call__(self, hidden_states, deterministic: bool = True):
  class FlaxGPTNeoBlock (line 296) | class FlaxGPTNeoBlock(nn.Module):
    method setup (line 301) | def setup(self):
    method __call__ (line 310) | def __call__(
  class FlaxGPTNeoPreTrainedModel (line 340) | class FlaxGPTNeoPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 350) | def __init__(
    method init_weights (line 362) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 382) | def init_cache(self, batch_size, max_length):
    method __call__ (line 402) | def __call__(
  class FlaxGPTNeoBlockCollection (line 472) | class FlaxGPTNeoBlockCollection(nn.Module):
    method setup (line 476) | def setup(self):
    method __call__ (line 482) | def __call__(
  class FlaxGPTNeoModule (line 517) | class FlaxGPTNeoModule(nn.Module):
    method setup (line 521) | def setup(self):
    method __call__ (line 538) | def __call__(
  class FlaxGPTNeoModel (line 591) | class FlaxGPTNeoModel(FlaxGPTNeoPreTrainedModel):
  class FlaxGPTNeoForCausalLMModule (line 598) | class FlaxGPTNeoForCausalLMModule(nn.Module):
    method setup (line 602) | def setup(self):
    method __call__ (line 611) | def __call__(
  class FlaxGPTNeoForCausalLM (line 654) | class FlaxGPTNeoForCausalLM(FlaxGPTNeoPreTrainedModel):
    method prepare_inputs_for_generation (line 657) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 678) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/gpt_neo/modeling_gpt_neo.py
  function load_tf_weights_in_gpt_neo (line 53) | def load_tf_weights_in_gpt_neo(model, config, gpt_neo_checkpoint_path):
  class GPTNeoSelfAttention (line 133) | class GPTNeoSelfAttention(nn.Module):
    method __init__ (line 134) | def __init__(self, config, attention_type):
    method _split_heads (line 168) | def _split_heads(self, tensor, num_heads, attn_head_size):
    method _merge_heads (line 176) | def _merge_heads(self, tensor, num_heads, attn_head_size):
    method _attn (line 184) | def _attn(self, query, key, value, attention_mask=None, head_mask=None):
    method forward (line 215) | def forward(
  class GPTNeoAttention (line 256) | class GPTNeoAttention(nn.Module):
    method __init__ (line 257) | def __init__(self, config, layer_id=0):
    method forward (line 271) | def forward(
  class GPTNeoMLP (line 290) | class GPTNeoMLP(nn.Module):
    method __init__ (line 291) | def __init__(self, intermediate_size, config):  # in MLP: intermediate...
    method forward (line 299) | def forward(self, hidden_states):
  class GPTNeoBlock (line 307) | class GPTNeoBlock(nn.Module):
    method __init__ (line 308) | def __init__(self, config, layer_id):
    method forward (line 317) | def forward(
  class GPTNeoPreTrainedModel (line 355) | class GPTNeoPreTrainedModel(PreTrainedModel):
    method __init__ (line 368) | def __init__(self, *inputs, **kwargs):
    method _init_weights (line 371) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 387) | def _set_gradient_checkpointing(self, module, value=False):
  class GPTNeoModel (line 477) | class GPTNeoModel(GPTNeoPreTrainedModel):
    method __init__ (line 478) | def __init__(self, config):
    method get_input_embeddings (line 492) | def get_input_embeddings(self):
    method set_input_embeddings (line 495) | def set_input_embeddings(self, new_embeddings):
    method forward (line 504) | def forward(
  class GPTNeoForCausalLM (line 665) | class GPTNeoForCausalLM(GPTNeoPreTrainedModel):
    method __init__ (line 673) | def __init__(self, config):
    method get_output_embeddings (line 681) | def get_output_embeddings(self):
    method set_output_embeddings (line 684) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 687) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method forward (line 721) | def forward(
    method _reorder_cache (line 792) | def _reorder_cache(
  class GPTNeoForSequenceClassification (line 821) | class GPTNeoForSequenceClassification(GPTNeoPreTrainedModel):
    method __init__ (line 824) | def __init__(self, config):
    method forward (line 839) | def forward(
  class GPTNeoForTokenClassification (line 941) | class GPTNeoForTokenClassification(GPTNeoPreTrainedModel):
    method __init__ (line 942) | def __init__(self, config):
    method forward (line 960) | def forward(
  class GPTNeoForQuestionAnswering (line 1026) | class GPTNeoForQuestionAnswering(GPTNeoPreTrainedModel):
    method __init__ (line 1029) | def __init__(self, config):
    method forward (line 1045) | def forward(

FILE: transformers/models/gpt_neox/configuration_gpt_neox.py
  class GPTNeoXConfig (line 29) | class GPTNeoXConfig(PretrainedConfig):
    method __init__ (line 92) | def __init__(

FILE: transformers/models/gpt_neox/modeling_gpt_neox.py
  class GPTNeoXPreTrainedModel (line 55) | class GPTNeoXPreTrainedModel(PreTrainedModel):
    method _init_weights (line 67) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 81) | def _set_gradient_checkpointing(self, module, value=False):
  class GPTNeoXAttention (line 86) | class GPTNeoXAttention(nn.Module):
    method __init__ (line 87) | def __init__(self, config):
    method forward (line 112) | def forward(
    method _split_heads (line 176) | def _split_heads(cls, tensor, num_attention_heads, attn_head_size):
    method _merge_heads (line 189) | def _merge_heads(cls, tensor, num_attention_heads, attn_head_size):
    method _attn (line 200) | def _attn(self, query, key, value, attention_mask=None, head_mask=None):
  function attention_mask_func (line 247) | def attention_mask_func(attention_scores, ltor_mask):
  class RotaryEmbedding (line 252) | class RotaryEmbedding(torch.nn.Module):
    method __init__ (line 253) | def __init__(self, dim, max_position_embeddings, base=10000, device=No...
    method forward (line 267) | def forward(self, x, seq_len=None):
  function rotate_half (line 281) | def rotate_half(x):
  function apply_rotary_pos_emb (line 288) | def apply_rotary_pos_emb(q, k, cos, sin, position_ids):
  class GPTNeoXMLP (line 298) | class GPTNeoXMLP(nn.Module):
    method __init__ (line 299) | def __init__(self, config):
    method forward (line 305) | def forward(self, hidden_states):
  class GPTNeoXLayer (line 312) | class GPTNeoXLayer(nn.Module):
    method __init__ (line 313) | def __init__(self, config):
    method forward (line 321) | def forward(
  class GPTNeoXModel (line 421) | class GPTNeoXModel(GPTNeoXPreTrainedModel):
    method __init__ (line 422) | def __init__(self, config):
    method get_input_embeddings (line 435) | def get_input_embeddings(self):
    method set_input_embeddings (line 438) | def set_input_embeddings(self, value):
    method forward (line 448) | def forward(
  class GPTNeoXForCausalLM (line 598) | class GPTNeoXForCausalLM(GPTNeoXPreTrainedModel):
    method __init__ (line 601) | def __init__(self, config):
    method get_output_embeddings (line 610) | def get_output_embeddings(self):
    method set_output_embeddings (line 613) | def set_output_embeddings(self, new_embeddings):
    method forward (line 618) | def forward(
    method prepare_inputs_for_generation (line 711) | def prepare_inputs_for_generation(
    method _reorder_cache (line 748) | def _reorder_cache(self, past_key_values, beam_idx):
  class GPTNeoXForSequenceClassification (line 772) | class GPTNeoXForSequenceClassification(GPTNeoXPreTrainedModel):
    method __init__ (line 775) | def __init__(self, config):
    method forward (line 790) | def forward(
  class GPTNeoXForTokenClassification (line 884) | class GPTNeoXForTokenClassification(GPTNeoXPreTrainedModel):
    method __init__ (line 885) | def __init__(self, config):
    method forward (line 903) | def forward(
  class GPTNeoXForQuestionAnswering (line 968) | class GPTNeoXForQuestionAnswering(GPTNeoXPreTrainedModel):
    method __init__ (line 971) | def __init__(self, config):
    method forward (line 987) | def forward(

FILE: transformers/models/gpt_neox/tokenization_gpt_neox_fast.py
  class GPTNeoXTokenizerFast (line 44) | class GPTNeoXTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 102) | def __init__(
    method save_vocabulary (line 132) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method _build_conversation_input_ids (line 136) | def _build_conversation_input_ids(self, conversation: "Conversation") ...

FILE: transformers/models/gpt_neox_japanese/configuration_gpt_neox_japanese.py
  class GPTNeoXJapaneseConfig (line 28) | class GPTNeoXJapaneseConfig(PretrainedConfig):
    method __init__ (line 86) | def __init__(

FILE: transformers/models/gpt_neox_japanese/modeling_gpt_neox_japanese.py
  class GPTNeoXJapanesePreTrainedModel (line 43) | class GPTNeoXJapanesePreTrainedModel(PreTrainedModel):
    method _init_weights (line 55) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 69) | def _set_gradient_checkpointing(self, module, value=False):
  class GPTNeoXJapaneseAttention (line 74) | class GPTNeoXJapaneseAttention(nn.Module):
    method __init__ (line 75) | def __init__(self, config, use_bias=False):
    method forward (line 95) | def forward(
    method _split_heads (line 160) | def _split_heads(cls, tensor, num_attention_heads, attn_head_size):
    method _merge_heads (line 173) | def _merge_heads(cls, tensor, num_attention_heads, attn_head_size):
    method _create_causal_mask (line 184) | def _create_causal_mask(self, key_length, query_length):
    method _attn (line 192) | def _attn(self, query, key, value, attention_mask=None, head_mask=None):
  class RotaryEmbedding (line 242) | class RotaryEmbedding(torch.nn.Module):
    method __init__ (line 243) | def __init__(self, dim, max_position_embeddings, base=10000, device=No...
    method forward (line 257) | def forward(self, x, seq_len=None):
  function rotate_half (line 271) | def rotate_half(x):
  function apply_rotary_pos_emb (line 278) | def apply_rotary_pos_emb(q, k, cos, sin, offset: int = 0):
  function bias_dropout_add (line 286) | def bias_dropout_add(x: Tensor, bias: Tensor, residual: Optional[Tensor]...
  class GPTNeoXJapaneseMLP (line 307) | class GPTNeoXJapaneseMLP(nn.Module):
    method __init__ (line 308) | def __init__(self, config):
    method forward (line 316) | def forward(self, hidden_states):
  class GPTNeoXJapaneseLayer (line 323) | class GPTNeoXJapaneseLayer(nn.Module):
    method __init__ (line 324) | def __init__(self, config, layer_number):
    method forward (line 334) | def forward(
  class GPTNeoXJapaneseModel (line 439) | class GPTNeoXJapaneseModel(GPTNeoXJapanesePreTrainedModel):
    method __init__ (line 440) | def __init__(self, config):
    method get_input_embeddings (line 453) | def get_input_embeddings(self):
    method set_input_embeddings (line 456) | def set_input_embeddings(self, value):
    method forward (line 461) | def forward(
  class GPTNeoXJapaneseForCausalLM (line 593) | class GPTNeoXJapaneseForCausalLM(GPTNeoXJapanesePreTrainedModel):
    method __init__ (line 596) | def __init__(self, config):
    method get_output_embeddings (line 606) | def get_output_embeddings(self):
    method set_output_embeddings (line 609) | def set_output_embeddings(self, new_embeddings):
    method forward (line 614) | def forward(
    method prepare_inputs_for_generation (line 707) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 720) | def _reorder_cache(self, past_key_values, beam_idx):

FILE: transformers/models/gpt_neox_japanese/tokenization_gpt_neox_japanese.py
  function load_vocab_and_emoji (line 50) | def load_vocab_and_emoji(vocab_file, emoji_file):
  class GPTNeoXJapaneseTokenizer (line 70) | class GPTNeoXJapaneseTokenizer(PreTrainedTokenizer):
    method __init__ (line 123) | def __init__(
    method vocab_size (line 159) | def vocab_size(self):
    method get_vocab (line 163) | def get_vocab(self):
    method _tokenize (line 166) | def _tokenize(self, text):
    method _convert_token_to_id (line 169) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 173) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 177) | def convert_tokens_to_string(self, tokens):
    method _build_conversation_input_ids (line 182) | def _build_conversation_input_ids(self, conversation: "Conversation") ...
    method save_vocabulary (line 192) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class SubWordJapaneseTokenizer (line 223) | class SubWordJapaneseTokenizer(object):
    method __init__ (line 247) | def __init__(self, vocab, ids_to_tokens, emoji):
    method __len__ (line 268) | def __len__(self):
    method clean_text (line 271) | def clean_text(self, content):
    method tokenize (line 283) | def tokenize(self, text, clean=False):
    method convert_id_to_token (line 350) | def convert_id_to_token(self, index, breakline="\n"):

FILE: transformers/models/gpt_sw3/convert_megatron_to_pytorch.py
  function recursive_print (line 25) | def recursive_print(name, val, spaces=0):
  function fix_query_key_value_ordering (line 45) | def fix_query_key_value_ordering(param, num_splits, num_heads, hidden_si...
  function convert_megatron_checkpoint (line 61) | def convert_megatron_checkpoint(sd_megatron, config):
  function copy_config (line 116) | def copy_config(config_hf, config_megatron):
  function main (line 147) | def main(args):

FILE: transformers/models/gpt_sw3/tokenization_gpt_sw3.py
  class GPTSw3Tokenizer (line 42) | class GPTSw3Tokenizer(PreTrainedTokenizer):
    method __init__ (line 108) | def __init__(
    method __getstate__ (line 172) | def __getstate__(self):
    method __setstate__ (line 178) | def __setstate__(self, d):
    method vocab_size (line 190) | def vocab_size(self) -> int:
    method preprocess_text (line 193) | def preprocess_text(self, text: str) -> str:
    method _tokenize (line 208) | def _tokenize(self, text: str, **kwargs) -> List[str]:
    method _convert_token_to_id (line 212) | def _convert_token_to_id(self, token: str) -> int:
    method _convert_id_to_token (line 216) | def _convert_id_to_token(self, index: int) -> str:
    method clean_up_tokenization (line 221) | def clean_up_tokenization(out_string: str) -> str:
    method convert_tokens_to_string (line 225) | def convert_tokens_to_string(self, tokens: List[str]) -> str:
    method get_vocab (line 246) | def get_vocab(self) -> Dict[str, int]:
    method save_vocabulary (line 252) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method encode_fast (line 269) | def encode_fast(
    method decode_fast (line 302) | def decode_fast(self, token_ids: Union[int, List[int]]) -> str:

FILE: transformers/models/gptj/configuration_gptj.py
  class GPTJConfig (line 33) | class GPTJConfig(PretrainedConfig):
    method __init__ (line 96) | def __init__(
  class GPTJOnnxConfig (line 141) | class GPTJOnnxConfig(OnnxConfigWithPast):
    method __init__ (line 142) | def __init__(
    method inputs (line 155) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method num_layers (line 166) | def num_layers(self) -> int:
    method num_attention_heads (line 170) | def num_attention_heads(self) -> int:
    method generate_dummy_inputs (line 173) | def generate_dummy_inputs(
    method default_onnx_opset (line 218) | def default_onnx_opset(self) -> int:

FILE: transformers/models/gptj/modeling_flax_gptj.py
  function create_sinusoidal_positions (line 109) | def create_sinusoidal_positions(num_pos, dim):
  function rotate_every_two (line 122) | def rotate_every_two(tensor):
  function apply_rotary_pos_emb (line 128) | def apply_rotary_pos_emb(tensor, sincos):
  class FlaxGPTJAttention (line 135) | class FlaxGPTJAttention(nn.Module):
    method setup (line 141) | def setup(self):
    method _split_heads (line 167) | def _split_heads(self, hidden_states):
    method _merge_heads (line 170) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 174) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 205) | def __call__(
  class FlaxGPTJMLP (line 294) | class FlaxGPTJMLP(nn.Module):
    method setup (line 299) | def setup(self):
    method __call__ (line 309) | def __call__(self, hidden_states, deterministic: bool = True):
  class FlaxGPTJBlock (line 317) | class FlaxGPTJBlock(nn.Module):
    method setup (line 321) | def setup(self):
    method __call__ (line 330) | def __call__(
  class FlaxGPTJPreTrainedModel (line 358) | class FlaxGPTJPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 368) | def __init__(
    method init_weights (line 380) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 415) | def init_cache(self, batch_size, max_length):
    method __call__ (line 435) | def __call__(
  class FlaxGPTJBlockCollection (line 505) | class FlaxGPTJBlockCollection(nn.Module):
    method setup (line 509) | def setup(self):
    method __call__ (line 514) | def __call__(
  class FlaxGPTJModule (line 551) | class FlaxGPTJModule(nn.Module):
    method setup (line 555) | def setup(self):
    method __call__ (line 567) | def __call__(
  class FlaxGPTJModel (line 616) | class FlaxGPTJModel(FlaxGPTJPreTrainedModel):
  class FlaxGPTJForCausalLMModule (line 628) | class FlaxGPTJForCausalLMModule(nn.Module):
    method setup (line 632) | def setup(self):
    method __call__ (line 640) | def __call__(
  class FlaxGPTJForCausalLM (line 682) | class FlaxGPTJForCausalLM(FlaxGPTJPreTrainedModel):
    method prepare_inputs_for_generation (line 685) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 706) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/gptj/modeling_gptj.py
  function create_sinusoidal_positions (line 58) | def create_sinusoidal_positions(num_pos: int, dim: int) -> torch.Tensor:
  function get_embed_positions (line 65) | def get_embed_positions(embed_positions, position_ids):
  function rotate_every_two (line 69) | def rotate_every_two(x: torch.Tensor) -> torch.Tensor:
  function apply_rotary_pos_emb (line 76) | def apply_rotary_pos_emb(tensor: torch.Tensor, sin: torch.Tensor, cos: t...
  class GPTJAttention (line 82) | class GPTJAttention(nn.Module):
    method __init__ (line 83) | def __init__(self, config):
    method _split_heads (line 117) | def _split_heads(self, tensor, num_attention_heads, attn_head_size, ro...
    method _merge_heads (line 132) | def _merge_heads(self, tensor, num_attention_heads, attn_head_size):
    method _attn (line 145) | def _attn(
    method _get_embed_positions (line 187) | def _get_embed_positions(self, position_ids):
    method forward (line 194) | def forward(
  class GPTJMLP (line 270) | class GPTJMLP(nn.Module):
    method __init__ (line 271) | def __init__(self, intermediate_size, config):  # in MLP: intermediate...
    method forward (line 281) | def forward(self, hidden_states: Optional[torch.FloatTensor]) -> torch...
  class GPTJBlock (line 289) | class GPTJBlock(nn.Module):
    method __init__ (line 290) | def __init__(self, config):
    method forward (line 297) | def forward(
  class GPTJPreTrainedModel (line 332) | class GPTJPreTrainedModel(PreTrainedModel):
    method __init__ (line 345) | def __init__(self, *inputs, **kwargs):
    method _init_weights (line 348) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 364) | def _set_gradient_checkpointing(self, module, value=False):
  class GPTJModel (line 482) | class GPTJModel(GPTJPreTrainedModel):
    method __init__ (line 483) | def __init__(self, config):
    method parallelize (line 502) | def parallelize(self, device_map=None):
    method deparallelize (line 528) | def deparallelize(self):
    method get_input_embeddings (line 543) | def get_input_embeddings(self):
    method set_input_embeddings (line 546) | def set_input_embeddings(self, new_embeddings):
    method forward (line 556) | def forward(
  class GPTJForCausalLM (line 736) | class GPTJForCausalLM(GPTJPreTrainedModel):
    method __init__ (line 739) | def __init__(self, config):
    method parallelize (line 752) | def parallelize(self, device_map=None):
    method deparallelize (line 771) | def deparallelize(self):
    method get_output_embeddings (line 782) | def get_output_embeddings(self):
    method set_output_embeddings (line 785) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 788) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method forward (line 831) | def forward(
    method _reorder_cache (line 905) | def _reorder_cache(
  class GPTJForSequenceClassification (line 934) | class GPTJForSequenceClassification(GPTJPreTrainedModel):
    method __init__ (line 937) | def __init__(self, config):
    method forward (line 957) | def forward(
  class GPTJForQuestionAnswering (line 1060) | class GPTJForQuestionAnswering(GPTJPreTrainedModel):
    method __init__ (line 1063) | def __init__(self, config):
    method forward (line 1083) | def forward(

FILE: transformers/models/gptj/modeling_tf_gptj.py
  function create_sinusoidal_positions (line 63) | def create_sinusoidal_positions(num_pos: int, dim: int) -> tf.Tensor:
  function rotate_every_two (line 71) | def rotate_every_two(x: tf.Tensor) -> tf.Tensor:
  function apply_rotary_pos_emb (line 78) | def apply_rotary_pos_emb(tensor: tf.Tensor, sincos: tf.Tensor) -> tf.Ten...
  class TFGPTJAttention (line 85) | class TFGPTJAttention(tf.keras.layers.Layer):
    method __init__ (line 86) | def __init__(self, config: GPTJConfig, **kwargs):
    method get_causal_mask (line 136) | def get_causal_mask(self, key_length, query_length) -> tf.Tensor:
    method get_masked_bias (line 140) | def get_masked_bias(dtype: tf.DType) -> tf.Tensor:
    method _split_heads (line 143) | def _split_heads(self, hidden_states: tf.Tensor, rotary: bool) -> tf.T...
    method _merge_heads (line 157) | def _merge_heads(self, hidden_states: tf.Tensor) -> tf.Tensor:
    method _attn (line 170) | def _attn(
    method call (line 207) | def call(
  class TFGPTJMLP (line 271) | class TFGPTJMLP(tf.keras.layers.Layer):
    method __init__ (line 272) | def __init__(self, intermediate_size: int, config: GPTJConfig, **kwargs):
    method call (line 286) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFGPTJBlock (line 294) | class TFGPTJBlock(tf.keras.layers.Layer):
    method __init__ (line 295) | def __init__(self, config: GPTJConfig, **kwargs):
    method call (line 302) | def call(
  class TFGPTJMainLayer (line 337) | class TFGPTJMainLayer(tf.keras.layers.Layer):
    method __init__ (line 340) | def __init__(self, config: GPTJConfig, *inputs, **kwargs):
    method get_input_embeddings (line 361) | def get_input_embeddings(self):
    method set_input_embeddings (line 364) | def set_input_embeddings(self, value: tf.Tensor):
    method _prune_heads (line 368) | def _prune_heads(self, heads_to_prune):
    method call (line 375) | def call(
  class TFGPTJPreTrainedModel (line 504) | class TFGPTJPreTrainedModel(TFPreTrainedModel):
  class TFGPTJModel (line 625) | class TFGPTJModel(TFGPTJPreTrainedModel):
    method __init__ (line 626) | def __init__(self, config, *inputs, **kwargs):
    method call (line 637) | def call(
  class TFGPTJForCausalLM (line 682) | class TFGPTJForCausalLM(TFGPTJPreTrainedModel, TFCausalLanguageModelingL...
    method __init__ (line 683) | def __init__(self, config, *inputs, **kwargs):
    method get_output_embeddings (line 690) | def get_output_embeddings(self):
    method set_output_embeddings (line 693) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 696) | def prepare_inputs_for_generation(self, inputs, past_key_values=None, ...
    method call (line 728) | def call(
  class TFGPTJForSequenceClassification (line 803) | class TFGPTJForSequenceClassification(TFGPTJPreTrainedModel, TFSequenceC...
    method __init__ (line 806) | def __init__(self, config, *inputs, **kwargs):
    method call (line 824) | def call(
  class TFGPTJForQuestionAnswering (line 919) | class TFGPTJForQuestionAnswering(TFGPTJPreTrainedModel, TFQuestionAnswer...
    method __init__ (line 922) | def __init__(self, config, *inputs, **kwargs):
    method call (line 937) | def call(

FILE: transformers/models/gptsan_japanese/configuration_gptsan_japanese.py
  class GPTSanJapaneseConfig (line 29) | class GPTSanJapaneseConfig(PretrainedConfig):
    method __init__ (line 100) | def __init__(

FILE: transformers/models/gptsan_japanese/convert_gptsan_tf_checkpoint_to_pytorch.py
  function convert_tf_gptsan_to_pt (line 28) | def convert_tf_gptsan_to_pt(args):

FILE: transformers/models/gptsan_japanese/modeling_gptsan_japanese.py
  function router_z_loss_func (line 54) | def router_z_loss_func(router_logits: torch.Tensor) -> float:
  function load_balancing_loss_func (line 75) | def load_balancing_loss_func(router_probs: torch.Tensor, expert_indices:...
  class GPTSanJapaneseDenseActDense (line 114) | class GPTSanJapaneseDenseActDense(nn.Module):
    method __init__ (line 124) | def __init__(self, config: GPTSanJapaneseConfig, ext_layer=False):
    method forward (line 132) | def forward(self, hidden_states):
  class GPTSanJapaneseTop1Router (line 149) | class GPTSanJapaneseTop1Router(nn.Module):
    method __init__ (line 160) | def __init__(self, config: GPTSanJapaneseConfig):
    method _compute_router_probabilities (line 169) | def _compute_router_probabilities(self, hidden_states: torch.Tensor) -...
    method _cast_classifier (line 211) | def _cast_classifier(self):
    method forward (line 219) | def forward(self, hidden_states: torch.Tensor) -> Tuple:
  class GPTSanJapaneseSparseMLP (line 252) | class GPTSanJapaneseSparseMLP(nn.Module):
    method __init__ (line 257) | def __init__(self, config: GPTSanJapaneseConfig, expert_class: nn.Modu...
    method forward (line 267) | def forward(self, hidden_states):
  class GPTSanJapaneseLayerSparseFF (line 295) | class GPTSanJapaneseLayerSparseFF(nn.Module):
    method __init__ (line 305) | def __init__(self, config: GPTSanJapaneseConfig):
    method forward (line 311) | def forward(self, hidden_states, output_router_logits):
  class GPTSanJapaneseLayerDenseFF (line 332) | class GPTSanJapaneseLayerDenseFF(nn.Module):
    method __init__ (line 342) | def __init__(self, config: GPTSanJapaneseConfig):
    method forward (line 348) | def forward(self, hidden_states):
  class GPTSanJapaneseAttention (line 363) | class GPTSanJapaneseAttention(nn.Module):
    method __init__ (line 366) | def __init__(
    method _shape (line 393) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 396) | def forward(
  class GPTSanJapaneseLayerSelfAttention (line 517) | class GPTSanJapaneseLayerSelfAttention(nn.Module):
    method __init__ (line 522) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 532) | def forward(
  class GPTSanJapaneseBlock (line 603) | class GPTSanJapaneseBlock(nn.Module):
    method __init__ (line 608) | def __init__(self, config, ext_layer=False):
    method forward (line 613) | def forward(
  class GPTSanJapanesePreTrainedModel (line 685) | class GPTSanJapanesePreTrainedModel(PreTrainedModel):
    method dummy_inputs (line 698) | def dummy_inputs(self):
    method _init_weights (line 707) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 762) | def _set_gradient_checkpointing(self, module, value=False):
    method _shift_right (line 767) | def _shift_right(self, input_ids):
  class GPTSanJapaneseModel (line 868) | class GPTSanJapaneseModel(GPTSanJapanesePreTrainedModel):
    method __init__ (line 869) | def __init__(self, config: GPTSanJapaneseConfig):
    method get_input_embeddings (line 896) | def get_input_embeddings(self):
    method set_input_embeddings (line 899) | def set_input_embeddings(self, new_embeddings):
    method forward (line 903) | def forward(
  class GPTSanJapaneseForConditionalGeneration (line 1113) | class GPTSanJapaneseForConditionalGeneration(GPTSanJapanesePreTrainedMod...
    method __init__ (line 1116) | def __init__(self, config: GPTSanJapaneseConfig):
    method forward (line 1125) | def forward(
    method prepare_inputs_for_generation (line 1282) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1312) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method resize_token_embeddings (line 1316) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_final_logits_bias (line 1322) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_input_embeddings (line 1331) | def get_input_embeddings(self):
    method set_input_embeddings (line 1334) | def set_input_embeddings(self, new_embeddings):
    method set_output_embeddings (line 1338) | def set_output_embeddings(self, new_embeddings):
    method get_output_embeddings (line 1342) | def get_output_embeddings(self):
    method _unpack_router_logits (line 1346) | def _unpack_router_logits(self, router_outputs):

FILE: transformers/models/gptsan_japanese/tokenization_gptsan_japanese.py
  function load_vocab_and_emoji (line 58) | def load_vocab_and_emoji(vocab_file, emoji_file):
  class GPTSanJapaneseTokenizer (line 78) | class GPTSanJapaneseTokenizer(PreTrainedTokenizer):
    method __init__ (line 161) | def __init__(
    method vocab_size (line 200) | def vocab_size(self):
    method get_vocab (line 205) | def get_vocab(self):
    method _tokenize (line 209) | def _tokenize(self, text):
    method _convert_token_to_id (line 213) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 218) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 222) | def convert_tokens_to_string(self, tokens):
    method _build_conversation_input_ids (line 262) | def _build_conversation_input_ids(self, conversation: "Conversation") ...
    method save_vocabulary (line 273) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method create_token_type_ids_from_sequences (line 303) | def create_token_type_ids_from_sequences(
    method prepare_for_tokenization (line 339) | def prepare_for_tokenization(self, text, prefix_text=None, add_sep_tok...
    method _batch_encode_plus (line 351) | def _batch_encode_plus(
  class SubWordJapaneseTokenizer (line 400) | class SubWordJapaneseTokenizer(object):
    method __init__ (line 429) | def __init__(self, vocab, ids_to_tokens, emoji):
    method __len__ (line 451) | def __len__(self):
    method clean_text (line 455) | def clean_text(self, content):
    method tokenize (line 468) | def tokenize(self, text, clean=False):
    method convert_id_to_token (line 535) | def convert_id_to_token(self, index):

FILE: transformers/models/graphormer/collating_graphormer.py
  function convert_to_single_emb (line 19) | def convert_to_single_emb(x, offset: int = 512):
  function preprocess_item (line 26) | def preprocess_item(item, keep_features=True):
  class GraphormerDataCollator (line 73) | class GraphormerDataCollator:
    method __init__ (line 74) | def __init__(self, spatial_pos_max=20, on_the_fly_processing=False):
    method __call__ (line 81) | def __call__(self, features: List[dict]) -> Dict[str, Any]:

FILE: transformers/models/graphormer/configuration_graphormer.py
  class GraphormerConfig (line 30) | class GraphormerConfig(PretrainedConfig):
    method __init__ (line 133) | def __init__(

FILE: transformers/models/graphormer/modeling_graphormer.py
  function quant_noise (line 47) | def quant_noise(module: nn.Module, p: float, block_size: int):
  class LayerDropModuleList (line 139) | class LayerDropModuleList(nn.ModuleList):
    method __init__ (line 166) | def __init__(self, p: float, modules: Optional[Iterable[nn.Module]] = ...
    method __iter__ (line 170) | def __iter__(self) -> Iterator[nn.Module]:
  class GraphormerGraphNodeFeature (line 177) | class GraphormerGraphNodeFeature(nn.Module):
    method __init__ (line 182) | def __init__(self, config: GraphormerConfig):
    method forward (line 197) | def forward(
  class GraphormerGraphAttnBias (line 218) | class GraphormerGraphAttnBias(nn.Module):
    method __init__ (line 223) | def __init__(self, config: GraphormerConfig):
    method forward (line 243) | def forward(
  class GraphormerMultiheadAttention (line 300) | class GraphormerMultiheadAttention(nn.Module):
    method __init__ (line 306) | def __init__(self, config: GraphormerConfig):
    method reset_parameters (line 351) | def reset_parameters(self):
    method forward (line 367) | def forward(
    method apply_sparse_mask (line 486) | def apply_sparse_mask(self, attn_weights: torch.Tensor, tgt_len: int, ...
  class GraphormerGraphEncoderLayer (line 490) | class GraphormerGraphEncoderLayer(nn.Module):
    method __init__ (line 491) | def __init__(self, config: GraphormerConfig) -> None:
    method build_fc (line 529) | def build_fc(
    method forward (line 534) | def forward(
  class GraphormerGraphEncoder (line 577) | class GraphormerGraphEncoder(nn.Module):
    method __init__ (line 578) | def __init__(self, config: GraphormerConfig):
    method forward (line 625) | def forward(
  class GraphormerDecoderHead (line 694) | class GraphormerDecoderHead(nn.Module):
    method __init__ (line 695) | def __init__(self, embedding_dim: int, num_classes: int):
    method forward (line 702) | def forward(self, input_nodes: torch.Tensor, **unused) -> torch.Tensor:
  class GraphormerPreTrainedModel (line 708) | class GraphormerPreTrainedModel(PreTrainedModel):
    method normal_ (line 721) | def normal_(self, data: torch.Tensor):
    method init_graphormer_params (line 726) | def init_graphormer_params(self, module: Union[nn.Linear, nn.Embedding...
    method _init_weights (line 743) | def _init_weights(
    method _set_gradient_checkpointing (line 777) | def _set_gradient_checkpointing(self, module, value=False):
  class GraphormerModel (line 782) | class GraphormerModel(GraphormerPreTrainedModel):
    method __init__ (line 790) | def __init__(self, config: GraphormerConfig):
    method reset_output_layer_parameters (line 808) | def reset_output_layer_parameters(self):
    method forward (line 811) | def forward(
    method max_nodes (line 848) | def max_nodes(self):
  class GraphormerForGraphClassification (line 853) | class GraphormerForGraphClassification(GraphormerPreTrainedModel):
    method __init__ (line 865) | def __init__(self, config: GraphormerConfig):
    method forward (line 876) | def forward(

FILE: transformers/models/groupvit/configuration_groupvit.py
  class GroupViTTextConfig (line 39) | class GroupViTTextConfig(PretrainedConfig):
    method __init__ (line 94) | def __init__(
    method from_pretrained (line 129) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class GroupViTVisionConfig (line 145) | class GroupViTVisionConfig(PretrainedConfig):
    method __init__ (line 203) | def __init__(
    method from_pretrained (line 252) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class GroupViTConfig (line 268) | class GroupViTConfig(PretrainedConfig):
    method __init__ (line 297) | def __init__(
    method from_text_vision_configs (line 395) | def from_text_vision_configs(cls, text_config: GroupViTTextConfig, vis...
    method to_dict (line 406) | def to_dict(self):
  class GroupViTOnnxConfig (line 420) | class GroupViTOnnxConfig(OnnxConfig):
    method inputs (line 422) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 432) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 443) | def atol_for_validation(self) -> float:
    method generate_dummy_inputs (line 446) | def generate_dummy_inputs(
    method default_onnx_opset (line 462) | def default_onnx_opset(self) -> int:

FILE: transformers/models/groupvit/convert_groupvit_nvlab_to_hf.py
  function rename_key (line 31) | def rename_key(name):
  function convert_state_dict (line 87) | def convert_state_dict(orig_state_dict, config):
  function prepare_img (line 150) | def prepare_img():
  function convert_groupvit_checkpoint (line 157) | def convert_groupvit_checkpoint(

FILE: transformers/models/groupvit/modeling_groupvit.py
  function _expand_mask (line 52) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  function contrastive_loss (line 68) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  function groupvit_loss (line 73) | def groupvit_loss(similarity: torch.Tensor) -> torch.Tensor:
  function hard_softmax (line 79) | def hard_softmax(logits: torch.Tensor, dim: int):
  function gumbel_softmax (line 89) | def gumbel_softmax(logits: torch.Tensor, tau: float = 1, hard: bool = Fa...
  function resize_attention_map (line 111) | def resize_attention_map(attentions, height, width, align_corners=False):
  function get_grouping_from_attentions (line 141) | def get_grouping_from_attentions(attentions, hw_shape):
  class GroupViTCrossAttentionLayer (line 170) | class GroupViTCrossAttentionLayer(nn.Module):
    method __init__ (line 171) | def __init__(self, config: GroupViTVisionConfig):
    method forward (line 178) | def forward(self, query, key):
  class GroupViTAssignAttention (line 186) | class GroupViTAssignAttention(nn.Module):
    method __init__ (line 187) | def __init__(self, config: GroupViTVisionConfig):
    method get_attn (line 197) | def get_attn(self, attn, gumbel=True, hard=True):
    method forward (line 208) | def forward(self, query, key):
  class GroupViTTokenAssign (line 234) | class GroupViTTokenAssign(nn.Module):
    method __init__ (line 235) | def __init__(self, config: GroupViTVisionConfig, num_group_token, num_...
    method project_group_token (line 256) | def project_group_token(self, group_tokens):
    method forward (line 269) | def forward(self, image_tokens, group_tokens):
  class GroupViTModelOutput (line 290) | class GroupViTModelOutput(ModelOutput):
    method to_tuple (line 333) | def to_tuple(self) -> Tuple[Any]:
  class GroupViTPatchEmbeddings (line 340) | class GroupViTPatchEmbeddings(nn.Module):
    method __init__ (line 345) | def __init__(
    method forward (line 362) | def forward(self, pixel_values: torch.Tensor, interpolate_pos_encoding...
  class GroupViTVisionEmbeddings (line 374) | class GroupViTVisionEmbeddings(nn.Module):
    method __init__ (line 375) | def __init__(self, config: GroupViTVisionConfig):
    method interpolate_pos_encoding (line 390) | def interpolate_pos_encoding(self, embeddings: torch.Tensor, height: i...
    method forward (line 424) | def forward(self, pixel_values: torch.Tensor, interpolate_pos_encoding...
  class GroupViTTextEmbeddings (line 444) | class GroupViTTextEmbeddings(nn.Module):
    method __init__ (line 445) | def __init__(self, config: GroupViTTextConfig):
    method forward (line 455) | def forward(
  class GroupViTStage (line 475) | class GroupViTStage(nn.Module):
    method __init__ (line 478) | def __init__(
    method with_group_token (line 514) | def with_group_token(self):
    method split_x (line 517) | def split_x(self, x):
    method concat_x (line 523) | def concat_x(self, x: torch.Tensor, group_token: Optional[torch.Tensor...
    method forward (line 528) | def forward(
  class GroupViTMLP (line 570) | class GroupViTMLP(nn.Module):
    method __init__ (line 571) | def __init__(
    method forward (line 587) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class GroupViTMixerMLP (line 594) | class GroupViTMixerMLP(GroupViTMLP):
    method forward (line 595) | def forward(self, x):
  class GroupViTAttention (line 600) | class GroupViTAttention(nn.Module):
    method __init__ (line 603) | def __init__(self, config):
    method _shape (line 622) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 625) | def forward(
  class GroupViTEncoderLayer (line 711) | class GroupViTEncoderLayer(nn.Module):
    method __init__ (line 712) | def __init__(self, config: GroupViTConfig):
    method forward (line 720) | def forward(
  class GroupViTPreTrainedModel (line 761) | class GroupViTPreTrainedModel(PreTrainedModel):
    method _init_weights (line 772) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 807) | def _set_gradient_checkpointing(self, module, value=False):
  class GroupViTVisionEncoder (line 908) | class GroupViTVisionEncoder(nn.Module):
    method __init__ (line 909) | def __init__(self, config: GroupViTVisionConfig) -> None:
    method forward (line 926) | def forward(
  class GroupViTTextEncoder (line 966) | class GroupViTTextEncoder(nn.Module):
    method __init__ (line 975) | def __init__(self, config: GroupViTTextConfig):
    method forward (line 981) | def forward(
  function _make_causal_mask (line 1070) | def _make_causal_mask(
  class GroupViTTextTransformer (line 1088) | class GroupViTTextTransformer(nn.Module):
    method __init__ (line 1089) | def __init__(self, config: GroupViTTextConfig):
    method forward (line 1099) | def forward(
  class GroupViTTextModel (line 1165) | class GroupViTTextModel(GroupViTPreTrainedModel):
    method __init__ (line 1168) | def __init__(self, config: GroupViTTextConfig):
    method get_input_embeddings (line 1174) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 1177) | def set_input_embeddings(self, value):
    method forward (line 1182) | def forward(
  class GroupViTVisionTransformer (line 1218) | class GroupViTVisionTransformer(nn.Module):
    method __init__ (line 1219) | def __init__(self, config: GroupViTVisionConfig):
    method forward (line 1230) | def forward(
  class GroupViTVisionModel (line 1276) | class GroupViTVisionModel(GroupViTPreTrainedModel):
    method __init__ (line 1280) | def __init__(self, config: GroupViTVisionConfig):
    method get_input_embeddings (line 1286) | def get_input_embeddings(self) -> GroupViTPatchEmbeddings:
    method forward (line 1291) | def forward(
  class GroupViTModel (line 1329) | class GroupViTModel(GroupViTPreTrainedModel):
    method __init__ (line 1332) | def __init__(self, config: GroupViTConfig):
    method get_text_features (line 1376) | def get_text_features(
    method get_image_features (line 1423) | def get_image_features(
    method forward (line 1473) | def forward(

FILE: transformers/models/groupvit/modeling_tf_groupvit.py
  function _expand_mask (line 78) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  function contrastive_loss (line 93) | def contrastive_loss(logits: tf.Tensor) -> tf.Tensor:
  function groupvit_loss (line 102) | def groupvit_loss(similarity: tf.Tensor) -> tf.Tensor:
  function hard_softmax (line 108) | def hard_softmax(logits: tf.Tensor, dim: int) -> tf.Tensor:
  function gumbel_softmax (line 125) | def gumbel_softmax(logits: tf.Tensor, tau: float = 1, hard: bool = False...
  function resize_attention_map (line 150) | def resize_attention_map(attentions: tf.Tensor, height: int, width: int,...
  function get_grouping_from_attentions (line 188) | def get_grouping_from_attentions(attentions: Tuple[tf.Tensor], hw_shape:...
  class TFGroupViTModelOutput (line 217) | class TFGroupViTModelOutput(ModelOutput):
    method to_tuple (line 260) | def to_tuple(self) -> Tuple[Any]:
  class TFGroupViTCrossAttentionLayer (line 267) | class TFGroupViTCrossAttentionLayer(tf.keras.layers.Layer):
    method __init__ (line 268) | def __init__(self, config: GroupViTVisionConfig, **kwargs):
    method call (line 275) | def call(self, query: tf.Tensor, key: tf.Tensor, training: bool = Fals...
  class TFGroupViTAssignAttention (line 283) | class TFGroupViTAssignAttention(tf.keras.layers.Layer):
    method __init__ (line 284) | def __init__(self, config: GroupViTVisionConfig, **kwargs):
    method get_attn (line 294) | def get_attn(self, attn: tf.Tensor, gumbel: bool = True, hard: bool = ...
    method call (line 305) | def call(self, query: tf.Tensor, key: tf.Tensor, training: bool = False):
  class TFGroupViTTokenAssign (line 331) | class TFGroupViTTokenAssign(tf.keras.layers.Layer):
    method __init__ (line 332) | def __init__(self, config: GroupViTVisionConfig, num_group_token: int,...
    method project_group_token (line 357) | def project_group_token(self, group_tokens: tf.Tensor) -> tf.Tensor:
    method call (line 370) | def call(self, image_tokens: tf.Tensor, group_tokens: tf.Tensor, train...
  class TFGroupViTPatchEmbeddings (line 391) | class TFGroupViTPatchEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 398) | def __init__(self, config: GroupViTConfig, **kwargs):
    method call (line 426) | def call(
  class TFGroupViTVisionEmbeddings (line 462) | class TFGroupViTVisionEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 468) | def __init__(self, config: GroupViTVisionConfig, **kwargs):
    method build (line 476) | def build(self, input_shape: tf.TensorShape):
    method interpolate_pos_encoding (line 487) | def interpolate_pos_encoding(self, embeddings, height, width) -> tf.Te...
    method call (line 514) | def call(
  class TFGroupViTTextEmbeddings (line 533) | class TFGroupViTTextEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 534) | def __init__(self, config: GroupViTTextConfig, **kwargs):
    method build (line 541) | def build(self, input_shape: tf.TensorShape = None):
    method call (line 560) | def call(
  class TFGroupViTStage (line 591) | class TFGroupViTStage(tf.keras.layers.Layer):
    method __init__ (line 594) | def __init__(
    method build (line 629) | def build(self, input_shape: tf.TensorShape):
    method with_group_token (line 642) | def with_group_token(self):
    method split_x (line 645) | def split_x(self, x: tf.Tensor) -> tf.Tensor:
    method concat_x (line 651) | def concat_x(self, x: tf.Tensor, group_token: tf.Tensor | None = None)...
    method call (line 656) | def call(
  class TFGroupViTMLP (line 706) | class TFGroupViTMLP(tf.keras.layers.Layer):
    method __init__ (line 707) | def __init__(
    method call (line 724) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFGroupViTMixerMLP (line 731) | class TFGroupViTMixerMLP(TFGroupViTMLP):
    method call (line 732) | def call(self, x, training: bool = False):
  class TFGroupViTAttention (line 738) | class TFGroupViTAttention(tf.keras.layers.Layer):
    method __init__ (line 741) | def __init__(self, config: GroupViTConfig, **kwargs):
    method transpose_for_scores (line 776) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 783) | def call(
  class TFGroupViTEncoderLayer (line 846) | class TFGroupViTEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 847) | def __init__(self, config: GroupViTConfig, **kwargs):
    method call (line 856) | def call(
  class TFGroupViTTextEncoder (line 899) | class TFGroupViTTextEncoder(tf.keras.layers.Layer):
    method __init__ (line 900) | def __init__(self, config: GroupViTTextConfig, **kwargs):
    method call (line 905) | def call(
  class TFGroupViTVisionEncoder (line 943) | class TFGroupViTVisionEncoder(tf.keras.layers.Layer):
    method __init__ (line 944) | def __init__(self, config: GroupViTVisionConfig, **kwargs) -> None:
    method call (line 959) | def call(
  class TFGroupViTTextTransformer (line 995) | class TFGroupViTTextTransformer(tf.keras.layers.Layer):
    method __init__ (line 996) | def __init__(self, config: GroupViTTextConfig, **kwargs):
    method call (line 1005) | def call(
    method _build_causal_attention_mask (line 1060) | def _build_causal_attention_mask(self, batch_size, seq_length, dtype=t...
  class TFGroupViTVisionTransformer (line 1080) | class TFGroupViTVisionTransformer(tf.keras.layers.Layer):
    method __init__ (line 1081) | def __init__(self, config: GroupViTVisionConfig, **kwargs):
    method call (line 1088) | def call(
  class TFGroupViTTextMainLayer (line 1124) | class TFGroupViTTextMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1127) | def __init__(self, config: GroupViTTextConfig, **kwargs):
    method get_input_embeddings (line 1132) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 1135) | def set_input_embeddings(self, value: tf.Variable):
    method call (line 1140) | def call(
  class TFGroupViTVisionMainLayer (line 1173) | class TFGroupViTVisionMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1176) | def __init__(self, config: GroupViTVisionConfig, **kwargs):
    method get_input_embeddings (line 1181) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method call (line 1185) | def call(
  class TFGroupViTMainLayer (line 1209) | class TFGroupViTMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1212) | def __init__(self, config: GroupViTConfig, **kwargs):
    method build (line 1253) | def build(self, input_shape: tf.TensorShape):
    method get_text_features (line 1264) | def get_text_features(
    method get_image_features (line 1300) | def get_image_features(
    method call (line 1327) | def call(
  class TFGroupViTPreTrainedModel (line 1454) | class TFGroupViTPreTrainedModel(TFPreTrainedModel):
  class TFGroupViTTextModel (line 1601) | class TFGroupViTTextModel(TFGroupViTPreTrainedModel):
    method __init__ (line 1605) | def __init__(self, config: GroupViTTextConfig, *inputs, **kwargs):
    method call (line 1613) | def call(
  class TFGroupViTVisionModel (line 1654) | class TFGroupViTVisionModel(TFGroupViTPreTrainedModel):
    method __init__ (line 1658) | def __init__(self, config: GroupViTVisionConfig, *inputs, **kwargs):
    method call (line 1666) | def call(
  class TFGroupViTModel (line 1709) | class TFGroupViTModel(TFGroupViTPreTrainedModel):
    method __init__ (line 1712) | def __init__(self, config: GroupViTConfig, *inputs, **kwargs):
    method get_text_features (line 1719) | def get_text_features(
    method get_image_features (line 1760) | def get_image_features(
    method call (line 1804) | def call(
    method serving_output (line 1858) | def serving_output(self, output: TFGroupViTModelOutput) -> TFGroupViTM...

FILE: transformers/models/herbert/tokenization_herbert.py
  function get_pairs (line 46) | def get_pairs(word):
  function replace_unicode_punct (line 60) | def replace_unicode_punct(text):
  function remove_non_printing_char (line 104) | def remove_non_printing_char(text):
  function whitespace_tokenize (line 118) | def whitespace_tokenize(text):
  class BasicTokenizer (line 128) | class BasicTokenizer(object):
    method __init__ (line 148) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 156) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 193) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 204) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 226) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 239) | def _is_chinese_char(self, cp):
    method _clean_text (line 263) | def _clean_text(self, text):
  class HerbertTokenizer (line 277) | class HerbertTokenizer(PreTrainedTokenizer):
    method __init__ (line 297) | def __init__(
    method do_lower_case (line 383) | def do_lower_case(self):
    method moses_punct_norm (line 387) | def moses_punct_norm(self, text, lang):
    method moses_tokenize (line 396) | def moses_tokenize(self, text, lang):
    method moses_pipeline (line 405) | def moses_pipeline(self, text, lang):
    method ja_tokenize (line 412) | def ja_tokenize(self, text):
    method vocab_size (line 435) | def vocab_size(self):
    method get_vocab (line 439) | def get_vocab(self):
    method bpe (line 443) | def bpe(self, token):
    method _tokenize (line 487) | def _tokenize(self, text):
    method _convert_token_to_id (line 498) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 503) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 508) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 514) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 542) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 571) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 601) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method __getstate__ (line 630) | def __getstate__(self):
    method __setstate__ (line 636) | def __setstate__(self, d):

FILE: transformers/models/herbert/tokenization_herbert_fast.py
  class HerbertTokenizerFast (line 40) | class HerbertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 65) | def __init__(
    method build_inputs_with_special_tokens (line 89) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 116) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 143) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 171) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/hubert/configuration_hubert.py
  class HubertConfig (line 32) | class HubertConfig(PretrainedConfig):
    method __init__ (line 162) | def __init__(
    method inputs_to_logits_ratio (line 259) | def inputs_to_logits_ratio(self):

FILE: transformers/models/hubert/convert_distilhubert_original_s3prl_checkpoint_to_pytorch.py
  function set_recursively (line 45) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function recursively_load_weights (line 73) | def recursively_load_weights(fairseq_model, hf_model):
  function load_conv_layer (line 117) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function convert_config (line 157) | def convert_config(model):
  function convert_hubert_checkpoint (line 191) | def convert_hubert_checkpoint(pytorch_dump_folder_path, config_path=None):

FILE: transformers/models/hubert/convert_hubert_original_pytorch_checkpoint_to_pytorch.py
  function set_recursively (line 58) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function recursively_load_weights (line 86) | def recursively_load_weights(fairseq_model, hf_model, is_finetuned):
  function load_conv_layer (line 130) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function convert_hubert_checkpoint (line 171) | def convert_hubert_checkpoint(

FILE: transformers/models/hubert/convert_hubert_original_s3prl_checkpoint_to_pytorch.py
  function convert_s3prl_checkpoint (line 32) | def convert_s3prl_checkpoint(base_model_name, config_path, checkpoint_pa...

FILE: transformers/models/hubert/modeling_hubert.py
  function _compute_mask_indices (line 68) | def _compute_mask_indices(
  class HubertNoLayerNormConvLayer (line 188) | class HubertNoLayerNormConvLayer(nn.Module):
    method __init__ (line 189) | def __init__(self, config, layer_id=0):
    method forward (line 203) | def forward(self, hidden_states):
  class HubertLayerNormConvLayer (line 210) | class HubertLayerNormConvLayer(nn.Module):
    method __init__ (line 211) | def __init__(self, config, layer_id=0):
    method forward (line 226) | def forward(self, hidden_states):
  class HubertGroupNormConvLayer (line 238) | class HubertGroupNormConvLayer(nn.Module):
    method __init__ (line 239) | def __init__(self, config, layer_id=0):
    method forward (line 255) | def forward(self, hidden_states):
  class HubertPositionalConvEmbedding (line 263) | class HubertPositionalConvEmbedding(nn.Module):
    method __init__ (line 264) | def __init__(self, config):
    method forward (line 291) | def forward(self, hidden_states):
  class HubertSamePadLayer (line 303) | class HubertSamePadLayer(nn.Module):
    method __init__ (line 304) | def __init__(self, num_conv_pos_embeddings):
    method forward (line 308) | def forward(self, hidden_states):
  class HubertFeatureEncoder (line 315) | class HubertFeatureEncoder(nn.Module):
    method __init__ (line 318) | def __init__(self, config):
    method _freeze_parameters (line 335) | def _freeze_parameters(self):
    method forward (line 340) | def forward(self, input_values):
  class HubertFeatureExtractor (line 366) | class HubertFeatureExtractor(HubertFeatureEncoder):
    method __init__ (line 367) | def __init__(self, config):
  class HubertFeatureProjection (line 377) | class HubertFeatureProjection(nn.Module):
    method __init__ (line 378) | def __init__(self, config):
    method forward (line 386) | def forward(self, hidden_states):
  class HubertAttention (line 396) | class HubertAttention(nn.Module):
    method __init__ (line 399) | def __init__(
    method _shape (line 426) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 429) | def forward(
  class HubertFeedForward (line 551) | class HubertFeedForward(nn.Module):
    method __init__ (line 552) | def __init__(self, config):
    method forward (line 565) | def forward(self, hidden_states):
  class HubertEncoderLayer (line 576) | class HubertEncoderLayer(nn.Module):
    method __init__ (line 577) | def __init__(self, config):
    method forward (line 590) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class HubertAttnAdapterLayer (line 611) | class HubertAttnAdapterLayer(nn.Module):
    method __init__ (line 612) | def __init__(self, config):
    method forward (line 626) | def forward(self, hidden_states: torch.FloatTensor):
  class HubertEncoderLayerStableLayerNorm (line 637) | class HubertEncoderLayerStableLayerNorm(nn.Module):
    method __init__ (line 638) | def __init__(self, config):
    method forward (line 656) | def forward(
  class HubertEncoder (line 683) | class HubertEncoder(nn.Module):
    method __init__ (line 684) | def __init__(self, config):
    method forward (line 693) | def forward(
  class HubertEncoderStableLayerNorm (line 771) | class HubertEncoderStableLayerNorm(nn.Module):
    method __init__ (line 772) | def __init__(self, config):
    method forward (line 783) | def forward(
  class HubertPreTrainedModel (line 862) | class HubertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 874) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 899) | def _set_gradient_checkpointing(self, module, value=False):
    method _get_feat_extract_output_lengths (line 903) | def _get_feat_extract_output_lengths(self, input_lengths: Union[torch....
    method _get_feature_vector_attention_mask (line 918) | def _get_feature_vector_attention_mask(self, feature_vector_length: in...
  class HubertModel (line 992) | class HubertModel(HubertPreTrainedModel):
    method __init__ (line 993) | def __init__(self, config: HubertConfig):
    method _mask_hidden_states (line 1011) | def _mask_hidden_states(
    method forward (line 1059) | def forward(
  class HubertForCTC (line 1136) | class HubertForCTC(HubertPreTrainedModel):
    method __init__ (line 1137) | def __init__(self, config, target_lang=None):
    method freeze_feature_extractor (line 1165) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1177) | def freeze_feature_encoder(self):
    method forward (line 1192) | def forward(
  class HubertForSequenceClassification (line 1272) | class HubertForSequenceClassification(HubertPreTrainedModel):
    method __init__ (line 1273) | def __init__(self, config):
    method freeze_feature_extractor (line 1290) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1302) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1309) | def freeze_base_model(self):
    method forward (line 1326) | def forward(

FILE: transformers/models/hubert/modeling_tf_hubert.py
  function _sample_without_replacement (line 56) | def _sample_without_replacement(distribution, num_samples):
  function _scatter_values_on_batch_indices (line 67) | def _scatter_values_on_batch_indices(values, batch_indices, output_shape):
  function _compute_mask_indices (line 83) | def _compute_mask_indices(
  function _expand_mask (line 158) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFHubertGroupNorm (line 172) | class TFHubertGroupNorm(tf.keras.layers.Layer):
    method __init__ (line 177) | def __init__(
    method build (line 207) | def build(self, input_shape):
    method call (line 218) | def call(self, inputs):
    method get_config (line 234) | def get_config(self):
    method compute_output_shape (line 251) | def compute_output_shape(self, input_shape):
    method _reshape_into_groups (line 254) | def _reshape_into_groups(self, inputs, input_shape, tensor_input_shape):
    method _apply_normalization (line 266) | def _apply_normalization(self, reshaped_inputs, input_shape):
    method _get_reshaped_weights (line 289) | def _get_reshaped_weights(self, input_shape):
    method _check_if_input_shape_is_none (line 300) | def _check_if_input_shape_is_none(self, input_shape):
    method _set_number_of_groups_for_instance_norm (line 311) | def _set_number_of_groups_for_instance_norm(self, input_shape):
    method _check_size_of_dimensions (line 317) | def _check_size_of_dimensions(self, input_shape):
    method _check_axis (line 337) | def _check_axis(self):
    method _create_input_spec (line 343) | def _create_input_spec(self, input_shape):
    method _add_gamma_weight (line 347) | def _add_gamma_weight(self, input_shape):
    method _add_beta_weight (line 362) | def _add_beta_weight(self, input_shape):
    method _create_broadcast_shape (line 377) | def _create_broadcast_shape(self, input_shape):
  class TFHubertWeightNormConv1D (line 389) | class TFHubertWeightNormConv1D(tf.keras.layers.Conv1D):
    method __init__ (line 392) | def __init__(self, filters, kernel_size, groups, explicit_padding, **k...
    method _init_norm (line 407) | def _init_norm(self):
    method _normalize_kernel (line 412) | def _normalize_kernel(self):
    method build (line 417) | def build(self, input_shape):
    method call (line 436) | def call(self, inputs):
  class TFHubertNoLayerNormConvLayer (line 450) | class TFHubertNoLayerNormConvLayer(tf.keras.layers.Layer):
    method __init__ (line 451) | def __init__(self, config: HubertConfig, layer_id: int = 0, **kwargs: ...
    method call (line 465) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFHubertLayerNormConvLayer (line 472) | class TFHubertLayerNormConvLayer(tf.keras.layers.Layer):
    method __init__ (line 473) | def __init__(self, config: HubertConfig, layer_id: int = 0, **kwargs: ...
    method call (line 488) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFHubertGroupNormConvLayer (line 496) | class TFHubertGroupNormConvLayer(tf.keras.layers.Layer):
    method __init__ (line 497) | def __init__(self, config: HubertConfig, layer_id: int = 0, **kwargs: ...
    method call (line 512) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFHubertPositionalConvEmbedding (line 520) | class TFHubertPositionalConvEmbedding(tf.keras.layers.Layer):
    method __init__ (line 521) | def __init__(self, config: HubertConfig, **kwargs: Any) -> None:
    method call (line 533) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFHubertSamePadLayer (line 541) | class TFHubertSamePadLayer(tf.keras.layers.Layer):
    method __init__ (line 542) | def __init__(self, num_conv_pos_embeddings, **kwargs):
    method call (line 546) | def call(self, hidden_states):
  class TFHubertFeatureEncoder (line 552) | class TFHubertFeatureEncoder(tf.keras.layers.Layer):
    method __init__ (line 553) | def __init__(self, config: HubertConfig, **kwargs: Any) -> None:
    method call (line 572) | def call(self, input_values):
  class TFHubertFeatureExtractor (line 579) | class TFHubertFeatureExtractor(TFHubertFeatureEncoder):
    method __init__ (line 580) | def __init__(self, config, **kwargs):
  class TFHubertFeatureProjection (line 590) | class TFHubertFeatureProjection(tf.keras.layers.Layer):
    method __init__ (line 591) | def __init__(self, config: HubertConfig, **kwargs):
    method call (line 603) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFHubertAttention (line 611) | class TFHubertAttention(tf.keras.layers.Layer):
    method __init__ (line 614) | def __init__(
    method _shape (line 642) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 645) | def call(
  class TFHubertFeedForward (line 765) | class TFHubertFeedForward(tf.keras.layers.Layer):
    method __init__ (line 766) | def __init__(self, config: HubertConfig, **kwargs):
    method call (line 787) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFHubertEncoderLayer (line 798) | class TFHubertEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 799) | def __init__(self, config: HubertConfig, **kwargs):
    method call (line 815) | def call(
  class TFHubertEncoderLayerStableLayerNorm (line 842) | class TFHubertEncoderLayerStableLayerNorm(tf.keras.layers.Layer):
    method __init__ (line 843) | def __init__(self, config: HubertConfig, **kwargs):
    method call (line 859) | def call(
  class TFHubertEncoder (line 884) | class TFHubertEncoder(tf.keras.layers.Layer):
    method __init__ (line 885) | def __init__(self, config: HubertConfig, **kwargs):
    method call (line 893) | def call(
  class TFHubertEncoderStableLayerNorm (line 950) | class TFHubertEncoderStableLayerNorm(tf.keras.layers.Layer):
    method __init__ (line 951) | def __init__(self, config: HubertConfig, **kwargs):
    method call (line 961) | def call(
  class TFHubertMainLayer (line 1018) | class TFHubertMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1021) | def __init__(self, config: HubertConfig, **kwargs):
    method build (line 1032) | def build(self, input_shape: tf.TensorShape):
    method _get_feat_extract_output_lengths (line 1039) | def _get_feat_extract_output_lengths(self, input_lengths: tf.Tensor):
    method _mask_hidden_states (line 1054) | def _mask_hidden_states(self, hidden_states: tf.Tensor, mask_time_indi...
    method call (line 1099) | def call(
  class TFHubertPreTrainedModel (line 1149) | class TFHubertPreTrainedModel(TFPreTrainedModel):
    method input_signature (line 1160) | def input_signature(self):
    method __init__ (line 1167) | def __init__(self, config, *inputs, **kwargs):
  class TFHubertModel (line 1277) | class TFHubertModel(TFHubertPreTrainedModel):
    method __init__ (line 1278) | def __init__(self, config: HubertConfig, *inputs, **kwargs):
    method call (line 1286) | def call(
  class TFHubertForCTC (line 1351) | class TFHubertForCTC(TFHubertPreTrainedModel):
    method __init__ (line 1352) | def __init__(self, config: HubertConfig, *inputs, **kwargs):
    method freeze_feature_extractor (line 1359) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1371) | def freeze_feature_encoder(self):
    method call (line 1381) | def call(

FILE: transformers/models/ibert/configuration_ibert.py
  class IBertConfig (line 37) | class IBertConfig(PretrainedConfig):
    method __init__ (line 93) | def __init__(
  class IBertOnnxConfig (line 134) | class IBertOnnxConfig(OnnxConfig):
    method inputs (line 136) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/ibert/modeling_ibert.py
  class IBertEmbeddings (line 57) | class IBertEmbeddings(nn.Module):
    method __init__ (line 62) | def __init__(self, config):
    method forward (line 112) | def forward(
    method create_position_ids_from_inputs_embeds (line 159) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class IBertSelfAttention (line 177) | class IBertSelfAttention(nn.Module):
    method __init__ (line 178) | def __init__(self, config):
    method transpose_for_scores (line 236) | def transpose_for_scores(self, x):
    method forward (line 241) | def forward(
  class IBertSelfOutput (line 319) | class IBertSelfOutput(nn.Module):
    method __init__ (line 320) | def __init__(self, config):
    method forward (line 349) | def forward(self, hidden_states, hidden_states_scaling_factor, input_t...
  class IBertAttention (line 366) | class IBertAttention(nn.Module):
    method __init__ (line 367) | def __init__(self, config):
    method prune_heads (line 374) | def prune_heads(self, heads):
    method forward (line 392) | def forward(
  class IBertIntermediate (line 415) | class IBertIntermediate(nn.Module):
    method __init__ (line 416) | def __init__(self, config):
    method forward (line 436) | def forward(self, hidden_states, hidden_states_scaling_factor):
  class IBertOutput (line 449) | class IBertOutput(nn.Module):
    method __init__ (line 450) | def __init__(self, config):
    method forward (line 479) | def forward(self, hidden_states, hidden_states_scaling_factor, input_t...
  class IBertLayer (line 496) | class IBertLayer(nn.Module):
    method __init__ (line 497) | def __init__(self, config):
    method forward (line 510) | def forward(
    method feed_forward_chunk (line 537) | def feed_forward_chunk(self, attention_output, attention_output_scalin...
  class IBertEncoder (line 554) | class IBertEncoder(nn.Module):
    method __init__ (line 555) | def __init__(self, config):
    method forward (line 561) | def forward(
  class IBertPooler (line 618) | class IBertPooler(nn.Module):
    method __init__ (line 619) | def __init__(self, config):
    method forward (line 625) | def forward(self, hidden_states):
  class IBertPreTrainedModel (line 634) | class IBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 643) | def _init_weights(self, module):
    method resize_token_embeddings (line 659) | def resize_token_embeddings(self, new_num_tokens=None):
  class IBertModel (line 733) | class IBertModel(IBertPreTrainedModel):
    method __init__ (line 745) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 758) | def get_input_embeddings(self):
    method set_input_embeddings (line 761) | def set_input_embeddings(self, value):
    method _prune_heads (line 764) | def _prune_heads(self, heads_to_prune):
    method forward (line 778) | def forward(
  class IBertForMaskedLM (line 856) | class IBertForMaskedLM(IBertPreTrainedModel):
    method __init__ (line 860) | def __init__(self, config):
    method get_output_embeddings (line 869) | def get_output_embeddings(self):
    method set_output_embeddings (line 872) | def set_output_embeddings(self, new_embeddings):
    method forward (line 882) | def forward(
  class IBertLMHead (line 936) | class IBertLMHead(nn.Module):
    method __init__ (line 939) | def __init__(self, config):
    method forward (line 948) | def forward(self, features, **kwargs):
    method _tie_weights (line 958) | def _tie_weights(self):
  class IBertForSequenceClassification (line 970) | class IBertForSequenceClassification(IBertPreTrainedModel):
    method __init__ (line 973) | def __init__(self, config):
    method forward (line 989) | def forward(
  class IBertForMultipleChoice (line 1065) | class IBertForMultipleChoice(IBertPreTrainedModel):
    method __init__ (line 1068) | def __init__(self, config):
    method forward (line 1084) | def forward(
  class IBertForTokenClassification (line 1157) | class IBertForTokenClassification(IBertPreTrainedModel):
    method __init__ (line 1161) | def __init__(self, config):
    method forward (line 1178) | def forward(
  class IBertClassificationHead (line 1231) | class IBertClassificationHead(nn.Module):
    method __init__ (line 1234) | def __init__(self, config):
    method forward (line 1240) | def forward(self, features, **kwargs):
  class IBertForQuestionAnswering (line 1257) | class IBertForQuestionAnswering(IBertPreTrainedModel):
    method __init__ (line 1261) | def __init__(self, config):
    method forward (line 1277) | def forward(
  function create_position_ids_from_input_ids (line 1352) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/ibert/quant_modules.py
  class QuantEmbedding (line 31) | class QuantEmbedding(nn.Module):
    method __init__ (line 44) | def __init__(
    method forward (line 77) | def forward(self, x, positions=None, incremental_state=None):
  class QuantAct (line 114) | class QuantAct(nn.Module):
    method __init__ (line 131) | def __init__(self, activation_bit, act_range_momentum=0.95, per_channe...
    method __repr__ (line 150) | def __repr__(self):
    method forward (line 157) | def forward(
  class QuantLinear (line 220) | class QuantLinear(nn.Module):
    method __init__ (line 235) | def __init__(
    method __repr__ (line 257) | def __repr__(self):
    method forward (line 262) | def forward(self, x, prev_act_scaling_factor=None):
  class IntGELU (line 300) | class IntGELU(nn.Module):
    method __init__ (line 311) | def __init__(self, quant_mode=True, force_dequant="none"):
    method int_erf (line 327) | def int_erf(self, x_int, scaling_factor):
    method forward (line 342) | def forward(self, x, scaling_factor=None):
  class IntSoftmax (line 357) | class IntSoftmax(nn.Module):
    method __init__ (line 370) | def __init__(self, output_bit, quant_mode=False, force_dequant="none"):
    method int_polynomial (line 387) | def int_polynomial(self, x_int, scaling_factor):
    method int_exp (line 395) | def int_exp(self, x_int, scaling_factor):
    method forward (line 407) | def forward(self, x, scaling_factor):
  class IntLayerNorm (line 428) | class IntLayerNorm(nn.Module):
    method __init__ (line 441) | def __init__(self, normalized_shape, eps, output_bit=8, quant_mode=Fal...
    method set_shift (line 460) | def set_shift(self, y_int):
    method overflow_fallback (line 469) | def overflow_fallback(self, y_int):
    method forward (line 480) | def forward(self, x, scaling_factor=None):
  function get_percentile_min_max (line 529) | def get_percentile_min_max(input, lower_percentile, upper_percentile, ou...
  function linear_quantize (line 565) | def linear_quantize(input, scale, zero_point, inplace=False):
  function symmetric_linear_quantization_params (line 600) | def symmetric_linear_quantization_params(num_bits, saturation_min, satur...
  class SymmetricQuantFunction (line 632) | class SymmetricQuantFunction(Function):
    method forward (line 638) | def forward(ctx, x, k, percentile_mode, scale):
    method backward (line 664) | def backward(ctx, grad_output):
  class floor_ste (line 677) | class floor_ste(Function):
    method forward (line 683) | def forward(ctx, x):
    method backward (line 687) | def backward(ctx, grad_output):
  class round_ste (line 691) | class round_ste(Function):
    method forward (line 697) | def forward(ctx, x):
    method backward (line 701) | def backward(ctx, grad_output):
  function batch_frexp (line 705) | def batch_frexp(inputs, max_bit=31):
  class FixedPointMul (line 739) | class FixedPointMul(Function):
    method forward (line 763) | def forward(
    method backward (line 816) | def backward(ctx, grad_output):

FILE: transformers/models/imagegpt/configuration_imagegpt.py
  class ImageGPTConfig (line 37) | class ImageGPTConfig(PretrainedConfig):
    method __init__ (line 110) | def __init__(
  class ImageGPTOnnxConfig (line 152) | class ImageGPTOnnxConfig(OnnxConfig):
    method inputs (line 154) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method generate_dummy_inputs (line 161) | def generate_dummy_inputs(

FILE: transformers/models/imagegpt/convert_imagegpt_original_tf2_to_pytorch.py
  function convert_imagegpt_checkpoint_to_pytorch (line 29) | def convert_imagegpt_checkpoint_to_pytorch(imagegpt_checkpoint_path, mod...

FILE: transformers/models/imagegpt/feature_extraction_imagegpt.py
  class ImageGPTFeatureExtractor (line 26) | class ImageGPTFeatureExtractor(ImageGPTImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/imagegpt/image_processing_imagegpt.py
  function squared_euclidean_distance (line 41) | def squared_euclidean_distance(a, b):
  function color_quantize (line 50) | def color_quantize(x, clusters):
  class ImageGPTImageProcessor (line 56) | class ImageGPTImageProcessor(BaseImageProcessor):
    method __init__ (line 82) | def __init__(
    method resize (line 103) | def resize(
    method normalize (line 131) | def normalize(
    method preprocess (line 149) | def preprocess(

FILE: transformers/models/imagegpt/modeling_imagegpt.py
  function load_tf_weights_in_imagegpt (line 53) | def load_tf_weights_in_imagegpt(model, config, imagegpt_checkpoint_path):
  class ImageGPTLayerNorm (line 161) | class ImageGPTLayerNorm(nn.Module):
    method __init__ (line 162) | def __init__(self, hidden_size: Tuple[int], eps: float = 1e-5):
    method forward (line 167) | def forward(self, tensor: torch.Tensor) -> tuple:
  class ImageGPTAttention (line 176) | class ImageGPTAttention(nn.Module):
    method __init__ (line 177) | def __init__(self, config, is_cross_attention: Optional[bool] = False,...
    method prune_heads (line 219) | def prune_heads(self, heads):
    method _attn (line 234) | def _attn(self, query, key, value, attention_mask=None, head_mask=None):
    method _upcast_and_reordered_attn (line 272) | def _upcast_and_reordered_attn(self, query, key, value, attention_mask...
    method _split_heads (line 324) | def _split_heads(self, tensor, num_heads, attn_head_size):
    method _merge_heads (line 332) | def _merge_heads(self, tensor, num_heads, attn_head_size):
    method forward (line 340) | def forward(
  class ImageGPTMLP (line 394) | class ImageGPTMLP(nn.Module):
    method __init__ (line 395) | def __init__(self, intermediate_size, config):
    method forward (line 403) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ImageGPTBlock (line 411) | class ImageGPTBlock(nn.Module):
    method __init__ (line 412) | def __init__(self, config, layer_idx=None):
    method forward (line 427) | def forward(
  class ImageGPTPreTrainedModel (line 486) | class ImageGPTPreTrainedModel(PreTrainedModel):
    method __init__ (line 498) | def __init__(self, *inputs, **kwargs):
    method _init_weights (line 501) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 527) | def _set_gradient_checkpointing(self, module, value=False):
  class ImageGPTModel (line 615) | class ImageGPTModel(ImageGPTPreTrainedModel):
    method __init__ (line 618) | def __init__(self, config: ImageGPTConfig):
    method get_input_embeddings (line 637) | def get_input_embeddings(self):
    method set_input_embeddings (line 640) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 643) | def _prune_heads(self, heads_to_prune):
    method forward (line 652) | def forward(
  class ImageGPTForCausalImageModeling (line 895) | class ImageGPTForCausalImageModeling(ImageGPTPreTrainedModel):
    method __init__ (line 898) | def __init__(self, config: ImageGPTConfig):
    method get_output_embeddings (line 909) | def get_output_embeddings(self):
    method set_output_embeddings (line 912) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 915) | def prepare_inputs_for_generation(self, input_ids: torch.Tensor, past_...
    method forward (line 945) | def forward(
    method _reorder_cache (line 1065) | def _reorder_cache(
  class ImageGPTForImageClassification (line 1086) | class ImageGPTForImageClassification(ImageGPTPreTrainedModel):
    method __init__ (line 1089) | def __init__(self, config: ImageGPTConfig):
    method forward (line 1100) | def forward(

FILE: transformers/models/informer/configuration_informer.py
  class InformerConfig (line 33) | class InformerConfig(PretrainedConfig):
    method __init__ (line 143) | def __init__(
    method _number_of_features (line 245) | def _number_of_features(self) -> int:

FILE: transformers/models/informer/modeling_informer.py
  class InformerFeatureEmbedder (line 50) | class InformerFeatureEmbedder(nn.Module):
    method __init__ (line 61) | def __init__(self, cardinalities: List[int], embedding_dims: List[int]...
    method forward (line 67) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class InformerStdScaler (line 85) | class InformerStdScaler(nn.Module):
    method __init__ (line 99) | def __init__(self, dim: int, keepdim: bool = False, minimum_scale: flo...
    method forward (line 108) | def forward(self, data: torch.Tensor, weights: torch.Tensor) -> Tuple[...
  class InformerMeanScaler (line 119) | class InformerMeanScaler(nn.Module):
    method __init__ (line 135) | def __init__(
    method forward (line 145) | def forward(
  class InformerNOPScaler (line 177) | class InformerNOPScaler(nn.Module):
    method __init__ (line 188) | def __init__(self, dim: int, keepdim: bool = False):
    method forward (line 193) | def forward(
  function weighted_average (line 202) | def weighted_average(input_tensor: torch.Tensor, weights: Optional[torch...
  function nll (line 227) | def nll(input: torch.distributions.Distribution, target: torch.Tensor) -...
  function _make_causal_mask (line 235) | def _make_causal_mask(
  function _expand_mask (line 253) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class InformerSinusoidalPositionalEmbedding (line 268) | class InformerSinusoidalPositionalEmbedding(nn.Embedding):
    method __init__ (line 271) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method _init_weight (line 276) | def _init_weight(out: nn.Parameter) -> nn.Parameter:
    method forward (line 293) | def forward(self, input_ids_shape: torch.Size, past_key_values_length:...
  class InformerValueEmbedding (line 303) | class InformerValueEmbedding(nn.Module):
    method __init__ (line 304) | def __init__(self, feature_size, d_model):
    method forward (line 308) | def forward(self, x):
  class InformerAttention (line 313) | class InformerAttention(nn.Module):
    method __init__ (line 316) | def __init__(
    method _shape (line 343) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 346) | def forward(
  class InformerProbSparseAttention (line 467) | class InformerProbSparseAttention(nn.Module):
    method __init__ (line 472) | def __init__(
    method _shape (line 501) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 504) | def forward(
  class InformerConvLayer (line 685) | class InformerConvLayer(nn.Module):
    method __init__ (line 686) | def __init__(self, c_in):
    method forward (line 699) | def forward(self, x):
  class InformerEncoderLayer (line 708) | class InformerEncoderLayer(nn.Module):
    method __init__ (line 709) | def __init__(self, config: InformerConfig):
    method forward (line 733) | def forward(
  class InformerDecoderLayer (line 784) | class InformerDecoderLayer(nn.Module):
    method __init__ (line 785) | def __init__(self, config: InformerConfig):
    method forward (line 820) | def forward(
  class InformerPreTrainedModel (line 910) | class InformerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 916) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 927) | def _set_gradient_checkpointing(self, module, value=False):
  class InformerEncoder (line 1103) | class InformerEncoder(InformerPreTrainedModel):
    method __init__ (line 1112) | def __init__(self, config: InformerConfig):
    method forward (line 1139) | def forward(
  class InformerDecoder (line 1256) | class InformerDecoder(InformerPreTrainedModel):
    method __init__ (line 1264) | def __init__(self, config: InformerConfig):
    method _prepare_decoder_attention_mask (line 1282) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 1305) | def forward(
  class InformerModel (line 1503) | class InformerModel(InformerPreTrainedModel):
    method __init__ (line 1504) | def __init__(self, config: InformerConfig):
    method _past_length (line 1528) | def _past_length(self) -> int:
    method get_lagged_subsequences (line 1531) | def get_lagged_subsequences(
    method create_network_inputs (line 1563) | def create_network_inputs(
    method get_encoder (line 1635) | def get_encoder(self):
    method get_decoder (line 1638) | def get_decoder(self):
    method forward (line 1643) | def forward(
  class InformerForPrediction (line 1765) | class InformerForPrediction(InformerPreTrainedModel):
    method __init__ (line 1766) | def __init__(self, config: InformerConfig):
    method output_params (line 1789) | def output_params(self, dec_output):
    method get_encoder (line 1792) | def get_encoder(self):
    method get_decoder (line 1795) | def get_decoder(self):
    method output_distribution (line 1799) | def output_distribution(self, params, loc=None, scale=None, trailing_n...
    method forward (line 1807) | def forward(
    method generate (line 1938) | def generate(

FILE: transformers/models/jukebox/configuration_jukebox.py
  function full_dense_attention (line 118) | def full_dense_attention(layer):
  function raw_column_previous_row_attention (line 122) | def raw_column_previous_row_attention(layer):
  function large_separated_enc_dec_w_lyrics (line 126) | def large_separated_enc_dec_w_lyrics(layer):
  function enc_dec_with_lyrics (line 130) | def enc_dec_with_lyrics(layer):
  class JukeboxPriorConfig (line 144) | class JukeboxPriorConfig(PretrainedConfig):
    method __init__ (line 260) | def __init__(
    method from_pretrained (line 353) | def from_pretrained(
    method to_dict (line 370) | def to_dict(self):
  class JukeboxVQVAEConfig (line 383) | class JukeboxVQVAEConfig(PretrainedConfig):
    method __init__ (line 440) | def __init__(
    method from_pretrained (line 488) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class JukeboxConfig (line 504) | class JukeboxConfig(PretrainedConfig):
    method __init__ (line 562) | def __init__(
    method from_configs (line 609) | def from_configs(cls, prior_configs: List[JukeboxPriorConfig], vqvae_c...
    method to_dict (line 620) | def to_dict(self):

FILE: transformers/models/jukebox/convert_jukebox.py
  function replace_key (line 50) | def replace_key(key):
  function fix_jukebox_keys (line 96) | def fix_jukebox_keys(state_dict, model_state_dict, key_prefix, mapping):
  function convert_openai_checkpoint (line 213) | def convert_openai_checkpoint(model_name=None, pytorch_dump_folder_path=...

FILE: transformers/models/jukebox/modeling_jukebox.py
  function filter_logits (line 43) | def filter_logits(logits, top_k=0, top_p=0.0, filter_value=-float("Inf")):
  function get_relevant_lyric_tokens (line 81) | def get_relevant_lyric_tokens(full_tokens, max_n_lyric_tokens, total_len...
  function get_starts (line 115) | def get_starts(total_length, n_ctx, hop_length):
  function get_alignment (line 125) | def get_alignment(music_tokens, labels, prior, config):
  function save_temp_audio (line 182) | def save_temp_audio(fname, lvl, metas, aud):
  function get_mask (line 193) | def get_mask(mask, query_length, key_value_length, blocks, spread, devic...
  class JukeboxConv1D (line 220) | class JukeboxConv1D(nn.Module):
    method __init__ (line 221) | def __init__(self, input_width, output_width):
    method forward (line 230) | def forward(self, hidden_states):
  class JukeboxResConv1DBlock (line 241) | class JukeboxResConv1DBlock(nn.Module):
    method __init__ (line 242) | def __init__(self, config, conv_width, depth=1, res_scale=1.0):
    method forward (line 253) | def forward(self, hidden_states):
  class JukeboxResnet1D (line 262) | class JukeboxResnet1D(nn.Module):
    method __init__ (line 263) | def __init__(self, config, conv_width, n_depth, reverse_dilation=False):
    method forward (line 277) | def forward(self, hidden_states):
  class JukeboxEncoderConvBlock (line 283) | class JukeboxEncoderConvBlock(nn.Module):
    method __init__ (line 284) | def __init__(self, config, embed_dim, hidden_dim, depth, down_t, strid...
    method forward (line 296) | def forward(self, hidden_states):
  class JukeboxEncoder (line 303) | class JukeboxEncoder(nn.Module):
    method __init__ (line 304) | def __init__(self, config, width, depth, levels, downs_t, strides_t):
    method forward (line 317) | def forward(self, hidden_states):
  class JukeboxDecoderConvBock (line 329) | class JukeboxDecoderConvBock(nn.Module):
    method __init__ (line 330) | def __init__(self, config, embed_dim, hidden_dim, depth, down_t, strid...
    method forward (line 348) | def forward(self, hidden_states):
  class JukeboxDecoder (line 355) | class JukeboxDecoder(nn.Module):
    method __init__ (line 356) | def __init__(self, config, hidden_dim, depth, levels, downs_t, strides...
    method forward (line 367) | def forward(self, hidden_states, all_levels=True):
  class JukeboxBottleneckBlock (line 382) | class JukeboxBottleneckBlock(nn.Module):
    method __init__ (line 383) | def __init__(self, config: JukeboxVQVAEConfig):
    method _tile (line 394) | def _tile(self, hidden_states):
    method init_codebook (line 403) | def init_codebook(self, hidden_states):
    method update_codebook (line 411) | def update_codebook(self, hidden_states, latent_states):
    method preprocess (line 441) | def preprocess(self, hidden_states):
    method postprocess (line 458) | def postprocess(self, latent_states, dequantised_states, x_shape):
    method quantise (line 464) | def quantise(self, latent_states):
    method dequantise (line 476) | def dequantise(self, music_tokens):
    method encode (line 480) | def encode(self, latent_states):
    method decode (line 493) | def decode(self, music_tokens):
    method forward (line 505) | def forward(self, hidden_states, update_codebook=True):
  class JukeboxBottleneck (line 536) | class JukeboxBottleneck(nn.Module):
    method __init__ (line 537) | def __init__(self, config, levels):
    method encode (line 544) | def encode(self, raw_audio):
    method decode (line 550) | def decode(self, music_tokens, start_level=0, end_level=None):
    method forward (line 558) | def forward(self, input_audio):
  class JukeboxVQVAE (line 602) | class JukeboxVQVAE(PreTrainedModel):
    method _init_weights (line 607) | def _init_weights(self, module):
    method __init__ (line 624) | def __init__(self, config: JukeboxVQVAEConfig):
    method _decode (line 663) | def _decode(self, music_tokens, start_level=0, end_level=None):
    method decode (line 674) | def decode(self, music_tokens, start_level=0, end_level=None, bs_chunk...
    method _encode (line 697) | def _encode(self, raw_audio, start_level=0, end_level=None):
    method encode (line 710) | def encode(self, input_audio, start_level=0, end_level=None, bs_chunks...
    method sample (line 733) | def sample(self, n_samples):
    method forward (line 740) | def forward(self, raw_audio: torch.FloatTensor) -> Tuple[torch.Tensor,...
  class JukeboxMLP (line 788) | class JukeboxMLP(nn.Module):
    method __init__ (line 789) | def __init__(self, config):
    method forward (line 800) | def forward(self, hidden_states):
  class JukeboxLayerNorm (line 808) | class JukeboxLayerNorm(FusedLayerNorm):
    method __init__ (line 809) | def __init__(self, normalized_shape, eps=1e-5, elementwise_affine=True):
    method forward (line 814) | def forward(self, input):
  class JukeboxAttention (line 821) | class JukeboxAttention(nn.Module):
    method __init__ (line 822) | def __init__(self, config, n_ctx, attn_func="dense_attn"):
    method _attn (line 876) | def _attn(self, query_states, key_states, value_states, sample):
    method merge_heads (line 910) | def merge_heads(self, hidden_states):
    method split_heads (line 915) | def split_heads(self, hidden_states, is_key=False):
    method dense_attn (line 927) | def dense_attn(self, query, key, value, sample):
    method block_attn (line 935) | def block_attn(self, query, key, value, sample):
    method transpose_block_attn (line 951) | def transpose_block_attn(self, query, key, value, sample):
    method prev_block_attn (line 980) | def prev_block_attn(self, query, key, value, sample):
    method summary_attn (line 1017) | def summary_attn(self, query, key, value, sample):
    method summary_spread_attn (line 1036) | def summary_spread_attn(self, query, key, value, sample):
    method prime_attn (line 1054) | def prime_attn(self, query, key, value, sample):
    method factored_qkv (line 1060) | def factored_qkv(self, hidden_states, last_encoder_hidden_states=None,...
    method prime_qkv (line 1083) | def prime_qkv(self, hidden_states, last_encoder_hidden_states=None, sa...
    method decode_qkv (line 1097) | def decode_qkv(self, hidden_states, last_encoder_hidden_states=None, s...
    method forward (line 1111) | def forward(self, hidden_states, last_encoder_hidden_states=None, samp...
    method _encoder_len (line 1125) | def _encoder_len(self):
    method _offset (line 1130) | def _offset(self, curr_ctx):
    method _pad_to_block_ctx (line 1135) | def _pad_to_block_ctx(self, hidden_states, query=False):
    method _cache_len (line 1145) | def _cache_len(self):
    method _suff_cache_len (line 1148) | def _suff_cache_len(self):
    method _slice_cache (line 1166) | def _slice_cache(self, start, end=None):
    method _append_cache (line 1170) | def _append_cache(self, key, value):
    method del_cache (line 1186) | def del_cache(self):
  class JukeboxBlock (line 1195) | class JukeboxBlock(nn.Module):
    method __init__ (line 1196) | def __init__(self, config, n_ctx, attn_func="dense_attn"):
    method forward (line 1207) | def forward(self, hidden_states, last_encoder_hidden_states, sample=Fa...
  class JukeboxLayerStack (line 1221) | class JukeboxLayerStack(nn.Module):
    method __init__ (line 1222) | def __init__(self, config, n_ctx):
    method set_record_attn (line 1242) | def set_record_attn(self, record_attn):
    method forward (line 1263) | def forward(self, hidden_states, last_encoder_hidden_states=None, samp...
    method del_cache (line 1276) | def del_cache(self):
  class JukeboxPositionalEmbedding (line 1281) | class JukeboxPositionalEmbedding(nn.Module):
    method __init__ (line 1282) | def __init__(self, embed_dim, width):
    method forward (line 1286) | def forward(self):
  class JukeboxConditionalAutoregressive (line 1291) | class JukeboxConditionalAutoregressive(nn.Module):
    method __init__ (line 1292) | def __init__(
    method forward (line 1355) | def forward(
    method get_emb (line 1425) | def get_emb(self, sample_t, n_samples, tokens, audio_conditioning, met...
    method sample (line 1444) | def sample(
    method split_chunks (line 1503) | def split_chunks(self, length, chunk_size):
    method primed_sample (line 1508) | def primed_sample(
  class JukeboxMusicTokenConditioner (line 1621) | class JukeboxMusicTokenConditioner(nn.Module):
    method __init__ (line 1627) | def __init__(self, config, level):
    method forward (line 1643) | def forward(self, music_tokens, raw_audio_conditionning=None):
  class JukeboxRangeEmbedding (line 1666) | class JukeboxRangeEmbedding(nn.Module):
    method __init__ (line 1676) | def __init__(self, n_time, embed_dim, range, out_width, clamp=False):
    method forward (line 1684) | def forward(self, pos_start, pos_end=None):
  class JukeboxLabelConditioner (line 1714) | class JukeboxLabelConditioner(nn.Module):
    method __init__ (line 1715) | def __init__(self, config, include_time_signal):
    method forward (line 1740) | def forward(self, metadata):
  class JukeboxPrior (line 1770) | class JukeboxPrior(PreTrainedModel):
    method _init_weights (line 1797) | def _init_weights(self, module):
    method __init__ (line 1824) | def __init__(self, config: JukeboxPriorConfig, level=None, nb_priors=3...
    method get_metadata (line 1914) | def get_metadata(self, labels, start, total_length, offset, get_indice...
    method set_metadata_lyric_tokens (line 1931) | def set_metadata_lyric_tokens(self, labels):
    method get_music_tokens_conds (line 1956) | def get_music_tokens_conds(self, music_tokens, start, end):
    method prior_preprocess (line 1972) | def prior_preprocess(self, tokens, conds):
    method prior_postprocess (line 1989) | def prior_postprocess(self, tokens):
    method embed_tokens (line 2007) | def embed_tokens(self, music_tokens_conds):
    method encode (line 2017) | def encode(self, hidden_states, start_level=None, end_level=None, bs_c...
    method decode (line 2032) | def decode(self, music_tokens, start_level=None, end_level=None, bs_ch...
    method get_cond (line 2046) | def get_cond(self, music_tokens_conds, metadata):
    method sample (line 2062) | def sample(
    method get_encoder_states (line 2157) | def get_encoder_states(self, lyric_tokens, sample=False):
    method get_encoder_loss (line 2172) | def get_encoder_loss(self, last_encoder_hidden_states, target_lyrics):
    method forward_tokens (line 2185) | def forward_tokens(
    method forward (line 2230) | def forward(
  class JukeboxPreTrainedModel (line 2266) | class JukeboxPreTrainedModel(PreTrainedModel):
    method _init_weights (line 2276) | def _init_weights(self, module):
    method __init__ (line 2280) | def __init__(self, *inputs, **kwargs):
  class JukeboxModel (line 2302) | class JukeboxModel(JukeboxPreTrainedModel):
    method __init__ (line 2305) | def __init__(self, config):
    method set_shared_params (line 2314) | def set_shared_params(self, model_config):
    method decode (line 2327) | def decode(self, music_tokens, start_level=0, end_level=None, bs_chunk...
    method encode (line 2330) | def encode(self, input_audio, start_level=0, end_level=None, bs_chunks...
    method split_batch (line 2333) | def split_batch(self, obj, n_samples, split_size):
    method sample_partial_window (line 2345) | def sample_partial_window(
    method sample_single_window (line 2362) | def sample_single_window(self, music_tokens, labels, offset, sampling_...
    method sample_level (line 2421) | def sample_level(
    method _sample (line 2438) | def _sample(
    method ancestral_sample (line 2591) | def ancestral_sample(self, labels, n_samples=1, **sampling_kwargs) -> ...
    method continue_sample (line 2632) | def continue_sample(self, music_tokens, labels, **sampling_kwargs) -> ...
    method upsample (line 2647) | def upsample(self, music_tokens, labels, **sampling_kwargs) -> List[to...
    method primed_sample (line 2664) | def primed_sample(self, raw_audio, labels, **sampling_kwargs) -> List[...

FILE: transformers/models/jukebox/tokenization_jukebox.py
  class JukeboxTokenizer (line 59) | class JukeboxTokenizer(PreTrainedTokenizer):
    method __init__ (line 119) | def __init__(
    method vocab_size (line 162) | def vocab_size(self):
    method get_vocab (line 165) | def get_vocab(self):
    method _convert_token_to_id (line 168) | def _convert_token_to_id(self, list_artists, list_genres, list_lyrics):
    method _tokenize (line 181) | def _tokenize(self, lyrics):
    method tokenize (line 191) | def tokenize(self, artist, genre, lyrics, **kwargs):
    method prepare_for_tokenization (line 199) | def prepare_for_tokenization(
    method _run_strip_accents (line 249) | def _run_strip_accents(self, text):
    method _normalize (line 260) | def _normalize(self, text: str) -> str:
    method convert_lyric_tokens_to_string (line 281) | def convert_lyric_tokens_to_string(self, lyrics: List[str]) -> str:
    method convert_to_tensors (line 284) | def convert_to_tensors(
    method __call__ (line 345) | def __call__(self, artist, genres, lyrics="", return_tensors="pt") -> ...
    method save_vocabulary (line 372) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method _convert_id_to_token (line 408) | def _convert_id_to_token(self, artists_index, genres_index, lyric_index):

FILE: transformers/models/layoutlm/configuration_layoutlm.py
  class LayoutLMConfig (line 36) | class LayoutLMConfig(PretrainedConfig):
    method __init__ (line 108) | def __init__(
  class LayoutLMOnnxConfig (line 146) | class LayoutLMOnnxConfig(OnnxConfig):
    method __init__ (line 147) | def __init__(
    method inputs (line 157) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method generate_dummy_inputs (line 167) | def generate_dummy_inputs(

FILE: transformers/models/layoutlm/modeling_layoutlm.py
  class LayoutLMEmbeddings (line 55) | class LayoutLMEmbeddings(nn.Module):
    method __init__ (line 58) | def __init__(self, config):
    method forward (line 73) | def forward(
  class LayoutLMSelfAttention (line 130) | class LayoutLMSelfAttention(nn.Module):
    method __init__ (line 131) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 157) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 162) | def forward(
  class LayoutLMSelfOutput (line 265) | class LayoutLMSelfOutput(nn.Module):
    method __init__ (line 266) | def __init__(self, config):
    method forward (line 272) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LayoutLMAttention (line 280) | class LayoutLMAttention(nn.Module):
    method __init__ (line 281) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 287) | def prune_heads(self, heads):
    method forward (line 305) | def forward(
  class LayoutLMIntermediate (line 330) | class LayoutLMIntermediate(nn.Module):
    method __init__ (line 331) | def __init__(self, config):
    method forward (line 339) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class LayoutLMOutput (line 346) | class LayoutLMOutput(nn.Module):
    method __init__ (line 347) | def __init__(self, config):
    method forward (line 353) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LayoutLMLayer (line 361) | class LayoutLMLayer(nn.Module):
    method __init__ (line 362) | def __init__(self, config):
    method forward (line 376) | def forward(
    method feed_forward_chunk (line 441) | def feed_forward_chunk(self, attention_output):
  class LayoutLMEncoder (line 448) | class LayoutLMEncoder(nn.Module):
    method __init__ (line 449) | def __init__(self, config):
    method forward (line 455) | def forward(
  class LayoutLMPooler (line 547) | class LayoutLMPooler(nn.Module):
    method __init__ (line 548) | def __init__(self, config):
    method forward (line 553) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class LayoutLMPredictionHeadTransform (line 563) | class LayoutLMPredictionHeadTransform(nn.Module):
    method __init__ (line 564) | def __init__(self, config):
    method forward (line 573) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class LayoutLMLMPredictionHead (line 581) | class LayoutLMLMPredictionHead(nn.Module):
    method __init__ (line 582) | def __init__(self, config):
    method forward (line 595) | def forward(self, hidden_states):
  class LayoutLMOnlyMLMHead (line 602) | class LayoutLMOnlyMLMHead(nn.Module):
    method __init__ (line 603) | def __init__(self, config):
    method forward (line 607) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class LayoutLMPreTrainedModel (line 612) | class LayoutLMPreTrainedModel(PreTrainedModel):
    method _init_weights (line 624) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 640) | def _set_gradient_checkpointing(self, module, value=False):
  class LayoutLMModel (line 711) | class LayoutLMModel(LayoutLMPreTrainedModel):
    method __init__ (line 712) | def __init__(self, config):
    method get_input_embeddings (line 723) | def get_input_embeddings(self):
    method set_input_embeddings (line 726) | def set_input_embeddings(self, value):
    method _prune_heads (line 729) | def _prune_heads(self, heads_to_prune):
    method forward (line 739) | def forward(
  class LayoutLMForMaskedLM (line 859) | class LayoutLMForMaskedLM(LayoutLMPreTrainedModel):
    method __init__ (line 866) | def __init__(self, config):
    method get_input_embeddings (line 875) | def get_input_embeddings(self):
    method get_output_embeddings (line 878) | def get_output_embeddings(self):
    method set_output_embeddings (line 881) | def set_output_embeddings(self, new_embeddings):
    method forward (line 886) | def forward(
  class LayoutLMForSequenceClassification (line 994) | class LayoutLMForSequenceClassification(LayoutLMPreTrainedModel):
    method __init__ (line 995) | def __init__(self, config):
    method get_input_embeddings (line 1005) | def get_input_embeddings(self):
    method forward (line 1010) | def forward(
  class LayoutLMForTokenClassification (line 1131) | class LayoutLMForTokenClassification(LayoutLMPreTrainedModel):
    method __init__ (line 1132) | def __init__(self, config):
    method get_input_embeddings (line 1142) | def get_input_embeddings(self):
    method forward (line 1147) | def forward(
  class LayoutLMForQuestionAnswering (line 1249) | class LayoutLMForQuestionAnswering(LayoutLMPreTrainedModel):
    method __init__ (line 1250) | def __init__(self, config, has_visual_segment_embedding=True):
    method get_input_embeddings (line 1260) | def get_input_embeddings(self):
    method forward (line 1264) | def forward(

FILE: transformers/models/layoutlm/modeling_tf_layoutlm.py
  class TFLayoutLMEmbeddings (line 62) | class TFLayoutLMEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 65) | def __init__(self, config: LayoutLMConfig, **kwargs):
    method build (line 76) | def build(self, input_shape: tf.TensorShape):
    method call (line 128) | def call(
  class TFLayoutLMSelfAttention (line 192) | class TFLayoutLMSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 193) | def __init__(self, config: LayoutLMConfig, **kwargs):
    method transpose_for_scores (line 220) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 227) | def call(
  class TFLayoutLMSelfOutput (line 310) | class TFLayoutLMSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 311) | def __init__(self, config: LayoutLMConfig, **kwargs):
    method call (line 320) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFLayoutLMAttention (line 329) | class TFLayoutLMAttention(tf.keras.layers.Layer):
    method __init__ (line 330) | def __init__(self, config: LayoutLMConfig, **kwargs):
    method prune_heads (line 336) | def prune_heads(self, heads):
    method call (line 339) | def call(
  class TFLayoutLMIntermediate (line 370) | class TFLayoutLMIntermediate(tf.keras.layers.Layer):
    method __init__ (line 371) | def __init__(self, config: LayoutLMConfig, **kwargs):
    method call (line 383) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFLayoutLMOutput (line 391) | class TFLayoutLMOutput(tf.keras.layers.Layer):
    method __init__ (line 392) | def __init__(self, config: LayoutLMConfig, **kwargs):
    method call (line 401) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFLayoutLMLayer (line 410) | class TFLayoutLMLayer(tf.keras.layers.Layer):
    method __init__ (line 411) | def __init__(self, config: LayoutLMConfig, **kwargs):
    method call (line 424) | def call(
  class TFLayoutLMEncoder (line 497) | class TFLayoutLMEncoder(tf.keras.layers.Layer):
    method __init__ (line 498) | def __init__(self, config: LayoutLMConfig, **kwargs):
    method call (line 503) | def call(
  class TFLayoutLMPooler (line 567) | class TFLayoutLMPooler(tf.keras.layers.Layer):
    method __init__ (line 568) | def __init__(self, config: LayoutLMConfig, **kwargs):
    method call (line 578) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFLayoutLMPredictionHeadTransform (line 588) | class TFLayoutLMPredictionHeadTransform(tf.keras.layers.Layer):
    method __init__ (line 589) | def __init__(self, config: LayoutLMConfig, **kwargs):
    method call (line 605) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFLayoutLMLMPredictionHead (line 614) | class TFLayoutLMLMPredictionHead(tf.keras.layers.Layer):
    method __init__ (line 615) | def __init__(self, config: LayoutLMConfig, input_embeddings: tf.keras....
    method build (line 627) | def build(self, input_shape: tf.TensorShape):
    method get_output_embeddings (line 632) | def get_output_embeddings(self) -> tf.keras.layers.Layer:
    method set_output_embeddings (line 635) | def set_output_embeddings(self, value: tf.Variable):
    method get_bias (line 639) | def get_bias(self) -> Dict[str, tf.Variable]:
    method set_bias (line 642) | def set_bias(self, value: tf.Variable):
    method call (line 646) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFLayoutLMMLMHead (line 658) | class TFLayoutLMMLMHead(tf.keras.layers.Layer):
    method __init__ (line 659) | def __init__(self, config: LayoutLMConfig, input_embeddings: tf.keras....
    method call (line 664) | def call(self, sequence_output: tf.Tensor) -> tf.Tensor:
  class TFLayoutLMMainLayer (line 671) | class TFLayoutLMMainLayer(tf.keras.layers.Layer):
    method __init__ (line 674) | def __init__(self, config: LayoutLMConfig, add_pooling_layer: bool = T...
    method get_input_embeddings (line 683) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 686) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 690) | def _prune_heads(self, heads_to_prune):
    method call (line 698) | def call(
  class TFLayoutLMPreTrainedModel (line 800) | class TFLayoutLMPreTrainedModel(TFPreTrainedModel):
  class TFLayoutLMModel (line 912) | class TFLayoutLMModel(TFLayoutLMPreTrainedModel):
    method __init__ (line 913) | def __init__(self, config: LayoutLMConfig, *inputs, **kwargs):
    method call (line 923) | def call(
  class TFLayoutLMForMaskedLM (line 991) | class TFLayoutLMForMaskedLM(TFLayoutLMPreTrainedModel, TFMaskedLanguageM...
    method __init__ (line 1000) | def __init__(self, config: LayoutLMConfig, *inputs, **kwargs):
    method get_lm_head (line 1012) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method get_prefix_bias_name (line 1015) | def get_prefix_bias_name(self) -> str:
    method call (line 1022) | def call(
  class TFLayoutLMForSequenceClassification (line 1118) | class TFLayoutLMForSequenceClassification(TFLayoutLMPreTrainedModel, TFS...
    method __init__ (line 1123) | def __init__(self, config: LayoutLMConfig, *inputs, **kwargs):
    method call (line 1139) | def call(
  class TFLayoutLMForTokenClassification (line 1236) | class TFLayoutLMForTokenClassification(TFLayoutLMPreTrainedModel, TFToke...
    method __init__ (line 1247) | def __init__(self, config: LayoutLMConfig, *inputs, **kwargs):
    method call (line 1263) | def call(
  class TFLayoutLMForQuestionAnswering (line 1359) | class TFLayoutLMForQuestionAnswering(TFLayoutLMPreTrainedModel, TFQuesti...
    method __init__ (line 1369) | def __init__(self, config: LayoutLMConfig, *inputs, **kwargs):
    method call (line 1383) | def call(

FILE: transformers/models/layoutlm/tokenization_layoutlm.py
  function load_vocab (line 53) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 65) | def whitespace_tokenize(text):
  class LayoutLMTokenizer (line 75) | class LayoutLMTokenizer(PreTrainedTokenizer):
    method __init__ (line 122) | def __init__(
    method do_lower_case (line 169) | def do_lower_case(self):
    method vocab_size (line 173) | def vocab_size(self):
    method get_vocab (line 176) | def get_vocab(self):
    method _tokenize (line 179) | def _tokenize(self, text):
    method _convert_token_to_id (line 192) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 196) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 200) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 205) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 230) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 258) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 287) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 309) | class BasicTokenizer(object):
    method __init__ (line 329) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 337) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 374) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 385) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 407) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 420) | def _is_chinese_char(self, cp):
    method _clean_text (line 444) | def _clean_text(self, text):
  class WordpieceTokenizer (line 459) | class WordpieceTokenizer(object):
    method __init__ (line 462) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 467) | def tokenize(self, text):

FILE: transformers/models/layoutlm/tokenization_layoutlm_fast.py
  class LayoutLMTokenizerFast (line 62) | class LayoutLMTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 108) | def __init__(
    method build_inputs_with_special_tokens (line 150) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 174) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 203) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/layoutlmv2/configuration_layoutlmv2.py
  class LayoutLMv2Config (line 34) | class LayoutLMv2Config(PretrainedConfig):
    method __init__ (line 119) | def __init__(
    method get_default_detectron2_config (line 184) | def get_default_detectron2_config(self):
    method get_detectron2_config (line 214) | def get_detectron2_config(self):

FILE: transformers/models/layoutlmv2/feature_extraction_layoutlmv2.py
  class LayoutLMv2FeatureExtractor (line 28) | class LayoutLMv2FeatureExtractor(LayoutLMv2ImageProcessor):
    method __init__ (line 29) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/layoutlmv2/image_processing_layoutlmv2.py
  function normalize_box (line 44) | def normalize_box(box, width, height):
  function apply_tesseract (line 53) | def apply_tesseract(image: np.ndarray, lang: Optional[str], tesseract_co...
  class LayoutLMv2ImageProcessor (line 87) | class LayoutLMv2ImageProcessor(BaseImageProcessor):
    method __init__ (line 113) | def __init__(
    method resize (line 134) | def resize(
    method preprocess (line 161) | def preprocess(

FILE: transformers/models/layoutlmv2/modeling_layoutlmv2.py
  class LayoutLMv2Embeddings (line 63) | class LayoutLMv2Embeddings(nn.Module):
    method __init__ (line 66) | def __init__(self, config):
    method _calc_spatial_position_embeddings (line 82) | def _calc_spatial_position_embeddings(self, bbox):
  class LayoutLMv2SelfAttention (line 108) | class LayoutLMv2SelfAttention(nn.Module):
    method __init__ (line 109) | def __init__(self, config):
    method transpose_for_scores (line 135) | def transpose_for_scores(self, x):
    method compute_qkv (line 140) | def compute_qkv(self, hidden_states):
    method forward (line 157) | def forward(
  class LayoutLMv2Attention (line 201) | class LayoutLMv2Attention(nn.Module):
    method __init__ (line 202) | def __init__(self, config):
    method forward (line 207) | def forward(
  class LayoutLMv2SelfOutput (line 229) | class LayoutLMv2SelfOutput(nn.Module):
    method __init__ (line 230) | def __init__(self, config):
    method forward (line 236) | def forward(self, hidden_states, input_tensor):
  class LayoutLMv2Intermediate (line 244) | class LayoutLMv2Intermediate(nn.Module):
    method __init__ (line 245) | def __init__(self, config):
    method forward (line 253) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class LayoutLMv2Output (line 260) | class LayoutLMv2Output(nn.Module):
    method __init__ (line 261) | def __init__(self, config):
    method forward (line 267) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LayoutLMv2Layer (line 274) | class LayoutLMv2Layer(nn.Module):
    method __init__ (line 275) | def __init__(self, config):
    method forward (line 283) | def forward(
    method feed_forward_chunk (line 311) | def feed_forward_chunk(self, attention_output):
  function relative_position_bucket (line 317) | def relative_position_bucket(relative_position, bidirectional=True, num_...
  class LayoutLMv2Encoder (line 361) | class LayoutLMv2Encoder(nn.Module):
    method __init__ (line 362) | def __init__(self, config):
    method _calculate_1d_position_embeddings (line 385) | def _calculate_1d_position_embeddings(self, hidden_states, position_ids):
    method _calculate_2d_position_embeddings (line 397) | def _calculate_2d_position_embeddings(self, hidden_states, bbox):
    method forward (line 421) | def forward(
  class LayoutLMv2PreTrainedModel (line 500) | class LayoutLMv2PreTrainedModel(PreTrainedModel):
    method _init_weights (line 511) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 527) | def _set_gradient_checkpointing(self, module, value=False):
  function my_convert_sync_batchnorm (line 532) | def my_convert_sync_batchnorm(module, process_group=None):
  class LayoutLMv2VisualBackbone (line 556) | class LayoutLMv2VisualBackbone(nn.Module):
    method __init__ (line 557) | def __init__(self, config):
    method forward (line 589) | def forward(self, images):
    method synchronize_batch_norm (line 596) | def synchronize_batch_norm(self):
  class LayoutLMv2Pooler (line 690) | class LayoutLMv2Pooler(nn.Module):
    method __init__ (line 691) | def __init__(self, config):
    method forward (line 696) | def forward(self, hidden_states):
  class LayoutLMv2Model (line 709) | class LayoutLMv2Model(LayoutLMv2PreTrainedModel):
    method __init__ (line 710) | def __init__(self, config):
    method get_input_embeddings (line 730) | def get_input_embeddings(self):
    method set_input_embeddings (line 733) | def set_input_embeddings(self, value):
    method _calc_text_embeddings (line 736) | def _calc_text_embeddings(self, input_ids, bbox, position_ids, token_t...
    method _calc_img_embeddings (line 761) | def _calc_img_embeddings(self, image, bbox, position_ids):
    method _calc_visual_bbox (line 772) | def _calc_visual_bbox(self, image_feature_pool_shape, bbox, device, fi...
    method _get_input_shape (line 809) | def _get_input_shape(self, input_ids=None, inputs_embeds=None):
    method forward (line 821) | def forward(
  class LayoutLMv2ForSequenceClassification (line 970) | class LayoutLMv2ForSequenceClassification(LayoutLMv2PreTrainedModel):
    method __init__ (line 971) | def __init__(self, config):
    method get_input_embeddings (line 981) | def get_input_embeddings(self):
    method forward (line 986) | def forward(
  class LayoutLMv2ForTokenClassification (line 1151) | class LayoutLMv2ForTokenClassification(LayoutLMv2PreTrainedModel):
    method __init__ (line 1152) | def __init__(self, config):
    method get_input_embeddings (line 1162) | def get_input_embeddings(self):
    method forward (line 1167) | def forward(
  class LayoutLMv2ForQuestionAnswering (line 1282) | class LayoutLMv2ForQuestionAnswering(LayoutLMv2PreTrainedModel):
    method __init__ (line 1283) | def __init__(self, config, has_visual_segment_embedding=True):
    method get_input_embeddings (line 1293) | def get_input_embeddings(self):
    method forward (line 1298) | def forward(

FILE: transformers/models/layoutlmv2/processing_layoutlmv2.py
  class LayoutLMv2Processor (line 27) | class LayoutLMv2Processor(ProcessorMixin):
    method __init__ (line 50) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 67) | def __call__(
    method get_overflowing_images (line 153) | def get_overflowing_images(self, images, overflow_to_sample_mapping):
    method batch_decode (line 167) | def batch_decode(self, *args, **kwargs):
    method decode (line 174) | def decode(self, *args, **kwargs):
    method model_input_names (line 182) | def model_input_names(self):
    method feature_extractor_class (line 186) | def feature_extractor_class(self):
    method feature_extractor (line 194) | def feature_extractor(self):

FILE: transformers/models/layoutlmv2/tokenization_layoutlmv2.py
  function load_vocab (line 170) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 181) | def whitespace_tokenize(text):
  function subfinder (line 193) | def subfinder(mylist, pattern):
  class LayoutLMv2Tokenizer (line 206) | class LayoutLMv2Tokenizer(PreTrainedTokenizer):
    method __init__ (line 225) | def __init__(
    method do_lower_case (line 293) | def do_lower_case(self):
    method vocab_size (line 297) | def vocab_size(self):
    method get_vocab (line 300) | def get_vocab(self):
    method _tokenize (line 303) | def _tokenize(self, text):
    method _convert_token_to_id (line 316) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 320) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 324) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 329) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 354) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 382) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 405) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method __call__ (line 426) | def __call__(
    method batch_encode_plus (line 575) | def batch_encode_plus(
    method _batch_encode_plus (line 633) | def _batch_encode_plus(
    method _batch_prepare_for_model (line 689) | def _batch_prepare_for_model(
    method encode (line 760) | def encode(
    method encode_plus (line 807) | def encode_plus(
    method _encode_plus (line 873) | def _encode_plus(
    method prepare_for_model (line 926) | def prepare_for_model(
    method truncate_sequences (line 1136) | def truncate_sequences(
    method _pad (line 1261) | def _pad(
  class BasicTokenizer (line 1347) | class BasicTokenizer(object):
    method __init__ (line 1367) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 1375) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 1412) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 1423) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 1445) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 1458) | def _is_chinese_char(self, cp):
    method _clean_text (line 1482) | def _clean_text(self, text):
  class WordpieceTokenizer (line 1497) | class WordpieceTokenizer(object):
    method __init__ (line 1500) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 1505) | def tokenize(self, text):

FILE: transformers/models/layoutlmv2/tokenization_layoutlmv2_fast.py
  class LayoutLMv2TokenizerFast (line 70) | class LayoutLMv2TokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 122) | def __init__(
    method __call__ (line 180) | def __call__(
    method batch_encode_plus (line 329) | def batch_encode_plus(
    method tokenize (line 387) | def tokenize(self, text: str, pair: Optional[str] = None, add_special_...
    method encode_plus (line 396) | def encode_plus(
    method _batch_encode_plus (line 462) | def _batch_encode_plus(
    method _encode_plus (line 618) | def _encode_plus(
    method _pad (line 684) | def _pad(
    method build_inputs_with_special_tokens (line 768) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 792) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 815) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/layoutlmv3/configuration_layoutlmv3.py
  class LayoutLMv3Config (line 40) | class LayoutLMv3Config(PretrainedConfig):
    method __init__ (line 124) | def __init__(
  class LayoutLMv3OnnxConfig (line 193) | class LayoutLMv3OnnxConfig(OnnxConfig):
    method inputs (line 197) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 219) | def atol_for_validation(self) -> float:
    method default_onnx_opset (line 223) | def default_onnx_opset(self) -> int:
    method generate_dummy_inputs (line 226) | def generate_dummy_inputs(

FILE: transformers/models/layoutlmv3/feature_extraction_layoutlmv3.py
  class LayoutLMv3FeatureExtractor (line 28) | class LayoutLMv3FeatureExtractor(LayoutLMv3ImageProcessor):
    method __init__ (line 29) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/layoutlmv3/image_processing_layoutlmv3.py
  function normalize_box (line 46) | def normalize_box(box, width, height):
  function apply_tesseract (line 55) | def apply_tesseract(image: np.ndarray, lang: Optional[str], tesseract_co...
  class LayoutLMv3ImageProcessor (line 88) | class LayoutLMv3ImageProcessor(BaseImageProcessor):
    method __init__ (line 129) | def __init__(
    method resize (line 160) | def resize(
    method rescale (line 187) | def rescale(
    method normalize (line 207) | def normalize(
    method preprocess (line 230) | def preprocess(

FILE: transformers/models/layoutlmv3/modeling_layoutlmv3.py
  class LayoutLMv3PatchEmbeddings (line 199) | class LayoutLMv3PatchEmbeddings(nn.Module):
    method __init__ (line 203) | def __init__(self, config):
    method forward (line 219) | def forward(self, pixel_values, position_embedding=None):
  class LayoutLMv3TextEmbeddings (line 234) | class LayoutLMv3TextEmbeddings(nn.Module):
    method __init__ (line 239) | def __init__(self, config):
    method calculate_spatial_position_embeddings (line 260) | def calculate_spatial_position_embeddings(self, bbox):
    method create_position_ids_from_input_ids (line 286) | def create_position_ids_from_input_ids(self, input_ids, padding_idx):
    method create_position_ids_from_inputs_embeds (line 296) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
    method forward (line 308) | def forward(
  class LayoutLMv3PreTrainedModel (line 350) | class LayoutLMv3PreTrainedModel(PreTrainedModel):
    method _init_weights (line 359) | def _init_weights(self, module):
  class LayoutLMv3SelfAttention (line 376) | class LayoutLMv3SelfAttention(nn.Module):
    method __init__ (line 377) | def __init__(self, config):
    method transpose_for_scores (line 397) | def transpose_for_scores(self, x):
    method cogview_attention (line 402) | def cogview_attention(self, attention_scores, alpha=32):
    method forward (line 414) | def forward(
  class LayoutLMv3SelfOutput (line 467) | class LayoutLMv3SelfOutput(nn.Module):
    method __init__ (line 468) | def __init__(self, config):
    method forward (line 474) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LayoutLMv3Attention (line 482) | class LayoutLMv3Attention(nn.Module):
    method __init__ (line 483) | def __init__(self, config):
    method forward (line 488) | def forward(
  class LayoutLMv3Layer (line 511) | class LayoutLMv3Layer(nn.Module):
    method __init__ (line 512) | def __init__(self, config):
    method forward (line 520) | def forward(
    method feed_forward_chunk (line 548) | def feed_forward_chunk(self, attention_output):
  class LayoutLMv3Encoder (line 554) | class LayoutLMv3Encoder(nn.Module):
    method __init__ (line 555) | def __init__(self, config):
    method relative_position_bucket (line 577) | def relative_position_bucket(self, relative_position, bidirectional=Tr...
    method _cal_1d_pos_emb (line 600) | def _cal_1d_pos_emb(self, hidden_states, position_ids):
    method _cal_2d_pos_emb (line 613) | def _cal_2d_pos_emb(self, hidden_states, bbox):
    method forward (line 637) | def forward(
  class LayoutLMv3Intermediate (line 718) | class LayoutLMv3Intermediate(nn.Module):
    method __init__ (line 719) | def __init__(self, config):
    method forward (line 727) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class LayoutLMv3Output (line 734) | class LayoutLMv3Output(nn.Module):
    method __init__ (line 735) | def __init__(self, config):
    method forward (line 741) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LayoutLMv3Model (line 752) | class LayoutLMv3Model(LayoutLMv3PreTrainedModel):
    method __init__ (line 755) | def __init__(self, config):
    method get_input_embeddings (line 784) | def get_input_embeddings(self):
    method set_input_embeddings (line 787) | def set_input_embeddings(self, value):
    method _prune_heads (line 790) | def _prune_heads(self, heads_to_prune):
    method init_visual_bbox (line 798) | def init_visual_bbox(self, image_size=(14, 14), max_len=1000):
    method calculate_visual_bbox (line 821) | def calculate_visual_bbox(self, device, dtype, batch_size):
    method forward_image (line 826) | def forward_image(self, pixel_values):
    method forward (line 847) | def forward(
  class LayoutLMv3ClassificationHead (line 1004) | class LayoutLMv3ClassificationHead(nn.Module):
    method __init__ (line 1009) | def __init__(self, config, pool_feature=False):
    method forward (line 1022) | def forward(self, x):
  class LayoutLMv3ForTokenClassification (line 1040) | class LayoutLMv3ForTokenClassification(LayoutLMv3PreTrainedModel):
    method __init__ (line 1044) | def __init__(self, config):
    method forward (line 1061) | def forward(
  class LayoutLMv3ForQuestionAnswering (line 1155) | class LayoutLMv3ForQuestionAnswering(LayoutLMv3PreTrainedModel):
    method __init__ (line 1159) | def __init__(self, config):
    method forward (line 1172) | def forward(
  class LayoutLMv3ForSequenceClassification (line 1288) | class LayoutLMv3ForSequenceClassification(LayoutLMv3PreTrainedModel):
    method __init__ (line 1291) | def __init__(self, config):
    method forward (line 1304) | def forward(

FILE: transformers/models/layoutlmv3/modeling_tf_layoutlmv3.py
  class TFLayoutLMv3PatchEmbeddings (line 68) | class TFLayoutLMv3PatchEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 71) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method call (line 91) | def call(self, pixel_values: tf.Tensor) -> tf.Tensor:
  class TFLayoutLMv3TextEmbeddings (line 101) | class TFLayoutLMv3TextEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 106) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method calculate_spatial_position_embeddings (line 155) | def calculate_spatial_position_embeddings(self, bbox: tf.Tensor) -> tf...
    method create_position_ids_from_inputs_embeds (line 196) | def create_position_ids_from_inputs_embeds(self, inputs_embds: tf.Tens...
    method create_position_ids_from_input_ids (line 211) | def create_position_ids_from_input_ids(self, input_ids: tf.Tensor) -> ...
    method create_position_ids (line 220) | def create_position_ids(self, input_ids: tf.Tensor, inputs_embeds: tf....
    method call (line 226) | def call(
  class TFLayoutLMv3SelfAttention (line 264) | class TFLayoutLMv3SelfAttention(tf.keras.layers.Layer):
    method __init__ (line 265) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method transpose_for_scores (line 298) | def transpose_for_scores(self, x: tf.Tensor):
    method cogview_attention (line 309) | def cogview_attention(self, attention_scores: tf.Tensor, alpha: Union[...
    method call (line 322) | def call(
  class TFLayoutLMv3SelfOutput (line 377) | class TFLayoutLMv3SelfOutput(tf.keras.layers.Layer):
    method __init__ (line 378) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method call (line 387) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFLayoutLMv3Attention (line 395) | class TFLayoutLMv3Attention(tf.keras.layers.Layer):
    method __init__ (line 396) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method call (line 401) | def call(
  class TFLayoutLMv3Intermediate (line 426) | class TFLayoutLMv3Intermediate(tf.keras.layers.Layer):
    method __init__ (line 427) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method call (line 439) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFLayoutLMv3Output (line 447) | class TFLayoutLMv3Output(tf.keras.layers.Layer):
    method __init__ (line 448) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method call (line 457) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFLayoutLMv3Layer (line 465) | class TFLayoutLMv3Layer(tf.keras.layers.Layer):
    method __init__ (line 466) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method call (line 472) | def call(
  class TFLayoutLMv3Encoder (line 499) | class TFLayoutLMv3Encoder(tf.keras.layers.Layer):
    method __init__ (line 500) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method relative_position_bucket (line 534) | def relative_position_bucket(self, relative_positions: tf.Tensor, num_...
    method _cal_pos_emb (line 559) | def _cal_pos_emb(
    method _cal_1d_pos_emb (line 575) | def _cal_1d_pos_emb(self, position_ids: tf.Tensor):
    method _cal_2d_pos_emb (line 578) | def _cal_2d_pos_emb(self, bbox: tf.Tensor):
    method call (line 596) | def call(
  class TFLayoutLMv3MainLayer (line 655) | class TFLayoutLMv3MainLayer(tf.keras.layers.Layer):
    method __init__ (line 658) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method build (line 679) | def build(self, input_shape: tf.TensorShape):
    method get_input_embeddings (line 699) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 702) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 706) | def _prune_heads(self, heads_to_prune):
    method init_visual_bbox (line 713) | def init_visual_bbox(self, image_size: Tuple[int, int], max_len: int =...
    method calculate_visual_bbox (line 736) | def calculate_visual_bbox(self, batch_size: int, dtype: tf.DType):
    method embed_image (line 742) | def embed_image(self, pixel_values: tf.Tensor) -> tf.Tensor:
    method get_extended_attention_mask (line 757) | def get_extended_attention_mask(self, attention_mask: tf.Tensor) -> tf...
    method get_head_mask (line 784) | def get_head_mask(self, head_mask: tf.Tensor | None) -> Union[tf.Tenso...
    method call (line 810) | def call(
  class TFLayoutLMv3PreTrainedModel (line 973) | class TFLayoutLMv3PreTrainedModel(TFPreTrainedModel):
    method input_signature (line 983) | def input_signature(self):
  class TFLayoutLMv3Model (line 1111) | class TFLayoutLMv3Model(TFLayoutLMv3PreTrainedModel):
    method __init__ (line 1115) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1122) | def call(
  class TFLayoutLMv3ClassificationHead (line 1184) | class TFLayoutLMv3ClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 1189) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method call (line 1210) | def call(self, inputs: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFLayoutLMv3ForSequenceClassification (line 1226) | class TFLayoutLMv3ForSequenceClassification(TFLayoutLMv3PreTrainedModel,...
    method __init__ (line 1230) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method call (line 1239) | def call(
  class TFLayoutLMv3ForTokenClassification (line 1330) | class TFLayoutLMv3ForTokenClassification(TFLayoutLMv3PreTrainedModel, TF...
    method __init__ (line 1334) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method call (line 1352) | def call(
  class TFLayoutLMv3ForQuestionAnswering (line 1452) | class TFLayoutLMv3ForQuestionAnswering(TFLayoutLMv3PreTrainedModel, TFQu...
    method __init__ (line 1456) | def __init__(self, config: LayoutLMv3Config, **kwargs):
    method call (line 1467) | def call(

FILE: transformers/models/layoutlmv3/processing_layoutlmv3.py
  class LayoutLMv3Processor (line 27) | class LayoutLMv3Processor(ProcessorMixin):
    method __init__ (line 50) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 67) | def __call__(
    method get_overflowing_images (line 151) | def get_overflowing_images(self, images, overflow_to_sample_mapping):
    method batch_decode (line 165) | def batch_decode(self, *args, **kwargs):
    method decode (line 172) | def decode(self, *args, **kwargs):
    method model_input_names (line 180) | def model_input_names(self):
    method feature_extractor_class (line 184) | def feature_extractor_class(self):
    method feature_extractor (line 192) | def feature_extractor(self):

FILE: transformers/models/layoutlmv3/tokenization_layoutlmv3.py
  function bytes_to_unicode (line 161) | def bytes_to_unicode():
  function get_pairs (line 186) | def get_pairs(word):
  class LayoutLMv3Tokenizer (line 200) | class LayoutLMv3Tokenizer(PreTrainedTokenizer):
    method __init__ (line 276) | def __init__(
    method vocab_size (line 349) | def vocab_size(self):
    method get_vocab (line 353) | def get_vocab(self):
    method bpe (line 357) | def bpe(self, token):
    method _tokenize (line 400) | def _tokenize(self, text):
    method _convert_token_to_id (line 411) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 416) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 421) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 428) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 458) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 484) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 512) | def create_token_type_ids_from_sequences(
    method prepare_for_tokenization (line 535) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method __call__ (line 549) | def __call__(
    method batch_encode_plus (line 699) | def batch_encode_plus(
    method _batch_encode_plus (line 758) | def _batch_encode_plus(
    method _batch_prepare_for_model (line 815) | def _batch_prepare_for_model(
    method encode (line 887) | def encode(
    method encode_plus (line 935) | def encode_plus(
    method _encode_plus (line 1002) | def _encode_plus(
    method prepare_for_model (line 1055) | def prepare_for_model(
    method truncate_sequences (line 1268) | def truncate_sequences(
    method _pad (line 1394) | def _pad(

FILE: transformers/models/layoutlmv3/tokenization_layoutlmv3_fast.py
  class LayoutLMv3TokenizerFast (line 65) | class LayoutLMv3TokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 139) | def __init__(
    method __call__ (line 226) | def __call__(
    method batch_encode_plus (line 376) | def batch_encode_plus(
    method tokenize (line 435) | def tokenize(self, text: str, pair: Optional[str] = None, add_special_...
    method encode_plus (line 445) | def encode_plus(
    method _batch_encode_plus (line 511) | def _batch_encode_plus(
    method _encode_plus (line 673) | def _encode_plus(
    method _pad (line 740) | def _pad(
    method save_vocabulary (line 825) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 829) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 836) | def create_token_type_ids_from_sequences(

FILE: transformers/models/layoutxlm/processing_layoutxlm.py
  class LayoutXLMProcessor (line 25) | class LayoutXLMProcessor(ProcessorMixin):
    method __call__ (line 47) | def __call__(
    method get_overflowing_images (line 134) | def get_overflowing_images(self, images, overflow_to_sample_mapping):
    method batch_decode (line 148) | def batch_decode(self, *args, **kwargs):
    method decode (line 155) | def decode(self, *args, **kwargs):
    method model_input_names (line 163) | def model_input_names(self):

FILE: transformers/models/layoutxlm/tokenization_layoutxlm.py
  class LayoutXLMTokenizer (line 149) | class LayoutXLMTokenizer(PreTrainedTokenizer):
    method __init__ (line 234) | def __init__(
    method __getstate__ (line 300) | def __getstate__(self):
    method __setstate__ (line 306) | def __setstate__(self, d):
    method build_inputs_with_special_tokens (line 316) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 342) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 370) | def create_token_type_ids_from_sequences(
    method vocab_size (line 396) | def vocab_size(self):
    method get_vocab (line 399) | def get_vocab(self):
    method _tokenize (line 404) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 407) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 416) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 422) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 427) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method __call__ (line 445) | def __call__(
    method _batch_encode_plus (line 593) | def _batch_encode_plus(
    method _batch_prepare_for_model (line 649) | def _batch_prepare_for_model(
    method _encode_plus (line 719) | def _encode_plus(
    method prepare_for_model (line 772) | def prepare_for_model(
    method truncate_sequences (line 969) | def truncate_sequences(
    method _pad (line 1094) | def _pad(

FILE: transformers/models/layoutxlm/tokenization_layoutxlm_fast.py
  class LayoutXLMTokenizerFast (line 152) | class LayoutXLMTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 220) | def __init__(
    method __call__ (line 270) | def __call__(
    method tokenize (line 418) | def tokenize(self, text: str, pair: Optional[str] = None, add_special_...
    method _batch_encode_plus (line 426) | def _batch_encode_plus(
    method _encode_plus (line 583) | def _encode_plus(
    method _pad (line 649) | def _pad(
    method build_inputs_with_special_tokens (line 733) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 759) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 784) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/led/configuration_led.py
  class LEDConfig (line 31) | class LEDConfig(PretrainedConfig):
    method __init__ (line 108) | def __init__(

FILE: transformers/models/led/modeling_led.py
  function shift_tokens_right (line 62) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 78) | def _make_causal_mask(input_ids_shape: torch.Size, dtype: torch.dtype, p...
  function _expand_mask (line 93) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class LEDLearnedPositionalEmbedding (line 111) | class LEDLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 116) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 119) | def forward(self, input_ids_shape: torch.Size, past_key_values_length:...
  class LEDEncoderSelfAttention (line 129) | class LEDEncoderSelfAttention(nn.Module):
    method __init__ (line 130) | def __init__(self, config, layer_id):
    method forward (line 165) | def forward(
    method _pad_and_transpose_last_two_dims (line 334) | def _pad_and_transpose_last_two_dims(hidden_states_padded, padding):
    method _pad_and_diagonalize (line 345) | def _pad_and_diagonalize(chunked_hidden_states):
    method _chunk (line 394) | def _chunk(hidden_states, window_overlap, onnx_export: bool = False):
    method _mask_invalid_locations (line 435) | def _mask_invalid_locations(input_tensor, affected_seq_len) -> torch.T...
    method _sliding_chunks_query_key_matmul (line 450) | def _sliding_chunks_query_key_matmul(self, query: torch.Tensor, key: t...
    method _sliding_chunks_matmul_attn_probs_value (line 516) | def _sliding_chunks_matmul_attn_probs_value(
    method _get_global_attn_indices (line 561) | def _get_global_attn_indices(is_index_global_attn):
    method _concat_with_global_key_attn_probs (line 589) | def _concat_with_global_key_attn_probs(
    method _compute_attn_output_with_global_indices (line 619) | def _compute_attn_output_with_global_indices(
    method _compute_global_attn_output_from_hidden (line 655) | def _compute_global_attn_output_from_hidden(
  class LEDEncoderAttention (line 764) | class LEDEncoderAttention(nn.Module):
    method __init__ (line 765) | def __init__(self, config, layer_id):
    method forward (line 770) | def forward(
  class LEDDecoderAttention (line 798) | class LEDDecoderAttention(nn.Module):
    method __init__ (line 801) | def __init__(
    method _shape (line 827) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 830) | def forward(
  class LEDEncoderLayer (line 941) | class LEDEncoderLayer(nn.Module):
    method __init__ (line 942) | def __init__(self, config: LEDConfig, layer_id: int):
    method forward (line 954) | def forward(
  class LEDDecoderLayer (line 1003) | class LEDDecoderLayer(nn.Module):
    method __init__ (line 1004) | def __init__(self, config: LEDConfig):
    method forward (line 1030) | def forward(
  class LEDClassificationHead (line 1119) | class LEDClassificationHead(nn.Module):
    method __init__ (line 1122) | def __init__(
    method forward (line 1134) | def forward(self, hidden_states: torch.Tensor):
  class LEDPreTrainedModel (line 1143) | class LEDPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1148) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1159) | def _set_gradient_checkpointing(self, module, value=False):
    method dummy_inputs (line 1164) | def dummy_inputs(self):
  class LEDEncoderBaseModelOutput (line 1176) | class LEDEncoderBaseModelOutput(ModelOutput):
  class LEDSeq2SeqModelOutput (line 1219) | class LEDSeq2SeqModelOutput(ModelOutput):
  class LEDSeq2SeqLMOutput (line 1287) | class LEDSeq2SeqLMOutput(ModelOutput):
  class LEDSeq2SeqSequenceClassifierOutput (line 1354) | class LEDSeq2SeqSequenceClassifierOutput(ModelOutput):
  class LEDSeq2SeqQuestionAnsweringModelOutput (line 1421) | class LEDSeq2SeqQuestionAnsweringModelOutput(ModelOutput):
  class LEDEncoder (line 1645) | class LEDEncoder(LEDPreTrainedModel):
    method __init__ (line 1655) | def __init__(self, config: LEDConfig, embed_tokens: Optional[nn.Embedd...
    method _merge_to_attention_mask (line 1694) | def _merge_to_attention_mask(self, attention_mask: torch.Tensor, globa...
    method _pad_to_window_size (line 1706) | def _pad_to_window_size(
    method forward (line 1749) | def forward(
  class LEDDecoder (line 1940) | class LEDDecoder(LEDPreTrainedModel):
    method __init__ (line 1949) | def __init__(self, config: LEDConfig, embed_tokens: Optional[nn.Embedd...
    method forward (line 1972) | def forward(
  class LEDModel (line 2211) | class LEDModel(LEDPreTrainedModel):
    method __init__ (line 2214) | def __init__(self, config: LEDConfig):
    method get_input_embeddings (line 2226) | def get_input_embeddings(self):
    method set_input_embeddings (line 2229) | def set_input_embeddings(self, value):
    method get_encoder (line 2234) | def get_encoder(self):
    method get_decoder (line 2237) | def get_decoder(self):
    method forward (line 2246) | def forward(
  class LEDForConditionalGeneration (line 2335) | class LEDForConditionalGeneration(LEDPreTrainedModel):
    method __init__ (line 2346) | def __init__(self, config: LEDConfig):
    method get_encoder (line 2355) | def get_encoder(self):
    method get_decoder (line 2358) | def get_decoder(self):
    method resize_token_embeddings (line 2361) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_final_logits_bias (line 2366) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_output_embeddings (line 2375) | def get_output_embeddings(self):
    method set_output_embeddings (line 2378) | def set_output_embeddings(self, new_embeddings):
    method forward (line 2384) | def forward(
    method prepare_inputs_for_generation (line 2479) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 2509) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 2513) | def _reorder_cache(past_key_values, beam_idx):
  class LEDForSequenceClassification (line 2530) | class LEDForSequenceClassification(LEDPreTrainedModel):
    method __init__ (line 2533) | def __init__(self, config: LEDConfig, **kwargs):
    method forward (line 2558) | def forward(
  class LEDForQuestionAnswering (line 2666) | class LEDForQuestionAnswering(LEDPreTrainedModel):
    method __init__ (line 2669) | def __init__(self, config):
    method forward (line 2687) | def forward(

FILE: transformers/models/led/modeling_tf_led.py
  function shift_tokens_right (line 61) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int, decoder_...
  function _make_causal_mask (line 86) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 104) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFLEDLearnedPositionalEmbedding (line 117) | class TFLEDLearnedPositionalEmbedding(tf.keras.layers.Embedding):
    method __init__ (line 122) | def __init__(self, num_embeddings: int, embedding_dim: int, **kwargs):
    method call (line 125) | def call(self, input_shape: tf.TensorShape, past_key_values_length: in...
  class TFLEDEncoderSelfAttention (line 135) | class TFLEDEncoderSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 136) | def __init__(self, config, layer_id, **kwargs):
    method build (line 195) | def build(self, input_shape=None):
    method call (line 205) | def call(
    method _sliding_chunks_query_key_matmul (line 396) | def _sliding_chunks_query_key_matmul(self, query, key, window_overlap):
    method _mask_invalid_locations (line 513) | def _mask_invalid_locations(input_tensor, window_overlap):
    method _sliding_chunks_matmul_attn_probs_value (line 542) | def _sliding_chunks_matmul_attn_probs_value(self, attn_probs, value, w...
    method _pad_and_transpose_last_two_dims (line 616) | def _pad_and_transpose_last_two_dims(hidden_states_padded, paddings):
    method _pad_and_diagonalize (line 627) | def _pad_and_diagonalize(chunked_hidden_states):
    method _chunk (line 679) | def _chunk(hidden_states, window_overlap):
    method _get_global_attn_indices (line 709) | def _get_global_attn_indices(is_index_global_attn):
    method _concat_with_global_key_attn_probs (line 739) | def _concat_with_global_key_attn_probs(
    method _compute_attn_output_with_global_indices (line 793) | def _compute_attn_output_with_global_indices(
    method _compute_global_attn_output_from_hidden (line 834) | def _compute_global_attn_output_from_hidden(
    method reshape_and_transpose (line 971) | def reshape_and_transpose(self, vector, batch_size):
  class TFLEDEncoderAttention (line 981) | class TFLEDEncoderAttention(tf.keras.layers.Layer):
    method __init__ (line 982) | def __init__(self, config, layer_id, **kwargs):
    method call (line 987) | def call(self, inputs, training=False):
  class TFLEDDecoderAttention (line 1008) | class TFLEDDecoderAttention(tf.keras.layers.Layer):
    method __init__ (line 1011) | def __init__(
    method _shape (line 1035) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 1038) | def call(
  class TFLEDEncoderLayer (line 1159) | class TFLEDEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 1160) | def __init__(self, config: LEDConfig, layer_id: int, **kwargs):
    method call (line 1172) | def call(
  class TFLEDDecoderLayer (line 1218) | class TFLEDDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 1219) | def __init__(self, config: LEDConfig, **kwargs):
    method call (line 1246) | def call(
  class TFLEDPreTrainedModel (line 1327) | class TFLEDPreTrainedModel(TFPreTrainedModel):
    method input_signature (line 1332) | def input_signature(self):
  class TFLEDEncoderBaseModelOutput (line 1340) | class TFLEDEncoderBaseModelOutput(ModelOutput):
  class TFLEDSeq2SeqModelOutput (line 1383) | class TFLEDSeq2SeqModelOutput(ModelOutput):
  class TFLEDSeq2SeqLMOutput (line 1451) | class TFLEDSeq2SeqLMOutput(ModelOutput):
  class TFLEDEncoder (line 1627) | class TFLEDEncoder(tf.keras.layers.Layer):
    method __init__ (line 1637) | def __init__(self, config: LEDConfig, embed_tokens: Optional[tf.keras....
    method get_embed_tokens (line 1666) | def get_embed_tokens(self):
    method set_embed_tokens (line 1669) | def set_embed_tokens(self, embed_tokens):
    method call (line 1673) | def call(
    method compute_hidden_states (line 1839) | def compute_hidden_states(self, hidden_states, padding_len):
    method _pad_to_window_size (line 1842) | def _pad_to_window_size(
  class TFLEDDecoder (line 1889) | class TFLEDDecoder(tf.keras.layers.Layer):
    method __init__ (line 1899) | def __init__(self, config: LEDConfig, embed_tokens: Optional[tf.keras....
    method set_embed_tokens (line 1917) | def set_embed_tokens(self, embed_tokens):
    method call (line 1921) | def call(
  class TFLEDMainLayer (line 2109) | class TFLEDMainLayer(tf.keras.layers.Layer):
    method __init__ (line 2112) | def __init__(self, config: LEDConfig, **kwargs):
    method get_input_embeddings (line 2127) | def get_input_embeddings(self):
    method set_input_embeddings (line 2130) | def set_input_embeddings(self, new_embeddings):
    method call (line 2136) | def call(
  class TFLEDModel (line 2218) | class TFLEDModel(TFLEDPreTrainedModel):
    method __init__ (line 2219) | def __init__(self, config, *inputs, **kwargs):
    method get_encoder (line 2224) | def get_encoder(self):
    method get_decoder (line 2227) | def get_decoder(self):
    method call (line 2237) | def call(
    method serving_output (line 2278) | def serving_output(self, output):
  class BiasLayer (line 2301) | class BiasLayer(tf.keras.layers.Layer):
    method __init__ (line 2307) | def __init__(self, shape, initializer, trainable, name, **kwargs):
    method call (line 2314) | def call(self, x):
  class TFLEDForConditionalGeneration (line 2322) | class TFLEDForConditionalGeneration(TFLEDPreTrainedModel):
    method __init__ (line 2328) | def __init__(self, config, *inputs, **kwargs):
    method get_decoder (line 2340) | def get_decoder(self):
    method get_encoder (line 2343) | def get_encoder(self):
    method get_bias (line 2346) | def get_bias(self):
    method set_bias (line 2349) | def set_bias(self, value):
    method get_output_embeddings (line 2357) | def get_output_embeddings(self):
    method set_output_embeddings (line 2360) | def set_output_embeddings(self, value):
    method call (line 2366) | def call(
    method serving_output (line 2450) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 2471) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 2497) | def prepare_decoder_input_ids_from_labels(self, labels: tf.Tensor):
    method hf_compute_loss (line 2500) | def hf_compute_loss(self, labels, logits):

FILE: transformers/models/led/tokenization_led.py
  function bytes_to_unicode (line 54) | def bytes_to_unicode():
  function get_pairs (line 79) | def get_pairs(word):
  class LEDTokenizer (line 93) | class LEDTokenizer(PreTrainedTokenizer):
    method __init__ (line 177) | def __init__(
    method vocab_size (line 233) | def vocab_size(self):
    method get_vocab (line 237) | def get_vocab(self):
    method bpe (line 241) | def bpe(self, token):
    method _tokenize (line 284) | def _tokenize(self, text):
    method _convert_token_to_id (line 295) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 300) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 305) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 312) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 342) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 368) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 396) | def create_token_type_ids_from_sequences(
    method prepare_for_tokenization (line 420) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method _pad (line 426) | def _pad(

FILE: transformers/models/led/tokenization_led_fast.py
  class LEDTokenizerFast (line 50) | class LEDTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 138) | def __init__(
    method mask_token (line 209) | def mask_token(self) -> str:
    method mask_token (line 224) | def mask_token(self, value):
    method _batch_encode_plus (line 236) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 248) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method save_vocabulary (line 260) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 265) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 273) | def create_token_type_ids_from_sequences(
    method _pad (line 297) | def _pad(

FILE: transformers/models/levit/configuration_levit.py
  class LevitConfig (line 35) | class LevitConfig(PretrainedConfig):
    method __init__ (line 92) | def __init__(
  class LevitOnnxConfig (line 132) | class LevitOnnxConfig(OnnxConfig):
    method inputs (line 136) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 144) | def atol_for_validation(self) -> float:

FILE: transformers/models/levit/convert_levit_timm_to_pytorch.py
  function convert_weight_and_push (line 36) | def convert_weight_and_push(
  function convert_weights_and_push (line 83) | def convert_weights_and_push(save_directory: Path, model_name: str = Non...

FILE: transformers/models/levit/feature_extraction_levit.py
  class LevitFeatureExtractor (line 26) | class LevitFeatureExtractor(LevitImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/levit/image_processing_levit.py
  class LevitImageProcessor (line 46) | class LevitImageProcessor(BaseImageProcessor):
    method __init__ (line 88) | def __init__(
    method resize (line 119) | def resize(
    method center_crop (line 164) | def center_crop(
    method rescale (line 187) | def rescale(
    method normalize (line 207) | def normalize(
    method preprocess (line 230) | def preprocess(

FILE: transformers/models/levit/modeling_levit.py
  class LevitForImageClassificationWithTeacherOutput (line 57) | class LevitForImageClassificationWithTeacherOutput(ModelOutput):
  class LevitConvEmbeddings (line 82) | class LevitConvEmbeddings(nn.Module):
    method __init__ (line 87) | def __init__(
    method forward (line 96) | def forward(self, embeddings):
  class LevitPatchEmbeddings (line 102) | class LevitPatchEmbeddings(nn.Module):
    method __init__ (line 108) | def __init__(self, config):
    method forward (line 130) | def forward(self, pixel_values):
  class MLPLayerWithBN (line 146) | class MLPLayerWithBN(nn.Module):
    method __init__ (line 147) | def __init__(self, input_dim, output_dim, bn_weight_init=1):
    method forward (line 152) | def forward(self, hidden_state):
  class LevitSubsample (line 158) | class LevitSubsample(nn.Module):
    method __init__ (line 159) | def __init__(self, stride, resolution):
    method forward (line 164) | def forward(self, hidden_state):
  class LevitAttention (line 172) | class LevitAttention(nn.Module):
    method __init__ (line 173) | def __init__(self, hidden_sizes, key_dim, num_attention_heads, attenti...
    method train (line 201) | def train(self, mode=True):
    method get_attention_biases (line 206) | def get_attention_biases(self, device):
    method forward (line 215) | def forward(self, hidden_state):
  class LevitAttentionSubsample (line 232) | class LevitAttentionSubsample(nn.Module):
    method __init__ (line 233) | def __init__(
    method train (line 277) | def train(self, mode=True):
    method get_attention_biases (line 282) | def get_attention_biases(self, device):
    method forward (line 291) | def forward(self, hidden_state):
  class LevitMLPLayer (line 313) | class LevitMLPLayer(nn.Module):
    method __init__ (line 318) | def __init__(self, input_dim, hidden_dim):
    method forward (line 324) | def forward(self, hidden_state):
  class LevitResidualLayer (line 331) | class LevitResidualLayer(nn.Module):
    method __init__ (line 336) | def __init__(self, module, drop_rate):
    method forward (line 341) | def forward(self, hidden_state):
  class LevitStage (line 352) | class LevitStage(nn.Module):
    method __init__ (line 357) | def __init__(
    method get_resolution (line 412) | def get_resolution(self):
    method forward (line 415) | def forward(self, hidden_state):
  class LevitEncoder (line 421) | class LevitEncoder(nn.Module):
    method __init__ (line 426) | def __init__(self, config):
    method forward (line 451) | def forward(self, hidden_state, output_hidden_states=False, return_dic...
  class LevitClassificationLayer (line 467) | class LevitClassificationLayer(nn.Module):
    method __init__ (line 472) | def __init__(self, input_dim, output_dim):
    method forward (line 477) | def forward(self, hidden_state):
  class LevitPreTrainedModel (line 483) | class LevitPreTrainedModel(PreTrainedModel):
    method _init_weights (line 494) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 506) | def _set_gradient_checkpointing(self, module, value=False):
  class LevitModel (line 540) | class LevitModel(LevitPreTrainedModel):
    method __init__ (line 541) | def __init__(self, config):
    method forward (line 557) | def forward(
  class LevitForImageClassification (line 600) | class LevitForImageClassification(LevitPreTrainedModel):
    method __init__ (line 601) | def __init__(self, config):
    method forward (line 624) | def forward(
  class LevitForImageClassificationWithTeacher (line 687) | class LevitForImageClassificationWithTeacher(LevitPreTrainedModel):
    method __init__ (line 688) | def __init__(self, config):
    method forward (line 716) | def forward(

FILE: transformers/models/lilt/configuration_lilt.py
  class LiltConfig (line 30) | class LiltConfig(PretrainedConfig):
    method __init__ (line 95) | def __init__(

FILE: transformers/models/lilt/modeling_lilt.py
  class LiltTextEmbeddings (line 49) | class LiltTextEmbeddings(nn.Module):
    method __init__ (line 50) | def __init__(self, config):
    method forward (line 71) | def forward(
    method create_position_ids_from_input_ids (line 107) | def create_position_ids_from_input_ids(self, input_ids, padding_idx):
    method create_position_ids_from_inputs_embeds (line 120) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class LiltLayoutEmbeddings (line 136) | class LiltLayoutEmbeddings(nn.Module):
    method __init__ (line 137) | def __init__(self, config):
    method forward (line 158) | def forward(self, bbox=None, position_ids=None):
  class LiltSelfAttention (line 192) | class LiltSelfAttention(nn.Module):
    method __init__ (line 193) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 229) | def transpose_for_scores(self, x, r=1):
    method forward (line 234) | def forward(
  class LiltSelfOutput (line 330) | class LiltSelfOutput(nn.Module):
    method __init__ (line 331) | def __init__(self, config):
    method forward (line 337) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LiltAttention (line 344) | class LiltAttention(nn.Module):
    method __init__ (line 345) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 357) | def prune_heads(self, heads):
    method forward (line 375) | def forward(
  class LiltIntermediate (line 397) | class LiltIntermediate(nn.Module):
    method __init__ (line 398) | def __init__(self, config):
    method forward (line 406) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class LiltOutput (line 413) | class LiltOutput(nn.Module):
    method __init__ (line 414) | def __init__(self, config):
    method forward (line 420) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LiltLayer (line 427) | class LiltLayer(nn.Module):
    method __init__ (line 428) | def __init__(self, config):
    method forward (line 445) | def forward(
    method feed_forward_chunk (line 476) | def feed_forward_chunk(self, attention_output):
    method layout_feed_forward_chunk (line 481) | def layout_feed_forward_chunk(self, attention_output):
  class LiltEncoder (line 487) | class LiltEncoder(nn.Module):
    method __init__ (line 489) | def __init__(self, config):
    method forward (line 495) | def forward(
  class LiltPooler (line 565) | class LiltPooler(nn.Module):
    method __init__ (line 566) | def __init__(self, config):
    method forward (line 571) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class LiltPreTrainedModel (line 581) | class LiltPreTrainedModel(PreTrainedModel):
    method _init_weights (line 593) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 609) | def _set_gradient_checkpointing(self, module, value=False):
    method update_keys_to_ignore (line 613) | def update_keys_to_ignore(self, config, del_keys_to_ignore):
  class LiltModel (line 699) | class LiltModel(LiltPreTrainedModel):
    method __init__ (line 702) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 715) | def get_input_embeddings(self):
    method set_input_embeddings (line 718) | def set_input_embeddings(self, value):
    method _prune_heads (line 721) | def _prune_heads(self, heads_to_prune):
    method forward (line 731) | def forward(
  class LiltForSequenceClassification (line 849) | class LiltForSequenceClassification(LiltPreTrainedModel):
    method __init__ (line 853) | def __init__(self, config):
    method forward (line 866) | def forward(
  class LiltForTokenClassification (line 969) | class LiltForTokenClassification(LiltPreTrainedModel):
    method __init__ (line 974) | def __init__(self, config):
    method forward (line 990) | def forward(
  class LiltClassificationHead (line 1069) | class LiltClassificationHead(nn.Module):
    method __init__ (line 1072) | def __init__(self, config):
    method forward (line 1081) | def forward(self, features, **kwargs):
  class LiltForQuestionAnswering (line 1098) | class LiltForQuestionAnswering(LiltPreTrainedModel):
    method __init__ (line 1103) | def __init__(self, config):
    method forward (line 1115) | def forward(

FILE: transformers/models/llama/configuration_llama.py
  class LlamaConfig (line 31) | class LlamaConfig(PretrainedConfig):
    method __init__ (line 84) | def __init__(

FILE: transformers/models/llama/convert_llama_weights_to_hf.py
  function compute_intermediate_size (line 71) | def compute_intermediate_size(n):
  function read_json (line 75) | def read_json(path):
  function write_json (line 80) | def write_json(text, path):
  function write_model (line 85) | def write_model(model_path, input_base_path, model_size):
  function write_tokenizer (line 244) | def write_tokenizer(tokenizer_path, input_tokenizer_path):
  function main (line 252) | def main():

FILE: transformers/models/llama/modeling_llama.py
  function _make_causal_mask (line 42) | def _make_causal_mask(
  function _expand_mask (line 60) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class LlamaRMSNorm (line 74) | class LlamaRMSNorm(nn.Module):
    method __init__ (line 75) | def __init__(self, hidden_size, eps=1e-6):
    method forward (line 83) | def forward(self, hidden_states):
  class LlamaRotaryEmbedding (line 91) | class LlamaRotaryEmbedding(torch.nn.Module):
    method __init__ (line 92) | def __init__(self, dim, max_position_embeddings=2048, base=10000, devi...
    method forward (line 106) | def forward(self, x, seq_len=None):
  function rotate_half (line 123) | def rotate_half(x):
  function apply_rotary_pos_emb (line 130) | def apply_rotary_pos_emb(q, k, cos, sin, position_ids):
  class LlamaMLP (line 141) | class LlamaMLP(nn.Module):
    method __init__ (line 142) | def __init__(
    method forward (line 154) | def forward(self, x, task_types=None):
  class LlamaAttention (line 161) | class LlamaAttention(nn.Module):
    method __init__ (line 164) | def __init__(self, config: LlamaConfig):
    method _shape (line 183) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 186) | def forward(
  class LlamaDecoderLayer (line 254) | class LlamaDecoderLayer(nn.Module):
    method __init__ (line 255) | def __init__(self, config: LlamaConfig):
    method forward (line 267) | def forward(
  class LlamaPreTrainedModel (line 344) | class LlamaPreTrainedModel(PreTrainedModel):
    method _init_weights (line 352) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 363) | def _set_gradient_checkpointing(self, module, value=False):
  class LlamaModel (line 436) | class LlamaModel(LlamaPreTrainedModel):
    method __init__ (line 444) | def __init__(self, config: LlamaConfig):
    method get_input_embeddings (line 457) | def get_input_embeddings(self):
    method set_input_embeddings (line 460) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 464) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 488) | def forward(
  class LlamaForCausalLM (line 621) | class LlamaForCausalLM(LlamaPreTrainedModel):
    method __init__ (line 622) | def __init__(self, config):
    method get_input_embeddings (line 631) | def get_input_embeddings(self):
    method set_input_embeddings (line 634) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 637) | def get_output_embeddings(self):
    method set_output_embeddings (line 640) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 643) | def set_decoder(self, decoder):
    method get_decoder (line 646) | def get_decoder(self):
    method forward (line 651) | def forward(
    method prepare_inputs_for_generation (line 740) | def prepare_inputs_for_generation(
    method _reorder_cache (line 771) | def _reorder_cache(past_key_values, beam_idx):
  class LlamaForSequenceClassification (line 793) | class LlamaForSequenceClassification(LlamaPreTrainedModel):
    method __init__ (line 796) | def __init__(self, config):
    method get_input_embeddings (line 805) | def get_input_embeddings(self):
    method set_input_embeddings (line 808) | def set_input_embeddings(self, value):
    method forward (line 812) | def forward(

FILE: transformers/models/llama/tokenization_llama.py
  class LlamaTokenizer (line 49) | class LlamaTokenizer(PreTrainedTokenizer):
    method __init__ (line 63) | def __init__(
    method __getstate__ (line 98) | def __getstate__(self):
    method __setstate__ (line 103) | def __setstate__(self, d):
    method vocab_size (line 109) | def vocab_size(self):
    method get_vocab (line 113) | def get_vocab(self):
    method _tokenize (line 119) | def _tokenize(self, text):
    method _convert_token_to_id (line 123) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 127) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 132) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 151) | def save_vocabulary(self, save_directory, filename_prefix: Optional[st...
    method build_inputs_with_special_tokens (line 178) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method get_special_tokens_mask (line 189) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 226) | def create_token_type_ids_from_sequences(

FILE: transformers/models/llama/tokenization_llama_fast.py
  class LlamaTokenizerFast (line 37) | class LlamaTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 81) | def __init__(
    method update_post_processor (line 109) | def update_post_processor(self):
    method add_eos_token (line 129) | def add_eos_token(self):
    method add_bos_token (line 133) | def add_bos_token(self):
    method add_eos_token (line 137) | def add_eos_token(self, value):
    method add_bos_token (line 142) | def add_bos_token(self, value):
    method save_vocabulary (line 146) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/longformer/configuration_longformer.py
  class LongformerConfig (line 46) | class LongformerConfig(PretrainedConfig):
    method __init__ (line 109) | def __init__(
  class LongformerOnnxConfig (line 153) | class LongformerOnnxConfig(OnnxConfig):
    method __init__ (line 154) | def __init__(self, config: "PretrainedConfig", task: str = "default", ...
    method inputs (line 159) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 173) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 180) | def atol_for_validation(self) -> float:
    method default_onnx_opset (line 190) | def default_onnx_opset(self) -> int:
    method generate_dummy_inputs (line 194) | def generate_dummy_inputs(

FILE: transformers/models/longformer/convert_longformer_original_pytorch_lightning_to_pytorch.py
  class LightningModel (line 27) | class LightningModel(pl.LightningModule):
    method __init__ (line 28) | def __init__(self, model):
    method forward (line 35) | def forward(self):
  function convert_longformer_qa_checkpoint_to_pytorch (line 39) | def convert_longformer_qa_checkpoint_to_pytorch(

FILE: transformers/models/longformer/modeling_longformer.py
  class LongformerBaseModelOutput (line 56) | class LongformerBaseModelOutput(ModelOutput):
  class LongformerBaseModelOutputWithPooling (line 99) | class LongformerBaseModelOutputWithPooling(ModelOutput):
  class LongformerMaskedLMOutput (line 147) | class LongformerMaskedLMOutput(ModelOutput):
  class LongformerQuestionAnsweringModelOutput (line 193) | class LongformerQuestionAnsweringModelOutput(ModelOutput):
  class LongformerSequenceClassifierOutput (line 242) | class LongformerSequenceClassifierOutput(ModelOutput):
  class LongformerMultipleChoiceModelOutput (line 288) | class LongformerMultipleChoiceModelOutput(ModelOutput):
  class LongformerTokenClassifierOutput (line 336) | class LongformerTokenClassifierOutput(ModelOutput):
  function _get_question_end_index (line 381) | def _get_question_end_index(input_ids, sep_token_id):
  function _compute_global_attention_mask (line 397) | def _compute_global_attention_mask(input_ids, sep_token_id, before_sep_t...
  function create_position_ids_from_input_ids (line 417) | def create_position_ids_from_input_ids(input_ids, padding_idx):
  class LongformerEmbeddings (line 433) | class LongformerEmbeddings(nn.Module):
    method __init__ (line 438) | def __init__(self, config):
    method forward (line 453) | def forward(self, input_ids=None, token_type_ids=None, position_ids=No...
    method create_position_ids_from_inputs_embeds (line 479) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class LongformerSelfAttention (line 497) | class LongformerSelfAttention(nn.Module):
    method __init__ (line 498) | def __init__(self, config, layer_id):
    method forward (line 533) | def forward(
    method _pad_and_transpose_last_two_dims (line 702) | def _pad_and_transpose_last_two_dims(hidden_states_padded, padding):
    method _pad_and_diagonalize (line 713) | def _pad_and_diagonalize(chunked_hidden_states):
    method _chunk (line 762) | def _chunk(hidden_states, window_overlap, onnx_export: bool = False):
    method _mask_invalid_locations (line 803) | def _mask_invalid_locations(input_tensor, affected_seq_len) -> torch.T...
    method _sliding_chunks_query_key_matmul (line 818) | def _sliding_chunks_query_key_matmul(self, query: torch.Tensor, key: t...
    method _sliding_chunks_matmul_attn_probs_value (line 884) | def _sliding_chunks_matmul_attn_probs_value(
    method _get_global_attn_indices (line 929) | def _get_global_attn_indices(is_index_global_attn):
    method _concat_with_global_key_attn_probs (line 957) | def _concat_with_global_key_attn_probs(
    method _compute_attn_output_with_global_indices (line 987) | def _compute_attn_output_with_global_indices(
    method _compute_global_attn_output_from_hidden (line 1023) | def _compute_global_attn_output_from_hidden(
  class LongformerSelfOutput (line 1133) | class LongformerSelfOutput(nn.Module):
    method __init__ (line 1134) | def __init__(self, config):
    method forward (line 1140) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LongformerAttention (line 1147) | class LongformerAttention(nn.Module):
    method __init__ (line 1148) | def __init__(self, config, layer_id=0):
    method prune_heads (line 1154) | def prune_heads(self, heads):
    method forward (line 1172) | def forward(
  class LongformerIntermediate (line 1197) | class LongformerIntermediate(nn.Module):
    method __init__ (line 1198) | def __init__(self, config):
    method forward (line 1206) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class LongformerOutput (line 1213) | class LongformerOutput(nn.Module):
    method __init__ (line 1214) | def __init__(self, config):
    method forward (line 1220) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LongformerLayer (line 1227) | class LongformerLayer(nn.Module):
    method __init__ (line 1228) | def __init__(self, config, layer_id=0):
    method forward (line 1236) | def forward(
    method ff_chunk (line 1264) | def ff_chunk(self, attn_output):
  class LongformerEncoder (line 1270) | class LongformerEncoder(nn.Module):
    method __init__ (line 1271) | def __init__(self, config):
    method forward (line 1277) | def forward(
  class LongformerPooler (line 1368) | class LongformerPooler(nn.Module):
    method __init__ (line 1369) | def __init__(self, config):
    method forward (line 1374) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class LongformerLMHead (line 1384) | class LongformerLMHead(nn.Module):
    method __init__ (line 1387) | def __init__(self, config):
    method forward (line 1396) | def forward(self, features, **kwargs):
    method _tie_weights (line 1406) | def _tie_weights(self):
  class LongformerPreTrainedModel (line 1415) | class LongformerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1427) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1443) | def _set_gradient_checkpointing(self, module, value=False):
  class LongformerModel (line 1535) | class LongformerModel(LongformerPreTrainedModel):
    method __init__ (line 1551) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 1572) | def get_input_embeddings(self):
    method set_input_embeddings (line 1575) | def set_input_embeddings(self, value):
    method _prune_heads (line 1578) | def _prune_heads(self, heads_to_prune):
    method _pad_to_window_size (line 1586) | def _pad_to_window_size(
    method _merge_to_attention_mask (line 1636) | def _merge_to_attention_mask(self, attention_mask: torch.Tensor, globa...
    method forward (line 1650) | def forward(
  class LongformerForMaskedLM (line 1772) | class LongformerForMaskedLM(LongformerPreTrainedModel):
    method __init__ (line 1776) | def __init__(self, config):
    method get_output_embeddings (line 1785) | def get_output_embeddings(self):
    method set_output_embeddings (line 1788) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1793) | def forward(
  class LongformerForSequenceClassification (line 1887) | class LongformerForSequenceClassification(LongformerPreTrainedModel):
    method __init__ (line 1890) | def __init__(self, config):
    method forward (line 1909) | def forward(
  class LongformerClassificationHead (line 1990) | class LongformerClassificationHead(nn.Module):
    method __init__ (line 1993) | def __init__(self, config):
    method forward (line 1999) | def forward(self, hidden_states, **kwargs):
  class LongformerForQuestionAnswering (line 2016) | class LongformerForQuestionAnswering(LongformerPreTrainedModel):
    method __init__ (line 2019) | def __init__(self, config):
    method forward (line 2031) | def forward(
  class LongformerForTokenClassification (line 2155) | class LongformerForTokenClassification(LongformerPreTrainedModel):
    method __init__ (line 2158) | def __init__(self, config):
    method forward (line 2180) | def forward(
  class LongformerForMultipleChoice (line 2245) | class LongformerForMultipleChoice(LongformerPreTrainedModel):
    method __init__ (line 2246) | def __init__(self, config):
    method forward (line 2264) | def forward(

FILE: transformers/models/longformer/modeling_tf_longformer.py
  class TFLongformerBaseModelOutput (line 69) | class TFLongformerBaseModelOutput(ModelOutput):
  class TFLongformerBaseModelOutputWithPooling (line 112) | class TFLongformerBaseModelOutputWithPooling(ModelOutput):
  class TFLongformerMaskedLMOutput (line 160) | class TFLongformerMaskedLMOutput(ModelOutput):
  class TFLongformerQuestionAnsweringModelOutput (line 206) | class TFLongformerQuestionAnsweringModelOutput(ModelOutput):
  class TFLongformerSequenceClassifierOutput (line 255) | class TFLongformerSequenceClassifierOutput(ModelOutput):
  class TFLongformerMultipleChoiceModelOutput (line 301) | class TFLongformerMultipleChoiceModelOutput(ModelOutput):
  class TFLongformerTokenClassifierOutput (line 349) | class TFLongformerTokenClassifierOutput(ModelOutput):
  function _compute_global_attention_mask (line 394) | def _compute_global_attention_mask(input_ids_shape, sep_token_indices, b...
  class TFLongformerLMHead (line 419) | class TFLongformerLMHead(tf.keras.layers.Layer):
    method __init__ (line 422) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 437) | def build(self, input_shape):
    method get_output_embeddings (line 442) | def get_output_embeddings(self):
    method set_output_embeddings (line 445) | def set_output_embeddings(self, value):
    method get_bias (line 449) | def get_bias(self):
    method set_bias (line 452) | def set_bias(self, value):
    method call (line 456) | def call(self, hidden_states):
  class TFLongformerEmbeddings (line 471) | class TFLongformerEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 476) | def __init__(self, config, **kwargs):
    method build (line 487) | def build(self, input_shape: tf.TensorShape):
    method create_position_ids_from_input_ids (line 511) | def create_position_ids_from_input_ids(self, input_ids, past_key_value...
    method call (line 525) | def call(
  class TFLongformerIntermediate (line 573) | class TFLongformerIntermediate(tf.keras.layers.Layer):
    method __init__ (line 574) | def __init__(self, config: LongformerConfig, **kwargs):
    method call (line 586) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFLongformerOutput (line 594) | class TFLongformerOutput(tf.keras.layers.Layer):
    method __init__ (line 595) | def __init__(self, config: LongformerConfig, **kwargs):
    method call (line 604) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFLongformerPooler (line 613) | class TFLongformerPooler(tf.keras.layers.Layer):
    method __init__ (line 614) | def __init__(self, config: LongformerConfig, **kwargs):
    method call (line 624) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFLongformerSelfOutput (line 634) | class TFLongformerSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 635) | def __init__(self, config: LongformerConfig, **kwargs):
    method call (line 644) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFLongformerSelfAttention (line 652) | class TFLongformerSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 653) | def __init__(self, config, layer_id, **kwargs):
    method build (line 712) | def build(self, input_shape=None):
    method call (line 722) | def call(
    method _sliding_chunks_query_key_matmul (line 913) | def _sliding_chunks_query_key_matmul(self, query, key, window_overlap):
    method _mask_invalid_locations (line 1030) | def _mask_invalid_locations(input_tensor, window_overlap):
    method _sliding_chunks_matmul_attn_probs_value (line 1059) | def _sliding_chunks_matmul_attn_probs_value(self, attn_probs, value, w...
    method _pad_and_transpose_last_two_dims (line 1133) | def _pad_and_transpose_last_two_dims(hidden_states_padded, paddings):
    method _pad_and_diagonalize (line 1144) | def _pad_and_diagonalize(chunked_hidden_states):
    method _chunk (line 1196) | def _chunk(hidden_states, window_overlap):
    method _get_global_attn_indices (line 1226) | def _get_global_attn_indices(is_index_global_attn):
    method _concat_with_global_key_attn_probs (line 1256) | def _concat_with_global_key_attn_probs(
    method _compute_attn_output_with_global_indices (line 1310) | def _compute_attn_output_with_global_indices(
    method _compute_global_attn_output_from_hidden (line 1351) | def _compute_global_attn_output_from_hidden(
    method reshape_and_transpose (line 1488) | def reshape_and_transpose(self, vector, batch_size):
  class TFLongformerAttention (line 1498) | class TFLongformerAttention(tf.keras.layers.Layer):
    method __init__ (line 1499) | def __init__(self, config, layer_id=0, **kwargs):
    method prune_heads (line 1505) | def prune_heads(self, heads):
    method call (line 1508) | def call(self, inputs, training=False):
  class TFLongformerLayer (line 1528) | class TFLongformerLayer(tf.keras.layers.Layer):
    method __init__ (line 1529) | def __init__(self, config, layer_id=0, **kwargs):
    method call (line 1536) | def call(self, inputs, training=False):
  class TFLongformerEncoder (line 1558) | class TFLongformerEncoder(tf.keras.layers.Layer):
    method __init__ (line 1559) | def __init__(self, config, **kwargs):
    method call (line 1566) | def call(
  class TFLongformerMainLayer (line 1637) | class TFLongformerMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1640) | def __init__(self, config, add_pooling_layer=True, **kwargs):
    method get_input_embeddings (line 1665) | def get_input_embeddings(self):
    method set_input_embeddings (line 1668) | def set_input_embeddings(self, value):
    method _prune_heads (line 1672) | def _prune_heads(self, heads_to_prune):
    method call (line 1680) | def call(
    method _pad_to_window_size (line 1800) | def _pad_to_window_size(
    method _merge_to_attention_mask (line 1855) | def _merge_to_attention_mask(attention_mask: tf.Tensor, global_attenti...
  class TFLongformerPreTrainedModel (line 1869) | class TFLongformerPreTrainedModel(TFPreTrainedModel):
    method input_signature (line 1879) | def input_signature(self):
  class TFLongformerModel (line 1999) | class TFLongformerModel(TFLongformerPreTrainedModel):
    method __init__ (line 2016) | def __init__(self, config, *inputs, **kwargs):
    method call (line 2023) | def call(
  class TFLongformerForMaskedLM (line 2058) | class TFLongformerForMaskedLM(TFLongformerPreTrainedModel, TFMaskedLangu...
    method __init__ (line 2062) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 2068) | def get_lm_head(self):
    method get_prefix_bias_name (line 2071) | def get_prefix_bias_name(self):
    method call (line 2085) | def call(
  class TFLongformerForQuestionAnswering (line 2145) | class TFLongformerForQuestionAnswering(TFLongformerPreTrainedModel, TFQu...
    method __init__ (line 2149) | def __init__(self, config, *inputs, **kwargs):
    method call (line 2169) | def call(
  class TFLongformerClassificationHead (line 2268) | class TFLongformerClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 2271) | def __init__(self, config, **kwargs):
    method call (line 2284) | def call(self, hidden_states, training=False):
  class TFLongformerForSequenceClassification (line 2300) | class TFLongformerForSequenceClassification(TFLongformerPreTrainedModel,...
    method __init__ (line 2304) | def __init__(self, config, *inputs, **kwargs):
    method call (line 2319) | def call(
  class TFLongformerForMultipleChoice (line 2403) | class TFLongformerForMultipleChoice(TFLongformerPreTrainedModel, TFMulti...
    method __init__ (line 2407) | def __init__(self, config, *inputs, **kwargs):
    method input_signature (line 2417) | def input_signature(self):
    method call (line 2433) | def call(
  class TFLongformerForTokenClassification (line 2517) | class TFLongformerForTokenClassification(TFLongformerPreTrainedModel, TF...
    method __init__ (line 2522) | def __init__(self, config, *inputs, **kwargs):
    method call (line 2539) | def call(

FILE: transformers/models/longformer/tokenization_longformer.py
  function bytes_to_unicode (line 76) | def bytes_to_unicode():
  function get_pairs (line 101) | def get_pairs(word):
  class LongformerTokenizer (line 116) | class LongformerTokenizer(PreTrainedTokenizer):
    method __init__ (line 199) | def __init__(
    method vocab_size (line 254) | def vocab_size(self):
    method get_vocab (line 257) | def get_vocab(self):
    method bpe (line 260) | def bpe(self, token):
    method _tokenize (line 302) | def _tokenize(self, text):
    method _convert_token_to_id (line 312) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 316) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 320) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 326) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 355) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 380) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 407) | def create_token_type_ids_from_sequences(
    method prepare_for_tokenization (line 430) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...

FILE: transformers/models/longformer/tokenization_longformer_fast.py
  class LongformerTokenizerFast (line 91) | class LongformerTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 178) | def __init__(
    method mask_token (line 247) | def mask_token(self) -> str:
    method mask_token (line 262) | def mask_token(self, value):
    method _batch_encode_plus (line 273) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 282) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method save_vocabulary (line 292) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 296) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 303) | def create_token_type_ids_from_sequences(

FILE: transformers/models/longt5/configuration_longt5.py
  class LongT5Config (line 33) | class LongT5Config(PretrainedConfig):
    method __init__ (line 89) | def __init__(
  class LongT5OnnxConfig (line 155) | class LongT5OnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 157) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method default_onnx_opset (line 176) | def default_onnx_opset(self) -> int:

FILE: transformers/models/longt5/convert_longt5x_checkpoint_to_flax.py
  function convert_t5x_checkpoint_to_flax (line 27) | def convert_t5x_checkpoint_to_flax(t5x_checkpoint_path, config_name, fla...

FILE: transformers/models/longt5/modeling_flax_longt5.py
  function shift_tokens_right (line 59) | def shift_tokens_right(input_ids: np.array, pad_token_id: int, decoder_s...
  function _pad_to_multiple (line 71) | def _pad_to_multiple(x: jnp.ndarray, block_len: int, axis: int, pad_valu...
  function _split_into_blocks (line 80) | def _split_into_blocks(x: jnp.ndarray, block_len: int, axis: int) -> jnp...
  function _concatenate_3_blocks (line 92) | def _concatenate_3_blocks(x: jnp.ndarray, block_axis: int, sequence_axis...
  function _make_3block_relative_position_ids (line 114) | def _make_3block_relative_position_ids(block_len: int) -> jnp.ndarray:
  function _mask_local_attention_mask (line 122) | def _mask_local_attention_mask(local_attention_mask: np.ndarray, block_l...
  function _get_local_attention_mask (line 130) | def _get_local_attention_mask(attention_mask: np.ndarray, block_len: int...
  function _make_global_fixed_block_ids (line 146) | def _make_global_fixed_block_ids(attention_mask: np.ndarray, global_bloc...
  function _make_side_relative_position_ids (line 188) | def _make_side_relative_position_ids(attention_mask: np.ndarray, global_...
  function _create_global_aggregates (line 197) | def _create_global_aggregates(hidden_states: np.ndarray, block_ids: np.n...
  class FlaxLongT5LayerNorm (line 205) | class FlaxLongT5LayerNorm(nn.Module):
    method setup (line 211) | def setup(self):
    method __call__ (line 214) | def __call__(self, hidden_states):
  class FlaxLongT5DenseActDense (line 226) | class FlaxLongT5DenseActDense(nn.Module):
    method setup (line 230) | def setup(self):
    method __call__ (line 249) | def __call__(self, hidden_states, deterministic=True):
  class FlaxLongT5DenseGatedActDense (line 258) | class FlaxLongT5DenseGatedActDense(nn.Module):
    method setup (line 262) | def setup(self):
    method __call__ (line 287) | def __call__(self, hidden_states, deterministic):
  class FlaxLongT5LayerFF (line 297) | class FlaxLongT5LayerFF(nn.Module):
    method setup (line 301) | def setup(self):
    method __call__ (line 312) | def __call__(self, hidden_states, deterministic=True):
  class FlaxLongT5Attention (line 320) | class FlaxLongT5Attention(nn.Module):
    method setup (line 326) | def setup(self):
    method _relative_position_bucket (line 373) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 408) | def compute_bias(self, query_length, key_length):
    method _split_heads (line 425) | def _split_heads(self, hidden_states):
    method _merge_heads (line 428) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 432) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method _create_position_bias (line 464) | def _create_position_bias(
    method __call__ (line 488) | def __call__(
  class FlaxLongT5LocalAttention (line 604) | class FlaxLongT5LocalAttention(nn.Module):
    method setup (line 609) | def setup(self):
    method _relative_position_bucket (line 658) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 693) | def compute_bias(self, block_length: int):
    method _split_heads (line 710) | def _split_heads(self, hidden_states):
    method _merge_heads (line 713) | def _merge_heads(self, hidden_states):
    method _create_position_bias (line 716) | def _create_position_bias(self, block_len: int, attention_mask: Option...
    method __call__ (line 727) | def __call__(
  class FlaxLongT5TransientGlobalAttention (line 815) | class FlaxLongT5TransientGlobalAttention(nn.Module):
    method setup (line 820) | def setup(self):
    method _relative_position_bucket (line 881) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 916) | def compute_bias(self, block_length: int):
    method compute_side_bias (line 933) | def compute_side_bias(self, attention_mask: np.ndarray, global_segment...
    method _split_heads (line 958) | def _split_heads(self, hidden_states):
    method _merge_heads (line 961) | def _merge_heads(self, hidden_states):
    method _create_position_bias (line 964) | def _create_position_bias(self, block_len: int, attention_mask: Option...
    method __call__ (line 975) | def __call__(
  class FlaxLongT5LayerLocalSelfAttention (line 1103) | class FlaxLongT5LayerLocalSelfAttention(nn.Module):
    method setup (line 1110) | def setup(self):
    method __call__ (line 1119) | def __call__(
  class FlaxLongT5LayerTransientGlobalSelfAttention (line 1141) | class FlaxLongT5LayerTransientGlobalSelfAttention(nn.Module):
    method setup (line 1148) | def setup(self):
    method __call__ (line 1157) | def __call__(
  class FlaxLongT5LayerSelfAttention (line 1180) | class FlaxLongT5LayerSelfAttention(nn.Module):
    method setup (line 1185) | def setup(self):
    method __call__ (line 1197) | def __call__(
  class FlaxLongT5LayerCrossAttention (line 1221) | class FlaxLongT5LayerCrossAttention(nn.Module):
    method setup (line 1225) | def setup(self):
    method __call__ (line 1234) | def __call__(
  class FlaxLongT5Block (line 1256) | class FlaxLongT5Block(nn.Module):
    method setup (line 1261) | def setup(self):
    method __call__ (line 1290) | def __call__(
  class FlaxLongT5LayerCollection (line 1342) | class FlaxLongT5LayerCollection(nn.Module):
    method setup (line 1347) | def setup(self):
    method __call__ (line 1352) | def __call__(
  class FlaxLongT5BlockCollection (line 1378) | class FlaxLongT5BlockCollection(nn.Module):
    method setup (line 1383) | def setup(self):
    method __call__ (line 1407) | def __call__(
  class FlaxLongT5Stack (line 1465) | class FlaxLongT5Stack(nn.Module):
    method setup (line 1471) | def setup(self):
    method __call__ (line 1482) | def __call__(
  class FlaxLongT5PreTrainedModel (line 1666) | class FlaxLongT5PreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 1676) | def __init__(
    method enable_gradient_checkpointing (line 1688) | def enable_gradient_checkpointing(self):
    method init_weights (line 1695) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 1725) | def __call__(
    method init_cache (line 1774) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 1812) | def encode(
    method decode (line 1870) | def decode(
  class FlaxLongT5Module (line 2018) | class FlaxLongT5Module(nn.Module):
    method _get_encoder_module (line 2023) | def _get_encoder_module(self):
    method _get_decoder_module (line 2026) | def _get_decoder_module(self):
    method setup (line 2029) | def setup(self):
    method __call__ (line 2056) | def __call__(
  class FlaxLongT5Model (line 2108) | class FlaxLongT5Model(FlaxLongT5PreTrainedModel):
  class FlaxLongT5ForConditionalGenerationModule (line 2143) | class FlaxLongT5ForConditionalGenerationModule(nn.Module):
    method _get_encoder_module (line 2148) | def _get_encoder_module(self):
    method _get_decoder_module (line 2151) | def _get_decoder_module(self):
    method setup (line 2154) | def setup(self):
    method __call__ (line 2187) | def __call__(
  class FlaxLongT5ForConditionalGeneration (line 2253) | class FlaxLongT5ForConditionalGeneration(FlaxLongT5PreTrainedModel):
    method decode (line 2258) | def decode(
    method prepare_inputs_for_generation (line 2387) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 2416) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/longt5/modeling_longt5.py
  function _pad_to_multiple (line 63) | def _pad_to_multiple(x: torch.Tensor, block_len: int, dim: int, pad_valu...
  function _split_into_blocks (line 79) | def _split_into_blocks(x: torch.Tensor, block_len: int, dim: int) -> tor...
  function _concatenate_3_blocks (line 94) | def _concatenate_3_blocks(x: torch.Tensor, block_dim: int, sequence_dim:...
  function _make_3block_relative_position_ids (line 119) | def _make_3block_relative_position_ids(block_len: int) -> torch.Tensor:
  function _mask_local_attention_mask (line 128) | def _mask_local_attention_mask(local_attention_mask: torch.Tensor, block...
  function _get_local_attention_mask (line 137) | def _get_local_attention_mask(attention_mask: torch.Tensor, block_len: i...
  function _make_global_fixed_block_ids (line 153) | def _make_global_fixed_block_ids(
  function _make_side_relative_position_ids (line 202) | def _make_side_relative_position_ids(attention_mask: torch.Tensor, globa...
  function _create_global_aggregates (line 211) | def _create_global_aggregates(
  class LongT5LayerNorm (line 224) | class LongT5LayerNorm(nn.Module):
    method __init__ (line 225) | def __init__(self, hidden_size, eps=1e-6):
    method forward (line 233) | def forward(self, hidden_states):
  class LongT5DenseActDense (line 266) | class LongT5DenseActDense(nn.Module):
    method __init__ (line 267) | def __init__(self, config: LongT5Config):
    method forward (line 274) | def forward(self, hidden_states):
  class LongT5DenseGatedActDense (line 288) | class LongT5DenseGatedActDense(nn.Module):
    method __init__ (line 289) | def __init__(self, config: LongT5Config):
    method forward (line 297) | def forward(self, hidden_states):
  class LongT5LayerFF (line 307) | class LongT5LayerFF(nn.Module):
    method __init__ (line 308) | def __init__(self, config: LongT5Config):
    method forward (line 318) | def forward(self, hidden_states):
  class LongT5Attention (line 326) | class LongT5Attention(nn.Module):
    method __init__ (line 327) | def __init__(self, config: LongT5Config, has_relative_attention_bias=F...
    method prune_heads (line 350) | def prune_heads(self, heads):
    method _relative_position_bucket (line 367) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 414) | def compute_bias(self, query_length, key_length, device=None):
    method forward (line 431) | def forward(
  class LongT5LocalAttention (line 561) | class LongT5LocalAttention(nn.Module):
    method __init__ (line 562) | def __init__(self, config: LongT5Config, has_relative_attention_bias: ...
    method prune_heads (line 588) | def prune_heads(self, heads):
    method _relative_position_bucket (line 606) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 653) | def compute_bias(self, block_length: int):
    method forward (line 677) | def forward(
  class LongT5TransientGlobalAttention (line 753) | class LongT5TransientGlobalAttention(nn.Module):
    method __init__ (line 754) | def __init__(self, config: LongT5Config, has_relative_attention_bias: ...
    method prune_heads (line 786) | def prune_heads(self, heads):
    method _relative_position_bucket (line 804) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 851) | def compute_bias(self, block_length: int):
    method compute_side_bias (line 875) | def compute_side_bias(self, mask: torch.Tensor, global_segment_ids: to...
    method forward (line 896) | def forward(
  class LongT5LayerSelfAttention (line 1018) | class LongT5LayerSelfAttention(nn.Module):
    method __init__ (line 1019) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 1025) | def forward(
  class LongT5LayerLocalSelfAttention (line 1050) | class LongT5LayerLocalSelfAttention(nn.Module):
    method __init__ (line 1053) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 1059) | def forward(
  class LongT5LayerTransientGlobalSelfAttention (line 1081) | class LongT5LayerTransientGlobalSelfAttention(nn.Module):
    method __init__ (line 1084) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 1092) | def forward(
  class LongT5LayerCrossAttention (line 1115) | class LongT5LayerCrossAttention(nn.Module):
    method __init__ (line 1116) | def __init__(self, config):
    method forward (line 1122) | def forward(
  class LongT5Block (line 1151) | class LongT5Block(nn.Module):
    method __init__ (line 1152) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 1173) | def forward(
  class LongT5PreTrainedModel (line 1274) | class LongT5PreTrainedModel(PreTrainedModel):
    method dummy_inputs (line 1287) | def dummy_inputs(self):
    method _init_weights (line 1297) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1344) | def _set_gradient_checkpointing(self, module, value=False):
    method _shift_right (line 1349) | def _shift_right(self, input_ids):
  class LongT5Stack (line 1377) | class LongT5Stack(LongT5PreTrainedModel):
    method __init__ (line 1378) | def __init__(self, config, embed_tokens=None):
    method get_input_embeddings (line 1401) | def get_input_embeddings(self):
    method set_input_embeddings (line 1405) | def set_input_embeddings(self, new_embeddings):
    method forward (line 1408) | def forward(
  class LongT5Model (line 1765) | class LongT5Model(LongT5PreTrainedModel):
    method __init__ (line 1774) | def __init__(self, config: LongT5Config):
    method get_input_embeddings (line 1793) | def get_input_embeddings(self):
    method set_input_embeddings (line 1796) | def set_input_embeddings(self, new_embeddings):
    method get_encoder (line 1801) | def get_encoder(self):
    method get_decoder (line 1804) | def get_decoder(self):
    method _prune_heads (line 1807) | def _prune_heads(self, heads_to_prune):
    method forward (line 1817) | def forward(
  class LongT5ForConditionalGeneration (line 1918) | class LongT5ForConditionalGeneration(LongT5PreTrainedModel):
    method __init__ (line 1928) | def __init__(self, config: LongT5Config):
    method get_input_embeddings (line 1951) | def get_input_embeddings(self):
    method set_input_embeddings (line 1954) | def set_input_embeddings(self, new_embeddings):
    method set_output_embeddings (line 1959) | def set_output_embeddings(self, new_embeddings):
    method get_output_embeddings (line 1962) | def get_output_embeddings(self):
    method get_encoder (line 1965) | def get_encoder(self):
    method get_decoder (line 1968) | def get_decoder(self):
    method forward (line 1973) | def forward(
    method prepare_inputs_for_generation (line 2101) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 2128) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 2131) | def _reorder_cache(self, past_key_values, beam_idx):
  class LongT5EncoderModel (line 2160) | class LongT5EncoderModel(LongT5PreTrainedModel):
    method __init__ (line 2163) | def __init__(self, config: LongT5Config):
    method get_input_embeddings (line 2175) | def get_input_embeddings(self):
    method set_input_embeddings (line 2178) | def set_input_embeddings(self, new_embeddings):
    method get_encoder (line 2182) | def get_encoder(self):
    method _prune_heads (line 2185) | def _prune_heads(self, heads_to_prune):
    method forward (line 2195) | def forward(

FILE: transformers/models/luke/configuration_luke.py
  class LukeConfig (line 29) | class LukeConfig(PretrainedConfig):
    method __init__ (line 96) | def __init__(

FILE: transformers/models/luke/convert_luke_original_pytorch_checkpoint_to_pytorch.py
  function convert_luke_checkpoint (line 28) | def convert_luke_checkpoint(checkpoint_path, metadata_path, entity_vocab...
  function load_entity_vocab (line 134) | def load_entity_vocab(entity_vocab_path):

FILE: transformers/models/luke/modeling_luke.py
  class BaseLukeModelOutputWithPooling (line 54) | class BaseLukeModelOutputWithPooling(BaseModelOutputWithPooling):
  class BaseLukeModelOutput (line 85) | class BaseLukeModelOutput(BaseModelOutput):
  class LukeMaskedLMOutput (line 116) | class LukeMaskedLMOutput(ModelOutput):
  class EntityClassificationOutput (line 159) | class EntityClassificationOutput(ModelOutput):
  class EntityPairClassificationOutput (line 190) | class EntityPairClassificationOutput(ModelOutput):
  class EntitySpanClassificationOutput (line 221) | class EntitySpanClassificationOutput(ModelOutput):
  class LukeSequenceClassifierOutput (line 252) | class LukeSequenceClassifierOutput(ModelOutput):
  class LukeTokenClassifierOutput (line 286) | class LukeTokenClassifierOutput(ModelOutput):
  class LukeQuestionAnsweringModelOutput (line 320) | class LukeQuestionAnsweringModelOutput(ModelOutput):
  class LukeMultipleChoiceModelOutput (line 357) | class LukeMultipleChoiceModelOutput(ModelOutput):
  class LukeEmbeddings (line 392) | class LukeEmbeddings(nn.Module):
    method __init__ (line 397) | def __init__(self, config):
    method forward (line 414) | def forward(
    method create_position_ids_from_inputs_embeds (line 447) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class LukeEntityEmbeddings (line 465) | class LukeEntityEmbeddings(nn.Module):
    method __init__ (line 466) | def __init__(self, config: LukeConfig):
    method forward (line 480) | def forward(
  class LukeSelfAttention (line 505) | class LukeSelfAttention(nn.Module):
    method __init__ (line 506) | def __init__(self, config):
    method transpose_for_scores (line 530) | def transpose_for_scores(self, x):
    method forward (line 535) | def forward(
  class LukeSelfOutput (line 619) | class LukeSelfOutput(nn.Module):
    method __init__ (line 620) | def __init__(self, config):
    method forward (line 626) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LukeAttention (line 633) | class LukeAttention(nn.Module):
    method __init__ (line 634) | def __init__(self, config):
    method prune_heads (line 640) | def prune_heads(self, heads):
    method forward (line 643) | def forward(
  class LukeIntermediate (line 681) | class LukeIntermediate(nn.Module):
    method __init__ (line 682) | def __init__(self, config):
    method forward (line 690) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class LukeOutput (line 697) | class LukeOutput(nn.Module):
    method __init__ (line 698) | def __init__(self, config):
    method forward (line 704) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class LukeLayer (line 711) | class LukeLayer(nn.Module):
    method __init__ (line 712) | def __init__(self, config):
    method forward (line 720) | def forward(
    method feed_forward_chunk (line 757) | def feed_forward_chunk(self, attention_output):
  class LukeEncoder (line 763) | class LukeEncoder(nn.Module):
    method __init__ (line 764) | def __init__(self, config):
    method forward (line 770) | def forward(
  class LukePooler (line 848) | class LukePooler(nn.Module):
    method __init__ (line 849) | def __init__(self, config):
    method forward (line 854) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class EntityPredictionHeadTransform (line 863) | class EntityPredictionHeadTransform(nn.Module):
    method __init__ (line 864) | def __init__(self, config):
    method forward (line 873) | def forward(self, hidden_states):
  class EntityPredictionHead (line 880) | class EntityPredictionHead(nn.Module):
    method __init__ (line 881) | def __init__(self, config):
    method forward (line 888) | def forward(self, hidden_states):
  class LukePreTrainedModel (line 895) | class LukePreTrainedModel(PreTrainedModel):
    method _init_weights (line 906) | def _init_weights(self, module: nn.Module):
    method _set_gradient_checkpointing (line 923) | def _set_gradient_checkpointing(self, module, value=False):
  class LukeModel (line 1024) | class LukeModel(LukePreTrainedModel):
    method __init__ (line 1027) | def __init__(self, config: LukeConfig, add_pooling_layer: bool = True):
    method get_input_embeddings (line 1040) | def get_input_embeddings(self):
    method set_input_embeddings (line 1043) | def set_input_embeddings(self, value):
    method get_entity_embeddings (line 1046) | def get_entity_embeddings(self):
    method set_entity_embeddings (line 1049) | def set_entity_embeddings(self, value):
    method _prune_heads (line 1052) | def _prune_heads(self, heads_to_prune):
    method forward (line 1057) | def forward(
    method get_extended_attention_mask (line 1194) | def get_extended_attention_mask(
  function create_position_ids_from_input_ids (line 1225) | def create_position_ids_from_input_ids(input_ids, padding_idx):
  class LukeLMHead (line 1242) | class LukeLMHead(nn.Module):
    method __init__ (line 1245) | def __init__(self, config):
    method forward (line 1254) | def forward(self, features, **kwargs):
    method _tie_weights (line 1264) | def _tie_weights(self):
  class LukeForMaskedLM (line 1280) | class LukeForMaskedLM(LukePreTrainedModel):
    method __init__ (line 1293) | def __init__(self, config):
    method tie_weights (line 1306) | def tie_weights(self):
    method get_output_embeddings (line 1310) | def get_output_embeddings(self):
    method set_output_embeddings (line 1313) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1318) | def forward(
  class LukeForEntityClassification (line 1425) | class LukeForEntityClassification(LukePreTrainedModel):
    method __init__ (line 1426) | def __init__(self, config):
    method forward (line 1440) | def forward(
  class LukeForEntityPairClassification (line 1540) | class LukeForEntityPairClassification(LukePreTrainedModel):
    method __init__ (line 1541) | def __init__(self, config):
    method forward (line 1555) | def forward(
  class LukeForEntitySpanClassification (line 1660) | class LukeForEntitySpanClassification(LukePreTrainedModel):
    method __init__ (line 1661) | def __init__(self, config):
    method forward (line 1675) | def forward(
  class LukeForSequenceClassification (line 1806) | class LukeForSequenceClassification(LukePreTrainedModel):
    method __init__ (line 1807) | def __init__(self, config):
    method forward (line 1825) | def forward(
  class LukeForTokenClassification (line 1920) | class LukeForTokenClassification(LukePreTrainedModel):
    method __init__ (line 1921) | def __init__(self, config):
    method forward (line 1940) | def forward(
  class LukeForQuestionAnswering (line 2016) | class LukeForQuestionAnswering(LukePreTrainedModel):
    method __init__ (line 2017) | def __init__(self, config):
    method forward (line 2034) | def forward(
  class LukeForMultipleChoice (line 2135) | class LukeForMultipleChoice(LukePreTrainedModel):
    method __init__ (line 2136) | def __init__(self, config):
    method forward (line 2154) | def forward(

FILE: transformers/models/luke/tokenization_luke.py
  function bytes_to_unicode (line 154) | def bytes_to_unicode():
  function get_pairs (line 179) | def get_pairs(word):
  class LukeTokenizer (line 193) | class LukeTokenizer(PreTrainedTokenizer):
    method __init__ (line 294) | def __init__(
    method vocab_size (line 412) | def vocab_size(self):
    method get_vocab (line 416) | def get_vocab(self):
    method bpe (line 420) | def bpe(self, token):
    method _tokenize (line 463) | def _tokenize(self, text):
    method _convert_token_to_id (line 474) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 479) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 484) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 491) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 517) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 545) | def create_token_type_ids_from_sequences(
    method prepare_for_tokenization (line 569) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method __call__ (line 576) | def __call__(
    method _encode_plus (line 721) | def _encode_plus(
    method _batch_encode_plus (line 801) | def _batch_encode_plus(
    method _check_entity_input_format (line 909) | def _check_entity_input_format(self, entities: Optional[EntityInput], ...
    method _create_input_sequence (line 927) | def _create_input_sequence(
    method _batch_prepare_for_model (line 1080) | def _batch_prepare_for_model(
    method prepare_for_model (line 1162) | def prepare_for_model(
    method pad (line 1391) | def pad(
    method _pad (line 1551) | def _pad(
    method save_vocabulary (line 1690) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/lxmert/configuration_lxmert.py
  class LxmertConfig (line 29) | class LxmertConfig(PretrainedConfig):
    method __init__ (line 118) | def __init__(

FILE: transformers/models/lxmert/convert_lxmert_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 29) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, config_file, py...

FILE: transformers/models/lxmert/modeling_lxmert.py
  class GeLU (line 51) | class GeLU(nn.Module):
    method __init__ (line 52) | def __init__(self):
    method forward (line 55) | def forward(self, x):
  class LxmertModelOutput (line 60) | class LxmertModelOutput(ModelOutput):
  class LxmertForQuestionAnsweringOutput (line 106) | class LxmertForQuestionAnsweringOutput(ModelOutput):
  class LxmertForPreTrainingOutput (line 146) | class LxmertForPreTrainingOutput(ModelOutput):
  function load_tf_weights_in_lxmert (line 193) | def load_tf_weights_in_lxmert(model, config, tf_checkpoint_path):
  class LxmertEmbeddings (line 272) | class LxmertEmbeddings(nn.Module):
    method __init__ (line 275) | def __init__(self, config):
    method forward (line 286) | def forward(self, input_ids, token_type_ids=None, inputs_embeds=None):
  class LxmertAttention (line 312) | class LxmertAttention(nn.Module):
    method __init__ (line 313) | def __init__(self, config, ctx_dim=None):
    method transpose_for_scores (line 333) | def transpose_for_scores(self, x):
    method forward (line 341) | def forward(self, hidden_states, context, attention_mask=None, output_...
  class LxmertAttentionOutput (line 373) | class LxmertAttentionOutput(nn.Module):
    method __init__ (line 374) | def __init__(self, config):
    method forward (line 380) | def forward(self, hidden_states, input_tensor):
  class LxmertCrossAttentionLayer (line 387) | class LxmertCrossAttentionLayer(nn.Module):
    method __init__ (line 388) | def __init__(self, config):
    method forward (line 393) | def forward(self, input_tensor, ctx_tensor, ctx_att_mask=None, output_...
  class LxmertSelfAttentionLayer (line 402) | class LxmertSelfAttentionLayer(nn.Module):
    method __init__ (line 403) | def __init__(self, config):
    method forward (line 408) | def forward(self, input_tensor, attention_mask, output_attentions=False):
  class LxmertIntermediate (line 423) | class LxmertIntermediate(nn.Module):
    method __init__ (line 424) | def __init__(self, config):
    method forward (line 429) | def forward(self, hidden_states):
  class LxmertOutput (line 435) | class LxmertOutput(nn.Module):
    method __init__ (line 436) | def __init__(self, config):
    method forward (line 442) | def forward(self, hidden_states, input_tensor):
  class LxmertLayer (line 449) | class LxmertLayer(nn.Module):
    method __init__ (line 450) | def __init__(self, config):
    method forward (line 456) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class LxmertXLayer (line 465) | class LxmertXLayer(nn.Module):
    method __init__ (line 466) | def __init__(self, config):
    method cross_att (line 481) | def cross_att(
    method self_att (line 504) | def self_att(self, lang_input, lang_attention_mask, visual_input, visu...
    method output_fc (line 510) | def output_fc(self, lang_input, visual_input):
    method forward (line 521) | def forward(
  class LxmertVisualFeatureEncoder (line 556) | class LxmertVisualFeatureEncoder(nn.Module):
    method __init__ (line 557) | def __init__(self, config):
    method forward (line 572) | def forward(self, visual_feats, visual_pos):
  class LxmertEncoder (line 583) | class LxmertEncoder(nn.Module):
    method __init__ (line 584) | def __init__(self, config):
    method forward (line 602) | def forward(
  class LxmertPooler (line 664) | class LxmertPooler(nn.Module):
    method __init__ (line 665) | def __init__(self, config):
    method forward (line 670) | def forward(self, hidden_states):
  class LxmertPredictionHeadTransform (line 679) | class LxmertPredictionHeadTransform(nn.Module):
    method __init__ (line 680) | def __init__(self, config):
    method forward (line 686) | def forward(self, hidden_states):
  class LxmertLMPredictionHead (line 693) | class LxmertLMPredictionHead(nn.Module):
    method __init__ (line 694) | def __init__(self, config, lxmert_model_embedding_weights):
    method forward (line 708) | def forward(self, hidden_states):
  class LxmertVisualAnswerHead (line 714) | class LxmertVisualAnswerHead(nn.Module):
    method __init__ (line 715) | def __init__(self, config, num_labels):
    method forward (line 725) | def forward(self, hidden_states):
  class LxmertVisualObjHead (line 729) | class LxmertVisualObjHead(nn.Module):
    method __init__ (line 730) | def __init__(self, config):
    method forward (line 752) | def forward(self, hidden_states):
  class LxmertPreTrainingHeads (line 760) | class LxmertPreTrainingHeads(nn.Module):
    method __init__ (line 761) | def __init__(self, config, lxmert_model_embedding_weights):
    method forward (line 766) | def forward(self, sequence_output, pooled_output):
  class LxmertPreTrainedModel (line 772) | class LxmertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 782) | def _init_weights(self, module):
  class LxmertModel (line 883) | class LxmertModel(LxmertPreTrainedModel):
    method __init__ (line 884) | def __init__(self, config):
    method get_input_embeddings (line 892) | def get_input_embeddings(self):
    method set_input_embeddings (line 895) | def set_input_embeddings(self, new_embeddings):
    method forward (line 904) | def forward(
  class LxmertForPreTraining (line 1020) | class LxmertForPreTraining(LxmertPreTrainedModel):
    method __init__ (line 1023) | def __init__(self, config):
    method resize_num_qa_labels (line 1078) | def resize_num_qa_labels(self, num_labels):
    method _resize_qa_labels (line 1102) | def _resize_qa_labels(self, num_labels):
    method get_qa_logit_layer (line 1108) | def get_qa_logit_layer(self) -> nn.Module:
    method _set_qa_logit_layer (line 1119) | def _set_qa_logit_layer(self, qa_logit_layer):
    method _get_resized_qa_labels (line 1122) | def _get_resized_qa_labels(self, cur_qa_logit_layer, num_labels):
    method forward (line 1151) | def forward(
  class LxmertForQuestionAnswering (line 1288) | class LxmertForQuestionAnswering(LxmertPreTrainedModel):
    method __init__ (line 1289) | def __init__(self, config):
    method resize_num_qa_labels (line 1308) | def resize_num_qa_labels(self, num_labels):
    method _resize_qa_labels (line 1332) | def _resize_qa_labels(self, num_labels):
    method get_qa_logit_layer (line 1338) | def get_qa_logit_layer(self) -> nn.Module:
    method _set_qa_logit_layer (line 1350) | def _set_qa_logit_layer(self, qa_logit_layer):
    method _get_resized_qa_labels (line 1353) | def _get_resized_qa_labels(self, cur_qa_logit_layer, num_labels):
    method forward (line 1386) | def forward(

FILE: transformers/models/lxmert/modeling_tf_lxmert.py
  class TFLxmertModelOutput (line 61) | class TFLxmertModelOutput(ModelOutput):
  class TFLxmertForPreTrainingOutput (line 107) | class TFLxmertForPreTrainingOutput(ModelOutput):
  class TFLxmertVisualFeatureEncoder (line 154) | class TFLxmertVisualFeatureEncoder(tf.keras.layers.Layer):
    method __init__ (line 155) | def __init__(self, config, **kwargs):
    method call (line 178) | def call(self, visn_input, training=False):
  class TFLxmertEmbeddings (line 191) | class TFLxmertEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 194) | def __init__(self, config, **kwargs):
    method build (line 204) | def build(self, input_shape):
    method call (line 228) | def call(self, input_ids=None, token_type_ids=None, inputs_embeds=None...
  class TFLxmertAttention (line 256) | class TFLxmertAttention(tf.keras.layers.Layer):
    method __init__ (line 257) | def __init__(self, config, **kwargs):
    method transpose_for_scores (line 288) | def transpose_for_scores(self, x, batch_size):
    method call (line 293) | def call(self, hidden_states, context, attention_mask, output_attentio...
  class TFLxmertIntermediate (line 332) | class TFLxmertIntermediate(tf.keras.layers.Layer):
    method __init__ (line 333) | def __init__(self, config, **kwargs):
    method call (line 345) | def call(self, hidden_states):
  class TFLxmertOutput (line 351) | class TFLxmertOutput(tf.keras.layers.Layer):
    method __init__ (line 352) | def __init__(self, config, **kwargs):
    method call (line 363) | def call(self, hidden_states, input_tensor, training=False):
  class TFLxmertAttentionOutput (line 370) | class TFLxmertAttentionOutput(tf.keras.layers.Layer):
    method __init__ (line 371) | def __init__(self, config, **kwargs):
    method call (line 381) | def call(self, hidden_states, input_tensor, training=False):
  class TFLxmertSelfAttentionLayer (line 388) | class TFLxmertSelfAttentionLayer(tf.keras.layers.Layer):
    method __init__ (line 389) | def __init__(self, config, **kwargs):
    method call (line 394) | def call(self, input_tensor, attention_mask, output_attentions, traini...
  class TFLxmertCrossAttentionLayer (line 403) | class TFLxmertCrossAttentionLayer(tf.keras.layers.Layer):
    method __init__ (line 404) | def __init__(self, config, **kwargs):
    method call (line 409) | def call(
  class TFLxmertLayer (line 425) | class TFLxmertLayer(tf.keras.layers.Layer):
    method __init__ (line 426) | def __init__(self, config, **kwargs):
    method call (line 432) | def call(self, hidden_states, attention_mask, output_attentions, train...
  class TFLxmertXLayer (line 441) | class TFLxmertXLayer(tf.keras.layers.Layer):
    method __init__ (line 442) | def __init__(self, config, **kwargs):
    method cross_att (line 456) | def cross_att(
    method self_att (line 489) | def self_att(
    method output_fc (line 503) | def output_fc(self, lang_input, visn_input, training=False):
    method call (line 513) | def call(
  class TFLxmertEncoder (line 546) | class TFLxmertEncoder(tf.keras.layers.Layer):
    method __init__ (line 547) | def __init__(self, config, **kwargs):
    method call (line 564) | def call(
  class TFLxmertMainLayer (line 636) | class TFLxmertMainLayer(tf.keras.layers.Layer):
    method __init__ (line 639) | def __init__(self, config, **kwargs):
    method get_input_embeddings (line 655) | def get_input_embeddings(self):
    method set_input_embeddings (line 658) | def set_input_embeddings(self, value):
    method _prune_heads (line 662) | def _prune_heads(self, heads_to_prune):
    method call (line 666) | def call(
  class TFLxmertPreTrainedModel (line 775) | class TFLxmertPreTrainedModel(TFPreTrainedModel):
    method dummy_inputs (line 785) | def dummy_inputs(self):
    method input_signature (line 805) | def input_signature(self):
  class TFLxmertModel (line 927) | class TFLxmertModel(TFLxmertPreTrainedModel):
    method __init__ (line 928) | def __init__(self, config, *inputs, **kwargs):
    method call (line 939) | def call(
  class TFLxmertPooler (line 970) | class TFLxmertPooler(tf.keras.layers.Layer):
    method __init__ (line 971) | def __init__(self, config, **kwargs):
    method call (line 980) | def call(self, hidden_states):
  class TFLxmertPredictionHeadTransform (line 989) | class TFLxmertPredictionHeadTransform(tf.keras.layers.Layer):
    method __init__ (line 990) | def __init__(self, config: LxmertConfig, **kwargs):
    method call (line 1006) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFLxmertLMPredictionHead (line 1015) | class TFLxmertLMPredictionHead(tf.keras.layers.Layer):
    method __init__ (line 1016) | def __init__(self, config: LxmertConfig, input_embeddings: tf.keras.la...
    method build (line 1028) | def build(self, input_shape: tf.TensorShape):
    method get_output_embeddings (line 1033) | def get_output_embeddings(self) -> tf.keras.layers.Layer:
    method set_output_embeddings (line 1036) | def set_output_embeddings(self, value: tf.Variable):
    method get_bias (line 1040) | def get_bias(self) -> Dict[str, tf.Variable]:
    method set_bias (line 1043) | def set_bias(self, value: tf.Variable):
    method call (line 1047) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFLxmertMLMHead (line 1059) | class TFLxmertMLMHead(tf.keras.layers.Layer):
    method __init__ (line 1060) | def __init__(self, config: LxmertConfig, input_embeddings: tf.keras.la...
    method call (line 1065) | def call(self, sequence_output: tf.Tensor) -> tf.Tensor:
  class TFLxmertPreTrainingHeads (line 1071) | class TFLxmertPreTrainingHeads(tf.keras.layers.Layer):
    method __init__ (line 1072) | def __init__(self, config, input_embeddings, **kwargs):
    method call (line 1082) | def call(self, sequence_output, pooled_output):
  class TFLxmertVisualAnswerHead (line 1088) | class TFLxmertVisualAnswerHead(tf.keras.layers.Layer):
    method __init__ (line 1089) | def __init__(self, config, num_labels, **kwargs):
    method call (line 1105) | def call(self, hidden_states):
  class TFLxmertVisualObjHead (line 1114) | class TFLxmertVisualObjHead(tf.keras.layers.Layer):
    method __init__ (line 1115) | def __init__(self, config, **kwargs):
    method call (line 1140) | def call(self, hidden_states):
  class TFLxmertForPreTraining (line 1149) | class TFLxmertForPreTraining(TFLxmertPreTrainedModel):
    method __init__ (line 1150) | def __init__(self, config, *inputs, **kwargs):
    method dummy_inputs (line 1202) | def dummy_inputs(self):
    method get_lm_head (line 1242) | def get_lm_head(self):
    method get_prefix_bias_name (line 1245) | def get_prefix_bias_name(self):
    method call (line 1252) | def call(

FILE: transformers/models/lxmert/tokenization_lxmert.py
  function load_vocab (line 45) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 57) | def whitespace_tokenize(text):
  class LxmertTokenizer (line 67) | class LxmertTokenizer(PreTrainedTokenizer):
    method __init__ (line 114) | def __init__(
    method do_lower_case (line 161) | def do_lower_case(self):
    method vocab_size (line 165) | def vocab_size(self):
    method get_vocab (line 168) | def get_vocab(self):
    method _tokenize (line 171) | def _tokenize(self, text):
    method _convert_token_to_id (line 184) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 188) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 192) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 197) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 222) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 250) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 279) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 301) | class BasicTokenizer(object):
    method __init__ (line 321) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 329) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 366) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 377) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 399) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 412) | def _is_chinese_char(self, cp):
    method _clean_text (line 436) | def _clean_text(self, text):
  class WordpieceTokenizer (line 451) | class WordpieceTokenizer(object):
    method __init__ (line 454) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 459) | def tokenize(self, text):

FILE: transformers/models/lxmert/tokenization_lxmert_fast.py
  class LxmertTokenizerFast (line 48) | class LxmertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 94) | def __init__(
    method build_inputs_with_special_tokens (line 136) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 160) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 189) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/m2m_100/configuration_m2m_100.py
  class M2M100Config (line 34) | class M2M100Config(PretrainedConfig):
    method __init__ (line 106) | def __init__(
  class M2M100OnnxConfig (line 163) | class M2M100OnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 165) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method _generate_dummy_inputs_for_sequence_classification_and_question_answering (line 188) | def _generate_dummy_inputs_for_sequence_classification_and_question_an...
    method _generate_dummy_inputs_for_default_and_seq2seq_lm (line 215) | def _generate_dummy_inputs_for_default_and_seq2seq_lm(

FILE: transformers/models/m2m_100/convert_m2m100_original_checkpoint_to_pytorch.py
  function remove_ignore_keys_ (line 23) | def remove_ignore_keys_(state_dict):
  function make_linear_from_emb (line 38) | def make_linear_from_emb(emb):
  function convert_fairseq_m2m100_checkpoint_from_disk (line 45) | def convert_fairseq_m2m100_checkpoint_from_disk(checkpoint_path):

FILE: transformers/models/m2m_100/modeling_m2m_100.py
  function shift_tokens_right (line 59) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 76) | def _make_causal_mask(
  function _expand_mask (line 94) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  function create_position_ids_from_input_ids (line 108) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...
  class M2M100SinusoidalPositionalEmbedding (line 119) | class M2M100SinusoidalPositionalEmbedding(nn.Module):
    method __init__ (line 122) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method make_weights (line 129) | def make_weights(self, num_embeddings: int, embedding_dim: int, paddin...
    method get_embedding (line 138) | def get_embedding(num_embeddings: int, embedding_dim: int, padding_idx...
    method forward (line 159) | def forward(
    method create_position_ids_from_inputs_embeds (line 179) | def create_position_ids_from_inputs_embeds(self, inputs_embeds, past_k...
  class M2M100Attention (line 198) | class M2M100Attention(nn.Module):
    method __init__ (line 201) | def __init__(
    method _shape (line 228) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 231) | def forward(
  class M2M100EncoderLayer (line 353) | class M2M100EncoderLayer(nn.Module):
    method __init__ (line 354) | def __init__(self, config: M2M100Config):
    method forward (line 370) | def forward(
  class M2M100DecoderLayer (line 422) | class M2M100DecoderLayer(nn.Module):
    method __init__ (line 423) | def __init__(self, config: M2M100Config):
    method forward (line 449) | def forward(
  class M2M100PreTrainedModel (line 539) | class M2M100PreTrainedModel(PreTrainedModel):
    method _init_weights (line 545) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 556) | def _set_gradient_checkpointing(self, module, value=False):
  class M2M100Encoder (line 684) | class M2M100Encoder(M2M100PreTrainedModel):
    method __init__ (line 694) | def __init__(self, config: M2M100Config, embed_tokens: Optional[nn.Emb...
    method forward (line 722) | def forward(
  class M2M100Decoder (line 864) | class M2M100Decoder(M2M100PreTrainedModel):
    method __init__ (line 873) | def __init__(self, config: M2M100Config, embed_tokens: Optional[nn.Emb...
    method forward (line 898) | def forward(
  class M2M100Model (line 1140) | class M2M100Model(M2M100PreTrainedModel):
    method __init__ (line 1150) | def __init__(self, config: M2M100Config):
    method get_input_embeddings (line 1162) | def get_input_embeddings(self):
    method set_input_embeddings (line 1165) | def set_input_embeddings(self, value):
    method get_encoder (line 1170) | def get_encoder(self):
    method get_decoder (line 1173) | def get_decoder(self):
    method forward (line 1182) | def forward(
  class M2M100ForConditionalGeneration (line 1259) | class M2M100ForConditionalGeneration(M2M100PreTrainedModel):
    method __init__ (line 1273) | def __init__(self, config: M2M100Config):
    method get_encoder (line 1281) | def get_encoder(self):
    method get_decoder (line 1284) | def get_decoder(self):
    method resize_token_embeddings (line 1287) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method get_output_embeddings (line 1291) | def get_output_embeddings(self):
    method set_output_embeddings (line 1294) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1300) | def forward(
    method prepare_inputs_for_generation (line 1377) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1406) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/m2m_100/tokenization_m2m_100.py
  class M2M100Tokenizer (line 64) | class M2M100Tokenizer(PreTrainedTokenizer):
    method __init__ (line 131) | def __init__(
    method vocab_size (line 196) | def vocab_size(self) -> int:
    method src_lang (line 200) | def src_lang(self) -> str:
    method src_lang (line 204) | def src_lang(self, new_src_lang: str) -> None:
    method _tokenize (line 208) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 211) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 216) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 222) | def convert_tokens_to_string(self, tokens):
    method get_special_tokens_mask (line 236) | def get_special_tokens_mask(
    method build_inputs_with_special_tokens (line 266) | def build_inputs_with_special_tokens(
    method get_vocab (line 293) | def get_vocab(self) -> Dict:
    method __getstate__ (line 298) | def __getstate__(self) -> Dict:
    method __setstate__ (line 303) | def __setstate__(self, d: Dict) -> None:
    method save_vocabulary (line 312) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method prepare_seq2seq_batch (line 334) | def prepare_seq2seq_batch(
    method _build_translation_inputs (line 347) | def _build_translation_inputs(self, raw_inputs, src_lang: Optional[str...
    method _switch_to_input_mode (line 357) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 360) | def _switch_to_target_mode(self):
    method set_src_lang_special_tokens (line 363) | def set_src_lang_special_tokens(self, src_lang: str) -> None:
    method set_tgt_lang_special_tokens (line 370) | def set_tgt_lang_special_tokens(self, tgt_lang: str) -> None:
    method get_lang_token (line 377) | def get_lang_token(self, lang: str) -> str:
    method get_lang_id (line 380) | def get_lang_id(self, lang: str) -> int:
  function load_spm (line 385) | def load_spm(path: str, sp_model_kwargs: Dict[str, Any]) -> sentencepiec...
  function load_json (line 391) | def load_json(path: str) -> Union[Dict, List]:
  function save_json (line 396) | def save_json(data, path: str) -> None:

FILE: transformers/models/marian/configuration_marian.py
  class MarianConfig (line 34) | class MarianConfig(PretrainedConfig):
    method __init__ (line 109) | def __init__(
  class MarianOnnxConfig (line 169) | class MarianOnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 172) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 217) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method _generate_dummy_inputs_for_default_and_seq2seq_lm (line 229) | def _generate_dummy_inputs_for_default_and_seq2seq_lm(
    method _generate_dummy_inputs_for_causal_lm (line 297) | def _generate_dummy_inputs_for_causal_lm(
    method _generate_dummy_inputs_for_encoder_and_decoder (line 337) | def _generate_dummy_inputs_for_encoder_and_decoder(
    method generate_dummy_inputs (line 363) | def generate_dummy_inputs(
    method _flatten_past_key_values_ (line 384) | def _flatten_past_key_values_(self, flattened_output, name, idx, t):
    method atol_for_validation (line 393) | def atol_for_validation(self) -> float:

FILE: transformers/models/marian/convert_marian_tatoeba_to_pytorch.py
  class TatoebaConverter (line 44) | class TatoebaConverter:
    method __init__ (line 58) | def __init__(self, save_dir="marian_converted"):
    method convert_models (line 76) | def convert_models(self, tatoeba_ids, dry_run=False):
    method expand_group_to_two_letter_codes (line 93) | def expand_group_to_two_letter_codes(self, grp_name):
    method is_group (line 96) | def is_group(self, code, name):
    method get_tags (line 99) | def get_tags(self, code, name):
    method resolve_lang_code (line 111) | def resolve_lang_code(self, src, tgt) -> Tuple[str, str]:
    method model_type_info_from_model_name (line 117) | def model_type_info_from_model_name(name):
    method write_model_card (line 131) | def write_model_card(self, model_dict, dry_run=False) -> str:
    method download_lang_info (line 279) | def download_lang_info(self):
    method parse_metadata (line 288) | def parse_metadata(self, model_name, repo_path=DEFAULT_MODEL_DIR, meth...
  function l2front_matter (line 1301) | def l2front_matter(langs):
  function dedup (line 1305) | def dedup(lst):

FILE: transformers/models/marian/convert_marian_to_pytorch.py
  function remove_suffix (line 34) | def remove_suffix(text: str, suffix: str):
  function remove_prefix (line 40) | def remove_prefix(text: str, prefix: str):
  function convert_encoder_layer (line 46) | def convert_encoder_layer(opus_dict, layer_prefix: str, converter: dict):
  function load_layers_ (line 57) | def load_layers_(layer_lst: nn.ModuleList, opus_state: dict, converter, ...
  function find_pretrained_model (line 64) | def find_pretrained_model(src_lang: str, tgt_lang: str) -> List[str]:
  function add_emb_entries (line 76) | def add_emb_entries(wemb, final_bias, n_special_tokens=1):
  function _cast_yaml_str (line 85) | def _cast_yaml_str(v):
  function cast_marian_config (line 97) | def cast_marian_config(raw_cfg: Dict[str, str]) -> Dict:
  function load_config_from_state_dict (line 104) | def load_config_from_state_dict(opus_dict):
  function find_model_file (line 112) | def find_model_file(dest_dir):  # this one better
  function convert_opus_name_to_hf_name (line 166) | def convert_opus_name_to_hf_name(x):
  function convert_hf_name_to_opus_name (line 173) | def convert_hf_name_to_opus_name(hf_model_name):
  function get_system_metadata (line 185) | def get_system_metadata(repo_root):
  function write_model_card (line 210) | def write_model_card(
  function make_registry (line 281) | def make_registry(repo_path="Opus-MT-train/models"):
  function convert_all_sentencepiece_models (line 298) | def convert_all_sentencepiece_models(model_list=None, repo_path=None, de...
  function lmap (line 318) | def lmap(f, x) -> List:
  function fetch_test_set (line 322) | def fetch_test_set(test_set_url):
  function convert_whole_dir (line 336) | def convert_whole_dir(path=Path("marian_ckpt/")):
  function _parse_readme (line 344) | def _parse_readme(lns):
  function save_tokenizer_config (line 367) | def save_tokenizer_config(dest_dir: Path, separate_vocabs=False):
  function add_to_vocab_ (line 373) | def add_to_vocab_(vocab: Dict[str, int], special_tokens: List[str]):
  function find_vocab_file (line 384) | def find_vocab_file(model_dir):
  function find_src_vocab_file (line 388) | def find_src_vocab_file(model_dir):
  function find_tgt_vocab_file (line 392) | def find_tgt_vocab_file(model_dir):
  function add_special_tokens_to_vocab (line 396) | def add_special_tokens_to_vocab(model_dir: Path, separate_vocab=False) -...
  function check_equal (line 417) | def check_equal(marian_cfg, k1, k2):
  function check_marian_cfg_assumptions (line 423) | def check_marian_cfg_assumptions(marian_cfg):
  class OpusState (line 477) | class OpusState:
    method __init__ (line 478) | def __init__(self, source_dir, eos_token_id=0):
    method _check_layer_entries (line 562) | def _check_layer_entries(self):
    method extra_keys (line 574) | def extra_keys(self):
    method sub_keys (line 587) | def sub_keys(self, layer_prefix):
    method load_tokenizer (line 590) | def load_tokenizer(self):
    method load_marian_model (line 595) | def load_marian_model(self) -> MarianMTModel:
  function download_and_unzip (line 648) | def download_and_unzip(url, dest_dir):
  function convert (line 659) | def convert(source_dir: Path, dest_dir):
  function load_yaml (line 677) | def load_yaml(path):
  function save_json (line 684) | def save_json(content: Union[Dict, List], path: str) -> None:
  function unzip (line 689) | def unzip(zip_path: str, dest_dir: str) -> None:

FILE: transformers/models/marian/modeling_flax_marian.py
  function create_sinusoidal_positions (line 219) | def create_sinusoidal_positions(n_pos, dim):
  function shift_tokens_right (line 230) | def shift_tokens_right(input_ids: jnp.ndarray, pad_token_id: int, decode...
  class FlaxMarianAttention (line 243) | class FlaxMarianAttention(nn.Module):
    method setup (line 252) | def setup(self) -> None:
    method _split_heads (line 278) | def _split_heads(self, hidden_states):
    method _merge_heads (line 281) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 285) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 316) | def __call__(
  class FlaxMarianEncoderLayer (line 411) | class FlaxMarianEncoderLayer(nn.Module):
    method setup (line 415) | def setup(self) -> None:
    method __call__ (line 438) | def __call__(
  class FlaxMarianEncoderLayerCollection (line 469) | class FlaxMarianEncoderLayerCollection(nn.Module):
    method setup (line 473) | def setup(self):
    method __call__ (line 480) | def __call__(
  class FlaxMarianDecoderLayer (line 524) | class FlaxMarianDecoderLayer(nn.Module):
    method setup (line 528) | def setup(self) -> None:
    method __call__ (line 561) | def __call__(
  class FlaxMarianDecoderLayerCollection (line 613) | class FlaxMarianDecoderLayerCollection(nn.Module):
    method setup (line 617) | def setup(self):
    method __call__ (line 624) | def __call__(
  class FlaxMarianEncoder (line 683) | class FlaxMarianEncoder(nn.Module):
    method setup (line 688) | def setup(self):
    method __call__ (line 698) | def __call__(
  class FlaxMarianDecoder (line 739) | class FlaxMarianDecoder(nn.Module):
    method setup (line 744) | def setup(self):
    method __call__ (line 754) | def __call__(
  class FlaxMarianModule (line 804) | class FlaxMarianModule(nn.Module):
    method setup (line 808) | def setup(self):
    method _get_encoder_module (line 818) | def _get_encoder_module(self):
    method _get_decoder_module (line 821) | def _get_decoder_module(self):
    method __call__ (line 824) | def __call__(
  class FlaxMarianPreTrainedModel (line 873) | class FlaxMarianPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 878) | def __init__(
    method init_weights (line 890) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 926) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 964) | def encode(
    method decode (line 1028) | def decode(
    method __call__ (line 1141) | def __call__(
  class FlaxMarianModel (line 1205) | class FlaxMarianModel(FlaxMarianPreTrainedModel):
  class FlaxMarianMTModule (line 1214) | class FlaxMarianMTModule(nn.Module):
    method setup (line 1219) | def setup(self):
    method _get_encoder_module (line 1229) | def _get_encoder_module(self):
    method _get_decoder_module (line 1232) | def _get_decoder_module(self):
    method __call__ (line 1235) | def __call__(
  class FlaxMarianMTModel (line 1289) | class FlaxMarianMTModel(FlaxMarianPreTrainedModel):
    method decode (line 1295) | def decode(
    method _adapt_logits_for_beam_search (line 1430) | def _adapt_logits_for_beam_search(self, logits):
    method prepare_inputs_for_generation (line 1435) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 1466) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/marian/modeling_marian.py
  function shift_tokens_right (line 61) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 78) | def _make_causal_mask(
  function _expand_mask (line 96) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class MarianSinusoidalPositionalEmbedding (line 110) | class MarianSinusoidalPositionalEmbedding(nn.Embedding):
    method __init__ (line 113) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method _init_weight (line 118) | def _init_weight(out: nn.Parameter) -> nn.Parameter:
    method forward (line 135) | def forward(self, input_ids_shape: torch.Size, past_key_values_length:...
  class MarianAttention (line 145) | class MarianAttention(nn.Module):
    method __init__ (line 148) | def __init__(
    method _shape (line 175) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 178) | def forward(
  class MarianEncoderLayer (line 300) | class MarianEncoderLayer(nn.Module):
    method __init__ (line 301) | def __init__(self, config: MarianConfig):
    method forward (line 317) | def forward(
  class MarianDecoderLayer (line 369) | class MarianDecoderLayer(nn.Module):
    method __init__ (line 370) | def __init__(self, config: MarianConfig):
    method forward (line 396) | def forward(
  class MarianPreTrainedModel (line 486) | class MarianPreTrainedModel(PreTrainedModel):
    method _init_weights (line 491) | def _init_weights(self, module: Union[nn.Linear, nn.Embedding, MarianS...
    method _set_gradient_checkpointing (line 504) | def _set_gradient_checkpointing(self, module, value=False):
    method dummy_inputs (line 509) | def dummy_inputs(self):
  class MarianEncoder (line 652) | class MarianEncoder(MarianPreTrainedModel):
    method __init__ (line 662) | def __init__(self, config: MarianConfig, embed_tokens: Optional[nn.Emb...
    method get_input_embeddings (line 687) | def get_input_embeddings(self):
    method set_input_embeddings (line 690) | def set_input_embeddings(self, value):
    method forward (line 693) | def forward(
  class MarianDecoder (line 822) | class MarianDecoder(MarianPreTrainedModel):
    method __init__ (line 831) | def __init__(self, config: MarianConfig, embed_tokens: Optional[nn.Emb...
    method get_input_embeddings (line 853) | def get_input_embeddings(self):
    method set_input_embeddings (line 856) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 860) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 883) | def forward(
  class MarianModel (line 1100) | class MarianModel(MarianPreTrainedModel):
    method __init__ (line 1103) | def __init__(self, config: MarianConfig):
    method get_input_embeddings (line 1125) | def get_input_embeddings(self):
    method set_input_embeddings (line 1129) | def set_input_embeddings(self, value):
    method get_decoder_input_embeddings (line 1137) | def get_decoder_input_embeddings(self):
    method set_decoder_input_embeddings (line 1145) | def set_decoder_input_embeddings(self, value):
    method get_encoder (line 1154) | def get_encoder(self):
    method get_decoder (line 1157) | def get_decoder(self):
    method resize_decoder_token_embeddings (line 1160) | def resize_decoder_token_embeddings(self, new_num_tokens: int) -> nn.E...
    method forward (line 1186) | def forward(
  class MarianMTModel (line 1286) | class MarianMTModel(MarianPreTrainedModel):
    method __init__ (line 1300) | def __init__(self, config: MarianConfig):
    method get_encoder (line 1311) | def get_encoder(self):
    method get_decoder (line 1314) | def get_decoder(self):
    method resize_token_embeddings (line 1317) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_token_embeddings (line 1323) | def _resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method resize_decoder_token_embeddings (line 1344) | def resize_decoder_token_embeddings(self, new_num_tokens):
    method _resize_final_logits_bias (line 1376) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_output_embeddings (line 1385) | def get_output_embeddings(self):
    method set_output_embeddings (line 1388) | def set_output_embeddings(self, new_embeddings: nn.Embedding):
    method tie_weights (line 1391) | def tie_weights(self):
    method forward (line 1416) | def forward(
    method prepare_inputs_for_generation (line 1495) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1523) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method adjust_logits_during_generation (line 1526) | def adjust_logits_during_generation(self, logits, cur_len):
    method _reorder_cache (line 1531) | def _reorder_cache(past_key_values, beam_idx):
  class MarianDecoderWrapper (line 1542) | class MarianDecoderWrapper(MarianPreTrainedModel):
    method __init__ (line 1548) | def __init__(self, config):
    method forward (line 1552) | def forward(self, *args, **kwargs):
  class MarianForCausalLM (line 1557) | class MarianForCausalLM(MarianPreTrainedModel):
    method __init__ (line 1560) | def __init__(self, config):
    method get_input_embeddings (line 1572) | def get_input_embeddings(self):
    method set_input_embeddings (line 1575) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1578) | def get_output_embeddings(self):
    method set_output_embeddings (line 1581) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 1584) | def set_decoder(self, decoder):
    method get_decoder (line 1587) | def get_decoder(self):
    method forward (line 1591) | def forward(
    method prepare_inputs_for_generation (line 1735) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1753) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/marian/modeling_tf_marian.py
  function shift_tokens_right (line 64) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int, decoder_...
  function _make_causal_mask (line 89) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 107) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFMarianSinusoidalPositionalEmbedding (line 120) | class TFMarianSinusoidalPositionalEmbedding(tf.keras.layers.Layer):
    method __init__ (line 123) | def __init__(self, num_positions: int, embedding_dim: int, **kwargs):
    method build (line 132) | def build(self, input_shape: tf.TensorShape):
    method _init_weight (line 151) | def _init_weight(n_pos: int, dim: int):
    method call (line 168) | def call(
  class TFMarianAttention (line 179) | class TFMarianAttention(tf.keras.layers.Layer):
    method __init__ (line 182) | def __init__(
    method _shape (line 210) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 213) | def call(
  class TFMarianEncoderLayer (line 333) | class TFMarianEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 334) | def __init__(self, config: MarianConfig, **kwargs):
    method call (line 348) | def call(
  class TFMarianDecoderLayer (line 390) | class TFMarianDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 391) | def __init__(self, config: MarianConfig, **kwargs):
    method call (line 418) | def call(
  class TFMarianPreTrainedModel (line 499) | class TFMarianPreTrainedModel(TFPreTrainedModel):
  class TFMarianEncoder (line 648) | class TFMarianEncoder(tf.keras.layers.Layer):
    method __init__ (line 658) | def __init__(self, config: MarianConfig, embed_tokens: Optional[tf.ker...
    method get_embed_tokens (line 675) | def get_embed_tokens(self):
    method set_embed_tokens (line 678) | def set_embed_tokens(self, embed_tokens):
    method call (line 682) | def call(
  class TFMarianDecoder (line 811) | class TFMarianDecoder(tf.keras.layers.Layer):
    method __init__ (line 821) | def __init__(self, config: MarianConfig, embed_tokens: Optional[tf.ker...
    method get_embed_tokens (line 837) | def get_embed_tokens(self):
    method set_embed_tokens (line 840) | def set_embed_tokens(self, embed_tokens):
    method call (line 844) | def call(
  class TFMarianMainLayer (line 1043) | class TFMarianMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1046) | def __init__(self, config: MarianConfig, **kwargs):
    method get_input_embeddings (line 1062) | def get_input_embeddings(self):
    method set_input_embeddings (line 1065) | def set_input_embeddings(self, new_embeddings):
    method call (line 1071) | def call(
  class TFMarianModel (line 1157) | class TFMarianModel(TFMarianPreTrainedModel):
    method __init__ (line 1158) | def __init__(self, config: MarianConfig, *inputs, **kwargs):
    method get_encoder (line 1163) | def get_encoder(self):
    method get_decoder (line 1166) | def get_decoder(self):
    method call (line 1176) | def call(
    method serving_output (line 1220) | def serving_output(self, output):
  class BiasLayer (line 1241) | class BiasLayer(tf.keras.layers.Layer):
    method __init__ (line 1247) | def __init__(self, shape, initializer, trainable, name, **kwargs):
    method call (line 1254) | def call(self, x):
  class TFMarianMTModel (line 1262) | class TFMarianMTModel(TFMarianPreTrainedModel, TFCausalLanguageModelingL...
    method __init__ (line 1268) | def __init__(self, config, *inputs, **kwargs):
    method get_decoder (line 1277) | def get_decoder(self):
    method get_encoder (line 1280) | def get_encoder(self):
    method get_output_embeddings (line 1283) | def get_output_embeddings(self):
    method set_output_embeddings (line 1286) | def set_output_embeddings(self, value):
    method get_bias (line 1289) | def get_bias(self):
    method set_bias (line 1292) | def set_bias(self, value):
    method call (line 1304) | def call(
    method serving_output (line 1386) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 1406) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1444) | def prepare_decoder_input_ids_from_labels(self, labels: tf.Tensor):
    method adjust_logits_during_generation (line 1447) | def adjust_logits_during_generation(

FILE: transformers/models/marian/tokenization_marian.py
  class MarianTokenizer (line 61) | class MarianTokenizer(PreTrainedTokenizer):
    method __init__ (line 127) | def __init__(
    method _setup_normalizer (line 188) | def _setup_normalizer(self):
    method normalize (line 197) | def normalize(self, x: str) -> str:
    method _convert_token_to_id (line 201) | def _convert_token_to_id(self, token):
    method remove_language_code (line 204) | def remove_language_code(self, text: str):
    method _tokenize (line 210) | def _tokenize(self, text: str) -> List[str]:
    method _convert_id_to_token (line 215) | def _convert_id_to_token(self, index: int) -> str:
    method batch_decode (line 219) | def batch_decode(self, sequences, **kwargs):
    method decode (line 242) | def decode(self, token_ids, **kwargs):
    method convert_tokens_to_string (line 268) | def convert_tokens_to_string(self, tokens: List[str]) -> str:
    method build_inputs_with_special_tokens (line 283) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method _switch_to_input_mode (line 290) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 294) | def _switch_to_target_mode(self):
    method vocab_size (line 300) | def vocab_size(self) -> int:
    method save_vocabulary (line 303) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method get_vocab (line 348) | def get_vocab(self) -> Dict:
    method get_src_vocab (line 351) | def get_src_vocab(self):
    method get_tgt_vocab (line 354) | def get_tgt_vocab(self):
    method __getstate__ (line 357) | def __getstate__(self) -> Dict:
    method __setstate__ (line 364) | def __setstate__(self, d: Dict) -> None:
    method num_special_tokens_to_add (line 375) | def num_special_tokens_to_add(self, *args, **kwargs):
    method _special_token_mask (line 379) | def _special_token_mask(self, seq):
    method get_special_tokens_mask (line 384) | def get_special_tokens_mask(
  function load_spm (line 396) | def load_spm(path: str, sp_model_kwargs: Dict[str, Any]) -> sentencepiec...
  function save_json (line 402) | def save_json(data, path: str) -> None:
  function load_json (line 407) | def load_json(path: str) -> Union[Dict, List]:

FILE: transformers/models/markuplm/configuration_markuplm.py
  class MarkupLMConfig (line 29) | class MarkupLMConfig(PretrainedConfig):
    method __init__ (line 102) | def __init__(

FILE: transformers/models/markuplm/feature_extraction_markuplm.py
  class MarkupLMFeatureExtractor (line 33) | class MarkupLMFeatureExtractor(FeatureExtractionMixin):
    method __init__ (line 43) | def __init__(self, **kwargs):
    method xpath_soup (line 47) | def xpath_soup(self, element):
    method get_three_from_single (line 62) | def get_three_from_single(self, html_string):
    method construct_xpath (line 91) | def construct_xpath(self, xpath_tags, xpath_subscripts):
    method __call__ (line 99) | def __call__(self, html_strings) -> BatchFeature:

FILE: transformers/models/markuplm/modeling_markuplm.py
  class XPathEmbeddings (line 61) | class XPathEmbeddings(nn.Module):
    method __init__ (line 67) | def __init__(self, config):
    method forward (line 93) | def forward(self, xpath_tags_seq=None, xpath_subs_seq=None):
  function create_position_ids_from_input_ids (line 112) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...
  class MarkupLMEmbeddings (line 128) | class MarkupLMEmbeddings(nn.Module):
    method __init__ (line 131) | def __init__(self, config):
    method create_position_ids_from_inputs_embeds (line 154) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
    method forward (line 171) | def forward(
  class MarkupLMSelfOutput (line 225) | class MarkupLMSelfOutput(nn.Module):
    method __init__ (line 226) | def __init__(self, config):
    method forward (line 232) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class MarkupLMIntermediate (line 240) | class MarkupLMIntermediate(nn.Module):
    method __init__ (line 241) | def __init__(self, config):
    method forward (line 249) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MarkupLMOutput (line 256) | class MarkupLMOutput(nn.Module):
    method __init__ (line 257) | def __init__(self, config):
    method forward (line 263) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class MarkupLMPooler (line 271) | class MarkupLMPooler(nn.Module):
    method __init__ (line 272) | def __init__(self, config):
    method forward (line 277) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MarkupLMPredictionHeadTransform (line 287) | class MarkupLMPredictionHeadTransform(nn.Module):
    method __init__ (line 288) | def __init__(self, config):
    method forward (line 297) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MarkupLMLMPredictionHead (line 305) | class MarkupLMLMPredictionHead(nn.Module):
    method __init__ (line 306) | def __init__(self, config):
    method forward (line 319) | def forward(self, hidden_states):
  class MarkupLMOnlyMLMHead (line 326) | class MarkupLMOnlyMLMHead(nn.Module):
    method __init__ (line 327) | def __init__(self, config):
    method forward (line 331) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class MarkupLMSelfAttention (line 337) | class MarkupLMSelfAttention(nn.Module):
    method __init__ (line 338) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 364) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 369) | def forward(
  class MarkupLMAttention (line 472) | class MarkupLMAttention(nn.Module):
    method __init__ (line 473) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 479) | def prune_heads(self, heads):
    method forward (line 497) | def forward(
  class MarkupLMLayer (line 522) | class MarkupLMLayer(nn.Module):
    method __init__ (line 523) | def __init__(self, config):
    method forward (line 537) | def forward(
    method feed_forward_chunk (line 602) | def feed_forward_chunk(self, attention_output):
  class MarkupLMEncoder (line 609) | class MarkupLMEncoder(nn.Module):
    method __init__ (line 610) | def __init__(self, config):
    method forward (line 616) | def forward(
  class MarkupLMPreTrainedModel (line 707) | class MarkupLMPreTrainedModel(PreTrainedModel):
    method _init_weights (line 719) | def _init_weights(self, module):
    method from_pretrained (line 736) | def from_pretrained(cls, pretrained_model_name_or_path: Optional[Union...
  class MarkupLMModel (line 806) | class MarkupLMModel(MarkupLMPreTrainedModel):
    method __init__ (line 808) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 820) | def get_input_embeddings(self):
    method set_input_embeddings (line 823) | def set_input_embeddings(self, value):
    method _prune_heads (line 826) | def _prune_heads(self, heads_to_prune):
    method forward (line 836) | def forward(
    method prepare_inputs_for_generation (line 939) | def prepare_inputs_for_generation(
    method _reorder_cache (line 959) | def _reorder_cache(self, past_key_values, beam_idx):
  class MarkupLMForQuestionAnswering (line 973) | class MarkupLMForQuestionAnswering(MarkupLMPreTrainedModel):
    method __init__ (line 977) | def __init__(self, config):
    method forward (line 989) | def forward(
  class MarkupLMForTokenClassification (line 1095) | class MarkupLMForTokenClassification(MarkupLMPreTrainedModel):
    method __init__ (line 1097) | def __init__(self, config):
    method forward (line 1113) | def forward(
  class MarkupLMForSequenceClassification (line 1201) | class MarkupLMForSequenceClassification(MarkupLMPreTrainedModel):
    method __init__ (line 1203) | def __init__(self, config):
    method forward (line 1220) | def forward(

FILE: transformers/models/markuplm/processing_markuplm.py
  class MarkupLMProcessor (line 25) | class MarkupLMProcessor(ProcessorMixin):
    method __call__ (line 48) | def __call__(
    method batch_decode (line 128) | def batch_decode(self, *args, **kwargs):
    method decode (line 135) | def decode(self, *args, **kwargs):
    method model_input_names (line 143) | def model_input_names(self):

FILE: transformers/models/markuplm/tokenization_markuplm.py
  function bytes_to_unicode (line 110) | def bytes_to_unicode():
  function get_pairs (line 132) | def get_pairs(word):
  class MarkupLMTokenizer (line 145) | class MarkupLMTokenizer(PreTrainedTokenizer):
    method __init__ (line 204) | def __init__(
    method get_xpath_seq (line 285) | def get_xpath_seq(self, xpath):
    method vocab_size (line 311) | def vocab_size(self):
    method get_vocab (line 314) | def get_vocab(self):
    method bpe (line 317) | def bpe(self, token):
    method _tokenize (line 359) | def _tokenize(self, text):
    method _convert_token_to_id (line 369) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 373) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 377) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 386) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method prepare_for_tokenization (line 417) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method build_inputs_with_special_tokens (line 423) | def build_inputs_with_special_tokens(
    method build_xpath_tags_with_special_tokens (line 446) | def build_xpath_tags_with_special_tokens(
    method build_xpath_subs_with_special_tokens (line 454) | def build_xpath_subs_with_special_tokens(
    method get_special_tokens_mask (line 462) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 487) | def create_token_type_ids_from_sequences(
    method __call__ (line 510) | def __call__(
    method batch_encode_plus (line 655) | def batch_encode_plus(
    method _batch_encode_plus (line 713) | def _batch_encode_plus(
    method _batch_prepare_for_model (line 769) | def _batch_prepare_for_model(
    method encode (line 840) | def encode(
    method encode_plus (line 887) | def encode_plus(
    method _encode_plus (line 953) | def _encode_plus(
    method prepare_for_model (line 1006) | def prepare_for_model(
    method truncate_sequences (line 1234) | def truncate_sequences(
    method _pad (line 1370) | def _pad(

FILE: transformers/models/markuplm/tokenization_markuplm_fast.py
  function bytes_to_unicode (line 64) | def bytes_to_unicode():
  function get_pairs (line 86) | def get_pairs(word):
  class MarkupLMTokenizerFast (line 99) | class MarkupLMTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 162) | def __init__(
    method get_xpath_seq (line 259) | def get_xpath_seq(self, xpath):
    method __call__ (line 285) | def __call__(
    method batch_encode_plus (line 430) | def batch_encode_plus(
    method tokenize (line 488) | def tokenize(self, text: str, pair: Optional[str] = None, add_special_...
    method encode_plus (line 497) | def encode_plus(
    method _batch_encode_plus (line 563) | def _batch_encode_plus(
    method _encode_plus (line 720) | def _encode_plus(
    method _pad (line 786) | def _pad(
    method build_inputs_with_special_tokens (line 879) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 902) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 924) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/mask2former/configuration_mask2former.py
  class Mask2FormerConfig (line 34) | class Mask2FormerConfig(PretrainedConfig):
    method __init__ (line 127) | def __init__(
    method from_backbone_config (line 211) | def from_backbone_config(cls, backbone_config: PretrainedConfig, **kwa...
    method to_dict (line 226) | def to_dict(self) -> Dict[str, any]:

FILE: transformers/models/mask2former/convert_mask2former_original_pytorch_checkpoint_to_pytorch.py
  class TrackedStateDict (line 55) | class TrackedStateDict:
    method __init__ (line 56) | def __init__(self, to_track: Dict):
    method __getitem__ (line 65) | def __getitem__(self, key: str) -> Any:
    method __setitem__ (line 68) | def __setitem__(self, key: str, item: Any):
    method diff (line 72) | def diff(self) -> List[str]:
    method copy (line 81) | def copy(self) -> Dict:
  function prepare_img (line 87) | def prepare_img():
  class Args (line 95) | class Args:
  function setup_cfg (line 101) | def setup_cfg(args: Args):
  class OriginalMask2FormerConfigToOursConverter (line 111) | class OriginalMask2FormerConfigToOursConverter:
    method __call__ (line 112) | def __call__(self, original_config: object) -> Mask2FormerConfig:
  class OriginalMask2FormerConfigToFeatureExtractorConverter (line 195) | class OriginalMask2FormerConfigToFeatureExtractorConverter:
    method __call__ (line 196) | def __call__(self, original_config: object) -> Mask2FormerImageProcessor:
  class OriginalMask2FormerCheckpointToOursConverter (line 211) | class OriginalMask2FormerCheckpointToOursConverter:
    method __init__ (line 212) | def __init__(self, original_model: nn.Module, config: Mask2FormerConfig):
    method pop_all (line 216) | def pop_all(self, renamed_keys: List[Tuple[str, str]], dst_state_dict:...
    method replace_maskformer_swin_backbone (line 220) | def replace_maskformer_swin_backbone(
    method replace_swin_backbone (line 379) | def replace_swin_backbone(self, dst_state_dict: StateDict, src_state_d...
    method replace_pixel_module (line 537) | def replace_pixel_module(self, dst_state_dict: StateDict, src_state_di...
    method rename_keys_in_masked_attention_decoder (line 621) | def rename_keys_in_masked_attention_decoder(self, dst_state_dict: Stat...
    method replace_masked_attention_decoder (line 718) | def replace_masked_attention_decoder(self, dst_state_dict: StateDict, ...
    method replace_keys_qkv_transformer_decoder (line 749) | def replace_keys_qkv_transformer_decoder(self, dst_state_dict: StateDi...
    method replace_transformer_module (line 768) | def replace_transformer_module(self, dst_state_dict: StateDict, src_st...
    method replace_universal_segmentation_module (line 783) | def replace_universal_segmentation_module(self, dst_state_dict: StateD...
    method convert (line 795) | def convert(self, mask2former: Mask2FormerModel) -> Mask2FormerModel:
    method convert_universal_segmentation (line 810) | def convert_universal_segmentation(
    method using_dirs (line 824) | def using_dirs(checkpoints_dir: Path, config_dir: Path) -> Iterator[Tu...
  function test (line 846) | def test(
  function get_model_name (line 907) | def get_model_name(checkpoint_file: Path):

FILE: transformers/models/mask2former/image_processing_mask2former.py
  function max_across_indices (line 62) | def max_across_indices(values: Iterable[Any]) -> List[Any]:
  function get_max_height_width (line 70) | def get_max_height_width(images: List[np.ndarray]) -> List[int]:
  function make_pixel_mask (line 86) | def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> ...
  function binary_mask_to_rle (line 103) | def binary_mask_to_rle(mask):
  function convert_segmentation_to_rle (line 126) | def convert_segmentation_to_rle(segmentation):
  function remove_low_and_no_objects (line 148) | def remove_low_and_no_objects(masks, scores, labels, object_mask_thresho...
  function check_segment_validity (line 177) | def check_segment_validity(mask_labels, mask_probs, k, mask_threshold=0....
  function compute_segments (line 196) | def compute_segments(
  function convert_segmentation_map_to_binary_masks (line 257) | def convert_segmentation_map_to_binary_masks(
  function get_mask2former_resize_output_image_size (line 293) | def get_mask2former_resize_output_image_size(
  class Mask2FormerImageProcessor (line 331) | class Mask2FormerImageProcessor(BaseImageProcessor):
    method __init__ (line 381) | def __init__(
    method from_dict (line 432) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method size_divisibility (line 445) | def size_divisibility(self):
    method max_size (line 454) | def max_size(self):
    method resize (line 462) | def resize(
    method rescale (line 505) | def rescale(
    method normalize (line 513) | def normalize(
    method convert_segmentation_map_to_binary_masks (line 525) | def convert_segmentation_map_to_binary_masks(
    method __call__ (line 541) | def __call__(self, images, segmentation_maps=None, **kwargs) -> BatchF...
    method _preprocess (line 544) | def _preprocess(
    method _preprocess_image (line 565) | def _preprocess_image(
    method _preprocess_mask (line 598) | def _preprocess_mask(
    method preprocess (line 629) | def preprocess(
    method _pad_image (line 724) | def _pad_image(
    method pad (line 746) | def pad(
    method encode_inputs (line 784) | def encode_inputs(
    method post_process_semantic_segmentation (line 883) | def post_process_semantic_segmentation(
    method post_process_instance_segmentation (line 938) | def post_process_instance_segmentation(
    method post_process_panoptic_segmentation (line 1056) | def post_process_panoptic_segmentation(

FILE: transformers/models/mask2former/modeling_mask2former.py
  class Mask2FormerPixelDecoderOutput (line 60) | class Mask2FormerPixelDecoderOutput(ModelOutput):
  class Mask2FormerMaskedAttentionDecoderOutput (line 84) | class Mask2FormerMaskedAttentionDecoderOutput(BaseModelOutputWithCrossAt...
  class Mask2FormerPixelLevelModuleOutput (line 116) | class Mask2FormerPixelLevelModuleOutput(ModelOutput):
  class Mask2FormerModelOutput (line 147) | class Mask2FormerModelOutput(ModelOutput):
  class Mask2FormerForUniversalSegmentationOutput (line 193) | class Mask2FormerForUniversalSegmentationOutput(ModelOutput):
  function _expand_mask (line 250) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, target_len: Opt...
  function sample_point (line 265) | def sample_point(
  function dice_loss (line 298) | def dice_loss(inputs: Tensor, labels: Tensor, num_masks: int) -> Tensor:
  function sigmoid_cross_entropy_loss (line 328) | def sigmoid_cross_entropy_loss(inputs: torch.Tensor, labels: torch.Tenso...
  function pair_wise_dice_loss (line 348) | def pair_wise_dice_loss(inputs: Tensor, labels: Tensor) -> Tensor:
  function pair_wise_sigmoid_cross_entropy_loss (line 370) | def pair_wise_sigmoid_cross_entropy_loss(inputs: torch.Tensor, labels: t...
  class Mask2FormerHungarianMatcher (line 399) | class Mask2FormerHungarianMatcher(nn.Module):
    method __init__ (line 407) | def __init__(
    method forward (line 434) | def forward(
  class Mask2FormerLoss (line 502) | class Mask2FormerLoss(nn.Module):
    method __init__ (line 503) | def __init__(self, config: Mask2FormerConfig, weight_dict: Dict[str, f...
    method _max_by_axis (line 538) | def _max_by_axis(self, sizes: List[List[int]]) -> List[int]:
    method _pad_images_to_max_in_batch (line 546) | def _pad_images_to_max_in_batch(self, tensors: List[Tensor]) -> Tuple[...
    method loss_labels (line 563) | def loss_labels(
    method loss_masks (line 597) | def loss_masks(
    method _get_predictions_permutation_indices (line 659) | def _get_predictions_permutation_indices(self, indices):
    method _get_targets_permutation_indices (line 665) | def _get_targets_permutation_indices(self, indices):
    method calculate_uncertainty (line 671) | def calculate_uncertainty(self, logits: torch.Tensor) -> torch.Tensor:
    method sample_points_using_uncertainty (line 688) | def sample_points_using_uncertainty(
    method forward (line 743) | def forward(
    method get_num_masks (line 798) | def get_num_masks(self, class_labels: torch.Tensor, device: torch.devi...
  function multi_scale_deformable_attention (line 808) | def multi_scale_deformable_attention(
  class Mask2FormerSinePositionEmbedding (line 848) | class Mask2FormerSinePositionEmbedding(nn.Module):
    method __init__ (line 854) | def __init__(
    method forward (line 865) | def forward(self, x: Tensor, mask: Optional[Tensor] = None) -> Tensor:
  class Mask2FormerPixelDecoderEncoderMultiscaleDeformableAttention (line 888) | class Mask2FormerPixelDecoderEncoderMultiscaleDeformableAttention(nn.Mod...
    method __init__ (line 893) | def __init__(self, embed_dim: int, num_heads: int, n_levels: int, n_po...
    method with_pos_embed (line 920) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 923) | def forward(
  class Mask2FormerPixelDecoderEncoderLayer (line 981) | class Mask2FormerPixelDecoderEncoderLayer(nn.Module):
    method __init__ (line 982) | def __init__(self, config: Mask2FormerConfig):
    method forward (line 1000) | def forward(
  class Mask2FormerPixelDecoderEncoderOnly (line 1071) | class Mask2FormerPixelDecoderEncoderOnly(nn.Module):
    method __init__ (line 1081) | def __init__(self, config: Mask2FormerConfig):
    method get_reference_points (line 1091) | def get_reference_points(spatial_shapes, valid_ratios, device):
    method forward (line 1122) | def forward(
  class Mask2FormerPixelDecoder (line 1200) | class Mask2FormerPixelDecoder(nn.Module):
    method __init__ (line 1201) | def __init__(self, config: Mask2FormerConfig, feature_channels):
    method get_valid_ratio (line 1271) | def get_valid_ratio(self, mask):
    method forward (line 1282) | def forward(
  class Mask2FormerPixelLevelModule (line 1379) | class Mask2FormerPixelLevelModule(nn.Module):
    method __init__ (line 1380) | def __init__(self, config: Mask2FormerConfig):
    method forward (line 1398) | def forward(self, pixel_values: Tensor, output_hidden_states: bool = F...
  class Mask2FormerAttention (line 1411) | class Mask2FormerAttention(nn.Module):
    method __init__ (line 1417) | def __init__(
    method _shape (line 1442) | def _shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method with_pos_embed (line 1445) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 1448) | def forward(
  class Mask2FormerMaskedAttentionDecoderLayer (line 1547) | class Mask2FormerMaskedAttentionDecoderLayer(nn.Module):
    method __init__ (line 1561) | def __init__(self, config: Mask2FormerConfig):
    method with_pos_embed (line 1584) | def with_pos_embed(self, tensor, pos: Optional[Tensor]):
    method forward_post (line 1587) | def forward_post(
    method forward_pre (line 1646) | def forward_pre(
    method forward (line 1707) | def forward(
  class Mask2FormerMaskedAttentionDecoder (line 1761) | class Mask2FormerMaskedAttentionDecoder(nn.Module):
    method __init__ (line 1774) | def __init__(self, config: Mask2FormerConfig):
    method forward (line 1797) | def forward(
  class Mask2FormerPredictionBlock (line 1939) | class Mask2FormerPredictionBlock(nn.Module):
    method __init__ (line 1940) | def __init__(self, in_dim: int, out_dim: int, activation: nn.Module) -...
    method forward (line 1947) | def forward(self, input: Tensor) -> Tensor:
  class Mask2FormerMLPPredictionHead (line 1954) | class Mask2FormerMLPPredictionHead(nn.Module):
    method __init__ (line 1955) | def __init__(self, input_dim: int, hidden_dim: int, output_dim: int, n...
    method forward (line 1986) | def forward(self, input: Tensor) -> Tensor:
  class Mask2FormerMaskPredictor (line 1993) | class Mask2FormerMaskPredictor(nn.Module):
    method __init__ (line 1994) | def __init__(self, hidden_size: int, num_heads: int, mask_feature_size...
    method forward (line 2015) | def forward(self, outputs: torch.Tensor, pixel_embeddings: torch.Tenso...
  class Mask2FormerTransformerModule (line 2032) | class Mask2FormerTransformerModule(nn.Module):
    method __init__ (line 2037) | def __init__(self, in_features: int, config: Mask2FormerConfig):
    method forward (line 2055) | def forward(
  class Mask2FormerPreTrainedModel (line 2132) | class Mask2FormerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 2137) | def _init_weights(self, module: nn.Module):
  class Mask2FormerModel (line 2211) | class Mask2FormerModel(Mask2FormerPreTrainedModel):
    method __init__ (line 2214) | def __init__(self, config: Mask2FormerConfig):
    method forward (line 2223) | def forward(
  class Mask2FormerForUniversalSegmentation (line 2315) | class Mask2FormerForUniversalSegmentation(Mask2FormerPreTrainedModel):
    method __init__ (line 2318) | def __init__(self, config: Mask2FormerConfig):
    method get_loss_dict (line 2333) | def get_loss_dict(
    method get_loss (line 2357) | def get_loss(self, loss_dict: Dict[str, Tensor]) -> Tensor:
    method get_auxiliary_logits (line 2360) | def get_auxiliary_logits(self, classes: torch.Tensor, output_masks: to...
    method forward (line 2370) | def forward(

FILE: transformers/models/maskformer/configuration_maskformer.py
  class MaskFormerConfig (line 36) | class MaskFormerConfig(PretrainedConfig):
    method __init__ (line 103) | def __init__(
    method from_backbone_and_decoder_configs (line 184) | def from_backbone_and_decoder_configs(
    method to_dict (line 205) | def to_dict(self) -> Dict[str, any]:

FILE: transformers/models/maskformer/configuration_maskformer_swin.py
  class MaskFormerSwinConfig (line 25) | class MaskFormerSwinConfig(BackboneConfigMixin, PretrainedConfig):
    method __init__ (line 100) | def __init__(

FILE: transformers/models/maskformer/convert_maskformer_original_pytorch_checkpoint_to_pytorch.py
  class TrackedStateDict (line 51) | class TrackedStateDict:
    method __init__ (line 52) | def __init__(self, to_track: Dict):
    method __getitem__ (line 61) | def __getitem__(self, key: str) -> Any:
    method __setitem__ (line 64) | def __setitem__(self, key: str, item: Any):
    method diff (line 68) | def diff(self) -> List[str]:
    method copy (line 77) | def copy(self) -> Dict:
  function prepare_img (line 83) | def prepare_img():
  class Args (line 91) | class Args:
  function setup_cfg (line 97) | def setup_cfg(args: Args):
  class OriginalMaskFormerConfigToOursConverter (line 107) | class OriginalMaskFormerConfigToOursConverter:
    method __call__ (line 108) | def __call__(self, original_config: object) -> MaskFormerConfig:
  class OriginalMaskFormerConfigToFeatureExtractorConverter (line 167) | class OriginalMaskFormerConfigToFeatureExtractorConverter:
    method __call__ (line 168) | def __call__(self, original_config: object) -> MaskFormerFeatureExtrac...
  class OriginalMaskFormerCheckpointToOursConverter (line 184) | class OriginalMaskFormerCheckpointToOursConverter:
    method __init__ (line 185) | def __init__(self, original_model: nn.Module, config: MaskFormerConfig):
    method pop_all (line 189) | def pop_all(self, renamed_keys: List[Tuple[str, str]], dst_state_dict:...
    method replace_backbone (line 193) | def replace_backbone(self, dst_state_dict: StateDict, src_state_dict: ...
    method replace_pixel_module (line 350) | def replace_pixel_module(self, dst_state_dict: StateDict, src_state_di...
    method rename_keys_in_detr_decoder (line 383) | def rename_keys_in_detr_decoder(self, dst_state_dict: StateDict, src_s...
    method replace_q_k_v_in_detr_decoder (line 440) | def replace_q_k_v_in_detr_decoder(self, dst_state_dict: StateDict, src...
    method replace_detr_decoder (line 467) | def replace_detr_decoder(self, dst_state_dict: StateDict, src_state_di...
    method replace_transformer_module (line 483) | def replace_transformer_module(self, dst_state_dict: StateDict, src_st...
    method replace_instance_segmentation_module (line 497) | def replace_instance_segmentation_module(self, dst_state_dict: StateDi...
    method convert (line 518) | def convert(self, mask_former: MaskFormerModel) -> MaskFormerModel:
    method convert_instance_segmentation (line 533) | def convert_instance_segmentation(
    method using_dirs (line 546) | def using_dirs(checkpoints_dir: Path, config_dir: Path) -> Iterator[Tu...
  function test (line 557) | def test(original_model, our_model: MaskFormerForInstanceSegmentation, f...
  function get_name (line 612) | def get_name(checkpoint_file: Path):

FILE: transformers/models/maskformer/convert_maskformer_resnet_to_pytorch.py
  function get_maskformer_config (line 37) | def get_maskformer_config(model_name: str):
  function create_rename_keys (line 80) | def create_rename_keys(config):
  function rename_key (line 221) | def rename_key(dct, old, new):
  function read_in_decoder_q_k_v (line 227) | def read_in_decoder_q_k_v(state_dict, config):
  function prepare_img (line 255) | def prepare_img() -> torch.Tensor:
  function convert_maskformer_checkpoint (line 262) | def convert_maskformer_checkpoint(

FILE: transformers/models/maskformer/convert_maskformer_swin_to_pytorch.py
  function get_maskformer_config (line 37) | def get_maskformer_config(model_name: str):
  function create_rename_keys (line 75) | def create_rename_keys(config):
  function rename_key (line 164) | def rename_key(dct, old, new):
  function read_in_swin_q_k_v (line 170) | def read_in_swin_q_k_v(state_dict, backbone_config):
  function read_in_decoder_q_k_v (line 196) | def read_in_decoder_q_k_v(state_dict, config):
  function prepare_img (line 224) | def prepare_img() -> torch.Tensor:
  function convert_maskformer_checkpoint (line 231) | def convert_maskformer_checkpoint(

FILE: transformers/models/maskformer/feature_extraction_maskformer.py
  class MaskFormerFeatureExtractor (line 26) | class MaskFormerFeatureExtractor(MaskFormerImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/maskformer/image_processing_maskformer.py
  function max_across_indices (line 66) | def max_across_indices(values: Iterable[Any]) -> List[Any]:
  function get_max_height_width (line 74) | def get_max_height_width(images: List[np.ndarray]) -> List[int]:
  function make_pixel_mask (line 90) | def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> ...
  function binary_mask_to_rle (line 107) | def binary_mask_to_rle(mask):
  function convert_segmentation_to_rle (line 130) | def convert_segmentation_to_rle(segmentation):
  function remove_low_and_no_objects (line 152) | def remove_low_and_no_objects(masks, scores, labels, object_mask_thresho...
  function check_segment_validity (line 181) | def check_segment_validity(mask_labels, mask_probs, k, mask_threshold=0....
  function compute_segments (line 200) | def compute_segments(
  function convert_segmentation_map_to_binary_masks (line 261) | def convert_segmentation_map_to_binary_masks(
  function get_maskformer_resize_output_image_size (line 297) | def get_maskformer_resize_output_image_size(
  class MaskFormerImageProcessor (line 335) | class MaskFormerImageProcessor(BaseImageProcessor):
    method __init__ (line 385) | def __init__(
    method from_dict (line 443) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method size_divisibility (line 456) | def size_divisibility(self):
    method max_size (line 465) | def max_size(self):
    method reduce_labels (line 474) | def reduce_labels(self):
    method resize (line 482) | def resize(
    method rescale (line 525) | def rescale(
    method normalize (line 533) | def normalize(
    method convert_segmentation_map_to_binary_masks (line 545) | def convert_segmentation_map_to_binary_masks(
    method __call__ (line 562) | def __call__(self, images, segmentation_maps=None, **kwargs) -> BatchF...
    method _preprocess (line 565) | def _preprocess(
    method _preprocess_image (line 586) | def _preprocess_image(
    method _preprocess_mask (line 619) | def _preprocess_mask(
    method preprocess (line 650) | def preprocess(
    method _pad_image (line 755) | def _pad_image(
    method pad (line 777) | def pad(
    method encode_inputs (line 815) | def encode_inputs(
    method post_process_segmentation (line 913) | def post_process_segmentation(
    method post_process_semantic_segmentation (line 961) | def post_process_semantic_segmentation(
    method post_process_instance_segmentation (line 1011) | def post_process_instance_segmentation(
    method post_process_panoptic_segmentation (line 1124) | def post_process_panoptic_segmentation(

FILE: transformers/models/maskformer/modeling_maskformer.py
  class DetrDecoderOutput (line 62) | class DetrDecoderOutput(BaseModelOutputWithCrossAttentions):
  class MaskFormerPixelLevelModuleOutput (line 92) | class MaskFormerPixelLevelModuleOutput(ModelOutput):
  class MaskFormerPixelDecoderOutput (line 122) | class MaskFormerPixelDecoderOutput(ModelOutput):
  class MaskFormerModelOutput (line 146) | class MaskFormerModelOutput(ModelOutput):
  class MaskFormerForInstanceSegmentationOutput (line 189) | class MaskFormerForInstanceSegmentationOutput(ModelOutput):
  function upsample_like (line 248) | def upsample_like(pixel_values: Tensor, like: Tensor, mode: str = "bilin...
  function dice_loss (line 269) | def dice_loss(inputs: Tensor, labels: Tensor, num_masks: int) -> Tensor:
  function sigmoid_focal_loss (line 300) | def sigmoid_focal_loss(
  function pair_wise_dice_loss (line 344) | def pair_wise_dice_loss(inputs: Tensor, labels: Tensor) -> Tensor:
  function pair_wise_sigmoid_focal_loss (line 367) | def pair_wise_sigmoid_focal_loss(inputs: Tensor, labels: Tensor, alpha: ...
  class DetrAttention (line 407) | class DetrAttention(nn.Module):
    method __init__ (line 414) | def __init__(
    method _shape (line 439) | def _shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method with_pos_embed (line 442) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 445) | def forward(
  class DetrDecoderLayer (line 539) | class DetrDecoderLayer(nn.Module):
    method __init__ (line 540) | def __init__(self, config: DetrConfig):
    method forward (line 566) | def forward(
  function _expand_mask (line 647) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, target_len: Opt...
  class DetrDecoder (line 661) | class DetrDecoder(nn.Module):
    method __init__ (line 676) | def __init__(self, config: DetrConfig):
    method forward (line 688) | def forward(
  class MaskFormerHungarianMatcher (line 840) | class MaskFormerHungarianMatcher(nn.Module):
    method __init__ (line 848) | def __init__(self, cost_class: float = 1.0, cost_mask: float = 1.0, co...
    method forward (line 867) | def forward(self, masks_queries_logits, class_queries_logits, mask_lab...
    method __repr__ (line 925) | def __repr__(self):
  class MaskFormerLoss (line 938) | class MaskFormerLoss(nn.Module):
    method __init__ (line 939) | def __init__(
    method _max_by_axis (line 972) | def _max_by_axis(self, the_list: List[List[int]]) -> List[int]:
    method _pad_images_to_max_in_batch (line 979) | def _pad_images_to_max_in_batch(self, tensors: List[Tensor]) -> Tuple[...
    method loss_labels (line 998) | def loss_labels(
    method loss_masks (line 1033) | def loss_masks(
    method _get_predictions_permutation_indices (line 1075) | def _get_predictions_permutation_indices(self, indices):
    method _get_targets_permutation_indices (line 1081) | def _get_targets_permutation_indices(self, indices):
    method forward (line 1087) | def forward(
    method get_num_masks (line 1141) | def get_num_masks(self, class_labels: torch.Tensor, device: torch.devi...
  class MaskFormerFPNConvLayer (line 1150) | class MaskFormerFPNConvLayer(nn.Module):
    method __init__ (line 1151) | def __init__(self, in_features: int, out_features: int, kernel_size: i...
    method forward (line 1176) | def forward(self, input: Tensor) -> Tensor:
  class MaskFormerFPNLayer (line 1183) | class MaskFormerFPNLayer(nn.Module):
    method __init__ (line 1184) | def __init__(self, in_features: int, lateral_features: int):
    method forward (line 1203) | def forward(self, down: Tensor, left: Tensor) -> Tensor:
  class MaskFormerFPNModel (line 1211) | class MaskFormerFPNModel(nn.Module):
    method __init__ (line 1212) | def __init__(self, in_features: int, lateral_widths: List[int], featur...
    method forward (line 1231) | def forward(self, features: List[Tensor]) -> List[Tensor]:
  class MaskFormerPixelDecoder (line 1242) | class MaskFormerPixelDecoder(nn.Module):
    method __init__ (line 1243) | def __init__(self, *args, feature_size: int = 256, mask_feature_size: ...
    method forward (line 1260) | def forward(self, features: List[Tensor], output_hidden_states: bool =...
  class MaskFormerSinePositionEmbedding (line 1271) | class MaskFormerSinePositionEmbedding(nn.Module):
    method __init__ (line 1277) | def __init__(
    method forward (line 1288) | def forward(self, x: Tensor, mask: Optional[Tensor] = None) -> Tensor:
  class PredictionBlock (line 1310) | class PredictionBlock(nn.Module):
    method __init__ (line 1311) | def __init__(self, in_dim: int, out_dim: int, activation: nn.Module) -...
    method forward (line 1318) | def forward(self, input: Tensor) -> Tensor:
  class MaskformerMLPPredictionHead (line 1325) | class MaskformerMLPPredictionHead(nn.Module):
    method __init__ (line 1326) | def __init__(self, input_dim: int, hidden_dim: int, output_dim: int, n...
    method forward (line 1357) | def forward(self, input: Tensor) -> Tensor:
  class MaskFormerPixelLevelModule (line 1364) | class MaskFormerPixelLevelModule(nn.Module):
    method __init__ (line 1365) | def __init__(self, config: MaskFormerConfig):
    method forward (line 1393) | def forward(self, pixel_values: Tensor, output_hidden_states: bool = F...
  class MaskFormerTransformerModule (line 1405) | class MaskFormerTransformerModule(nn.Module):
    method __init__ (line 1410) | def __init__(self, in_features: int, config: MaskFormerConfig):
    method forward (line 1419) | def forward(
  class MaskFormerPreTrainedModel (line 1482) | class MaskFormerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1487) | def _init_weights(self, module: nn.Module):
    method _set_gradient_checkpointing (line 1526) | def _set_gradient_checkpointing(self, module, value=False):
  class MaskFormerModel (line 1537) | class MaskFormerModel(MaskFormerPreTrainedModel):
    method __init__ (line 1538) | def __init__(self, config: MaskFormerConfig):
    method forward (line 1549) | def forward(
  class MaskFormerForInstanceSegmentation (line 1634) | class MaskFormerForInstanceSegmentation(MaskFormerPreTrainedModel):
    method __init__ (line 1635) | def __init__(self, config: MaskFormerConfig):
    method get_loss_dict (line 1662) | def get_loss_dict(
    method get_loss (line 1681) | def get_loss(self, loss_dict: Dict[str, Tensor]) -> Tensor:
    method get_logits (line 1684) | def get_logits(self, outputs: MaskFormerModelOutput) -> Tuple[Tensor, ...
    method forward (line 1717) | def forward(

FILE: transformers/models/maskformer/modeling_maskformer_swin.py
  class MaskFormerSwinModelOutputWithPooling (line 37) | class MaskFormerSwinModelOutputWithPooling(ModelOutput):
  class MaskFormerSwinBaseModelOutput (line 71) | class MaskFormerSwinBaseModelOutput(ModelOutput):
  function window_partition (line 102) | def window_partition(input_feature, window_size):
  function window_reverse (line 115) | def window_reverse(windows, window_size, height, width):
  function drop_path (line 126) | def drop_path(input, drop_prob=0.0, training=False, scale_by_keep=True):
  class MaskFormerSwinEmbeddings (line 146) | class MaskFormerSwinEmbeddings(nn.Module):
    method __init__ (line 151) | def __init__(self, config):
    method forward (line 166) | def forward(self, pixel_values):
  class MaskFormerSwinPatchEmbeddings (line 179) | class MaskFormerSwinPatchEmbeddings(nn.Module):
    method __init__ (line 186) | def __init__(self, config):
    method maybe_pad (line 201) | def maybe_pad(self, pixel_values, height, width):
    method forward (line 210) | def forward(self, pixel_values: Optional[torch.FloatTensor]) -> Tuple[...
  class MaskFormerSwinPatchMerging (line 227) | class MaskFormerSwinPatchMerging(nn.Module):
    method __init__ (line 240) | def __init__(self, input_resolution: Tuple[int], dim: int, norm_layer:...
    method maybe_pad (line 247) | def maybe_pad(self, input_feature, height, width):
    method forward (line 255) | def forward(self, input_feature: torch.Tensor, input_dimensions: Tuple...
  class MaskFormerSwinDropPath (line 282) | class MaskFormerSwinDropPath(nn.Module):
    method __init__ (line 285) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 289) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 292) | def extra_repr(self) -> str:
  class MaskFormerSwinSelfAttention (line 297) | class MaskFormerSwinSelfAttention(nn.Module):
    method __init__ (line 298) | def __init__(self, config, dim, num_heads, window_size):
    method transpose_for_scores (line 335) | def transpose_for_scores(self, x):
    method forward (line 340) | def forward(
  class MaskFormerSwinSelfOutput (line 398) | class MaskFormerSwinSelfOutput(nn.Module):
    method __init__ (line 399) | def __init__(self, config, dim):
    method forward (line 404) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class MaskFormerSwinAttention (line 412) | class MaskFormerSwinAttention(nn.Module):
    method __init__ (line 413) | def __init__(self, config, dim, num_heads, window_size):
    method prune_heads (line 419) | def prune_heads(self, heads):
    method forward (line 437) | def forward(
  class MaskFormerSwinIntermediate (line 451) | class MaskFormerSwinIntermediate(nn.Module):
    method __init__ (line 452) | def __init__(self, config, dim):
    method forward (line 460) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MaskFormerSwinOutput (line 467) | class MaskFormerSwinOutput(nn.Module):
    method __init__ (line 468) | def __init__(self, config, dim):
    method forward (line 473) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MaskFormerSwinLayer (line 479) | class MaskFormerSwinLayer(nn.Module):
    method __init__ (line 480) | def __init__(self, config, dim, input_resolution, num_heads, shift_siz...
    method get_attn_mask (line 494) | def get_attn_mask(self, input_resolution):
    method maybe_pad (line 523) | def maybe_pad(self, hidden_states, height, width):
    method forward (line 531) | def forward(self, hidden_states, input_dimensions, head_mask=None, out...
  class MaskFormerSwinStage (line 591) | class MaskFormerSwinStage(nn.Module):
    method __init__ (line 593) | def __init__(self, config, dim, input_resolution, depth, num_heads, dr...
    method forward (line 618) | def forward(
  class MaskFormerSwinEncoder (line 647) | class MaskFormerSwinEncoder(nn.Module):
    method __init__ (line 649) | def __init__(self, config, grid_size):
    method forward (line 671) | def forward(
  class MaskFormerSwinPreTrainedModel (line 732) | class MaskFormerSwinPreTrainedModel(PreTrainedModel):
    method _init_weights (line 743) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 755) | def _set_gradient_checkpointing(self, module, value=False):
  class MaskFormerSwinModel (line 760) | class MaskFormerSwinModel(MaskFormerSwinPreTrainedModel):
    method __init__ (line 761) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 773) | def get_input_embeddings(self):
    method _prune_heads (line 776) | def _prune_heads(self, heads_to_prune):
    method forward (line 784) | def forward(
  class MaskFormerSwinBackbone (line 841) | class MaskFormerSwinBackbone(MaskFormerSwinPreTrainedModel, BackboneMixin):
    method __init__ (line 853) | def __init__(self, config: MaskFormerSwinConfig):
    method forward (line 868) | def forward(

FILE: transformers/models/mbart/configuration_mbart.py
  class MBartConfig (line 34) | class MBartConfig(PretrainedConfig):
    method __init__ (line 111) | def __init__(
  class MBartOnnxConfig (line 170) | class MBartOnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 172) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 216) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method _generate_dummy_inputs_for_default_and_seq2seq_lm (line 228) | def _generate_dummy_inputs_for_default_and_seq2seq_lm(
    method _generate_dummy_inputs_for_causal_lm (line 296) | def _generate_dummy_inputs_for_causal_lm(
    method _generate_dummy_inputs_for_sequence_classification_and_question_answering (line 334) | def _generate_dummy_inputs_for_sequence_classification_and_question_an...
    method generate_dummy_inputs (line 360) | def generate_dummy_inputs(
    method _flatten_past_key_values_ (line 384) | def _flatten_past_key_values_(self, flattened_output, name, idx, t):

FILE: transformers/models/mbart/convert_mbart_original_checkpoint_to_pytorch.py
  function remove_ignore_keys_ (line 23) | def remove_ignore_keys_(state_dict):
  function make_linear_from_emb (line 36) | def make_linear_from_emb(emb):
  function convert_fairseq_mbart_checkpoint_from_disk (line 43) | def convert_fairseq_mbart_checkpoint_from_disk(

FILE: transformers/models/mbart/modeling_flax_mbart.py
  function shift_tokens_right (line 220) | def shift_tokens_right(input_ids: jnp.ndarray, pad_token_id: int) -> jnp...
  class FlaxMBartAttention (line 244) | class FlaxMBartAttention(nn.Module):
    method setup (line 253) | def setup(self) -> None:
    method _split_heads (line 279) | def _split_heads(self, hidden_states):
    method _merge_heads (line 282) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 286) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 317) | def __call__(
  class FlaxMBartEncoderLayer (line 411) | class FlaxMBartEncoderLayer(nn.Module):
    method setup (line 415) | def setup(self) -> None:
    method __call__ (line 438) | def __call__(
  class FlaxMBartEncoderLayerCollection (line 468) | class FlaxMBartEncoderLayerCollection(nn.Module):
    method setup (line 472) | def setup(self):
    method __call__ (line 479) | def __call__(
  class FlaxMBartDecoderLayer (line 522) | class FlaxMBartDecoderLayer(nn.Module):
    method setup (line 526) | def setup(self) -> None:
    method __call__ (line 559) | def __call__(
  class FlaxMBartDecoderLayerCollection (line 611) | class FlaxMBartDecoderLayerCollection(nn.Module):
    method setup (line 615) | def setup(self):
    method __call__ (line 622) | def __call__(
  class FlaxMBartClassificationHead (line 682) | class FlaxMBartClassificationHead(nn.Module):
    method setup (line 691) | def setup(self):
    method __call__ (line 702) | def __call__(self, hidden_states: jnp.ndarray, deterministic: bool):
  class FlaxMBartEncoder (line 711) | class FlaxMBartEncoder(nn.Module):
    method setup (line 716) | def setup(self):
    method __call__ (line 736) | def __call__(
  class FlaxMBartDecoder (line 786) | class FlaxMBartDecoder(nn.Module):
    method setup (line 791) | def setup(self):
    method __call__ (line 812) | def __call__(
  class FlaxMBartModule (line 872) | class FlaxMBartModule(nn.Module):
    method setup (line 876) | def setup(self):
    method _get_encoder_module (line 887) | def _get_encoder_module(self):
    method _get_decoder_module (line 890) | def _get_decoder_module(self):
    method __call__ (line 893) | def __call__(
  class FlaxMBartPreTrainedModel (line 942) | class FlaxMBartPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 947) | def __init__(
    method init_weights (line 959) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 996) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 1039) | def encode(
    method decode (line 1102) | def decode(
    method __call__ (line 1214) | def __call__(
  class FlaxMBartModel (line 1276) | class FlaxMBartModel(FlaxMBartPreTrainedModel):
  class FlaxMBartForConditionalGenerationModule (line 1286) | class FlaxMBartForConditionalGenerationModule(nn.Module):
    method setup (line 1291) | def setup(self):
    method _get_encoder_module (line 1301) | def _get_encoder_module(self):
    method _get_decoder_module (line 1304) | def _get_decoder_module(self):
    method __call__ (line 1307) | def __call__(
  class FlaxMBartForConditionalGeneration (line 1361) | class FlaxMBartForConditionalGeneration(FlaxMBartPreTrainedModel):
    method decode (line 1367) | def decode(
    method prepare_inputs_for_generation (line 1501) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 1532) | def update_inputs_for_generation(self, model_outputs, model_kwargs):
  class FlaxMBartForSequenceClassificationModule (line 1587) | class FlaxMBartForSequenceClassificationModule(nn.Module):
    method setup (line 1592) | def setup(self):
    method _get_encoder_module (line 1601) | def _get_encoder_module(self):
    method _get_decoder_module (line 1604) | def _get_decoder_module(self):
    method __call__ (line 1607) | def __call__(
  class FlaxMBartForSequenceClassification (line 1674) | class FlaxMBartForSequenceClassification(FlaxMBartPreTrainedModel):
  class FlaxMBartForQuestionAnsweringModule (line 1688) | class FlaxMBartForQuestionAnsweringModule(nn.Module):
    method setup (line 1693) | def setup(self):
    method _get_encoder_module (line 1699) | def _get_encoder_module(self):
    method _get_decoder_module (line 1702) | def _get_decoder_module(self):
    method __call__ (line 1705) | def __call__(
  class FlaxMBartForQuestionAnswering (line 1761) | class FlaxMBartForQuestionAnswering(FlaxMBartPreTrainedModel):

FILE: transformers/models/mbart/modeling_mbart.py
  function shift_tokens_right (line 62) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int):
  function _make_causal_mask (line 83) | def _make_causal_mask(
  function _expand_mask (line 101) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class MBartLearnedPositionalEmbedding (line 116) | class MBartLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 121) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 127) | def forward(self, input_ids: torch.Tensor, past_key_values_length: int...
  class MBartAttention (line 139) | class MBartAttention(nn.Module):
    method __init__ (line 142) | def __init__(
    method _shape (line 169) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 172) | def forward(
  class MBartEncoderLayer (line 293) | class MBartEncoderLayer(nn.Module):
    method __init__ (line 294) | def __init__(self, config: MBartConfig):
    method forward (line 310) | def forward(
  class MBartDecoderLayer (line 361) | class MBartDecoderLayer(nn.Module):
    method __init__ (line 362) | def __init__(self, config: MBartConfig):
    method forward (line 388) | def forward(
  class MBartClassificationHead (line 479) | class MBartClassificationHead(nn.Module):
    method __init__ (line 482) | def __init__(
    method forward (line 494) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MBartPreTrainedModel (line 503) | class MBartPreTrainedModel(PreTrainedModel):
    method _init_weights (line 509) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 520) | def _set_gradient_checkpointing(self, module, value=False):
    method dummy_inputs (line 525) | def dummy_inputs(self):
  class MBartEncoder (line 687) | class MBartEncoder(MBartPreTrainedModel):
    method __init__ (line 697) | def __init__(self, config: MBartConfig, embed_tokens: Optional[nn.Embe...
    method _backward_compatibility_gradient_checkpointing (line 725) | def _backward_compatibility_gradient_checkpointing(self):
    method forward (line 730) | def forward(
  class MBartDecoder (line 865) | class MBartDecoder(MBartPreTrainedModel):
    method __init__ (line 874) | def __init__(self, config: MBartConfig, embed_tokens: Optional[nn.Embe...
    method get_input_embeddings (line 899) | def get_input_embeddings(self):
    method set_input_embeddings (line 902) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 906) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 929) | def forward(
  class MBartModel (line 1153) | class MBartModel(MBartPreTrainedModel):
    method __init__ (line 1156) | def __init__(self, config: MBartConfig):
    method get_input_embeddings (line 1168) | def get_input_embeddings(self):
    method set_input_embeddings (line 1171) | def set_input_embeddings(self, value):
    method get_encoder (line 1176) | def get_encoder(self):
    method get_decoder (line 1179) | def get_decoder(self):
    method forward (line 1189) | def forward(
  class MBartForConditionalGeneration (line 1272) | class MBartForConditionalGeneration(MBartPreTrainedModel):
    method __init__ (line 1283) | def __init__(self, config: MBartConfig):
    method get_encoder (line 1292) | def get_encoder(self):
    method get_decoder (line 1295) | def get_decoder(self):
    method resize_token_embeddings (line 1298) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_final_logits_bias (line 1303) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_output_embeddings (line 1312) | def get_output_embeddings(self):
    method set_output_embeddings (line 1315) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1321) | def forward(
    method prepare_inputs_for_generation (line 1398) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1426) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 1430) | def _reorder_cache(past_key_values, beam_idx):
  class MBartForSequenceClassification (line 1447) | class MBartForSequenceClassification(MBartPreTrainedModel):
    method __init__ (line 1450) | def __init__(self, config: MBartConfig, **kwargs):
    method forward (line 1470) | def forward(
  class MBartForQuestionAnswering (line 1576) | class MBartForQuestionAnswering(MBartPreTrainedModel):
    method __init__ (line 1579) | def __init__(self, config):
    method forward (line 1598) | def forward(
  class MBartDecoderWrapper (line 1694) | class MBartDecoderWrapper(MBartPreTrainedModel):
    method __init__ (line 1700) | def __init__(self, config):
    method forward (line 1704) | def forward(self, *args, **kwargs):
  class MBartForCausalLM (line 1709) | class MBartForCausalLM(MBartPreTrainedModel):
    method __init__ (line 1712) | def __init__(self, config):
    method get_input_embeddings (line 1724) | def get_input_embeddings(self):
    method set_input_embeddings (line 1727) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1730) | def get_output_embeddings(self):
    method set_output_embeddings (line 1733) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 1736) | def set_decoder(self, decoder):
    method get_decoder (line 1739) | def get_decoder(self):
    method forward (line 1743) | def forward(
    method prepare_inputs_for_generation (line 1887) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1905) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/mbart/modeling_tf_mbart.py
  function shift_tokens_right (line 63) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int):
  function _make_causal_mask (line 88) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 106) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFMBartLearnedPositionalEmbedding (line 120) | class TFMBartLearnedPositionalEmbedding(tf.keras.layers.Embedding):
    method __init__ (line 125) | def __init__(self, num_embeddings: int, embedding_dim: int, **kwargs):
    method call (line 131) | def call(
  class TFMBartAttention (line 148) | class TFMBartAttention(tf.keras.layers.Layer):
    method __init__ (line 151) | def __init__(
    method _shape (line 179) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 182) | def call(
  class TFMBartEncoderLayer (line 301) | class TFMBartEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 302) | def __init__(self, config: MBartConfig, **kwargs):
    method call (line 316) | def call(
  class TFMBartDecoderLayer (line 357) | class TFMBartDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 358) | def __init__(self, config: MBartConfig, **kwargs):
    method call (line 385) | def call(
  class TFMBartPreTrainedModel (line 466) | class TFMBartPreTrainedModel(TFPreTrainedModel):
  class TFMBartEncoder (line 638) | class TFMBartEncoder(tf.keras.layers.Layer):
    method __init__ (line 648) | def __init__(self, config: MBartConfig, embed_tokens: Optional[tf.kera...
    method get_embed_tokens (line 667) | def get_embed_tokens(self):
    method set_embed_tokens (line 670) | def set_embed_tokens(self, embed_tokens):
    method call (line 674) | def call(
  class TFMBartDecoder (line 806) | class TFMBartDecoder(tf.keras.layers.Layer):
    method __init__ (line 816) | def __init__(self, config: MBartConfig, embed_tokens: Optional[tf.kera...
    method get_embed_tokens (line 834) | def get_embed_tokens(self):
    method set_embed_tokens (line 837) | def set_embed_tokens(self, embed_tokens):
    method call (line 841) | def call(
  class TFMBartMainLayer (line 1045) | class TFMBartMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1048) | def __init__(self, config: MBartConfig, **kwargs):
    method get_input_embeddings (line 1064) | def get_input_embeddings(self):
    method set_input_embeddings (line 1067) | def set_input_embeddings(self, new_embeddings):
    method call (line 1073) | def call(
  class TFMBartModel (line 1162) | class TFMBartModel(TFMBartPreTrainedModel):
    method __init__ (line 1163) | def __init__(self, config: MBartConfig, *inputs, **kwargs):
    method get_encoder (line 1168) | def get_encoder(self):
    method get_decoder (line 1171) | def get_decoder(self):
    method call (line 1181) | def call(
    method serving_output (line 1225) | def serving_output(self, output):
  class BiasLayer (line 1246) | class BiasLayer(tf.keras.layers.Layer):
    method __init__ (line 1252) | def __init__(self, shape, initializer, trainable, name, **kwargs):
    method call (line 1259) | def call(self, x):
  class TFMBartForConditionalGeneration (line 1267) | class TFMBartForConditionalGeneration(TFMBartPreTrainedModel, TFCausalLa...
    method __init__ (line 1273) | def __init__(self, config, *inputs, **kwargs):
    method get_decoder (line 1282) | def get_decoder(self):
    method get_encoder (line 1285) | def get_encoder(self):
    method get_output_embeddings (line 1288) | def get_output_embeddings(self):
    method set_output_embeddings (line 1291) | def set_output_embeddings(self, value):
    method get_bias (line 1294) | def get_bias(self):
    method set_bias (line 1297) | def set_bias(self, value):
    method call (line 1309) | def call(
    method serving_output (line 1389) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 1409) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1447) | def prepare_decoder_input_ids_from_labels(self, labels: tf.Tensor):

FILE: transformers/models/mbart/tokenization_mbart.py
  class MBartTokenizer (line 53) | class MBartTokenizer(PreTrainedTokenizer):
    method __init__ (line 82) | def __init__(
    method __getstate__ (line 158) | def __getstate__(self):
    method __setstate__ (line 164) | def __setstate__(self, d):
    method vocab_size (line 175) | def vocab_size(self):
    method src_lang (line 179) | def src_lang(self) -> str:
    method src_lang (line 183) | def src_lang(self, new_src_lang: str) -> None:
    method get_special_tokens_mask (line 187) | def get_special_tokens_mask(
    method build_inputs_with_special_tokens (line 217) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 244) | def create_token_type_ids_from_sequences(
    method _build_translation_inputs (line 269) | def _build_translation_inputs(
    method get_vocab (line 281) | def get_vocab(self):
    method _tokenize (line 286) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 289) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 298) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 304) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 309) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method prepare_seq2seq_batch (line 326) | def prepare_seq2seq_batch(
    method _switch_to_input_mode (line 338) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 341) | def _switch_to_target_mode(self):
    method set_src_lang_special_tokens (line 344) | def set_src_lang_special_tokens(self, src_lang) -> None:
    method set_tgt_lang_special_tokens (line 350) | def set_tgt_lang_special_tokens(self, lang: str) -> None:

FILE: transformers/models/mbart/tokenization_mbart_fast.py
  class MBartTokenizerFast (line 63) | class MBartTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 96) | def __init__(
    method src_lang (line 153) | def src_lang(self) -> str:
    method src_lang (line 157) | def src_lang(self, new_src_lang: str) -> None:
    method build_inputs_with_special_tokens (line 161) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 190) | def create_token_type_ids_from_sequences(
    method _build_translation_inputs (line 215) | def _build_translation_inputs(
    method prepare_seq2seq_batch (line 227) | def prepare_seq2seq_batch(
    method _switch_to_input_mode (line 239) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 242) | def _switch_to_target_mode(self):
    method set_src_lang_special_tokens (line 245) | def set_src_lang_special_tokens(self, src_lang) -> None:
    method set_tgt_lang_special_tokens (line 260) | def set_tgt_lang_special_tokens(self, lang: str) -> None:
    method save_vocabulary (line 275) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/mbart50/tokenization_mbart50.py
  class MBart50Tokenizer (line 49) | class MBart50Tokenizer(PreTrainedTokenizer):
    method __init__ (line 116) | def __init__(
    method vocab_size (line 185) | def vocab_size(self) -> int:
    method src_lang (line 189) | def src_lang(self) -> str:
    method src_lang (line 193) | def src_lang(self, new_src_lang: str) -> None:
    method __getstate__ (line 197) | def __getstate__(self) -> Dict:
    method __setstate__ (line 202) | def __setstate__(self, d: Dict) -> None:
    method get_vocab (line 212) | def get_vocab(self) -> Dict:
    method _tokenize (line 217) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 220) | def _convert_token_to_id(self, token: str) -> int:
    method _convert_id_to_token (line 229) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 235) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 254) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method get_special_tokens_mask (line 271) | def get_special_tokens_mask(
    method build_inputs_with_special_tokens (line 301) | def build_inputs_with_special_tokens(
    method _build_translation_inputs (line 328) | def _build_translation_inputs(
    method prepare_seq2seq_batch (line 340) | def prepare_seq2seq_batch(
    method _switch_to_input_mode (line 352) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 355) | def _switch_to_target_mode(self):
    method set_src_lang_special_tokens (line 358) | def set_src_lang_special_tokens(self, src_lang: str) -> None:
    method set_tgt_lang_special_tokens (line 364) | def set_tgt_lang_special_tokens(self, tgt_lang: str) -> None:

FILE: transformers/models/mbart50/tokenization_mbart50_fast.py
  class MBart50TokenizerFast (line 59) | class MBart50TokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 113) | def __init__(
    method src_lang (line 162) | def src_lang(self) -> str:
    method src_lang (line 166) | def src_lang(self, new_src_lang: str) -> None:
    method build_inputs_with_special_tokens (line 170) | def build_inputs_with_special_tokens(
    method prepare_seq2seq_batch (line 199) | def prepare_seq2seq_batch(
    method _switch_to_input_mode (line 211) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 214) | def _switch_to_target_mode(self):
    method set_src_lang_special_tokens (line 217) | def set_src_lang_special_tokens(self, src_lang: str) -> None:
    method set_tgt_lang_special_tokens (line 232) | def set_tgt_lang_special_tokens(self, tgt_lang: str) -> None:
    method _build_translation_inputs (line 247) | def _build_translation_inputs(
    method save_vocabulary (line 259) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/mctct/configuration_mctct.py
  class MCTCTConfig (line 29) | class MCTCTConfig(PretrainedConfig):
    method __init__ (line 119) | def __init__(

FILE: transformers/models/mctct/feature_extraction_mctct.py
  class MCTCTFeatureExtractor (line 34) | class MCTCTFeatureExtractor(SequenceFeatureExtractor):
    method __init__ (line 71) | def __init__(
    method _extract_mfsc_features (line 108) | def _extract_mfsc_features(self, one_waveform: np.array) -> np.ndarray:
    method _normalize_one (line 141) | def _normalize_one(self, x, input_length, padding_value):
    method normalize (line 158) | def normalize(
    method __call__ (line 164) | def __call__(

FILE: transformers/models/mctct/modeling_mctct.py
  function _expand_mask (line 62) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class MCTCTConv1dSubsampler (line 76) | class MCTCTConv1dSubsampler(nn.Module):
    method __init__ (line 82) | def __init__(self, config):
    method forward (line 121) | def forward(self, input_features):
  class MCTCTEmbeddings (line 137) | class MCTCTEmbeddings(nn.Module):
    method __init__ (line 140) | def __init__(self, config):
    method forward (line 160) | def forward(
  class MCTCTSelfAttention (line 193) | class MCTCTSelfAttention(nn.Module):
    method __init__ (line 194) | def __init__(self, config):
    method transpose_for_scores (line 217) | def transpose_for_scores(self, x):
    method reshape_fortran (line 222) | def reshape_fortran(self, x, shape):
    method relative_position_embedding_rotate (line 227) | def relative_position_embedding_rotate(self, scores):
    method forward (line 252) | def forward(
  class MCTCTLayerNorm (line 301) | class MCTCTLayerNorm(nn.Module):
    method __init__ (line 302) | def __init__(self):
    method forward (line 307) | def forward(self, hidden_states):
  class MCTCTSelfOutput (line 311) | class MCTCTSelfOutput(nn.Module):
    method __init__ (line 312) | def __init__(self, config):
    method forward (line 319) | def forward(self, hidden_states, input_tensor):
  class MCTCTAttention (line 326) | class MCTCTAttention(nn.Module):
    method __init__ (line 327) | def __init__(self, config):
    method prune_heads (line 333) | def prune_heads(self, heads):
    method forward (line 351) | def forward(
  class MCTCTIntermediate (line 370) | class MCTCTIntermediate(nn.Module):
    method __init__ (line 371) | def __init__(self, config):
    method forward (line 379) | def forward(self, hidden_states):
  class MCTCTOutput (line 385) | class MCTCTOutput(nn.Module):
    method __init__ (line 386) | def __init__(self, config):
    method forward (line 392) | def forward(self, hidden_states, input_tensor):
  class MCTCTLayer (line 399) | class MCTCTLayer(nn.Module):
    method __init__ (line 400) | def __init__(self, config: MCTCTConfig):
    method forward (line 411) | def forward(
    method feed_forward_chunk (line 432) | def feed_forward_chunk(self, attention_output):
  class MCTCTPreTrainedModel (line 438) | class MCTCTPreTrainedModel(PreTrainedModel):
    method _init_weights (line 450) | def _init_weights(self, module):
    method _get_feat_extract_output_lengths (line 474) | def _get_feat_extract_output_lengths(self, input_lengths: torch.LongTe...
    method _get_feature_vector_attention_mask (line 488) | def _get_feature_vector_attention_mask(self, feature_vector_length, at...
    method _set_gradient_checkpointing (line 507) | def _set_gradient_checkpointing(self, module, value=False):
  class MCTCTEncoder (line 555) | class MCTCTEncoder(MCTCTPreTrainedModel):
    method __init__ (line 556) | def __init__(self, config: MCTCTConfig):
    method forward (line 566) | def forward(
  class MCTCTModel (line 661) | class MCTCTModel(MCTCTPreTrainedModel):
    method __init__ (line 662) | def __init__(self, config):
    method forward (line 679) | def forward(
  class MCTCTForCTC (line 721) | class MCTCTForCTC(MCTCTPreTrainedModel):
    method __init__ (line 722) | def __init__(self, config):
    method forward (line 749) | def forward(

FILE: transformers/models/mctct/processing_mctct.py
  class MCTCTProcessor (line 24) | class MCTCTProcessor(ProcessorMixin):
    method __init__ (line 40) | def __init__(self, feature_extractor, tokenizer):
    method __call__ (line 45) | def __call__(self, *args, **kwargs):
    method batch_decode (line 83) | def batch_decode(self, *args, **kwargs):
    method pad (line 90) | def pad(self, *args, **kwargs):
    method decode (line 120) | def decode(self, *args, **kwargs):
    method as_target_processor (line 128) | def as_target_processor(self):

FILE: transformers/models/mega/configuration_mega.py
  class MegaConfig (line 31) | class MegaConfig(PretrainedConfig):
    method __init__ (line 150) | def __init__(
  class MegaOnnxConfig (line 230) | class MegaOnnxConfig(OnnxConfig):
    method inputs (line 232) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/mega/convert_mega_original_pytorch_checkpoint_to_pytorch.py
  class MegaLM (line 51) | class MegaLM(nn.Module):
    method __init__ (line 54) | def __init__(self, mega_args, depth, vocab_size):
    method forward (line 61) | def forward(self, input_ids, attention_mask, batch_first=True, ignore_...
  class OriginalMegaForMaskedLM (line 101) | class OriginalMegaForMaskedLM(nn.Module):
    method __init__ (line 104) | def __init__(self, mega_args, depth, vocab_size):
    method forward (line 110) | def forward(self, input_ids, attention_mask, batch_first=True, ignore_...
  function convert_checkpoint_to_huggingface (line 123) | def convert_checkpoint_to_huggingface(pretrained_checkpoint_path, output...

FILE: transformers/models/mega/modeling_mega.py
  class MegaEmbeddings (line 59) | class MegaEmbeddings(nn.Module):
    method __init__ (line 65) | def __init__(self, config: MegaConfig):
    method forward (line 79) | def forward(self, input_ids=None, token_type_ids=None, inputs_embeds=N...
  class MegaSimpleRelativePositionalBias (line 113) | class MegaSimpleRelativePositionalBias(nn.Module):
    method __init__ (line 118) | def __init__(self, config: MegaConfig):
    method forward (line 124) | def forward(self, seq_len):
  class MegaRotaryRelativePositionalBias (line 143) | class MegaRotaryRelativePositionalBias(nn.Module):
    method __init__ (line 152) | def __init__(self, config: MegaConfig):
    method get_sinusoid_embeddings (line 169) | def get_sinusoid_embeddings(max_positions: int, embedding_dim: int):
    method rotary (line 176) | def rotary(self, input):
    method forward (line 189) | def forward(self, seq_len):
  class MegaDropout (line 196) | class MegaDropout(nn.Module):
    method __init__ (line 205) | def __init__(self, dropout_probability, is_featurewise=False):
    method forward (line 210) | def forward(self, input, batch_first: bool = False):
  class MegaRMSNorm (line 234) | class MegaRMSNorm(nn.Module):
    method __init__ (line 240) | def __init__(self, number_features, eps=1e-6, affine=True):
    method forward (line 250) | def forward(self, input):
  class MegaScaleNorm (line 259) | class MegaScaleNorm(nn.Module):
    method __init__ (line 265) | def __init__(self, dim, eps=1e-6, affine=True):
    method forward (line 275) | def forward(self, input):
  class MegaSequenceNorm (line 284) | class MegaSequenceNorm(nn.Module):
    method __init__ (line 290) | def __init__(self, norm_type, embedding_dim, eps=1e-5, affine=True, ex...
    method forward (line 305) | def forward(self, input):
  class MegaMultiDimensionDampedEma (line 320) | class MegaMultiDimensionDampedEma(nn.Module):
    method __init__ (line 327) | def __init__(self, config: MegaConfig):
    method _compute_ema_coefficients (line 351) | def _compute_ema_coefficients(self):
    method _compute_efficient_ema_kernel (line 359) | def _compute_efficient_ema_kernel(self, length: int):
    method get_ema_coefficients (line 371) | def get_ema_coefficients(self):
    method get_ema_kernel (line 379) | def get_ema_kernel(self, length: int):
    method fft_convolution (line 388) | def fft_convolution(self, inputs, kernel, length):
    method ema_step (line 395) | def ema_step(self, inputs, length, past_state=None):
    method one_ema_step (line 437) | def one_ema_step(self, inputs, past_state=None):
    method forward (line 449) | def forward(
  class MegaGatedCrossAttention (line 529) | class MegaGatedCrossAttention(nn.Module):
    method __init__ (line 536) | def __init__(self, config: MegaConfig):
    method element_attention (line 574) | def element_attention(self, query, key, key_padding_mask, pidx):
    method softmax_attention (line 604) | def softmax_attention(self, query, key, key_padding_mask, pidx):
    method forward (line 630) | def forward(
  class MegaMovingAverageGatedAttention (line 789) | class MegaMovingAverageGatedAttention(nn.Module):
    method __init__ (line 798) | def __init__(self, config: MegaConfig):
    method element_attention (line 839) | def element_attention(self, query, key, padding_mask, causal_mask):
    method softmax_attention (line 878) | def softmax_attention(self, query, key, padding_mask, causal_mask):
    method forward (line 916) | def forward(
  class MegaNormalizedFeedForwardNetwork (line 1125) | class MegaNormalizedFeedForwardNetwork(nn.Module):
    method __init__ (line 1131) | def __init__(self, config: MegaConfig):
    method forward (line 1152) | def forward(self, inputs):
  class MegaBlock (line 1170) | class MegaBlock(nn.Module):
    method __init__ (line 1171) | def __init__(self, config: MegaConfig):
    method forward (line 1185) | def forward(
  class MegaPooler (line 1320) | class MegaPooler(nn.Module):
    method __init__ (line 1321) | def __init__(self, config):
    method forward (line 1326) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MegaPreTrainedModel (line 1335) | class MegaPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1346) | def _init_weights(self, module):
    method update_keys_to_ignore (line 1390) | def update_keys_to_ignore(self, config, del_keys_to_ignore):
  class MegaModel (line 1460) | class MegaModel(MegaPreTrainedModel):
    method __init__ (line 1479) | def __init__(self, config: MegaConfig, add_pooling_layer=True):
    method get_input_embeddings (line 1491) | def get_input_embeddings(self):
    method set_input_embeddings (line 1494) | def set_input_embeddings(self, value):
    method forward (line 1503) | def forward(
  class MegaForCausalLM (line 1658) | class MegaForCausalLM(MegaPreTrainedModel):
    method __init__ (line 1663) | def __init__(self, config: MegaConfig):
    method get_output_embeddings (line 1686) | def get_output_embeddings(self):
    method set_output_embeddings (line 1689) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1694) | def forward(
    method prepare_inputs_for_generation (line 1802) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1814) | def _reorder_cache(self, past_key_values, beam_idx):
  class MegaForMaskedLM (line 1822) | class MegaForMaskedLM(MegaPreTrainedModel):
    method __init__ (line 1827) | def __init__(self, config: MegaConfig):
    method get_output_embeddings (line 1852) | def get_output_embeddings(self):
    method set_output_embeddings (line 1855) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1867) | def forward(
  class MegaForSequenceClassification (line 1931) | class MegaForSequenceClassification(MegaPreTrainedModel):
    method __init__ (line 1934) | def __init__(self, config):
    method forward (line 1951) | def forward(
  class MegaForMultipleChoice (line 2024) | class MegaForMultipleChoice(MegaPreTrainedModel):
    method __init__ (line 2027) | def __init__(self, config):
    method forward (line 2043) | def forward(
  class MegaForTokenClassification (line 2111) | class MegaForTokenClassification(MegaPreTrainedModel):
    method __init__ (line 2115) | def __init__(self, config):
    method forward (line 2135) | def forward(
  class MegaClassificationHead (line 2185) | class MegaClassificationHead(nn.Module):
    method __init__ (line 2188) | def __init__(self, config):
    method forward (line 2197) | def forward(self, features, **kwargs):
  class MegaForQuestionAnswering (line 2214) | class MegaForQuestionAnswering(MegaPreTrainedModel):
    method __init__ (line 2218) | def __init__(self, config):
    method forward (line 2234) | def forward(

FILE: transformers/models/megatron_bert/configuration_megatron_bert.py
  class MegatronBertConfig (line 28) | class MegatronBertConfig(PretrainedConfig):
    method __init__ (line 95) | def __init__(

FILE: transformers/models/megatron_bert/convert_megatron_bert_checkpoint.py
  function recursive_print (line 48) | def recursive_print(name, val, spaces=0):
  function fix_query_key_value_ordering (line 68) | def fix_query_key_value_ordering(param, checkpoint_version, num_splits, ...
  function convert_megatron_checkpoint (line 94) | def convert_megatron_checkpoint(args, input_state_dict, config):
  function main (line 275) | def main():

FILE: transformers/models/megatron_bert/modeling_megatron_bert.py
  function load_tf_weights_in_megatron_bert (line 66) | def load_tf_weights_in_megatron_bert(model, config, tf_checkpoint_path):
  class MegatronBertEmbeddings (line 135) | class MegatronBertEmbeddings(nn.Module):
    method __init__ (line 138) | def __init__(self, config):
    method forward (line 155) | def forward(
  class MegatronBertSelfAttention (line 192) | class MegatronBertSelfAttention(nn.Module):
    method __init__ (line 193) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 219) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 224) | def forward(
  class MegatronBertSelfOutput (line 327) | class MegatronBertSelfOutput(nn.Module):
    method __init__ (line 328) | def __init__(self, config):
    method forward (line 333) | def forward(self, hidden_states: torch.Tensor, residual: torch.Tensor)...
  class MegatronBertAttention (line 340) | class MegatronBertAttention(nn.Module):
    method __init__ (line 341) | def __init__(self, config):
    method prune_heads (line 348) | def prune_heads(self, heads):
    method forward (line 366) | def forward(
  class MegatronBertIntermediate (line 392) | class MegatronBertIntermediate(nn.Module):
    method __init__ (line 393) | def __init__(self, config):
    method forward (line 401) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MegatronBertOutput (line 408) | class MegatronBertOutput(nn.Module):
    method __init__ (line 409) | def __init__(self, config):
    method forward (line 414) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class MegatronBertLayer (line 421) | class MegatronBertLayer(nn.Module):
    method __init__ (line 422) | def __init__(self, config):
    method forward (line 437) | def forward(
    method feed_forward_chunk (line 502) | def feed_forward_chunk(self, attention_output):
  class MegatronBertEncoder (line 509) | class MegatronBertEncoder(nn.Module):
    method __init__ (line 510) | def __init__(self, config):
    method forward (line 520) | def forward(
  class MegatronBertPooler (line 617) | class MegatronBertPooler(nn.Module):
    method __init__ (line 618) | def __init__(self, config):
    method forward (line 623) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MegatronBertPredictionHeadTransform (line 633) | class MegatronBertPredictionHeadTransform(nn.Module):
    method __init__ (line 634) | def __init__(self, config):
    method forward (line 643) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MegatronBertLMPredictionHead (line 651) | class MegatronBertLMPredictionHead(nn.Module):
    method __init__ (line 652) | def __init__(self, config):
    method forward (line 665) | def forward(self, hidden_states):
  class MegatronBertOnlyMLMHead (line 672) | class MegatronBertOnlyMLMHead(nn.Module):
    method __init__ (line 673) | def __init__(self, config):
    method forward (line 677) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class MegatronBertOnlyNSPHead (line 683) | class MegatronBertOnlyNSPHead(nn.Module):
    method __init__ (line 684) | def __init__(self, config):
    method forward (line 688) | def forward(self, pooled_output):
  class MegatronBertPreTrainingHeads (line 694) | class MegatronBertPreTrainingHeads(nn.Module):
    method __init__ (line 695) | def __init__(self, config):
    method forward (line 700) | def forward(self, sequence_output, pooled_output):
  class MegatronBertPreTrainedModel (line 706) | class MegatronBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 718) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 730) | def _set_gradient_checkpointing(self, module, value=False):
  class MegatronBertForPreTrainingOutput (line 737) | class MegatronBertForPreTrainingOutput(ModelOutput):
  class MegatronBertModel (line 840) | class MegatronBertModel(MegatronBertPreTrainedModel):
    method __init__ (line 853) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 865) | def get_input_embeddings(self):
    method set_input_embeddings (line 868) | def set_input_embeddings(self, value):
    method _prune_heads (line 871) | def _prune_heads(self, heads_to_prune):
    method forward (line 885) | def forward(
  class MegatronBertForPreTraining (line 1016) | class MegatronBertForPreTraining(MegatronBertPreTrainedModel):
    method __init__ (line 1019) | def __init__(self, config, add_binary_head=True):
    method get_output_embeddings (line 1028) | def get_output_embeddings(self):
    method set_output_embeddings (line 1031) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1036) | def forward(
  class MegatronBertForCausalLM (line 1122) | class MegatronBertForCausalLM(MegatronBertPreTrainedModel):
    method __init__ (line 1126) | def __init__(self, config):
    method get_output_embeddings (line 1138) | def get_output_embeddings(self):
    method set_output_embeddings (line 1141) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1146) | def forward(
    method prepare_inputs_for_generation (line 1247) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1259) | def _reorder_cache(self, past_key_values, beam_idx):
  class MegatronBertForMaskedLM (line 1267) | class MegatronBertForMaskedLM(MegatronBertPreTrainedModel):
    method __init__ (line 1271) | def __init__(self, config):
    method get_output_embeddings (line 1286) | def get_output_embeddings(self):
    method set_output_embeddings (line 1289) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1298) | def forward(
    method prepare_inputs_for_generation (line 1355) | def prepare_inputs_for_generation(self, input_ids, attention_mask=None...
  class MegatronBertForNextSentencePrediction (line 1375) | class MegatronBertForNextSentencePrediction(MegatronBertPreTrainedModel):
    method __init__ (line 1378) | def __init__(self, config):
    method forward (line 1389) | def forward(
  class MegatronBertForSequenceClassification (line 1481) | class MegatronBertForSequenceClassification(MegatronBertPreTrainedModel):
    method __init__ (line 1482) | def __init__(self, config):
    method forward (line 1499) | def forward(
  class MegatronBertForMultipleChoice (line 1578) | class MegatronBertForMultipleChoice(MegatronBertPreTrainedModel):
    method __init__ (line 1579) | def __init__(self, config):
    method forward (line 1597) | def forward(
  class MegatronBertForTokenClassification (line 1671) | class MegatronBertForTokenClassification(MegatronBertPreTrainedModel):
    method __init__ (line 1674) | def __init__(self, config):
    method forward (line 1691) | def forward(
  class MegatronBertForQuestionAnswering (line 1751) | class MegatronBertForQuestionAnswering(MegatronBertPreTrainedModel):
    method __init__ (line 1754) | def __init__(self, config):
    method forward (line 1770) | def forward(

FILE: transformers/models/megatron_gpt2/checkpoint_reshaping_and_interoperability.py
  function add_checkpointing_args (line 28) | def add_checkpointing_args(parser):
  function add_megatron_checkpoint_args (line 54) | def add_megatron_checkpoint_args(parser):
  function add_transformers_checkpoint_args (line 112) | def add_transformers_checkpoint_args(parser):
  function recursive_print (line 168) | def recursive_print(name, val, spaces=0):
  function megatron_to_transformers_fix_query_key_value_ordering (line 196) | def megatron_to_transformers_fix_query_key_value_ordering(
  function transformers_to_megatron_fix_query_key_value_ordering (line 230) | def transformers_to_megatron_fix_query_key_value_ordering(
  function merge_transformers_sharded_states (line 264) | def merge_transformers_sharded_states(path, num_checkpoints):
  function get_megatron_sharded_states (line 280) | def get_megatron_sharded_states(args, tp_size, pp_size, pp_rank):
  function get_element_from_dict_by_path (line 301) | def get_element_from_dict_by_path(d, path):
  function convert_checkpoint_from_megatron_to_transformers (line 317) | def convert_checkpoint_from_megatron_to_transformers(args):
  function convert_checkpoint_from_transformers_to_megatron (line 591) | def convert_checkpoint_from_transformers_to_megatron(args):
  function main (line 882) | def main():

FILE: transformers/models/megatron_gpt2/convert_megatron_gpt2_checkpoint.py
  function recursive_print (line 48) | def recursive_print(name, val, spaces=0):
  function fix_query_key_value_ordering (line 68) | def fix_query_key_value_ordering(param, checkpoint_version, num_splits, ...
  function convert_megatron_checkpoint (line 94) | def convert_megatron_checkpoint(args, input_state_dict, config):
  function main (line 240) | def main():

FILE: transformers/models/mgp_str/configuration_mgp_str.py
  class MgpstrConfig (line 28) | class MgpstrConfig(PretrainedConfig):
    method __init__ (line 94) | def __init__(

FILE: transformers/models/mgp_str/modeling_mgp_str.py
  function drop_path (line 54) | def drop_path(input, drop_prob: float = 0.0, training: bool = False):
  class MgpstrDropPath (line 75) | class MgpstrDropPath(nn.Module):
    method __init__ (line 78) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 82) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 85) | def extra_repr(self) -> str:
  class MgpstrModelOutput (line 90) | class MgpstrModelOutput(ModelOutput):
  class MgpstrEmbeddings (line 127) | class MgpstrEmbeddings(nn.Module):
    method __init__ (line 130) | def __init__(self, config: MgpstrConfig):
    method forward (line 155) | def forward(self, pixel_values):
  class MgpstrMlp (line 173) | class MgpstrMlp(nn.Module):
    method __init__ (line 176) | def __init__(self, config: MgpstrConfig, hidden_features):
    method forward (line 184) | def forward(self, hidden_states):
  class MgpstrAttention (line 193) | class MgpstrAttention(nn.Module):
    method __init__ (line 194) | def __init__(self, config: MgpstrConfig):
    method forward (line 205) | def forward(self, hidden_states):
  class MgpstrLayer (line 224) | class MgpstrLayer(nn.Module):
    method __init__ (line 225) | def __init__(self, config: MgpstrConfig, drop_path=None):
    method forward (line 235) | def forward(self, hidden_states):
  class MgpstrEncoder (line 250) | class MgpstrEncoder(nn.Module):
    method __init__ (line 251) | def __init__(self, config: MgpstrConfig):
    method forward (line 260) | def forward(self, hidden_states, output_attentions=False, output_hidde...
  class MgpstrA3Module (line 286) | class MgpstrA3Module(nn.Module):
    method __init__ (line 287) | def __init__(self, config: MgpstrConfig):
    method forward (line 299) | def forward(self, hidden_states):
  class MgpstrPreTrainedModel (line 314) | class MgpstrPreTrainedModel(PreTrainedModel):
    method _init_weights (line 323) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 336) | def _set_gradient_checkpointing(self, module: MgpstrEncoder, value: bo...
  class MgpstrModel (line 372) | class MgpstrModel(MgpstrPreTrainedModel):
    method __init__ (line 373) | def __init__(self, config: MgpstrConfig):
    method get_input_embeddings (line 379) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 383) | def forward(self, pixel_values, output_attentions=None, output_hidden_...
  class MgpstrForSceneTextRecognition (line 418) | class MgpstrForSceneTextRecognition(MgpstrPreTrainedModel):
    method __init__ (line 422) | def __init__(self, config: MgpstrConfig) -> None:
    method forward (line 438) | def forward(

FILE: transformers/models/mgp_str/processing_mgp_str.py
  class DecodeType (line 30) | class DecodeType(ExplicitEnum):
  class MgpstrProcessor (line 39) | class MgpstrProcessor(ProcessorMixin):
    method __init__ (line 56) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 77) | def __call__(self, text=None, images=None, return_tensors=None, **kwar...
    method batch_decode (line 100) | def batch_decode(self, sequences):
    method _decode_helper (line 142) | def _decode_helper(self, pred_logits, format):
    method char_decode (line 192) | def char_decode(self, sequences):
    method bpe_decode (line 205) | def bpe_decode(self, sequences):
    method wp_decode (line 217) | def wp_decode(self, sequences):

FILE: transformers/models/mgp_str/tokenization_mgp_str.py
  class MgpstrTokenizer (line 38) | class MgpstrTokenizer(PreTrainedTokenizer):
    method __init__ (line 64) | def __init__(self, vocab_file, unk_token="[GO]", bos_token="[GO]", eos...
    method vocab_size (line 78) | def vocab_size(self):
    method get_vocab (line 81) | def get_vocab(self):
    method _tokenize (line 84) | def _tokenize(self, text):
    method _convert_token_to_id (line 91) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 95) | def _convert_id_to_token(self, index):
    method save_vocabulary (line 99) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/mluke/convert_mluke_original_pytorch_checkpoint_to_pytorch.py
  function convert_luke_checkpoint (line 29) | def convert_luke_checkpoint(checkpoint_path, metadata_path, entity_vocab...
  function load_original_entity_vocab (line 186) | def load_original_entity_vocab(entity_vocab_path):

FILE: transformers/models/mluke/tokenization_mluke.py
  class MLukeTokenizer (line 147) | class MLukeTokenizer(PreTrainedTokenizer):
    method __init__ (line 237) | def __init__(
    method __getstate__ (line 348) | def __getstate__(self):
    method __setstate__ (line 354) | def __setstate__(self, d):
    method __call__ (line 366) | def __call__(
    method _encode_plus (line 512) | def _encode_plus(
    method _batch_encode_plus (line 593) | def _batch_encode_plus(
    method _check_entity_input_format (line 702) | def _check_entity_input_format(self, entities: Optional[EntityInput], ...
    method _create_input_sequence (line 721) | def _create_input_sequence(
    method _batch_prepare_for_model (line 875) | def _batch_prepare_for_model(
    method prepare_for_model (line 958) | def prepare_for_model(
    method pad (line 1188) | def pad(
    method _pad (line 1349) | def _pad(
    method save_vocabulary (line 1488) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 1510) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 1537) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 1566) | def create_token_type_ids_from_sequences(
    method vocab_size (line 1593) | def vocab_size(self):
    method get_vocab (line 1597) | def get_vocab(self):
    method _tokenize (line 1603) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 1607) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 1616) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 1622) | def convert_tokens_to_string(self, tokens):

FILE: transformers/models/mmbt/configuration_mmbt.py
  class MMBTConfig (line 24) | class MMBTConfig(object):
    method __init__ (line 38) | def __init__(self, config, num_labels=None, modal_hidden_size=2048):

FILE: transformers/models/mmbt/modeling_mmbt.py
  class ModalEmbeddings (line 33) | class ModalEmbeddings(nn.Module):
    method __init__ (line 36) | def __init__(self, config, encoder, embeddings):
    method forward (line 47) | def forward(self, input_modal, start_token=None, end_token=None, posit...
  class MMBTModel (line 181) | class MMBTModel(nn.Module, ModuleUtilsMixin):
    method __init__ (line 182) | def __init__(self, config, transformer, encoder):
    method forward (line 190) | def forward(
    method get_input_embeddings (line 299) | def get_input_embeddings(self):
    method set_input_embeddings (line 302) | def set_input_embeddings(self, value):
  class MMBTForClassification (line 313) | class MMBTForClassification(nn.Module):
    method __init__ (line 343) | def __init__(self, config, transformer, encoder):
    method forward (line 351) | def forward(

FILE: transformers/models/mobilebert/configuration_mobilebert.py
  class MobileBertConfig (line 31) | class MobileBertConfig(PretrainedConfig):
    method __init__ (line 114) | def __init__(
  class MobileBertOnnxConfig (line 174) | class MobileBertOnnxConfig(OnnxConfig):
    method inputs (line 176) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/mobilebert/convert_mobilebert_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 26) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, mobilebert_conf...

FILE: transformers/models/mobilebert/modeling_mobilebert.py
  function load_tf_weights_in_mobilebert (line 82) | def load_tf_weights_in_mobilebert(model, config, tf_checkpoint_path):
  class NoNorm (line 160) | class NoNorm(nn.Module):
    method __init__ (line 161) | def __init__(self, feat_size, eps=None):
    method forward (line 166) | def forward(self, input_tensor: torch.Tensor) -> torch.Tensor:
  class MobileBertEmbeddings (line 173) | class MobileBertEmbeddings(nn.Module):
    method __init__ (line 176) | def __init__(self, config):
    method forward (line 196) | def forward(
  class MobileBertSelfAttention (line 247) | class MobileBertSelfAttention(nn.Module):
    method __init__ (line 248) | def __init__(self, config):
    method transpose_for_scores (line 261) | def transpose_for_scores(self, x):
    method forward (line 266) | def forward(
  class MobileBertSelfOutput (line 305) | class MobileBertSelfOutput(nn.Module):
    method __init__ (line 306) | def __init__(self, config):
    method forward (line 314) | def forward(self, hidden_states: torch.Tensor, residual_tensor: torch....
  class MobileBertAttention (line 322) | class MobileBertAttention(nn.Module):
    method __init__ (line 323) | def __init__(self, config):
    method prune_heads (line 329) | def prune_heads(self, heads):
    method forward (line 347) | def forward(
  class MobileBertIntermediate (line 372) | class MobileBertIntermediate(nn.Module):
    method __init__ (line 373) | def __init__(self, config):
    method forward (line 381) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class OutputBottleneck (line 387) | class OutputBottleneck(nn.Module):
    method __init__ (line 388) | def __init__(self, config):
    method forward (line 394) | def forward(self, hidden_states: torch.Tensor, residual_tensor: torch....
  class MobileBertOutput (line 401) | class MobileBertOutput(nn.Module):
    method __init__ (line 402) | def __init__(self, config):
    method forward (line 412) | def forward(
  class BottleneckLayer (line 425) | class BottleneckLayer(nn.Module):
    method __init__ (line 426) | def __init__(self, config):
    method forward (line 431) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Bottleneck (line 437) | class Bottleneck(nn.Module):
    method __init__ (line 438) | def __init__(self, config):
    method forward (line 446) | def forward(self, hidden_states: torch.Tensor) -> Tuple[torch.Tensor]:
  class FFNOutput (line 473) | class FFNOutput(nn.Module):
    method __init__ (line 474) | def __init__(self, config):
    method forward (line 479) | def forward(self, hidden_states: torch.Tensor, residual_tensor: torch....
  class FFNLayer (line 485) | class FFNLayer(nn.Module):
    method __init__ (line 486) | def __init__(self, config):
    method forward (line 491) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileBertLayer (line 497) | class MobileBertLayer(nn.Module):
    method __init__ (line 498) | def __init__(self, config):
    method forward (line 511) | def forward(
  class MobileBertEncoder (line 560) | class MobileBertEncoder(nn.Module):
    method __init__ (line 561) | def __init__(self, config):
    method forward (line 565) | def forward(
  class MobileBertPooler (line 602) | class MobileBertPooler(nn.Module):
    method __init__ (line 603) | def __init__(self, config):
    method forward (line 609) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileBertPredictionHeadTransform (line 621) | class MobileBertPredictionHeadTransform(nn.Module):
    method __init__ (line 622) | def __init__(self, config):
    method forward (line 631) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileBertLMPredictionHead (line 638) | class MobileBertLMPredictionHead(nn.Module):
    method __init__ (line 639) | def __init__(self, config):
    method forward (line 650) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileBertOnlyMLMHead (line 657) | class MobileBertOnlyMLMHead(nn.Module):
    method __init__ (line 658) | def __init__(self, config):
    method forward (line 662) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class MobileBertPreTrainingHeads (line 667) | class MobileBertPreTrainingHeads(nn.Module):
    method __init__ (line 668) | def __init__(self, config):
    method forward (line 673) | def forward(self, sequence_output: torch.Tensor, pooled_output: torch....
  class MobileBertPreTrainedModel (line 679) | class MobileBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 691) | def _init_weights(self, module):
  class MobileBertForPreTrainingOutput (line 709) | class MobileBertForPreTrainingOutput(ModelOutput):
  class MobileBertModel (line 812) | class MobileBertModel(MobileBertPreTrainedModel):
    method __init__ (line 817) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 828) | def get_input_embeddings(self):
    method set_input_embeddings (line 831) | def set_input_embeddings(self, value):
    method _prune_heads (line 834) | def _prune_heads(self, heads_to_prune):
    method forward (line 848) | def forward(
  class MobileBertForPreTraining (line 925) | class MobileBertForPreTraining(MobileBertPreTrainedModel):
    method __init__ (line 932) | def __init__(self, config):
    method get_output_embeddings (line 940) | def get_output_embeddings(self):
    method set_output_embeddings (line 943) | def set_output_embeddings(self, new_embeddigs):
    method resize_token_embeddings (line 946) | def resize_token_embeddings(self, new_num_tokens: Optional[int] = None...
    method forward (line 956) | def forward(
  class MobileBertForMaskedLM (line 1037) | class MobileBertForMaskedLM(MobileBertPreTrainedModel):
    method __init__ (line 1045) | def __init__(self, config):
    method get_output_embeddings (line 1054) | def get_output_embeddings(self):
    method set_output_embeddings (line 1057) | def set_output_embeddings(self, new_embeddigs):
    method resize_token_embeddings (line 1060) | def resize_token_embeddings(self, new_num_tokens: Optional[int] = None...
    method forward (line 1075) | def forward(
  class MobileBertOnlyNSPHead (line 1128) | class MobileBertOnlyNSPHead(nn.Module):
    method __init__ (line 1129) | def __init__(self, config):
    method forward (line 1133) | def forward(self, pooled_output: torch.Tensor) -> torch.Tensor:
  class MobileBertForNextSentencePrediction (line 1142) | class MobileBertForNextSentencePrediction(MobileBertPreTrainedModel):
    method __init__ (line 1143) | def __init__(self, config):
    method forward (line 1154) | def forward(
  class MobileBertForSequenceClassification (line 1246) | class MobileBertForSequenceClassification(MobileBertPreTrainedModel):
    method __init__ (line 1247) | def __init__(self, config):
    method forward (line 1270) | def forward(
  class MobileBertForQuestionAnswering (line 1350) | class MobileBertForQuestionAnswering(MobileBertPreTrainedModel):
    method __init__ (line 1353) | def __init__(self, config):
    method forward (line 1373) | def forward(
  class MobileBertForMultipleChoice (line 1456) | class MobileBertForMultipleChoice(MobileBertPreTrainedModel):
    method __init__ (line 1457) | def __init__(self, config):
    method forward (line 1478) | def forward(
  class MobileBertForTokenClassification (line 1553) | class MobileBertForTokenClassification(MobileBertPreTrainedModel):
    method __init__ (line 1556) | def __init__(self, config):
    method forward (line 1578) | def forward(

FILE: transformers/models/mobilebert/modeling_tf_mobilebert.py
  class TFMobileBertPreTrainingLoss (line 93) | class TFMobileBertPreTrainingLoss:
    method hf_compute_loss (line 100) | def hf_compute_loss(self, labels: tf.Tensor, logits: tf.Tensor) -> tf....
  class TFMobileBertIntermediate (line 123) | class TFMobileBertIntermediate(tf.keras.layers.Layer):
    method __init__ (line 124) | def __init__(self, config, **kwargs):
    method call (line 134) | def call(self, hidden_states):
  class TFLayerNorm (line 141) | class TFLayerNorm(tf.keras.layers.LayerNormalization):
    method __init__ (line 142) | def __init__(self, feat_size, *args, **kwargs):
  class TFNoNorm (line 146) | class TFNoNorm(tf.keras.layers.Layer):
    method __init__ (line 147) | def __init__(self, feat_size, epsilon=None, **kwargs):
    method build (line 151) | def build(self, input_shape):
    method call (line 156) | def call(self, inputs: tf.Tensor):
  class TFMobileBertEmbeddings (line 163) | class TFMobileBertEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 166) | def __init__(self, config, **kwargs):
    method build (line 184) | def build(self, input_shape):
    method call (line 208) | def call(self, input_ids=None, position_ids=None, token_type_ids=None,...
  class TFMobileBertSelfAttention (line 258) | class TFMobileBertSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 259) | def __init__(self, config, **kwargs):
    method transpose_for_scores (line 285) | def transpose_for_scores(self, x, batch_size):
    method call (line 290) | def call(
  class TFMobileBertSelfOutput (line 336) | class TFMobileBertSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 337) | def __init__(self, config, **kwargs):
    method call (line 349) | def call(self, hidden_states, residual_tensor, training=False):
  class TFMobileBertAttention (line 357) | class TFMobileBertAttention(tf.keras.layers.Layer):
    method __init__ (line 358) | def __init__(self, config, **kwargs):
    method prune_heads (line 363) | def prune_heads(self, heads):
    method call (line 366) | def call(
  class TFOutputBottleneck (line 386) | class TFOutputBottleneck(tf.keras.layers.Layer):
    method __init__ (line 387) | def __init__(self, config, **kwargs):
    method call (line 395) | def call(self, hidden_states, residual_tensor, training=False):
  class TFMobileBertOutput (line 402) | class TFMobileBertOutput(tf.keras.layers.Layer):
    method __init__ (line 403) | def __init__(self, config, **kwargs):
    method call (line 417) | def call(self, hidden_states, residual_tensor_1, residual_tensor_2, tr...
  class TFBottleneckLayer (line 428) | class TFBottleneckLayer(tf.keras.layers.Layer):
    method __init__ (line 429) | def __init__(self, config, **kwargs):
    method call (line 436) | def call(self, inputs):
  class TFBottleneck (line 442) | class TFBottleneck(tf.keras.layers.Layer):
    method __init__ (line 443) | def __init__(self, config, **kwargs):
    method call (line 451) | def call(self, hidden_states):
  class TFFFNOutput (line 478) | class TFFFNOutput(tf.keras.layers.Layer):
    method __init__ (line 479) | def __init__(self, config, **kwargs):
    method call (line 486) | def call(self, hidden_states, residual_tensor):
  class TFFFNLayer (line 492) | class TFFFNLayer(tf.keras.layers.Layer):
    method __init__ (line 493) | def __init__(self, config, **kwargs):
    method call (line 498) | def call(self, hidden_states):
  class TFMobileBertLayer (line 504) | class TFMobileBertLayer(tf.keras.layers.Layer):
    method __init__ (line 505) | def __init__(self, config, **kwargs):
    method call (line 518) | def call(self, hidden_states, attention_mask, head_mask, output_attent...
  class TFMobileBertEncoder (line 564) | class TFMobileBertEncoder(tf.keras.layers.Layer):
    method __init__ (line 565) | def __init__(self, config, **kwargs):
    method call (line 571) | def call(
  class TFMobileBertPooler (line 607) | class TFMobileBertPooler(tf.keras.layers.Layer):
    method __init__ (line 608) | def __init__(self, config, **kwargs):
    method call (line 619) | def call(self, hidden_states):
  class TFMobileBertPredictionHeadTransform (line 630) | class TFMobileBertPredictionHeadTransform(tf.keras.layers.Layer):
    method __init__ (line 631) | def __init__(self, config, **kwargs):
    method call (line 642) | def call(self, hidden_states):
  class TFMobileBertLMPredictionHead (line 649) | class TFMobileBertLMPredictionHead(tf.keras.layers.Layer):
    method __init__ (line 650) | def __init__(self, config, **kwargs):
    method build (line 655) | def build(self, input_shape):
    method get_output_embeddings (line 671) | def get_output_embeddings(self):
    method set_output_embeddings (line 674) | def set_output_embeddings(self, value):
    method get_bias (line 678) | def get_bias(self):
    method set_bias (line 681) | def set_bias(self, value):
    method call (line 685) | def call(self, hidden_states):
  class TFMobileBertMLMHead (line 692) | class TFMobileBertMLMHead(tf.keras.layers.Layer):
    method __init__ (line 693) | def __init__(self, config, **kwargs):
    method call (line 697) | def call(self, sequence_output):
  class TFMobileBertMainLayer (line 703) | class TFMobileBertMainLayer(tf.keras.layers.Layer):
    method __init__ (line 706) | def __init__(self, config, add_pooling_layer=True, **kwargs):
    method get_input_embeddings (line 719) | def get_input_embeddings(self):
    method set_input_embeddings (line 722) | def set_input_embeddings(self, value):
    method _prune_heads (line 726) | def _prune_heads(self, heads_to_prune):
    method call (line 734) | def call(
  class TFMobileBertPreTrainedModel (line 818) | class TFMobileBertPreTrainedModel(TFPreTrainedModel):
  class TFMobileBertForPreTrainingOutput (line 829) | class TFMobileBertForPreTrainingOutput(ModelOutput):
  class TFMobileBertModel (line 961) | class TFMobileBertModel(TFMobileBertPreTrainedModel):
    method __init__ (line 962) | def __init__(self, config, *inputs, **kwargs):
    method call (line 973) | def call(
  class TFMobileBertForPreTraining (line 1009) | class TFMobileBertForPreTraining(TFMobileBertPreTrainedModel, TFMobileBe...
    method __init__ (line 1010) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 1016) | def get_lm_head(self):
    method get_prefix_bias_name (line 1019) | def get_prefix_bias_name(self):
    method call (line 1026) | def call(
  class TFMobileBertForMaskedLM (line 1093) | class TFMobileBertForMaskedLM(TFMobileBertPreTrainedModel, TFMaskedLangu...
    method __init__ (line 1101) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 1107) | def get_lm_head(self):
    method get_prefix_bias_name (line 1110) | def get_prefix_bias_name(self):
    method call (line 1123) | def call(
  class TFMobileBertOnlyNSPHead (line 1172) | class TFMobileBertOnlyNSPHead(tf.keras.layers.Layer):
    method __init__ (line 1173) | def __init__(self, config, **kwargs):
    method call (line 1177) | def call(self, pooled_output):
  class TFMobileBertForNextSentencePrediction (line 1186) | class TFMobileBertForNextSentencePrediction(TFMobileBertPreTrainedModel,...
    method __init__ (line 1190) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1199) | def call(
  class TFMobileBertForSequenceClassification (line 1271) | class TFMobileBertForSequenceClassification(TFMobileBertPreTrainedModel,...
    method __init__ (line 1281) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1303) | def call(
  class TFMobileBertForQuestionAnswering (line 1361) | class TFMobileBertForQuestionAnswering(TFMobileBertPreTrainedModel, TFQu...
    method __init__ (line 1371) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1391) | def call(
  class TFMobileBertForMultipleChoice (line 1460) | class TFMobileBertForMultipleChoice(TFMobileBertPreTrainedModel, TFMulti...
    method __init__ (line 1470) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1488) | def call(
  class TFMobileBertForTokenClassification (line 1561) | class TFMobileBertForTokenClassification(TFMobileBertPreTrainedModel, TF...
    method __init__ (line 1572) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1594) | def call(

FILE: transformers/models/mobilebert/tokenization_mobilebert.py
  function load_vocab (line 43) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 55) | def whitespace_tokenize(text):
  class MobileBertTokenizer (line 65) | class MobileBertTokenizer(PreTrainedTokenizer):
    method __init__ (line 112) | def __init__(
    method do_lower_case (line 159) | def do_lower_case(self):
    method vocab_size (line 163) | def vocab_size(self):
    method get_vocab (line 166) | def get_vocab(self):
    method _tokenize (line 169) | def _tokenize(self, text):
    method _convert_token_to_id (line 182) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 186) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 190) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 195) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 220) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 248) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 277) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 299) | class BasicTokenizer(object):
    method __init__ (line 319) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 327) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 364) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 375) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 397) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 410) | def _is_chinese_char(self, cp):
    method _clean_text (line 434) | def _clean_text(self, text):
  class WordpieceTokenizer (line 449) | class WordpieceTokenizer(object):
    method __init__ (line 452) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 457) | def tokenize(self, text):

FILE: transformers/models/mobilebert/tokenization_mobilebert_fast.py
  class MobileBertTokenizerFast (line 46) | class MobileBertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 92) | def __init__(
    method build_inputs_with_special_tokens (line 134) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 158) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 187) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/mobilenet_v1/configuration_mobilenet_v1.py
  class MobileNetV1Config (line 36) | class MobileNetV1Config(PretrainedConfig):
    method __init__ (line 83) | def __init__(
  class MobileNetV1OnnxConfig (line 112) | class MobileNetV1OnnxConfig(OnnxConfig):
    method inputs (line 116) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 120) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 127) | def atol_for_validation(self) -> float:

FILE: transformers/models/mobilenet_v1/convert_original_tf_checkpoint_to_pytorch.py
  function get_mobilenet_v1_config (line 41) | def get_mobilenet_v1_config(model_name):
  function prepare_img (line 67) | def prepare_img():
  function convert_movilevit_checkpoint (line 74) | def convert_movilevit_checkpoint(model_name, checkpoint_path, pytorch_du...

FILE: transformers/models/mobilenet_v1/feature_extraction_mobilenet_v1.py
  class MobileNetV1FeatureExtractor (line 26) | class MobileNetV1FeatureExtractor(MobileNetV1ImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/mobilenet_v1/image_processing_mobilenet_v1.py
  class MobileNetV1ImageProcessor (line 46) | class MobileNetV1ImageProcessor(BaseImageProcessor):
    method __init__ (line 87) | def __init__(
    method resize (line 117) | def resize(
    method center_crop (line 145) | def center_crop(
    method rescale (line 167) | def rescale(
    method normalize (line 189) | def normalize(
    method preprocess (line 218) | def preprocess(

FILE: transformers/models/mobilenet_v1/modeling_mobilenet_v1.py
  function _build_tf_to_pytorch_map (line 53) | def _build_tf_to_pytorch_map(model, config, tf_weights=None):
  function load_tf_weights_in_mobilenet_v1 (line 100) | def load_tf_weights_in_mobilenet_v1(model, config, tf_checkpoint_path):
  function apply_tf_padding (line 156) | def apply_tf_padding(features: torch.Tensor, conv_layer: nn.Conv2d) -> t...
  class MobileNetV1ConvLayer (line 184) | class MobileNetV1ConvLayer(nn.Module):
    method __init__ (line 185) | def __init__(
    method forward (line 239) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileNetV1PreTrainedModel (line 250) | class MobileNetV1PreTrainedModel(PreTrainedModel):
    method _init_weights (line 262) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d]) -> None:
  class MobileNetV1Model (line 301) | class MobileNetV1Model(MobileNetV1PreTrainedModel):
    method __init__ (line 302) | def __init__(self, config: MobileNetV1Config, add_pooling_layer: bool ...
    method _prune_heads (line 352) | def _prune_heads(self, heads_to_prune):
    method forward (line 363) | def forward(
  class MobileNetV1ForImageClassification (line 411) | class MobileNetV1ForImageClassification(MobileNetV1PreTrainedModel):
    method __init__ (line 412) | def __init__(self, config: MobileNetV1Config) -> None:
    method forward (line 434) | def forward(

FILE: transformers/models/mobilenet_v2/configuration_mobilenet_v2.py
  class MobileNetV2Config (line 38) | class MobileNetV2Config(PretrainedConfig):
    method __init__ (line 101) | def __init__(
  class MobileNetV2OnnxConfig (line 142) | class MobileNetV2OnnxConfig(OnnxConfig):
    method inputs (line 146) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 150) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 157) | def atol_for_validation(self) -> float:

FILE: transformers/models/mobilenet_v2/convert_original_tf_checkpoint_to_pytorch.py
  function get_mobilenet_v2_config (line 42) | def get_mobilenet_v2_config(model_name):
  function prepare_img (line 79) | def prepare_img():
  function convert_movilevit_checkpoint (line 86) | def convert_movilevit_checkpoint(model_name, checkpoint_path, pytorch_du...

FILE: transformers/models/mobilenet_v2/feature_extraction_mobilenet_v2.py
  class MobileNetV2FeatureExtractor (line 26) | class MobileNetV2FeatureExtractor(MobileNetV2ImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/mobilenet_v2/image_processing_mobilenet_v2.py
  class MobileNetV2ImageProcessor (line 50) | class MobileNetV2ImageProcessor(BaseImageProcessor):
    method __init__ (line 91) | def __init__(
    method resize (line 121) | def resize(
    method center_crop (line 149) | def center_crop(
    method rescale (line 173) | def rescale(
    method normalize (line 195) | def normalize(
    method preprocess (line 224) | def preprocess(
    method post_process_semantic_segmentation (line 335) | def post_process_semantic_segmentation(self, outputs, target_sizes: Li...

FILE: transformers/models/mobilenet_v2/modeling_mobilenet_v2.py
  function _build_tf_to_pytorch_map (line 65) | def _build_tf_to_pytorch_map(model, config, tf_weights=None):
  function load_tf_weights_in_mobilenet_v2 (line 177) | def load_tf_weights_in_mobilenet_v2(model, config, tf_checkpoint_path):
  function make_divisible (line 234) | def make_divisible(value: int, divisor: int = 8, min_value: Optional[int...
  function apply_depth_multiplier (line 249) | def apply_depth_multiplier(config: MobileNetV2Config, channels: int) -> ...
  function apply_tf_padding (line 253) | def apply_tf_padding(features: torch.Tensor, conv_layer: nn.Conv2d) -> t...
  class MobileNetV2ConvLayer (line 288) | class MobileNetV2ConvLayer(nn.Module):
    method __init__ (line 289) | def __init__(
    method forward (line 346) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileNetV2InvertedResidual (line 357) | class MobileNetV2InvertedResidual(nn.Module):
    method __init__ (line 358) | def __init__(
    method forward (line 394) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileNetV2Stem (line 404) | class MobileNetV2Stem(nn.Module):
    method __init__ (line 405) | def __init__(self, config: MobileNetV2Config, in_channels: int, expand...
    method forward (line 442) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileNetV2PreTrainedModel (line 451) | class MobileNetV2PreTrainedModel(PreTrainedModel):
    method _init_weights (line 463) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d]) -> None:
  class MobileNetV2Model (line 502) | class MobileNetV2Model(MobileNetV2PreTrainedModel):
    method __init__ (line 503) | def __init__(self, config: MobileNetV2Config, add_pooling_layer: bool ...
    method _prune_heads (line 563) | def _prune_heads(self, heads_to_prune):
    method forward (line 574) | def forward(
  class MobileNetV2ForImageClassification (line 622) | class MobileNetV2ForImageClassification(MobileNetV2PreTrainedModel):
    method __init__ (line 623) | def __init__(self, config: MobileNetV2Config) -> None:
    method forward (line 645) | def forward(
  class MobileNetV2DeepLabV3Plus (line 700) | class MobileNetV2DeepLabV3Plus(nn.Module):
    method __init__ (line 706) | def __init__(self, config: MobileNetV2Config) -> None:
    method forward (line 756) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileNetV2ForSemanticSegmentation (line 781) | class MobileNetV2ForSemanticSegmentation(MobileNetV2PreTrainedModel):
    method __init__ (line 782) | def __init__(self, config: MobileNetV2Config) -> None:
    method forward (line 794) | def forward(

FILE: transformers/models/mobilevit/configuration_mobilevit.py
  class MobileViTConfig (line 46) | class MobileViTConfig(PretrainedConfig):
    method __init__ (line 116) | def __init__(
  class MobileViTOnnxConfig (line 168) | class MobileViTOnnxConfig(OnnxConfig):
    method inputs (line 172) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 176) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 183) | def atol_for_validation(self) -> float:

FILE: transformers/models/mobilevit/convert_mlcvnets_to_pytorch.py
  function get_mobilevit_config (line 40) | def get_mobilevit_config(mobilevit_name):
  function rename_key (line 74) | def rename_key(name, base_model=False):
  function convert_state_dict (line 153) | def convert_state_dict(orig_state_dict, model, base_model=False):
  function prepare_img (line 189) | def prepare_img():
  function convert_movilevit_checkpoint (line 196) | def convert_movilevit_checkpoint(mobilevit_name, checkpoint_path, pytorc...

FILE: transformers/models/mobilevit/feature_extraction_mobilevit.py
  class MobileViTFeatureExtractor (line 26) | class MobileViTFeatureExtractor(MobileViTImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/mobilevit/image_processing_mobilevit.py
  class MobileViTImageProcessor (line 51) | class MobileViTImageProcessor(BaseImageProcessor):
    method __init__ (line 85) | def __init__(
    method resize (line 112) | def resize(
    method center_crop (line 140) | def center_crop(
    method rescale (line 164) | def rescale(
    method flip_channel_order (line 184) | def flip_channel_order(
    method preprocess (line 198) | def preprocess(
    method post_process_semantic_segmentation (line 300) | def post_process_semantic_segmentation(self, outputs, target_sizes: Li...

FILE: transformers/models/mobilevit/modeling_mobilevit.py
  function make_divisible (line 73) | def make_divisible(value: int, divisor: int = 8, min_value: Optional[int...
  class MobileViTConvLayer (line 88) | class MobileViTConvLayer(nn.Module):
    method __init__ (line 89) | def __init__(
    method forward (line 143) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTInvertedResidual (line 152) | class MobileViTInvertedResidual(nn.Module):
    method __init__ (line 157) | def __init__(
    method forward (line 190) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTMobileNetLayer (line 200) | class MobileViTMobileNetLayer(nn.Module):
    method __init__ (line 201) | def __init__(
    method forward (line 217) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTSelfAttention (line 223) | class MobileViTSelfAttention(nn.Module):
    method __init__ (line 224) | def __init__(self, config: MobileViTConfig, hidden_size: int) -> None:
    method transpose_for_scores (line 243) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 248) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTSelfOutput (line 274) | class MobileViTSelfOutput(nn.Module):
    method __init__ (line 275) | def __init__(self, config: MobileViTConfig, hidden_size: int) -> None:
    method forward (line 280) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTAttention (line 286) | class MobileViTAttention(nn.Module):
    method __init__ (line 287) | def __init__(self, config: MobileViTConfig, hidden_size: int) -> None:
    method prune_heads (line 293) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 311) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTIntermediate (line 317) | class MobileViTIntermediate(nn.Module):
    method __init__ (line 318) | def __init__(self, config: MobileViTConfig, hidden_size: int, intermed...
    method forward (line 326) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTOutput (line 332) | class MobileViTOutput(nn.Module):
    method __init__ (line 333) | def __init__(self, config: MobileViTConfig, hidden_size: int, intermed...
    method forward (line 338) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class MobileViTTransformerLayer (line 345) | class MobileViTTransformerLayer(nn.Module):
    method __init__ (line 346) | def __init__(self, config: MobileViTConfig, hidden_size: int, intermed...
    method forward (line 354) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTTransformer (line 364) | class MobileViTTransformer(nn.Module):
    method __init__ (line 365) | def __init__(self, config: MobileViTConfig, hidden_size: int, num_stag...
    method forward (line 377) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTLayer (line 383) | class MobileViTLayer(nn.Module):
    method __init__ (line 388) | def __init__(
    method unfolding (line 446) | def unfolding(self, features: torch.Tensor) -> Tuple[torch.Tensor, Dict]:
    method folding (line 489) | def folding(self, patches: torch.Tensor, info_dict: Dict) -> torch.Ten...
    method forward (line 518) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTEncoder (line 544) | class MobileViTEncoder(nn.Module):
    method __init__ (line 545) | def __init__(self, config: MobileViTConfig) -> None:
    method forward (line 619) | def forward(
  class MobileViTPreTrainedModel (line 652) | class MobileViTPreTrainedModel(PreTrainedModel):
    method _init_weights (line 663) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 675) | def _set_gradient_checkpointing(self, module, value=False):
  class MobileViTModel (line 708) | class MobileViTModel(MobileViTPreTrainedModel):
    method __init__ (line 709) | def __init__(self, config: MobileViTConfig, expand_output: bool = True):
    method _prune_heads (line 735) | def _prune_heads(self, heads_to_prune):
    method forward (line 753) | def forward(
  class MobileViTForImageClassification (line 802) | class MobileViTForImageClassification(MobileViTPreTrainedModel):
    method __init__ (line 803) | def __init__(self, config: MobileViTConfig) -> None:
    method forward (line 825) | def forward(
  class MobileViTASPPPooling (line 880) | class MobileViTASPPPooling(nn.Module):
    method __init__ (line 881) | def __init__(self, config: MobileViTConfig, in_channels: int, out_chan...
    method forward (line 896) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTASPP (line 904) | class MobileViTASPP(nn.Module):
    method __init__ (line 909) | def __init__(self, config: MobileViTConfig) -> None:
    method forward (line 952) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTDeepLabV3 (line 963) | class MobileViTDeepLabV3(nn.Module):
    method __init__ (line 968) | def __init__(self, config: MobileViTConfig) -> None:
    method forward (line 984) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTForSemanticSegmentation (line 997) | class MobileViTForSemanticSegmentation(MobileViTPreTrainedModel):
    method __init__ (line 998) | def __init__(self, config: MobileViTConfig) -> None:
    method forward (line 1010) | def forward(

FILE: transformers/models/mobilevit/modeling_tf_mobilevit.py
  function make_divisible (line 69) | def make_divisible(value: int, divisor: int = 8, min_value: Optional[int...
  class TFMobileViTConvLayer (line 84) | class TFMobileViTConvLayer(tf.keras.layers.Layer):
    method __init__ (line 85) | def __init__(
    method call (line 136) | def call(self, features: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFMobileViTInvertedResidual (line 146) | class TFMobileViTInvertedResidual(tf.keras.layers.Layer):
    method __init__ (line 151) | def __init__(
    method call (line 184) | def call(self, features: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFMobileViTMobileNetLayer (line 194) | class TFMobileViTMobileNetLayer(tf.keras.layers.Layer):
    method __init__ (line 195) | def __init__(
    method call (line 218) | def call(self, features: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFMobileViTSelfAttention (line 224) | class TFMobileViTSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 225) | def __init__(self, config: MobileViTConfig, hidden_size: int, **kwargs...
    method transpose_for_scores (line 246) | def transpose_for_scores(self, x: tf.Tensor) -> tf.Tensor:
    method call (line 251) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFMobileViTSelfOutput (line 276) | class TFMobileViTSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 277) | def __init__(self, config: MobileViTConfig, hidden_size: int, **kwargs...
    method call (line 282) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFMobileViTAttention (line 288) | class TFMobileViTAttention(tf.keras.layers.Layer):
    method __init__ (line 289) | def __init__(self, config: MobileViTConfig, hidden_size: int, **kwargs...
    method prune_heads (line 294) | def prune_heads(self, heads):
    method call (line 297) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFMobileViTIntermediate (line 303) | class TFMobileViTIntermediate(tf.keras.layers.Layer):
    method __init__ (line 304) | def __init__(self, config: MobileViTConfig, hidden_size: int, intermed...
    method call (line 312) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFMobileViTOutput (line 318) | class TFMobileViTOutput(tf.keras.layers.Layer):
    method __init__ (line 319) | def __init__(self, config: MobileViTConfig, hidden_size: int, intermed...
    method call (line 324) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFMobileViTTransformerLayer (line 331) | class TFMobileViTTransformerLayer(tf.keras.layers.Layer):
    method __init__ (line 332) | def __init__(self, config: MobileViTConfig, hidden_size: int, intermed...
    method call (line 344) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFMobileViTTransformer (line 354) | class TFMobileViTTransformer(tf.keras.layers.Layer):
    method __init__ (line 355) | def __init__(self, config: MobileViTConfig, hidden_size: int, num_stag...
    method call (line 368) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFMobileViTLayer (line 374) | class TFMobileViTLayer(tf.keras.layers.Layer):
    method __init__ (line 379) | def __init__(
    method unfolding (line 434) | def unfolding(self, features: tf.Tensor) -> Tuple[tf.Tensor, Dict]:
    method folding (line 478) | def folding(self, patches: tf.Tensor, info_dict: Dict) -> tf.Tensor:
    method call (line 506) | def call(self, features: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFMobileViTEncoder (line 532) | class TFMobileViTEncoder(tf.keras.layers.Layer):
    method __init__ (line 533) | def __init__(self, config: MobileViTConfig, **kwargs) -> None:
    method call (line 611) | def call(
  class TFMobileViTMainLayer (line 633) | class TFMobileViTMainLayer(tf.keras.layers.Layer):
    method __init__ (line 636) | def __init__(self, config: MobileViTConfig, expand_output: bool = True...
    method _prune_heads (line 658) | def _prune_heads(self, heads_to_prune):
    method call (line 666) | def call(
  class TFMobileViTPreTrainedModel (line 728) | class TFMobileViTPreTrainedModel(TFPreTrainedModel):
  class TFMobileViTModel (line 800) | class TFMobileViTModel(TFMobileViTPreTrainedModel):
    method __init__ (line 801) | def __init__(self, config: MobileViTConfig, expand_output: bool = True...
    method call (line 817) | def call(
  class TFMobileViTForImageClassification (line 835) | class TFMobileViTForImageClassification(TFMobileViTPreTrainedModel, TFSe...
    method __init__ (line 836) | def __init__(self, config: MobileViTConfig, *inputs, **kwargs) -> None:
    method call (line 856) | def call(
  class TFMobileViTASPPPooling (line 888) | class TFMobileViTASPPPooling(tf.keras.layers.Layer):
    method __init__ (line 889) | def __init__(self, config: MobileViTConfig, out_channels: int, **kwarg...
    method call (line 904) | def call(self, features: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFMobileViTASPP (line 912) | class TFMobileViTASPP(tf.keras.layers.Layer):
    method __init__ (line 917) | def __init__(self, config: MobileViTConfig, **kwargs) -> None:
    method call (line 963) | def call(self, features: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFMobileViTDeepLabV3 (line 977) | class TFMobileViTDeepLabV3(tf.keras.layers.Layer):
    method __init__ (line 982) | def __init__(self, config: MobileViTConfig, **kwargs) -> None:
    method call (line 998) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFMobileViTForSemanticSegmentation (line 1011) | class TFMobileViTForSemanticSegmentation(TFMobileViTPreTrainedModel):
    method __init__ (line 1012) | def __init__(self, config: MobileViTConfig, **kwargs) -> None:
    method hf_compute_loss (line 1019) | def hf_compute_loss(self, logits, labels):
    method call (line 1042) | def call(

FILE: transformers/models/mobilevitv2/configuration_mobilevitv2.py
  class MobileViTV2Config (line 34) | class MobileViTV2Config(PretrainedConfig):
    method __init__ (line 102) | def __init__(
  class MobileViTV2OnnxConfig (line 152) | class MobileViTV2OnnxConfig(OnnxConfig):
    method inputs (line 156) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 160) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 167) | def atol_for_validation(self) -> float:

FILE: transformers/models/mobilevitv2/convert_mlcvnets_to_pytorch.py
  function load_orig_config_file (line 42) | def load_orig_config_file(orig_cfg_file):
  function get_mobilevitv2_config (line 68) | def get_mobilevitv2_config(task_name, orig_cfg_file):
  function rename_key (line 126) | def rename_key(dct, old, new):
  function create_rename_keys (line 131) | def create_rename_keys(state_dict, base_model=False):
  function remove_unused_keys (line 215) | def remove_unused_keys(state_dict):
  function prepare_img (line 226) | def prepare_img():
  function convert_mobilevitv2_checkpoint (line 234) | def convert_mobilevitv2_checkpoint(task_name, checkpoint_path, orig_conf...

FILE: transformers/models/mobilevitv2/modeling_mobilevitv2.py
  function make_divisible (line 67) | def make_divisible(value: int, divisor: int = 8, min_value: Optional[int...
  function clip (line 82) | def clip(value: float, min_val: float = float("-inf"), max_val: float = ...
  class MobileViTV2ConvLayer (line 87) | class MobileViTV2ConvLayer(nn.Module):
    method __init__ (line 88) | def __init__(
    method forward (line 142) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTV2InvertedResidual (line 152) | class MobileViTV2InvertedResidual(nn.Module):
    method __init__ (line 157) | def __init__(
    method forward (line 190) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTV2MobileNetLayer (line 201) | class MobileViTV2MobileNetLayer(nn.Module):
    method __init__ (line 202) | def __init__(
    method forward (line 218) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTV2LinearSelfAttention (line 224) | class MobileViTV2LinearSelfAttention(nn.Module):
    method __init__ (line 236) | def __init__(self, config: MobileViTV2Config, embed_dim: int) -> None:
    method forward (line 261) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTV2FFN (line 287) | class MobileViTV2FFN(nn.Module):
    method __init__ (line 288) | def __init__(
    method forward (line 320) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTV2TransformerLayer (line 328) | class MobileViTV2TransformerLayer(nn.Module):
    method __init__ (line 329) | def __init__(
    method forward (line 343) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTV2Transformer (line 355) | class MobileViTV2Transformer(nn.Module):
    method __init__ (line 356) | def __init__(self, config: MobileViTV2Config, n_layers: int, d_model: ...
    method forward (line 373) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTV2Layer (line 379) | class MobileViTV2Layer(nn.Module):
    method __init__ (line 384) | def __init__(
    method unfolding (line 445) | def unfolding(self, feature_map: torch.Tensor) -> Tuple[torch.Tensor, ...
    method folding (line 456) | def folding(self, patches: torch.Tensor, output_size: Tuple[int, int])...
    method forward (line 469) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTV2Encoder (line 493) | class MobileViTV2Encoder(nn.Module):
    method __init__ (line 494) | def __init__(self, config: MobileViTV2Config) -> None:
    method forward (line 575) | def forward(
  class MobileViTV2PreTrainedModel (line 609) | class MobileViTV2PreTrainedModel(PreTrainedModel):
    method _init_weights (line 620) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 632) | def _set_gradient_checkpointing(self, module, value=False):
  class MobileViTV2Model (line 665) | class MobileViTV2Model(MobileViTV2PreTrainedModel):
    method __init__ (line 666) | def __init__(self, config: MobileViTV2Config, expand_output: bool = Tr...
    method _prune_heads (line 689) | def _prune_heads(self, heads_to_prune):
    method forward (line 707) | def forward(
  class MobileViTV2ForImageClassification (line 756) | class MobileViTV2ForImageClassification(MobileViTV2PreTrainedModel):
    method __init__ (line 757) | def __init__(self, config: MobileViTV2Config) -> None:
    method forward (line 781) | def forward(
  class MobileViTV2ASPPPooling (line 837) | class MobileViTV2ASPPPooling(nn.Module):
    method __init__ (line 838) | def __init__(self, config: MobileViTV2Config, in_channels: int, out_ch...
    method forward (line 853) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTV2ASPP (line 861) | class MobileViTV2ASPP(nn.Module):
    method __init__ (line 866) | def __init__(self, config: MobileViTV2Config) -> None:
    method forward (line 910) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class MobileViTV2DeepLabV3 (line 922) | class MobileViTV2DeepLabV3(nn.Module):
    method __init__ (line 927) | def __init__(self, config: MobileViTV2Config) -> None:
    method forward (line 943) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MobileViTV2ForSemanticSegmentation (line 957) | class MobileViTV2ForSemanticSegmentation(MobileViTV2PreTrainedModel):
    method __init__ (line 958) | def __init__(self, config: MobileViTV2Config) -> None:
    method forward (line 970) | def forward(

FILE: transformers/models/mpnet/configuration_mpnet.py
  class MPNetConfig (line 29) | class MPNetConfig(PretrainedConfig):
    method __init__ (line 84) | def __init__(

FILE: transformers/models/mpnet/modeling_mpnet.py
  class MPNetPreTrainedModel (line 53) | class MPNetPreTrainedModel(PreTrainedModel):
    method _init_weights (line 58) | def _init_weights(self, module):
  class MPNetEmbeddings (line 75) | class MPNetEmbeddings(nn.Module):
    method __init__ (line 76) | def __init__(self, config):
    method forward (line 88) | def forward(self, input_ids=None, position_ids=None, inputs_embeds=Non...
    method create_position_ids_from_inputs_embeds (line 114) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class MPNetSelfAttention (line 132) | class MPNetSelfAttention(nn.Module):
    method __init__ (line 133) | def __init__(self, config):
    method transpose_for_scores (line 152) | def transpose_for_scores(self, x):
    method forward (line 157) | def forward(
  class MPNetAttention (line 205) | class MPNetAttention(nn.Module):
    method __init__ (line 206) | def __init__(self, config):
    method prune_heads (line 214) | def prune_heads(self, heads):
    method forward (line 230) | def forward(
  class MPNetIntermediate (line 252) | class MPNetIntermediate(nn.Module):
    method __init__ (line 253) | def __init__(self, config):
    method forward (line 261) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MPNetOutput (line 268) | class MPNetOutput(nn.Module):
    method __init__ (line 269) | def __init__(self, config):
    method forward (line 275) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class MPNetLayer (line 282) | class MPNetLayer(nn.Module):
    method __init__ (line 283) | def __init__(self, config):
    method forward (line 289) | def forward(
  class MPNetEncoder (line 314) | class MPNetEncoder(nn.Module):
    method __init__ (line 315) | def __init__(self, config):
    method forward (line 322) | def forward(
    method compute_position_bias (line 364) | def compute_position_bias(self, x, position_ids=None, num_buckets=32):
    method relative_position_bucket (line 383) | def relative_position_bucket(relative_position, num_buckets=32, max_di...
  class MPNetPooler (line 404) | class MPNetPooler(nn.Module):
    method __init__ (line 405) | def __init__(self, config):
    method forward (line 410) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MPNetModel (line 481) | class MPNetModel(MPNetPreTrainedModel):
    method __init__ (line 484) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 495) | def get_input_embeddings(self):
    method set_input_embeddings (line 498) | def set_input_embeddings(self, value):
    method _prune_heads (line 501) | def _prune_heads(self, heads_to_prune):
    method forward (line 515) | def forward(
  class MPNetForMaskedLM (line 572) | class MPNetForMaskedLM(MPNetPreTrainedModel):
    method __init__ (line 576) | def __init__(self, config):
    method get_output_embeddings (line 585) | def get_output_embeddings(self):
    method set_output_embeddings (line 588) | def set_output_embeddings(self, new_embeddings):
    method forward (line 597) | def forward(
  class MPNetLMHead (line 648) | class MPNetLMHead(nn.Module):
    method __init__ (line 651) | def __init__(self, config):
    method forward (line 662) | def forward(self, features, **kwargs):
  class MPNetForSequenceClassification (line 680) | class MPNetForSequenceClassification(MPNetPreTrainedModel):
    method __init__ (line 683) | def __init__(self, config):
    method forward (line 699) | def forward(
  class MPNetForMultipleChoice (line 774) | class MPNetForMultipleChoice(MPNetPreTrainedModel):
    method __init__ (line 777) | def __init__(self, config):
    method forward (line 793) | def forward(
  class MPNetForTokenClassification (line 864) | class MPNetForTokenClassification(MPNetPreTrainedModel):
    method __init__ (line 868) | def __init__(self, config):
    method forward (line 885) | def forward(
  class MPNetClassificationHead (line 937) | class MPNetClassificationHead(nn.Module):
    method __init__ (line 940) | def __init__(self, config):
    method forward (line 946) | def forward(self, features, **kwargs):
  class MPNetForQuestionAnswering (line 963) | class MPNetForQuestionAnswering(MPNetPreTrainedModel):
    method __init__ (line 967) | def __init__(self, config):
    method forward (line 983) | def forward(
  function create_position_ids_from_input_ids (line 1057) | def create_position_ids_from_input_ids(input_ids, padding_idx):

FILE: transformers/models/mpnet/modeling_tf_mpnet.py
  class TFMPNetPreTrainedModel (line 70) | class TFMPNetPreTrainedModel(TFPreTrainedModel):
  class TFMPNetEmbeddings (line 80) | class TFMPNetEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 83) | def __init__(self, config, **kwargs):
    method build (line 94) | def build(self, input_shape: tf.TensorShape):
    method create_position_ids_from_input_ids (line 111) | def create_position_ids_from_input_ids(self, input_ids):
    method call (line 125) | def call(self, input_ids=None, position_ids=None, inputs_embeds=None, ...
  class TFMPNetPooler (line 158) | class TFMPNetPooler(tf.keras.layers.Layer):
    method __init__ (line 159) | def __init__(self, config: MPNetConfig, **kwargs):
    method call (line 169) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFMPNetSelfAttention (line 178) | class TFMPNetSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 179) | def __init__(self, config, **kwargs):
    method transpose_for_scores (line 207) | def transpose_for_scores(self, x, batch_size):
    method call (line 213) | def call(self, hidden_states, attention_mask, head_mask, output_attent...
  class TFMPNetAttention (line 251) | class TFMPNetAttention(tf.keras.layers.Layer):
    method __init__ (line 252) | def __init__(self, config, **kwargs):
    method prune_heads (line 259) | def prune_heads(self, heads):
    method call (line 262) | def call(self, input_tensor, attention_mask, head_mask, output_attenti...
  class TFMPNetIntermediate (line 272) | class TFMPNetIntermediate(tf.keras.layers.Layer):
    method __init__ (line 273) | def __init__(self, config: MPNetConfig, **kwargs):
    method call (line 285) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFMPNetOutput (line 293) | class TFMPNetOutput(tf.keras.layers.Layer):
    method __init__ (line 294) | def __init__(self, config: MPNetConfig, **kwargs):
    method call (line 303) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFMPNetLayer (line 311) | class TFMPNetLayer(tf.keras.layers.Layer):
    method __init__ (line 312) | def __init__(self, config, **kwargs):
    method call (line 319) | def call(self, hidden_states, attention_mask, head_mask, output_attent...
  class TFMPNetEncoder (line 333) | class TFMPNetEncoder(tf.keras.layers.Layer):
    method __init__ (line 334) | def __init__(self, config, **kwargs):
    method build (line 347) | def build(self, input_shape):
    method call (line 357) | def call(
    method _relative_position_bucket (line 400) | def _relative_position_bucket(relative_position, num_buckets=32, max_d...
    method compute_position_bias (line 421) | def compute_position_bias(self, x, position_ids=None):
  class TFMPNetMainLayer (line 445) | class TFMPNetMainLayer(tf.keras.layers.Layer):
    method __init__ (line 448) | def __init__(self, config, **kwargs):
    method get_input_embeddings (line 463) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 467) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 472) | def _prune_heads(self, heads_to_prune):
    method call (line 480) | def call(
  class TFMPNetModel (line 659) | class TFMPNetModel(TFMPNetPreTrainedModel):
    method __init__ (line 660) | def __init__(self, config, *inputs, **kwargs):
    method call (line 671) | def call(
  class TFMPNetLMHead (line 697) | class TFMPNetLMHead(tf.keras.layers.Layer):
    method __init__ (line 700) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 715) | def build(self, input_shape):
    method get_output_embeddings (line 720) | def get_output_embeddings(self):
    method set_output_embeddings (line 723) | def set_output_embeddings(self, value):
    method get_bias (line 727) | def get_bias(self):
    method set_bias (line 730) | def set_bias(self, value):
    method call (line 734) | def call(self, hidden_states):
  class TFMPNetForMaskedLM (line 750) | class TFMPNetForMaskedLM(TFMPNetPreTrainedModel, TFMaskedLanguageModelin...
    method __init__ (line 753) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 759) | def get_lm_head(self):
    method get_prefix_bias_name (line 762) | def get_prefix_bias_name(self):
    method call (line 773) | def call(
  class TFMPNetClassificationHead (line 820) | class TFMPNetClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 823) | def __init__(self, config, **kwargs):
    method call (line 836) | def call(self, features, training=False):
  class TFMPNetForSequenceClassification (line 852) | class TFMPNetForSequenceClassification(TFMPNetPreTrainedModel, TFSequenc...
    method __init__ (line 855) | def __init__(self, config, *inputs, **kwargs):
    method call (line 869) | def call(
  class TFMPNetForMultipleChoice (line 924) | class TFMPNetForMultipleChoice(TFMPNetPreTrainedModel, TFMultipleChoiceL...
    method __init__ (line 925) | def __init__(self, config, *inputs, **kwargs):
    method call (line 941) | def call(
  class TFMPNetForTokenClassification (line 1010) | class TFMPNetForTokenClassification(TFMPNetPreTrainedModel, TFTokenClass...
    method __init__ (line 1013) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1030) | def call(
  class TFMPNetForQuestionAnswering (line 1084) | class TFMPNetForQuestionAnswering(TFMPNetPreTrainedModel, TFQuestionAnsw...
    method __init__ (line 1087) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1103) | def call(

FILE: transformers/models/mpnet/tokenization_mpnet.py
  function load_vocab (line 46) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 57) | def whitespace_tokenize(text):
  class MPNetTokenizer (line 66) | class MPNetTokenizer(PreTrainedTokenizer):
    method __init__ (line 133) | def __init__(
    method do_lower_case (line 194) | def do_lower_case(self):
    method vocab_size (line 198) | def vocab_size(self):
    method get_vocab (line 201) | def get_vocab(self):
    method _tokenize (line 204) | def _tokenize(self, text):
    method _convert_token_to_id (line 217) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 221) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 225) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 230) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 255) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 282) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 305) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 327) | class BasicTokenizer(object):
    method __init__ (line 347) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 355) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 392) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 403) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 425) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 438) | def _is_chinese_char(self, cp):
    method _clean_text (line 462) | def _clean_text(self, text):
  class WordpieceTokenizer (line 477) | class WordpieceTokenizer(object):
    method __init__ (line 480) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 485) | def tokenize(self, text):

FILE: transformers/models/mpnet/tokenization_mpnet_fast.py
  class MPNetTokenizerFast (line 51) | class MPNetTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 113) | def __init__(
    method mask_token (line 158) | def mask_token(self) -> str:
    method mask_token (line 173) | def mask_token(self, value):
    method build_inputs_with_special_tokens (line 184) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 191) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 214) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/mt5/configuration_mt5.py
  class MT5Config (line 26) | class MT5Config(PretrainedConfig):
    method __init__ (line 72) | def __init__(
    method hidden_size (line 138) | def hidden_size(self):
    method num_attention_heads (line 142) | def num_attention_heads(self):
    method num_hidden_layers (line 146) | def num_hidden_layers(self):
  class MT5OnnxConfig (line 150) | class MT5OnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 153) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method default_onnx_opset (line 173) | def default_onnx_opset(self) -> int:
    method atol_for_validation (line 177) | def atol_for_validation(self) -> float:

FILE: transformers/models/mt5/modeling_flax_mt5.py
  function shift_tokens_right (line 30) | def shift_tokens_right(input_ids: jnp.array, pad_token_id: int, decoder_...
  class FlaxMT5Model (line 42) | class FlaxMT5Model(FlaxT5Model):
  class FlaxMT5EncoderModel (line 68) | class FlaxMT5EncoderModel(FlaxT5EncoderModel):
  class FlaxMT5ForConditionalGeneration (line 94) | class FlaxMT5ForConditionalGeneration(FlaxT5ForConditionalGeneration):

FILE: transformers/models/mt5/modeling_mt5.py
  class MT5LayerNorm (line 110) | class MT5LayerNorm(nn.Module):
    method __init__ (line 111) | def __init__(self, hidden_size, eps=1e-6):
    method forward (line 119) | def forward(self, hidden_states):
  class MT5DenseActDense (line 136) | class MT5DenseActDense(nn.Module):
    method __init__ (line 137) | def __init__(self, config: MT5Config):
    method forward (line 144) | def forward(self, hidden_states):
  class MT5DenseGatedActDense (line 159) | class MT5DenseGatedActDense(nn.Module):
    method __init__ (line 160) | def __init__(self, config: MT5Config):
    method forward (line 168) | def forward(self, hidden_states):
  class MT5LayerFF (line 189) | class MT5LayerFF(nn.Module):
    method __init__ (line 190) | def __init__(self, config: MT5Config):
    method forward (line 200) | def forward(self, hidden_states):
  class MT5Attention (line 208) | class MT5Attention(nn.Module):
    method __init__ (line 209) | def __init__(self, config: MT5Config, has_relative_attention_bias=False):
    method prune_heads (line 232) | def prune_heads(self, heads):
    method _relative_position_bucket (line 249) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 296) | def compute_bias(self, query_length, key_length, device=None):
    method forward (line 313) | def forward(
  class MT5LayerSelfAttention (line 444) | class MT5LayerSelfAttention(nn.Module):
    method __init__ (line 445) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 451) | def forward(
  class MT5LayerCrossAttention (line 477) | class MT5LayerCrossAttention(nn.Module):
    method __init__ (line 478) | def __init__(self, config):
    method forward (line 484) | def forward(
  class MT5Block (line 514) | class MT5Block(nn.Module):
    method __init__ (line 515) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 525) | def forward(
  function load_tf_weights_in_mt5 (line 638) | def load_tf_weights_in_mt5(model, config, tf_checkpoint_path):
  class MT5PreTrainedModel (line 745) | class MT5PreTrainedModel(PreTrainedModel):
    method dummy_inputs (line 760) | def dummy_inputs(self):
    method _init_weights (line 770) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 814) | def _set_gradient_checkpointing(self, module, value=False):
    method _shift_right (line 818) | def _shift_right(self, input_ids):
  class MT5Stack (line 847) | class MT5Stack(MT5PreTrainedModel):
    method __init__ (line 848) | def __init__(self, config, embed_tokens=None):
    method parallelize (line 868) | def parallelize(self, device_map=None):
    method deparallelize (line 896) | def deparallelize(self):
    method get_input_embeddings (line 911) | def get_input_embeddings(self):
    method set_input_embeddings (line 914) | def set_input_embeddings(self, new_embeddings):
    method forward (line 917) | def forward(
  class MT5Model (line 1296) | class MT5Model(MT5PreTrainedModel):
    method __init__ (line 1329) | def __init__(self, config: MT5Config):
    method parallelize (line 1354) | def parallelize(self, device_map=None):
    method deparallelize (line 1374) | def deparallelize(self):
    method get_input_embeddings (line 1388) | def get_input_embeddings(self):
    method set_input_embeddings (line 1392) | def set_input_embeddings(self, new_embeddings):
    method get_encoder (line 1398) | def get_encoder(self):
    method get_decoder (line 1402) | def get_decoder(self):
    method _prune_heads (line 1406) | def _prune_heads(self, heads_to_prune):
    method forward (line 1417) | def forward(
  class MT5ForConditionalGeneration (line 1531) | class MT5ForConditionalGeneration(MT5PreTrainedModel):
    method __init__ (line 1561) | def __init__(self, config: MT5Config):
    method parallelize (line 1590) | def parallelize(self, device_map=None):
    method deparallelize (line 1611) | def deparallelize(self):
    method get_input_embeddings (line 1626) | def get_input_embeddings(self):
    method set_input_embeddings (line 1630) | def set_input_embeddings(self, new_embeddings):
    method set_output_embeddings (line 1636) | def set_output_embeddings(self, new_embeddings):
    method get_output_embeddings (line 1640) | def get_output_embeddings(self):
    method get_encoder (line 1644) | def get_encoder(self):
    method get_decoder (line 1648) | def get_decoder(self):
    method forward (line 1654) | def forward(
    method prepare_inputs_for_generation (line 1808) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1838) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 1842) | def _reorder_cache(self, past_key_values, beam_idx):
  class MT5EncoderModel (line 1877) | class MT5EncoderModel(MT5PreTrainedModel):
    method __init__ (line 1903) | def __init__(self, config: MT5Config):
    method parallelize (line 1921) | def parallelize(self, device_map=None):
    method deparallelize (line 1940) | def deparallelize(self):
    method get_input_embeddings (line 1952) | def get_input_embeddings(self):
    method set_input_embeddings (line 1956) | def set_input_embeddings(self, new_embeddings):
    method get_encoder (line 1961) | def get_encoder(self):
    method _prune_heads (line 1965) | def _prune_heads(self, heads_to_prune):
    method forward (line 1976) | def forward(

FILE: transformers/models/mt5/modeling_tf_mt5.py
  class TFMT5Model (line 27) | class TFMT5Model(TFT5Model):
  class TFMT5ForConditionalGeneration (line 51) | class TFMT5ForConditionalGeneration(TFT5ForConditionalGeneration):
  class TFMT5EncoderModel (line 75) | class TFMT5EncoderModel(TFT5EncoderModel):

FILE: transformers/models/mvp/configuration_mvp.py
  class MvpConfig (line 29) | class MvpConfig(PretrainedConfig):
    method __init__ (line 111) | def __init__(

FILE: transformers/models/mvp/modeling_mvp.py
  function shift_tokens_right (line 77) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 94) | def _make_causal_mask(
  function _expand_mask (line 112) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class MvpLearnedPositionalEmbedding (line 127) | class MvpLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 132) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 138) | def forward(self, input_ids: torch.Tensor, past_key_values_length: int...
  class MvpAttention (line 149) | class MvpAttention(nn.Module):
    method __init__ (line 152) | def __init__(
    method _shape (line 179) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 182) | def forward(
  class MvpEncoderLayer (line 304) | class MvpEncoderLayer(nn.Module):
    method __init__ (line 305) | def __init__(self, config: MvpConfig):
    method forward (line 321) | def forward(
  class MvpDecoderLayer (line 376) | class MvpDecoderLayer(nn.Module):
    method __init__ (line 377) | def __init__(self, config: MvpConfig):
    method forward (line 403) | def forward(
  class MvpClassificationHead (line 502) | class MvpClassificationHead(nn.Module):
    method __init__ (line 505) | def __init__(
    method forward (line 517) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class MvpPrompt (line 526) | class MvpPrompt(nn.Module):
    method __init__ (line 529) | def __init__(self, config, num_layers, num_heads):
    method forward (line 543) | def forward(self, prompt_ids: torch.Tensor) -> Tuple[torch.Tensor]:
  class MvpPreTrainedModel (line 551) | class MvpPreTrainedModel(PreTrainedModel):
    method _init_weights (line 557) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 568) | def _set_gradient_checkpointing(self, module, value=False):
    method dummy_inputs (line 573) | def dummy_inputs(self):
  class MvpEncoder (line 790) | class MvpEncoder(MvpPreTrainedModel):
    method __init__ (line 801) | def __init__(
    method get_input_embeddings (line 839) | def get_input_embeddings(self):
    method set_input_embeddings (line 842) | def set_input_embeddings(self, value):
    method forward (line 845) | def forward(
  class MvpDecoder (line 987) | class MvpDecoder(MvpPreTrainedModel):
    method __init__ (line 997) | def __init__(
    method get_input_embeddings (line 1037) | def get_input_embeddings(self):
    method set_input_embeddings (line 1040) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 1043) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 1064) | def forward(
  class MvpModel (line 1297) | class MvpModel(MvpPreTrainedModel):
    method __init__ (line 1301) | def __init__(self, config: MvpConfig):
    method get_input_embeddings (line 1314) | def get_input_embeddings(self):
    method set_input_embeddings (line 1317) | def set_input_embeddings(self, value):
    method get_encoder (line 1322) | def get_encoder(self):
    method get_decoder (line 1325) | def get_decoder(self):
    method set_lightweight_tuning (line 1328) | def set_lightweight_tuning(self):
    method forward (line 1343) | def forward(
  class MvpForConditionalGeneration (line 1434) | class MvpForConditionalGeneration(MvpPreTrainedModel):
    method __init__ (line 1437) | def __init__(self, config: MvpConfig):
    method get_encoder (line 1446) | def get_encoder(self):
    method get_decoder (line 1449) | def get_decoder(self):
    method resize_token_embeddings (line 1452) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_final_logits_bias (line 1457) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_output_embeddings (line 1466) | def get_output_embeddings(self):
    method set_output_embeddings (line 1469) | def set_output_embeddings(self, new_embeddings):
    method set_lightweight_tuning (line 1472) | def set_lightweight_tuning(self):
    method forward (line 1479) | def forward(
    method prepare_inputs_for_generation (line 1557) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1585) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 1589) | def _reorder_cache(past_key_values, beam_idx):
  class MvpForSequenceClassification (line 1606) | class MvpForSequenceClassification(MvpPreTrainedModel):
    method __init__ (line 1610) | def __init__(self, config: MvpConfig, **kwargs):
    method set_lightweight_tuning (line 1623) | def set_lightweight_tuning(self):
    method forward (line 1629) | def forward(
  class MvpForQuestionAnswering (line 1734) | class MvpForQuestionAnswering(MvpPreTrainedModel):
    method __init__ (line 1738) | def __init__(self, config):
    method set_lightweight_tuning (line 1750) | def set_lightweight_tuning(self):
    method forward (line 1756) | def forward(
  class MvpDecoderWrapper (line 1852) | class MvpDecoderWrapper(MvpPreTrainedModel):
    method __init__ (line 1858) | def __init__(self, config):
    method forward (line 1862) | def forward(self, *args, **kwargs):
  class MvpForCausalLM (line 1866) | class MvpForCausalLM(MvpPreTrainedModel):
    method __init__ (line 1869) | def __init__(self, config):
    method get_input_embeddings (line 1881) | def get_input_embeddings(self):
    method set_input_embeddings (line 1884) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1887) | def get_output_embeddings(self):
    method set_output_embeddings (line 1890) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 1893) | def set_decoder(self, decoder):
    method get_decoder (line 1896) | def get_decoder(self):
    method set_lightweight_tuning (line 1899) | def set_lightweight_tuning(self):
    method forward (line 1904) | def forward(
    method prepare_inputs_for_generation (line 2046) | def prepare_inputs_for_generation(
    method _reorder_cache (line 2064) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/mvp/tokenization_mvp.py
  function bytes_to_unicode (line 51) | def bytes_to_unicode():
  function get_pairs (line 75) | def get_pairs(word):
  class MvpTokenizer (line 89) | class MvpTokenizer(PreTrainedTokenizer):
    method __init__ (line 172) | def __init__(
    method vocab_size (line 226) | def vocab_size(self):
    method get_vocab (line 229) | def get_vocab(self):
    method bpe (line 232) | def bpe(self, token):
    method _tokenize (line 274) | def _tokenize(self, text):
    method _convert_token_to_id (line 284) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 288) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 292) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 298) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 327) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 352) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 379) | def create_token_type_ids_from_sequences(
    method prepare_for_tokenization (line 402) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...

FILE: transformers/models/mvp/tokenization_mvp_fast.py
  class MvpTokenizerFast (line 53) | class MvpTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 139) | def __init__(
    method mask_token (line 209) | def mask_token(self) -> str:
    method mask_token (line 224) | def mask_token(self, value):
    method _batch_encode_plus (line 235) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 246) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method save_vocabulary (line 257) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 261) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 268) | def create_token_type_ids_from_sequences(

FILE: transformers/models/nat/configuration_nat.py
  class NatConfig (line 30) | class NatConfig(BackboneConfigMixin, PretrainedConfig):
    method __init__ (line 102) | def __init__(

FILE: transformers/models/nat/modeling_nat.py
  function natten2dqkrpb (line 50) | def natten2dqkrpb(*args, **kwargs):
  function natten2dav (line 53) | def natten2dav(*args, **kwargs):
  class NatEncoderOutput (line 80) | class NatEncoderOutput(ModelOutput):
  class NatModelOutput (line 113) | class NatModelOutput(ModelOutput):
  class NatImageClassifierOutput (line 149) | class NatImageClassifierOutput(ModelOutput):
  class NatEmbeddings (line 184) | class NatEmbeddings(nn.Module):
    method __init__ (line 189) | def __init__(self, config):
    method forward (line 197) | def forward(self, pixel_values: Optional[torch.FloatTensor]) -> Tuple[...
  class NatPatchEmbeddings (line 206) | class NatPatchEmbeddings(nn.Module):
    method __init__ (line 213) | def __init__(self, config):
    method forward (line 230) | def forward(self, pixel_values: Optional[torch.FloatTensor]) -> torch....
  class NatDownsampler (line 242) | class NatDownsampler(nn.Module):
    method __init__ (line 253) | def __init__(self, dim: int, norm_layer: nn.Module = nn.LayerNorm) -> ...
    method forward (line 259) | def forward(self, input_feature: torch.Tensor) -> torch.Tensor:
  function drop_path (line 266) | def drop_path(input, drop_prob=0.0, training=False, scale_by_keep=True):
  class NatDropPath (line 287) | class NatDropPath(nn.Module):
    method __init__ (line 290) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 294) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 297) | def extra_repr(self) -> str:
  class NeighborhoodAttention (line 301) | class NeighborhoodAttention(nn.Module):
    method __init__ (line 302) | def __init__(self, config, dim, num_heads, kernel_size):
    method transpose_for_scores (line 323) | def transpose_for_scores(self, x):
    method forward (line 328) | def forward(
  class NeighborhoodAttentionOutput (line 362) | class NeighborhoodAttentionOutput(nn.Module):
    method __init__ (line 363) | def __init__(self, config, dim):
    method forward (line 368) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class NeighborhoodAttentionModule (line 375) | class NeighborhoodAttentionModule(nn.Module):
    method __init__ (line 376) | def __init__(self, config, dim, num_heads, kernel_size):
    method prune_heads (line 382) | def prune_heads(self, heads):
    method forward (line 400) | def forward(
  class NatIntermediate (line 411) | class NatIntermediate(nn.Module):
    method __init__ (line 412) | def __init__(self, config, dim):
    method forward (line 420) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class NatOutput (line 426) | class NatOutput(nn.Module):
    method __init__ (line 427) | def __init__(self, config, dim):
    method forward (line 432) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class NatLayer (line 438) | class NatLayer(nn.Module):
    method __init__ (line 439) | def __init__(self, config, dim, num_heads, drop_path_rate=0.0):
    method maybe_pad (line 455) | def maybe_pad(self, hidden_states, height, width):
    method forward (line 466) | def forward(
  class NatStage (line 505) | class NatStage(nn.Module):
    method __init__ (line 506) | def __init__(self, config, dim, depth, num_heads, drop_path_rate, down...
    method forward (line 530) | def forward(
  class NatEncoder (line 551) | class NatEncoder(nn.Module):
    method __init__ (line 552) | def __init__(self, config):
    method forward (line 571) | def forward(
  class NatPreTrainedModel (line 620) | class NatPreTrainedModel(PreTrainedModel):
    method _init_weights (line 630) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 642) | def _set_gradient_checkpointing(self, module: NatEncoder, value: bool ...
  class NatModel (line 678) | class NatModel(NatPreTrainedModel):
    method __init__ (line 679) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 697) | def get_input_embeddings(self):
    method _prune_heads (line 700) | def _prune_heads(self, heads_to_prune):
    method forward (line 716) | def forward(
  class NatForImageClassification (line 770) | class NatForImageClassification(NatPreTrainedModel):
    method __init__ (line 771) | def __init__(self, config):
    method forward (line 794) | def forward(
  class NatBackbone (line 861) | class NatBackbone(NatPreTrainedModel, BackboneMixin):
    method __init__ (line 862) | def __init__(self, config):
    method get_input_embeddings (line 881) | def get_input_embeddings(self):
    method forward (line 886) | def forward(

FILE: transformers/models/nezha/configuration_nezha.py
  class NezhaConfig (line 9) | class NezhaConfig(PretrainedConfig):
    method __init__ (line 69) | def __init__(

FILE: transformers/models/nezha/modeling_nezha.py
  function load_tf_weights_in_nezha (line 67) | def load_tf_weights_in_nezha(model, config, tf_checkpoint_path):
  class NezhaRelativePositionsEncoding (line 140) | class NezhaRelativePositionsEncoding(nn.Module):
    method __init__ (line 143) | def __init__(self, length, depth, max_relative_position=127):
    method forward (line 168) | def forward(self, length):
  class NezhaEmbeddings (line 172) | class NezhaEmbeddings(nn.Module):
    method __init__ (line 175) | def __init__(self, config):
    method forward (line 188) | def forward(
  class NezhaSelfAttention (line 223) | class NezhaSelfAttention(nn.Module):
    method __init__ (line 224) | def __init__(self, config):
    method transpose_for_scores (line 248) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 253) | def forward(
  class NezhaSelfOutput (line 359) | class NezhaSelfOutput(nn.Module):
    method __init__ (line 360) | def __init__(self, config):
    method forward (line 366) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class NezhaAttention (line 373) | class NezhaAttention(nn.Module):
    method __init__ (line 374) | def __init__(self, config):
    method prune_heads (line 380) | def prune_heads(self, heads):
    method forward (line 398) | def forward(
  class NezhaIntermediate (line 423) | class NezhaIntermediate(nn.Module):
    method __init__ (line 424) | def __init__(self, config):
    method forward (line 432) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class NezhaOutput (line 439) | class NezhaOutput(nn.Module):
    method __init__ (line 440) | def __init__(self, config):
    method forward (line 446) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class NezhaLayer (line 453) | class NezhaLayer(nn.Module):
    method __init__ (line 454) | def __init__(self, config):
    method forward (line 468) | def forward(
    method feed_forward_chunk (line 533) | def feed_forward_chunk(self, attention_output):
  class NezhaEncoder (line 540) | class NezhaEncoder(nn.Module):
    method __init__ (line 541) | def __init__(self, config):
    method forward (line 547) | def forward(
  class NezhaPooler (line 639) | class NezhaPooler(nn.Module):
    method __init__ (line 640) | def __init__(self, config):
    method forward (line 645) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class NezhaPredictionHeadTransform (line 655) | class NezhaPredictionHeadTransform(nn.Module):
    method __init__ (line 656) | def __init__(self, config):
    method forward (line 665) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class NezhaLMPredictionHead (line 673) | class NezhaLMPredictionHead(nn.Module):
    method __init__ (line 674) | def __init__(self, config):
    method forward (line 687) | def forward(self, hidden_states):
  class NezhaOnlyMLMHead (line 694) | class NezhaOnlyMLMHead(nn.Module):
    method __init__ (line 695) | def __init__(self, config):
    method forward (line 699) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class NezhaOnlyNSPHead (line 705) | class NezhaOnlyNSPHead(nn.Module):
    method __init__ (line 706) | def __init__(self, config):
    method forward (line 710) | def forward(self, pooled_output):
  class NezhaPreTrainingHeads (line 716) | class NezhaPreTrainingHeads(nn.Module):
    method __init__ (line 717) | def __init__(self, config):
    method forward (line 722) | def forward(self, sequence_output, pooled_output):
  class NezhaPreTrainedModel (line 728) | class NezhaPreTrainedModel(PreTrainedModel):
    method _init_weights (line 740) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 756) | def _set_gradient_checkpointing(self, module, value=False):
  class NezhaForPreTrainingOutput (line 762) | class NezhaForPreTrainingOutput(ModelOutput):
  class NezhaModel (line 860) | class NezhaModel(NezhaPreTrainedModel):
    method __init__ (line 873) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 885) | def get_input_embeddings(self):
    method set_input_embeddings (line 888) | def set_input_embeddings(self, value):
    method _prune_heads (line 891) | def _prune_heads(self, heads_to_prune):
    method forward (line 905) | def forward(
  class NezhaForPreTraining (line 1039) | class NezhaForPreTraining(NezhaPreTrainedModel):
    method __init__ (line 1042) | def __init__(self, config):
    method get_output_embeddings (line 1051) | def get_output_embeddings(self):
    method set_output_embeddings (line 1054) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1059) | def forward(
  class NezhaForMaskedLM (line 1141) | class NezhaForMaskedLM(NezhaPreTrainedModel):
    method __init__ (line 1145) | def __init__(self, config):
    method get_output_embeddings (line 1160) | def get_output_embeddings(self):
    method set_output_embeddings (line 1163) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1172) | def forward(
    method prepare_inputs_for_generation (line 1227) | def prepare_inputs_for_generation(self, input_ids, attention_mask=None...
  class NezhaForNextSentencePrediction (line 1248) | class NezhaForNextSentencePrediction(NezhaPreTrainedModel):
    method __init__ (line 1249) | def __init__(self, config):
    method forward (line 1260) | def forward(
  class NezhaForSequenceClassification (line 1351) | class NezhaForSequenceClassification(NezhaPreTrainedModel):
    method __init__ (line 1352) | def __init__(self, config):
    method forward (line 1373) | def forward(
  class NezhaForMultipleChoice (line 1450) | class NezhaForMultipleChoice(NezhaPreTrainedModel):
    method __init__ (line 1451) | def __init__(self, config):
    method forward (line 1470) | def forward(
  class NezhaForTokenClassification (line 1542) | class NezhaForTokenClassification(NezhaPreTrainedModel):
    method __init__ (line 1545) | def __init__(self, config):
    method forward (line 1565) | def forward(
  class NezhaForQuestionAnswering (line 1623) | class NezhaForQuestionAnswering(NezhaPreTrainedModel):
    method __init__ (line 1626) | def __init__(self, config):
    method forward (line 1642) | def forward(

FILE: transformers/models/nllb/tokenization_nllb.py
  class NllbTokenizer (line 49) | class NllbTokenizer(PreTrainedTokenizer):
    method __init__ (line 128) | def __init__(
    method __getstate__ (line 206) | def __getstate__(self):
    method __setstate__ (line 212) | def __setstate__(self, d):
    method vocab_size (line 223) | def vocab_size(self):
    method src_lang (line 227) | def src_lang(self) -> str:
    method src_lang (line 231) | def src_lang(self, new_src_lang: str) -> None:
    method get_special_tokens_mask (line 235) | def get_special_tokens_mask(
    method build_inputs_with_special_tokens (line 265) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 292) | def create_token_type_ids_from_sequences(
    method _build_translation_inputs (line 317) | def _build_translation_inputs(
    method get_vocab (line 329) | def get_vocab(self):
    method _tokenize (line 334) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 337) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 346) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 352) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 357) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method prepare_seq2seq_batch (line 374) | def prepare_seq2seq_batch(
    method _switch_to_input_mode (line 386) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 389) | def _switch_to_target_mode(self):
    method set_src_lang_special_tokens (line 392) | def set_src_lang_special_tokens(self, src_lang) -> None:
    method set_tgt_lang_special_tokens (line 405) | def set_tgt_lang_special_tokens(self, lang: str) -> None:

FILE: transformers/models/nllb/tokenization_nllb_fast.py
  class NllbTokenizerFast (line 61) | class NllbTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 140) | def __init__(
    method src_lang (line 199) | def src_lang(self) -> str:
    method src_lang (line 203) | def src_lang(self, new_src_lang: str) -> None:
    method build_inputs_with_special_tokens (line 207) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 236) | def create_token_type_ids_from_sequences(
    method _build_translation_inputs (line 261) | def _build_translation_inputs(
    method prepare_seq2seq_batch (line 273) | def prepare_seq2seq_batch(
    method _switch_to_input_mode (line 285) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 288) | def _switch_to_target_mode(self):
    method set_src_lang_special_tokens (line 291) | def set_src_lang_special_tokens(self, src_lang) -> None:
    method set_tgt_lang_special_tokens (line 314) | def set_tgt_lang_special_tokens(self, lang: str) -> None:
    method save_vocabulary (line 336) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/nllb_moe/configuration_nllb_moe.py
  class NllbMoeConfig (line 27) | class NllbMoeConfig(PretrainedConfig):
    method __init__ (line 132) | def __init__(

FILE: transformers/models/nllb_moe/convert_nllb_moe_sharded_original_checkpoint_to_pytorch.py
  function remove_ignore_keys_ (line 26) | def remove_ignore_keys_(state_dict):
  function make_linear_from_emb (line 41) | def make_linear_from_emb(emb):
  function rename_fairseq_keys (line 48) | def rename_fairseq_keys(state_dict, expert_idx=None):
  function shard_on_the_fly (line 73) | def shard_on_the_fly(switch_checkpoint_path, dump_path, num_experts, dty...

FILE: transformers/models/nllb_moe/modeling_nllb_moe.py
  function shift_tokens_right (line 64) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 81) | def _make_causal_mask(
  function _expand_mask (line 99) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  function create_position_ids_from_input_ids (line 114) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...
  function load_balancing_loss_func (line 131) | def load_balancing_loss_func(router_probs: torch.Tensor, expert_indices:...
  class NllbMoeSinusoidalPositionalEmbedding (line 171) | class NllbMoeSinusoidalPositionalEmbedding(nn.Module):
    method __init__ (line 174) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method make_weights (line 181) | def make_weights(self, num_embeddings: int, embedding_dim: int, paddin...
    method get_embedding (line 190) | def get_embedding(num_embeddings: int, embedding_dim: int, padding_idx...
    method forward (line 211) | def forward(
    method create_position_ids_from_inputs_embeds (line 231) | def create_position_ids_from_inputs_embeds(self, inputs_embeds, past_k...
  class NllbMoeTop2Router (line 249) | class NllbMoeTop2Router(nn.Module):
    method __init__ (line 261) | def __init__(self, config: NllbMoeConfig):
    method _cast_classifier (line 274) | def _cast_classifier(self):
    method normalize_router_probabilities (line 282) | def normalize_router_probabilities(self, router_probs, top_1_mask, top...
    method route_tokens (line 290) | def route_tokens(
    method forward (line 375) | def forward(self, hidden_states: torch.Tensor, padding_mask: Optional[...
  class NllbMoeDenseActDense (line 404) | class NllbMoeDenseActDense(nn.Module):
    method __init__ (line 405) | def __init__(self, config: NllbMoeConfig, ffn_dim: int):
    method forward (line 412) | def forward(self, hidden_states):
  class NllbMoeSparseMLP (line 426) | class NllbMoeSparseMLP(nn.Module):
    method __init__ (line 431) | def __init__(self, config: NllbMoeConfig, ffn_dim: int, expert_class: ...
    method forward (line 442) | def forward(self, hidden_states: torch.Tensor, padding_mask: Optional[...
  class NllbMoeAttention (line 492) | class NllbMoeAttention(nn.Module):
    method __init__ (line 495) | def __init__(
    method _shape (line 522) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 525) | def forward(
  class NllbMoeEncoderLayer (line 646) | class NllbMoeEncoderLayer(nn.Module):
    method __init__ (line 647) | def __init__(self, config: NllbMoeConfig, is_sparse: bool = False):
    method forward (line 665) | def forward(
  class NllbMoeDecoderLayer (line 725) | class NllbMoeDecoderLayer(nn.Module):
    method __init__ (line 726) | def __init__(self, config: NllbMoeConfig, is_sparse: bool = False):
    method forward (line 752) | def forward(
  class NllbMoePreTrainedModel (line 855) | class NllbMoePreTrainedModel(PreTrainedModel):
    method _init_weights (line 861) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 873) | def _set_gradient_checkpointing(self, module, value=False):
  class NllbMoeEncoder (line 1004) | class NllbMoeEncoder(NllbMoePreTrainedModel):
    method __init__ (line 1016) | def __init__(self, config: NllbMoeConfig, embed_tokens: Optional[nn.Em...
    method forward (line 1049) | def forward(
  class NllbMoeDecoder (line 1199) | class NllbMoeDecoder(NllbMoePreTrainedModel):
    method __init__ (line 1210) | def __init__(self, config: NllbMoeConfig, embed_tokens: Optional[nn.Em...
    method forward (line 1241) | def forward(
  class NllbMoeModel (line 1503) | class NllbMoeModel(NllbMoePreTrainedModel):
    method __init__ (line 1513) | def __init__(self, config: NllbMoeConfig):
    method get_input_embeddings (line 1525) | def get_input_embeddings(self):
    method set_input_embeddings (line 1528) | def set_input_embeddings(self, value):
    method get_encoder (line 1533) | def get_encoder(self):
    method get_decoder (line 1536) | def get_decoder(self):
    method forward (line 1542) | def forward(
  class NllbMoeForConditionalGeneration (line 1642) | class NllbMoeForConditionalGeneration(NllbMoePreTrainedModel):
    method __init__ (line 1656) | def __init__(self, config: NllbMoeConfig):
    method get_encoder (line 1666) | def get_encoder(self):
    method get_decoder (line 1669) | def get_decoder(self):
    method resize_token_embeddings (line 1672) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method get_output_embeddings (line 1676) | def get_output_embeddings(self):
    method set_output_embeddings (line 1679) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1685) | def forward(
    method _unpack_router_logits (line 1800) | def _unpack_router_logits(self, router_outputs):
    method prepare_inputs_for_generation (line 1815) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1844) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/nystromformer/configuration_nystromformer.py
  class NystromformerConfig (line 29) | class NystromformerConfig(PretrainedConfig):
    method __init__ (line 94) | def __init__(

FILE: transformers/models/nystromformer/convert_nystromformer_original_pytorch_checkpoint_to_pytorch.py
  function rename_key (line 25) | def rename_key(orig_key):
  function convert_checkpoint_helper (line 63) | def convert_checkpoint_helper(config, orig_state_dict):
  function convert_nystromformer_checkpoint (line 80) | def convert_nystromformer_checkpoint(checkpoint_path, nystromformer_conf...

FILE: transformers/models/nystromformer/modeling_nystromformer.py
  class NystromformerEmbeddings (line 52) | class NystromformerEmbeddings(nn.Module):
    method __init__ (line 55) | def __init__(self, config):
    method forward (line 75) | def forward(self, input_ids=None, token_type_ids=None, position_ids=No...
  class NystromformerSelfAttention (line 110) | class NystromformerSelfAttention(nn.Module):
    method __init__ (line 111) | def __init__(self, config, position_embedding_type=None):
    method iterative_inv (line 152) | def iterative_inv(self, mat, n_iter=6):
    method transpose_for_scores (line 173) | def transpose_for_scores(self, layer):
    method forward (line 178) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class NystromformerSelfOutput (line 241) | class NystromformerSelfOutput(nn.Module):
    method __init__ (line 242) | def __init__(self, config):
    method forward (line 248) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class NystromformerAttention (line 255) | class NystromformerAttention(nn.Module):
    method __init__ (line 256) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 262) | def prune_heads(self, heads):
    method forward (line 280) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class NystromformerIntermediate (line 288) | class NystromformerIntermediate(nn.Module):
    method __init__ (line 289) | def __init__(self, config):
    method forward (line 297) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class NystromformerOutput (line 304) | class NystromformerOutput(nn.Module):
    method __init__ (line 305) | def __init__(self, config):
    method forward (line 311) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class NystromformerLayer (line 318) | class NystromformerLayer(nn.Module):
    method __init__ (line 319) | def __init__(self, config):
    method forward (line 328) | def forward(self, hidden_states, attention_mask=None, output_attention...
    method feed_forward_chunk (line 341) | def feed_forward_chunk(self, attention_output):
  class NystromformerEncoder (line 347) | class NystromformerEncoder(nn.Module):
    method __init__ (line 348) | def __init__(self, config):
    method forward (line 354) | def forward(
  class NystromformerPredictionHeadTransform (line 403) | class NystromformerPredictionHeadTransform(nn.Module):
    method __init__ (line 404) | def __init__(self, config):
    method forward (line 413) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class NystromformerLMPredictionHead (line 421) | class NystromformerLMPredictionHead(nn.Module):
    method __init__ (line 422) | def __init__(self, config):
    method forward (line 435) | def forward(self, hidden_states):
  class NystromformerOnlyMLMHead (line 442) | class NystromformerOnlyMLMHead(nn.Module):
    method __init__ (line 443) | def __init__(self, config):
    method forward (line 447) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class NystromformerPreTrainedModel (line 452) | class NystromformerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 463) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 479) | def _set_gradient_checkpointing(self, module, value=False):
  class NystromformerModel (line 549) | class NystromformerModel(NystromformerPreTrainedModel):
    method __init__ (line 550) | def __init__(self, config):
    method get_input_embeddings (line 560) | def get_input_embeddings(self):
    method set_input_embeddings (line 563) | def set_input_embeddings(self, value):
    method _prune_heads (line 566) | def _prune_heads(self, heads_to_prune):
    method forward (line 580) | def forward(
  class NystromformerForMaskedLM (line 660) | class NystromformerForMaskedLM(NystromformerPreTrainedModel):
    method __init__ (line 663) | def __init__(self, config):
    method get_output_embeddings (line 672) | def get_output_embeddings(self):
    method set_output_embeddings (line 675) | def set_output_embeddings(self, new_embeddings):
    method forward (line 684) | def forward(
  class NystromformerClassificationHead (line 737) | class NystromformerClassificationHead(nn.Module):
    method __init__ (line 740) | def __init__(self, config):
    method forward (line 748) | def forward(self, features, **kwargs):
  class NystromformerForSequenceClassification (line 765) | class NystromformerForSequenceClassification(NystromformerPreTrainedModel):
    method __init__ (line 766) | def __init__(self, config):
    method forward (line 781) | def forward(
  class NystromformerForMultipleChoice (line 858) | class NystromformerForMultipleChoice(NystromformerPreTrainedModel):
    method __init__ (line 859) | def __init__(self, config):
    method forward (line 877) | def forward(
  class NystromformerForTokenClassification (line 953) | class NystromformerForTokenClassification(NystromformerPreTrainedModel):
    method __init__ (line 954) | def __init__(self, config):
    method forward (line 971) | def forward(
  class NystromformerForQuestionAnswering (line 1031) | class NystromformerForQuestionAnswering(NystromformerPreTrainedModel):
    method __init__ (line 1032) | def __init__(self, config):
    method forward (line 1050) | def forward(

FILE: transformers/models/oneformer/configuration_oneformer.py
  class OneFormerConfig (line 34) | class OneFormerConfig(PretrainedConfig):
    method __init__ (line 144) | def __init__(
    method to_dict (line 254) | def to_dict(self) -> Dict[str, any]:

FILE: transformers/models/oneformer/convert_to_hf_oneformer.py
  class TrackedStateDict (line 61) | class TrackedStateDict:
    method __init__ (line 62) | def __init__(self, to_track: Dict):
    method __getitem__ (line 71) | def __getitem__(self, key: str) -> Any:
    method __setitem__ (line 74) | def __setitem__(self, key: str, item: Any):
    method diff (line 78) | def diff(self) -> List[str]:
    method copy (line 87) | def copy(self) -> Dict:
  function prepare_img (line 93) | def prepare_img():
  class Args (line 101) | class Args:
  function setup_cfg (line 107) | def setup_cfg(args: Args):
  class OriginalOneFormerConfigToOursConverter (line 120) | class OriginalOneFormerConfigToOursConverter:
    method __call__ (line 121) | def __call__(self, original_config: object, is_swin: bool) -> OneForme...
  class OriginalOneFormerConfigToProcessorConverter (line 205) | class OriginalOneFormerConfigToProcessorConverter:
    method __call__ (line 206) | def __call__(self, original_config: object, model_repo: str) -> OneFor...
  class OriginalOneFormerCheckpointToOursConverter (line 240) | class OriginalOneFormerCheckpointToOursConverter:
    method __init__ (line 241) | def __init__(self, original_model: nn.Module, config: OneFormerConfig):
    method pop_all (line 245) | def pop_all(self, renamed_keys: List[Tuple[str, str]], dst_state_dict:...
    method replace_swin_backbone (line 250) | def replace_swin_backbone(self, dst_state_dict: StateDict, src_state_d...
    method replace_dinat_backbone (line 409) | def replace_dinat_backbone(self, dst_state_dict: StateDict, src_state_...
    method replace_pixel_module (line 546) | def replace_pixel_module(self, dst_state_dict: StateDict, src_state_di...
    method replace_keys_qkv_transformer_decoder (line 634) | def replace_keys_qkv_transformer_decoder(self, dst_state_dict: StateDi...
    method replace_transformer_module (line 653) | def replace_transformer_module(self, dst_state_dict: StateDict, src_st...
    method replace_task_mlp (line 815) | def replace_task_mlp(self, dst_state_dict: StateDict, src_state_dict: ...
    method replace_text_projector (line 834) | def replace_text_projector(self, dst_state_dict: StateDict, src_state_...
    method replace_text_mapper (line 851) | def replace_text_mapper(self, dst_state_dict: StateDict, src_state_dic...
    method convert (line 905) | def convert(self, oneformer: OneFormerModel, is_swin: bool) -> OneForm...
    method using_dirs (line 924) | def using_dirs(checkpoints_dir: Path, config_dir: Path) -> Iterator[Tu...
  function post_process_sem_seg_output (line 935) | def post_process_sem_seg_output(outputs: OneFormerForUniversalSegmentati...
  function test (line 961) | def test(
  function get_name (line 1059) | def get_name(checkpoint_file: Path):

FILE: transformers/models/oneformer/image_processing_oneformer.py
  function max_across_indices (line 63) | def max_across_indices(values: Iterable[Any]) -> List[Any]:
  function get_max_height_width (line 71) | def get_max_height_width(images: List[np.ndarray]) -> List[int]:
  function make_pixel_mask (line 87) | def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> ...
  function binary_mask_to_rle (line 104) | def binary_mask_to_rle(mask):
  function convert_segmentation_to_rle (line 127) | def convert_segmentation_to_rle(segmentation):
  function remove_low_and_no_objects (line 149) | def remove_low_and_no_objects(masks, scores, labels, object_mask_thresho...
  function check_segment_validity (line 178) | def check_segment_validity(mask_labels, mask_probs, k, mask_threshold=0....
  function compute_segments (line 197) | def compute_segments(
  function convert_segmentation_map_to_binary_masks (line 258) | def convert_segmentation_map_to_binary_masks(
  function get_oneformer_resize_output_image_size (line 294) | def get_oneformer_resize_output_image_size(
  function prepare_metadata (line 322) | def prepare_metadata(repo_path, class_info_file):
  class OneFormerImageProcessor (line 338) | class OneFormerImageProcessor(BaseImageProcessor):
    method __init__ (line 391) | def __init__(
    method resize (line 440) | def resize(
    method rescale (line 482) | def rescale(
    method normalize (line 491) | def normalize(
    method convert_segmentation_map_to_binary_masks (line 504) | def convert_segmentation_map_to_binary_masks(
    method __call__ (line 521) | def __call__(self, images, task_inputs=None, segmentation_maps=None, *...
    method _preprocess (line 524) | def _preprocess(
    method _preprocess_image (line 544) | def _preprocess_image(
    method _preprocess_mask (line 575) | def _preprocess_mask(
    method preprocess (line 604) | def preprocess(
    method _pad_image (line 718) | def _pad_image(
    method pad (line 740) | def pad(
    method get_semantic_annotations (line 778) | def get_semantic_annotations(self, label, num_class_obj):
    method get_instance_annotations (line 813) | def get_instance_annotations(self, label, num_class_obj):
    method get_panoptic_annotations (line 845) | def get_panoptic_annotations(self, label, num_class_obj):
    method encode_inputs (line 875) | def encode_inputs(
    method post_process_semantic_segmentation (line 1005) | def post_process_semantic_segmentation(
    method post_process_instance_segmentation (line 1055) | def post_process_instance_segmentation(
    method post_process_panoptic_segmentation (line 1174) | def post_process_panoptic_segmentation(

FILE: transformers/models/oneformer/modeling_oneformer.py
  function _get_clones (line 59) | def _get_clones(module, N):
  function multi_scale_deformable_attention (line 64) | def multi_scale_deformable_attention(
  function dice_loss (line 104) | def dice_loss(inputs: Tensor, labels: Tensor, num_masks: int) -> Tensor:
  function sigmoid_cross_entropy_loss (line 135) | def sigmoid_cross_entropy_loss(inputs: torch.Tensor, labels: torch.Tenso...
  function pair_wise_dice_loss (line 155) | def pair_wise_dice_loss(inputs: Tensor, labels: Tensor) -> Tensor:
  function pair_wise_sigmoid_cross_entropy_loss (line 178) | def pair_wise_sigmoid_cross_entropy_loss(inputs: torch.Tensor, labels: t...
  function sample_point (line 207) | def sample_point(
  class OneFormerHungarianMatcher (line 240) | class OneFormerHungarianMatcher(nn.Module):
    method __init__ (line 241) | def __init__(
    method forward (line 269) | def forward(self, masks_queries_logits, class_queries_logits, mask_lab...
  class OneFormerLoss (line 349) | class OneFormerLoss(nn.Module):
    method __init__ (line 350) | def __init__(
    method _max_by_axis (line 404) | def _max_by_axis(self, the_list: List[List[int]]) -> List[int]:
    method _pad_images_to_max_in_batch (line 411) | def _pad_images_to_max_in_batch(self, tensors: List[Tensor]) -> Tuple[...
    method loss_contrastive (line 430) | def loss_contrastive(self, contrastive_queries_logits: Tensor, text_qu...
    method loss_labels (line 467) | def loss_labels(
    method loss_masks (line 502) | def loss_masks(
    method calculate_uncertainty (line 560) | def calculate_uncertainty(self, logits: torch.Tensor) -> torch.Tensor:
    method sample_points_using_uncertainty (line 578) | def sample_points_using_uncertainty(
    method _get_predictions_permutation_indices (line 633) | def _get_predictions_permutation_indices(self, indices):
    method _get_targets_permutation_indices (line 639) | def _get_targets_permutation_indices(self, indices):
    method forward (line 645) | def forward(
    method get_num_masks (line 720) | def get_num_masks(self, class_labels: torch.Tensor, device: torch.devi...
  class OneFormerTransformerDecoderOutput (line 730) | class OneFormerTransformerDecoderOutput(BaseModelOutput):
  class OneFormerPixelDecoderOutput (line 757) | class OneFormerPixelDecoderOutput(ModelOutput):
  class OneFormerPixelLevelModuleOutput (line 781) | class OneFormerPixelLevelModuleOutput(ModelOutput):
  class OneFormerModelOutput (line 804) | class OneFormerModelOutput(ModelOutput):
  class OneFormerForUniversalSegmentationOutput (line 854) | class OneFormerForUniversalSegmentationOutput(ModelOutput):
  class OneFormerPixelDecoderFrozenBatchNorm2d (line 923) | class OneFormerPixelDecoderFrozenBatchNorm2d(nn.Module):
    method __init__ (line 931) | def __init__(self, n):
    method _load_from_state_dict (line 938) | def _load_from_state_dict(
    method forward (line 949) | def forward(self, x):
  class OneFormerPixelDecoderEncoderMultiscaleDeformableAttention (line 961) | class OneFormerPixelDecoderEncoderMultiscaleDeformableAttention(nn.Module):
    method __init__ (line 966) | def __init__(self, embed_dim: int, num_heads: int, n_levels: int, n_po...
    method with_pos_embed (line 993) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 996) | def forward(
  class OneFormerPixelDecoderEncoderLayer (line 1054) | class OneFormerPixelDecoderEncoderLayer(nn.Module):
    method __init__ (line 1055) | def __init__(self, config: OneFormerConfig):
    method forward (line 1075) | def forward(
  class OneFormerPixelDecoderEncoderOnly (line 1146) | class OneFormerPixelDecoderEncoderOnly(nn.Module):
    method __init__ (line 1157) | def __init__(self, config: OneFormerConfig):
    method get_reference_points (line 1165) | def get_reference_points(spatial_shapes, valid_ratios, device):
    method forward (line 1193) | def forward(
  class OneFormerPixelDecoder (line 1269) | class OneFormerPixelDecoder(nn.Module):
    method __init__ (line 1270) | def __init__(self, config: OneFormerConfig, feature_channels):
    method get_valid_ratio (line 1355) | def get_valid_ratio(self, mask):
    method forward (line 1366) | def forward(
  class OneFormerPixelLevelModule (line 1472) | class OneFormerPixelLevelModule(nn.Module):
    method __init__ (line 1473) | def __init__(self, config: OneFormerConfig):
    method forward (line 1488) | def forward(self, pixel_values: Tensor, output_hidden_states: bool = F...
  class OneFormerAttention (line 1499) | class OneFormerAttention(nn.Module):
    method __init__ (line 1505) | def __init__(
    method _shape (line 1530) | def _shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method with_pos_embed (line 1533) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 1536) | def forward(
  class OneFormerTransformerDecoderSelfAttentionLayer (line 1635) | class OneFormerTransformerDecoderSelfAttentionLayer(nn.Module):
    method __init__ (line 1636) | def __init__(
    method with_pos_embed (line 1648) | def with_pos_embed(self, tensor, pos: Optional[Tensor]):
    method forward_post (line 1651) | def forward_post(
    method forward_pre (line 1666) | def forward_pre(
    method forward (line 1681) | def forward(
  class OneFormerTransformerDecoderCrossAttentionLayer (line 1693) | class OneFormerTransformerDecoderCrossAttentionLayer(nn.Module):
    method __init__ (line 1694) | def __init__(
    method with_pos_embed (line 1706) | def with_pos_embed(self, tensor, pos: Optional[Tensor]):
    method forward_post (line 1709) | def forward_post(
    method forward_pre (line 1730) | def forward_pre(
    method forward (line 1751) | def forward(
  class OneFormerTransformerDecoderFFNLayer (line 1765) | class OneFormerTransformerDecoderFFNLayer(nn.Module):
    method __init__ (line 1766) | def __init__(
    method with_pos_embed (line 1786) | def with_pos_embed(self, tensor, pos: Optional[Tensor]):
    method forward_post (line 1789) | def forward_post(self, output):
    method forward_pre (line 1795) | def forward_pre(self, output):
    method forward (line 1801) | def forward(self, output):
  class OneFormerMLPPredictionHead (line 1807) | class OneFormerMLPPredictionHead(nn.Module):
    method __init__ (line 1808) | def __init__(self, input_dim: int, hidden_dim: int, output_dim: int, n...
    method forward (line 1834) | def forward(self, input: Tensor) -> Tensor:
  class OneFormerTransformerDecoderLayer (line 1839) | class OneFormerTransformerDecoderLayer(nn.Module):
    method __init__ (line 1840) | def __init__(self, config: OneFormerConfig):
    method forward (line 1869) | def forward(
  class OneFormerTransformerDecoderQueryTransformerDecoder (line 1926) | class OneFormerTransformerDecoderQueryTransformerDecoder(nn.Module):
    method __init__ (line 1927) | def __init__(self, decoder_layer, num_layers, norm=None, return_interm...
    method forward (line 1934) | def forward(
  class OneFormerTransformerDecoderQueryTransformerDecoderLayer (line 1973) | class OneFormerTransformerDecoderQueryTransformerDecoderLayer(nn.Module):
    method __init__ (line 1974) | def __init__(
    method with_pos_embed (line 2002) | def with_pos_embed(self, tensor, pos: Optional[Tensor]):
    method forward_post (line 2005) | def forward_post(
    method forward_pre (line 2036) | def forward_pre(
    method forward (line 2067) | def forward(
  class OneFormerTransformerDecoderQueryTransformer (line 2101) | class OneFormerTransformerDecoderQueryTransformer(nn.Module):
    method __init__ (line 2102) | def __init__(
    method forward (line 2130) | def forward(self, src, mask, query_embed, pos_embed, task_token=None):
  class OneFormerTransformerDecoder (line 2147) | class OneFormerTransformerDecoder(nn.Module):
    method __init__ (line 2152) | def __init__(self, in_channels: int, config: OneFormerConfig):
    method forward (line 2191) | def forward(
    method forward_prediction_heads (line 2275) | def forward_prediction_heads(self, output, mask_features, attention_ma...
    method _get_aux_predictions (line 2296) | def _get_aux_predictions(self, outputs_class, outputs_seg_masks):
  class OneFormerTransformerModule (line 2307) | class OneFormerTransformerModule(nn.Module):
    method __init__ (line 2312) | def __init__(self, in_features: int, config: OneFormerConfig):
    method forward (line 2329) | def forward(
  class OneFormerSinePositionEmbedding (line 2379) | class OneFormerSinePositionEmbedding(nn.Module):
    method __init__ (line 2385) | def __init__(
    method forward (line 2396) | def forward(self, x: Tensor, mask: Optional[Tensor] = None) -> Tensor:
  class PredictionBlock (line 2419) | class PredictionBlock(nn.Module):
    method __init__ (line 2420) | def __init__(self, in_dim: int, out_dim: int, activation: nn.Module) -...
    method forward (line 2427) | def forward(self, input: Tensor) -> Tensor:
  class OneFormerTextMapperAttention (line 2434) | class OneFormerTextMapperAttention(nn.Module):
    method __init__ (line 2435) | def __init__(self, dim, num_heads=8, qkv_bias=False, qk_scale=None, at...
    method forward (line 2450) | def forward(self, q, k, v):
  class OneFormerTextTransformerDecoderLayer (line 2470) | class OneFormerTextTransformerDecoderLayer(nn.Module):
    method __init__ (line 2471) | def __init__(
    method forward (line 2491) | def forward(self, hidden_state, mem):
  class OneFormerTextContextDecoder (line 2500) | class OneFormerTextContextDecoder(nn.Module):
    method __init__ (line 2501) | def __init__(
    method forward (line 2535) | def forward(self, text, visual):
  class OneFormerTextMLP (line 2545) | class OneFormerTextMLP(nn.Module):
    method __init__ (line 2546) | def __init__(
    method forward (line 2560) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class OneFormerTextTransformerLayer (line 2567) | class OneFormerTextTransformerLayer(nn.Module):
    method __init__ (line 2568) | def __init__(self, width: int, heads: int, attn_mask: torch.Tensor, la...
    method forward (line 2576) | def forward(
  class OneFormerTextTransformer (line 2601) | class OneFormerTextTransformer(nn.Module):
    method __init__ (line 2602) | def __init__(
    method forward (line 2619) | def forward(self, hidden_states: torch.Tensor):
  class OneFormerTextEncoder (line 2628) | class OneFormerTextEncoder(nn.Module):
    method __init__ (line 2629) | def __init__(
    method build_attention_mask (line 2655) | def build_attention_mask(self):
    method forward (line 2663) | def forward(self, text):
  class OneFormerTextMapper (line 2675) | class OneFormerTextMapper(nn.Module):
    method __init__ (line 2676) | def __init__(self, config: OneFormerConfig):
    method forward (line 2700) | def forward(
    method encode_text (line 2708) | def encode_text(self, text):
  class OneFormerTaskModel (line 2736) | class OneFormerTaskModel(nn.Module):
    method __init__ (line 2737) | def __init__(self, config: OneFormerConfig):
    method forward (line 2746) | def forward(self, inputs: Tensor) -> Tensor:
  class OneFormerPreTrainedModel (line 2786) | class OneFormerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 2791) | def _init_weights(self, module: nn.Module):
  class OneFormerModel (line 2905) | class OneFormerModel(OneFormerPreTrainedModel):
    method __init__ (line 2908) | def __init__(self, config: OneFormerConfig):
    method forward (line 2924) | def forward(
  class OneFormerForUniversalSegmentation (line 3034) | class OneFormerForUniversalSegmentation(OneFormerPreTrainedModel):
    method __init__ (line 3037) | def __init__(self, config: OneFormerConfig):
    method get_loss_dict (line 3068) | def get_loss_dict(
    method get_loss (line 3098) | def get_loss(self, loss_dict: Dict[str, Tensor]) -> Tensor:
    method forward (line 3103) | def forward(

FILE: transformers/models/oneformer/processing_oneformer.py
  class OneFormerProcessor (line 29) | class OneFormerProcessor(ProcessorMixin):
    method __init__ (line 49) | def __init__(
    method _preprocess_text (line 62) | def _preprocess_text(self, text_list=None, max_length=77):
    method __call__ (line 78) | def __call__(self, images=None, task_inputs=None, segmentation_maps=No...
    method encode_inputs (line 145) | def encode_inputs(self, images=None, task_inputs=None, segmentation_ma...
    method post_process_semantic_segmentation (line 185) | def post_process_semantic_segmentation(self, *args, **kwargs):
    method post_process_instance_segmentation (line 192) | def post_process_instance_segmentation(self, *args, **kwargs):
    method post_process_panoptic_segmentation (line 199) | def post_process_panoptic_segmentation(self, *args, **kwargs):

FILE: transformers/models/open_llama/configuration_open_llama.py
  class OpenLlamaConfig (line 33) | class OpenLlamaConfig(PretrainedConfig):
    method __init__ (line 86) | def __init__(

FILE: transformers/models/open_llama/modeling_open_llama.py
  function _make_causal_mask (line 51) | def _make_causal_mask(
  function _expand_mask (line 69) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class OpenLlamaRMSNorm (line 84) | class OpenLlamaRMSNorm(nn.Module):
    method __init__ (line 85) | def __init__(self, hidden_size, eps=1e-6):
    method forward (line 93) | def forward(self, hidden_states):
  class OpenLlamaRotaryEmbedding (line 102) | class OpenLlamaRotaryEmbedding(torch.nn.Module):
    method __init__ (line 103) | def __init__(self, dim, max_position_embeddings=2048, base=10000, devi...
    method forward (line 117) | def forward(self, x, seq_len=None):
  function rotate_half (line 134) | def rotate_half(x):
  function apply_rotary_pos_emb (line 141) | def apply_rotary_pos_emb(q, k, cos, sin, position_ids):
  class OpenLlamaMLP (line 151) | class OpenLlamaMLP(nn.Module):
    method __init__ (line 152) | def __init__(
    method forward (line 166) | def forward(self, x):
  class OpenLlamaAttention (line 171) | class OpenLlamaAttention(nn.Module):
    method __init__ (line 174) | def __init__(self, config: OpenLlamaConfig):
    method _shape (line 194) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 197) | def forward(
  class OpenLlamaDecoderLayer (line 274) | class OpenLlamaDecoderLayer(nn.Module):
    method __init__ (line 275) | def __init__(self, config: OpenLlamaConfig):
    method forward (line 289) | def forward(
  class OpenLlamaPreTrainedModel (line 365) | class OpenLlamaPreTrainedModel(PreTrainedModel):
    method _init_weights (line 372) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 386) | def _set_gradient_checkpointing(self, module, value=False):
  class OpenLlamaModel (line 459) | class OpenLlamaModel(OpenLlamaPreTrainedModel):
    method __init__ (line 467) | def __init__(self, config: OpenLlamaConfig):
    method get_input_embeddings (line 484) | def get_input_embeddings(self):
    method set_input_embeddings (line 487) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 491) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 515) | def forward(
  class OpenLlamaForCausalLM (line 647) | class OpenLlamaForCausalLM(OpenLlamaPreTrainedModel):
    method __init__ (line 648) | def __init__(self, config):
    method get_input_embeddings (line 659) | def get_input_embeddings(self):
    method set_input_embeddings (line 662) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 665) | def get_output_embeddings(self):
    method set_output_embeddings (line 668) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 671) | def set_decoder(self, decoder):
    method get_decoder (line 674) | def get_decoder(self):
    method forward (line 679) | def forward(
    method prepare_inputs_for_generation (line 768) | def prepare_inputs_for_generation(
    method _reorder_cache (line 799) | def _reorder_cache(past_key_values, beam_idx):
  class OpenLlamaForSequenceClassification (line 822) | class OpenLlamaForSequenceClassification(OpenLlamaPreTrainedModel):
    method __init__ (line 825) | def __init__(self, config):
    method get_input_embeddings (line 834) | def get_input_embeddings(self):
    method set_input_embeddings (line 837) | def set_input_embeddings(self, value):
    method forward (line 841) | def forward(

FILE: transformers/models/openai/configuration_openai.py
  class OpenAIGPTConfig (line 27) | class OpenAIGPTConfig(PretrainedConfig):
    method __init__ (line 121) | def __init__(

FILE: transformers/models/openai/convert_openai_original_tf_checkpoint_to_pytorch.py
  function convert_openai_checkpoint_to_pytorch (line 29) | def convert_openai_checkpoint_to_pytorch(openai_checkpoint_folder_path, ...

FILE: transformers/models/openai/modeling_openai.py
  function load_tf_weights_in_openai_gpt (line 55) | def load_tf_weights_in_openai_gpt(model, config, openai_checkpoint_folde...
  class Attention (line 136) | class Attention(nn.Module):
    method __init__ (line 137) | def __init__(self, nx, n_positions, config, scale=False):
    method prune_heads (line 156) | def prune_heads(self, heads):
    method _attn (line 171) | def _attn(self, q, k, v, attention_mask=None, head_mask=None, output_a...
    method merge_heads (line 196) | def merge_heads(self, x):
    method split_heads (line 201) | def split_heads(self, x, k=False):
    method forward (line 209) | def forward(self, x, attention_mask=None, head_mask=None, output_atten...
  class MLP (line 227) | class MLP(nn.Module):
    method __init__ (line 228) | def __init__(self, n_state, config):  # in MLP: n_state=3072 (4 * n_embd)
    method forward (line 236) | def forward(self, x):
  class Block (line 242) | class Block(nn.Module):
    method __init__ (line 243) | def __init__(self, n_positions, config, scale=False):
    method forward (line 251) | def forward(self, x, attention_mask=None, head_mask=None, output_atten...
  class OpenAIGPTPreTrainedModel (line 268) | class OpenAIGPTPreTrainedModel(PreTrainedModel):
    method _init_weights (line 279) | def _init_weights(self, module):
  class OpenAIGPTDoubleHeadsModelOutput (line 297) | class OpenAIGPTDoubleHeadsModelOutput(ModelOutput):
  class OpenAIGPTModel (line 401) | class OpenAIGPTModel(OpenAIGPTPreTrainedModel):
    method __init__ (line 402) | def __init__(self, config):
    method get_input_embeddings (line 414) | def get_input_embeddings(self):
    method set_input_embeddings (line 417) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 420) | def _prune_heads(self, heads_to_prune):
    method forward (line 433) | def forward(
  class OpenAIGPTLMHeadModel (line 531) | class OpenAIGPTLMHeadModel(OpenAIGPTPreTrainedModel):
    method __init__ (line 534) | def __init__(self, config):
    method get_output_embeddings (line 542) | def get_output_embeddings(self):
    method set_output_embeddings (line 545) | def set_output_embeddings(self, new_embeddings):
    method forward (line 554) | def forward(
    method prepare_inputs_for_generation (line 609) | def prepare_inputs_for_generation(self, input_ids: torch.LongTensor, *...
  class OpenAIGPTDoubleHeadsModel (line 622) | class OpenAIGPTDoubleHeadsModel(OpenAIGPTPreTrainedModel):
    method __init__ (line 625) | def __init__(self, config):
    method get_output_embeddings (line 636) | def get_output_embeddings(self):
    method set_output_embeddings (line 639) | def set_output_embeddings(self, new_embeddings):
    method forward (line 644) | def forward(
  class OpenAIGPTForSequenceClassification (line 750) | class OpenAIGPTForSequenceClassification(OpenAIGPTPreTrainedModel):
    method __init__ (line 751) | def __init__(self, config):
    method forward (line 766) | def forward(

FILE: transformers/models/openai/modeling_tf_openai.py
  class TFAttention (line 63) | class TFAttention(tf.keras.layers.Layer):
    method __init__ (line 64) | def __init__(self, nx, config, scale=False, **kwargs):
    method prune_heads (line 83) | def prune_heads(self, heads):
    method causal_attention_mask (line 87) | def causal_attention_mask(nd, ns):
    method _attn (line 97) | def _attn(self, q, k, v, attention_mask, head_mask, output_attentions,...
    method merge_heads (line 127) | def merge_heads(self, x):
    method split_heads (line 133) | def split_heads(self, x):
    method call (line 139) | def call(self, x, attention_mask, head_mask, output_attentions, traini...
  class TFMLP (line 157) | class TFMLP(tf.keras.layers.Layer):
    method __init__ (line 158) | def __init__(self, n_state, config, **kwargs):
    method call (line 166) | def call(self, x, training=False):
  class TFBlock (line 173) | class TFBlock(tf.keras.layers.Layer):
    method __init__ (line 174) | def __init__(self, config, scale=False, **kwargs):
    method call (line 182) | def call(self, x, attention_mask, head_mask, output_attentions, traini...
  class TFOpenAIGPTMainLayer (line 195) | class TFOpenAIGPTMainLayer(tf.keras.layers.Layer):
    method __init__ (line 198) | def __init__(self, config, *inputs, **kwargs):
    method build (line 216) | def build(self, input_shape):
    method get_input_embeddings (line 226) | def get_input_embeddings(self):
    method set_input_embeddings (line 229) | def set_input_embeddings(self, value):
    method _prune_heads (line 233) | def _prune_heads(self, heads_to_prune):
    method call (line 240) | def call(
  class TFOpenAIGPTPreTrainedModel (line 351) | class TFOpenAIGPTPreTrainedModel(TFPreTrainedModel):
  class TFOpenAIGPTDoubleHeadsModelOutput (line 362) | class TFOpenAIGPTDoubleHeadsModelOutput(ModelOutput):
  class TFOpenAIGPTModel (line 492) | class TFOpenAIGPTModel(TFOpenAIGPTPreTrainedModel):
    method __init__ (line 493) | def __init__(self, config, *inputs, **kwargs):
    method call (line 504) | def call(
  class TFOpenAIGPTLMHeadModel (line 539) | class TFOpenAIGPTLMHeadModel(TFOpenAIGPTPreTrainedModel, TFCausalLanguag...
    method __init__ (line 540) | def __init__(self, config, *inputs, **kwargs):
    method get_output_embeddings (line 546) | def get_output_embeddings(self):
    method set_output_embeddings (line 549) | def set_output_embeddings(self, value):
    method call (line 559) | def call(
    method prepare_inputs_for_generation (line 613) | def prepare_inputs_for_generation(self, inputs, **kwargs):
  class TFOpenAIGPTDoubleHeadsModel (line 626) | class TFOpenAIGPTDoubleHeadsModel(TFOpenAIGPTPreTrainedModel):
    method __init__ (line 627) | def __init__(self, config, *inputs, **kwargs):
    method call (line 638) | def call(
    method input_signature (line 730) | def input_signature(self):
  class TFOpenAIGPTForSequenceClassification (line 753) | class TFOpenAIGPTForSequenceClassification(TFOpenAIGPTPreTrainedModel, T...
    method __init__ (line 754) | def __init__(self, config, *inputs, **kwargs):
    method call (line 772) | def call(

FILE: transformers/models/openai/tokenization_openai.py
  function whitespace_tokenize (line 46) | def whitespace_tokenize(text):
  class BasicTokenizer (line 56) | class BasicTokenizer(object):
    method __init__ (line 76) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 84) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 121) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 132) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 154) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 167) | def _is_chinese_char(self, cp):
    method _clean_text (line 191) | def _clean_text(self, text):
  function get_pairs (line 205) | def get_pairs(word):
  function text_standardize (line 218) | def text_standardize(text):
  class OpenAIGPTTokenizer (line 233) | class OpenAIGPTTokenizer(PreTrainedTokenizer):
    method __init__ (line 259) | def __init__(self, vocab_file, merges_file, unk_token="<unk>", **kwargs):
    method do_lower_case (line 284) | def do_lower_case(self):
    method vocab_size (line 288) | def vocab_size(self):
    method get_vocab (line 291) | def get_vocab(self):
    method bpe (line 294) | def bpe(self, token):
    method _tokenize (line 338) | def _tokenize(self, text):
    method _convert_token_to_id (line 353) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 357) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 361) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 366) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/openai/tokenization_openai_fast.py
  class OpenAIGPTTokenizerFast (line 40) | class OpenAIGPTTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 67) | def __init__(self, vocab_file=None, merges_file=None, tokenizer_file=N...
    method do_lower_case (line 71) | def do_lower_case(self):
    method save_vocabulary (line 74) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/opt/configuration_opt.py
  class OPTConfig (line 32) | class OPTConfig(PretrainedConfig):
    method __init__ (line 99) | def __init__(

FILE: transformers/models/opt/convert_opt_original_pytorch_checkpoint_to_pytorch.py
  function load_checkpoint (line 31) | def load_checkpoint(checkpoint_path):
  function convert_opt_checkpoint (line 81) | def convert_opt_checkpoint(checkpoint_path, pytorch_dump_folder_path, co...

FILE: transformers/models/opt/modeling_flax_opt.py
  class FlaxOPTAttention (line 108) | class FlaxOPTAttention(nn.Module):
    method setup (line 117) | def setup(self) -> None:
    method _split_heads (line 143) | def _split_heads(self, hidden_states):
    method _merge_heads (line 146) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 150) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 181) | def __call__(
  class FlaxOPTDecoderLayer (line 275) | class FlaxOPTDecoderLayer(nn.Module):
    method setup (line 279) | def setup(self) -> None:
    method __call__ (line 304) | def __call__(
  class FlaxOPTDecoderLayerCollection (line 360) | class FlaxOPTDecoderLayerCollection(nn.Module):
    method setup (line 364) | def setup(self):
    method __call__ (line 371) | def __call__(
  class FlaxOPTLearnedPositionalEmbedding (line 404) | class FlaxOPTLearnedPositionalEmbedding(nn.Embed):
    method setup (line 409) | def setup(self):
    method __call__ (line 415) | def __call__(self, positions):
  class FlaxOPTDecoder (line 421) | class FlaxOPTDecoder(nn.Module):
    method setup (line 426) | def setup(self):
    method __call__ (line 465) | def __call__(
  class FlaxOPTPreTrainedModel (line 517) | class FlaxOPTPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 522) | def __init__(
    method init_weights (line 534) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 564) | def init_cache(self, batch_size, max_length):
    method __call__ (line 583) | def __call__(
  class FlaxOPTModule (line 647) | class FlaxOPTModule(nn.Module):
    method setup (line 651) | def setup(self):
    method _get_decoder_module (line 654) | def _get_decoder_module(self):
    method __call__ (line 657) | def __call__(
  class FlaxOPTModel (line 690) | class FlaxOPTModel(FlaxOPTPreTrainedModel):
  class FlaxOPTForCausalLMModule (line 703) | class FlaxOPTForCausalLMModule(nn.Module):
    method setup (line 707) | def setup(self):
    method __call__ (line 716) | def __call__(
  class FlaxOPTForCausalLM (line 763) | class FlaxOPTForCausalLM(FlaxOPTPreTrainedModel):
    method prepare_inputs_for_generation (line 766) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 788) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/opt/modeling_opt.py
  function _make_causal_mask (line 68) | def _make_causal_mask(
  function _expand_mask (line 85) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class OPTLearnedPositionalEmbedding (line 99) | class OPTLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 104) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 110) | def forward(self, attention_mask: torch.LongTensor, past_key_values_le...
  class OPTAttention (line 123) | class OPTAttention(nn.Module):
    method __init__ (line 126) | def __init__(
    method _shape (line 153) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 156) | def forward(
  class OPTDecoderLayer (line 277) | class OPTDecoderLayer(nn.Module):
    method __init__ (line 278) | def __init__(self, config: OPTConfig):
    method forward (line 299) | def forward(
  class OPTPreTrainedModel (line 398) | class OPTPreTrainedModel(PreTrainedModel):
    method _init_weights (line 405) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 416) | def _set_gradient_checkpointing(self, module, value=False):
  class OPTDecoder (line 483) | class OPTDecoder(OPTPreTrainedModel):
    method __init__ (line 491) | def __init__(self, config: OPTConfig):
    method get_input_embeddings (line 528) | def get_input_embeddings(self):
    method set_input_embeddings (line 531) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 535) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 558) | def forward(
  class OPTModel (line 753) | class OPTModel(OPTPreTrainedModel):
    method __init__ (line 754) | def __init__(self, config: OPTConfig):
    method get_input_embeddings (line 760) | def get_input_embeddings(self):
    method set_input_embeddings (line 763) | def set_input_embeddings(self, value):
    method get_decoder (line 766) | def get_decoder(self):
    method forward (line 776) | def forward(
  class OPTForCausalLM (line 819) | class OPTForCausalLM(OPTPreTrainedModel):
    method __init__ (line 822) | def __init__(self, config):
    method get_input_embeddings (line 832) | def get_input_embeddings(self):
    method set_input_embeddings (line 835) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 838) | def get_output_embeddings(self):
    method set_output_embeddings (line 841) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 844) | def set_decoder(self, decoder):
    method get_decoder (line 847) | def get_decoder(self):
    method forward (line 851) | def forward(
    method prepare_inputs_for_generation (line 982) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1004) | def _reorder_cache(past_key_values, beam_idx):
  class OPTForSequenceClassification (line 1026) | class OPTForSequenceClassification(OPTPreTrainedModel):
    method __init__ (line 1029) | def __init__(self, config: OPTConfig):
    method forward (line 1046) | def forward(
    method get_input_embeddings (line 1134) | def get_input_embeddings(self):
    method set_input_embeddings (line 1137) | def set_input_embeddings(self, value):
  class OPTForQuestionAnswering (line 1148) | class OPTForQuestionAnswering(OPTPreTrainedModel):
    method __init__ (line 1151) | def __init__(self, config: OPTConfig):
    method forward (line 1161) | def forward(
    method get_input_embeddings (line 1267) | def get_input_embeddings(self):
    method set_input_embeddings (line 1270) | def set_input_embeddings(self, value):

FILE: transformers/models/opt/modeling_tf_opt.py
  function _make_causal_mask (line 63) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 81) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFOPTLearnedPositionalEmbedding (line 94) | class TFOPTLearnedPositionalEmbedding(TFSharedEmbeddings):
    method __init__ (line 99) | def __init__(self, num_embeddings: int, embedding_dim: int, **kwargs):
    method call (line 105) | def call(self, attention_mask, past_key_values_length: int = 0):
  class TFOPTAttention (line 119) | class TFOPTAttention(tf.keras.layers.Layer):
    method __init__ (line 122) | def __init__(
    method _shape (line 150) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 153) | def call(
  class TFOPTDecoderLayer (line 272) | class TFOPTDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 273) | def __init__(self, config: OPTConfig, **kwargs):
    method call (line 292) | def call(
  class TFOPTPreTrainedModel (line 404) | class TFOPTPreTrainedModel(TFPreTrainedModel):
  class TFOPTDecoder (line 464) | class TFOPTDecoder(tf.keras.layers.Layer):
    method __init__ (line 467) | def __init__(self, config: OPTConfig, **kwargs):
    method get_embed_tokens (line 501) | def get_embed_tokens(self):
    method set_embed_tokens (line 504) | def set_embed_tokens(self, embed_tokens):
    method set_input_embeddings (line 507) | def set_input_embeddings(self, new_embeddings):
    method get_input_embeddings (line 511) | def get_input_embeddings(self):
    method _prepare_decoder_attention_mask (line 514) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method call (line 531) | def call(
  class TFOPTMainLayer (line 696) | class TFOPTMainLayer(tf.keras.layers.Layer):
    method __init__ (line 699) | def __init__(self, config: OPTConfig, **kwargs):
    method get_input_embeddings (line 704) | def get_input_embeddings(self):
    method set_input_embeddings (line 707) | def set_input_embeddings(self, new_embeddings):
    method call (line 711) | def call(
  class TFOPTModel (line 761) | class TFOPTModel(TFOPTPreTrainedModel):
    method __init__ (line 764) | def __init__(self, config: OPTConfig, **kwargs):
    method get_input_embeddings (line 769) | def get_input_embeddings(self):
    method set_input_embeddings (line 772) | def set_input_embeddings(self, new_embeddings):
    method call (line 783) | def call(
    method serving_output (line 827) | def serving_output(self, output):
  class TFOPTForCausalLM (line 847) | class TFOPTForCausalLM(TFOPTPreTrainedModel, TFCausalLanguageModelingLoss):
    method __init__ (line 850) | def __init__(self, config: OPTConfig, **kwargs):
    method get_output_embeddings (line 855) | def get_output_embeddings(self):
    method prepare_inputs_for_generation (line 858) | def prepare_inputs_for_generation(self, inputs, past_key_values=None, ...
    method call (line 880) | def call(
    method serving_output (line 992) | def serving_output(self, output):

FILE: transformers/models/owlvit/configuration_owlvit.py
  class OwlViTTextConfig (line 41) | class OwlViTTextConfig(PretrainedConfig):
    method __init__ (line 96) | def __init__(
    method from_pretrained (line 129) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class OwlViTVisionConfig (line 145) | class OwlViTVisionConfig(PretrainedConfig):
    method __init__ (line 200) | def __init__(
    method from_pretrained (line 232) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class OwlViTConfig (line 248) | class OwlViTConfig(PretrainedConfig):
    method __init__ (line 275) | def __init__(
    method from_pretrained (line 303) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
    method from_text_vision_configs (line 315) | def from_text_vision_configs(cls, text_config: Dict, vision_config: Di...
    method to_dict (line 329) | def to_dict(self):
  class OwlViTOnnxConfig (line 343) | class OwlViTOnnxConfig(OnnxConfig):
    method inputs (line 345) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 355) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 366) | def atol_for_validation(self) -> float:
    method generate_dummy_inputs (line 369) | def generate_dummy_inputs(
    method default_onnx_opset (line 385) | def default_onnx_opset(self) -> int:

FILE: transformers/models/owlvit/convert_owlvit_original_flax_to_hf.py
  function flatten_nested_dict (line 79) | def flatten_nested_dict(params, parent_key="", sep="/"):
  function to_f32 (line 92) | def to_f32(params):
  function copy_attn_layer (line 96) | def copy_attn_layer(hf_attn_layer, pt_attn_layer):
  function copy_mlp (line 116) | def copy_mlp(hf_mlp, pt_mlp):
  function copy_linear (line 121) | def copy_linear(hf_linear, pt_linear):
  function copy_layer (line 126) | def copy_layer(hf_layer, pt_layer):
  function copy_layers (line 138) | def copy_layers(hf_layers, pt_layers):
  function copy_encoder (line 143) | def copy_encoder(hf_encoder, pt_model):
  function copy_text_model_and_projection (line 155) | def copy_text_model_and_projection(hf_model, pt_model):
  function copy_vision_model_and_projection (line 163) | def copy_vision_model_and_projection(hf_model, pt_model):
  function copy_class_merge_token (line 180) | def copy_class_merge_token(hf_model, flax_params):
  function copy_class_box_heads (line 189) | def copy_class_box_heads(hf_model, flax_params):
  function copy_flax_attn_params (line 227) | def copy_flax_attn_params(hf_backbone, flax_attn_params):
  function _convert_attn_layers (line 257) | def _convert_attn_layers(params):
  function convert_clip_backbone (line 276) | def convert_clip_backbone(flax_params, torch_config):
  function convert_owlvit_checkpoint (line 326) | def convert_owlvit_checkpoint(pt_backbone, flax_params, attn_params, pyt...

FILE: transformers/models/owlvit/feature_extraction_owlvit.py
  class OwlViTFeatureExtractor (line 26) | class OwlViTFeatureExtractor(OwlViTImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/owlvit/image_processing_owlvit.py
  function _upcast (line 52) | def _upcast(t):
  function box_area (line 60) | def box_area(boxes):
  function box_iou (line 75) | def box_iou(boxes1, boxes2):
  class OwlViTImageProcessor (line 91) | class OwlViTImageProcessor(BaseImageProcessor):
    method __init__ (line 129) | def __init__(
    method resize (line 168) | def resize(
    method center_crop (line 185) | def center_crop(
    method rescale (line 201) | def rescale(
    method normalize (line 213) | def normalize(
    method preprocess (line 226) | def preprocess(
    method post_process (line 338) | def post_process(self, outputs, target_sizes):
    method post_process_object_detection (line 384) | def post_process_object_detection(
    method post_process_image_guided_detection (line 440) | def post_process_image_guided_detection(self, outputs, threshold=0.6, ...

FILE: transformers/models/owlvit/modeling_owlvit.py
  function _expand_mask (line 58) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  function contrastive_loss (line 73) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  function owlvit_loss (line 78) | def owlvit_loss(similarity: torch.Tensor) -> torch.Tensor:
  class OwlViTOutput (line 85) | class OwlViTOutput(ModelOutput):
    method to_tuple (line 115) | def to_tuple(self) -> Tuple[Any]:
  function _upcast (line 123) | def _upcast(t: torch.Tensor) -> torch.Tensor:
  function box_area (line 132) | def box_area(boxes: torch.Tensor) -> torch.Tensor:
  function box_iou (line 149) | def box_iou(boxes1: torch.Tensor, boxes2: torch.Tensor) -> torch.Tensor:
  function generalized_box_iou (line 166) | def generalized_box_iou(boxes1, boxes2):
  class OwlViTObjectDetectionOutput (line 191) | class OwlViTObjectDetectionOutput(ModelOutput):
    method to_tuple (line 233) | def to_tuple(self) -> Tuple[Any]:
  class OwlViTImageGuidedObjectDetectionOutput (line 241) | class OwlViTImageGuidedObjectDetectionOutput(ModelOutput):
    method to_tuple (line 282) | def to_tuple(self) -> Tuple[Any]:
  class OwlViTVisionEmbeddings (line 289) | class OwlViTVisionEmbeddings(nn.Module):
    method __init__ (line 290) | def __init__(self, config: OwlViTVisionConfig):
    method forward (line 309) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class OwlViTTextEmbeddings (line 321) | class OwlViTTextEmbeddings(nn.Module):
    method __init__ (line 322) | def __init__(self, config: OwlViTTextConfig):
    method forward (line 330) | def forward(
  class OwlViTAttention (line 350) | class OwlViTAttention(nn.Module):
    method __init__ (line 353) | def __init__(self, config):
    method _shape (line 372) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 375) | def forward(
  class OwlViTMLP (line 458) | class OwlViTMLP(nn.Module):
    method __init__ (line 459) | def __init__(self, config):
    method forward (line 466) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class OwlViTEncoderLayer (line 474) | class OwlViTEncoderLayer(nn.Module):
    method __init__ (line 475) | def __init__(self, config: OwlViTConfig):
    method forward (line 483) | def forward(
  class OwlViTPreTrainedModel (line 524) | class OwlViTPreTrainedModel(PreTrainedModel):
    method _init_weights (line 536) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 578) | def _set_gradient_checkpointing(self, module, value=False):
  class OwlViTEncoder (line 691) | class OwlViTEncoder(nn.Module):
    method __init__ (line 700) | def __init__(self, config: OwlViTConfig):
    method forward (line 705) | def forward(
  function _make_causal_mask (line 787) | def _make_causal_mask(
  class OwlViTTextTransformer (line 804) | class OwlViTTextTransformer(nn.Module):
    method __init__ (line 805) | def __init__(self, config: OwlViTTextConfig):
    method forward (line 815) | def forward(
  class OwlViTTextModel (line 876) | class OwlViTTextModel(OwlViTPreTrainedModel):
    method __init__ (line 879) | def __init__(self, config: OwlViTTextConfig):
    method get_input_embeddings (line 885) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 888) | def set_input_embeddings(self, value):
    method forward (line 893) | def forward(
  class OwlViTVisionTransformer (line 928) | class OwlViTVisionTransformer(nn.Module):
    method __init__ (line 929) | def __init__(self, config: OwlViTVisionConfig):
    method forward (line 940) | def forward(
  class OwlViTVisionModel (line 986) | class OwlViTVisionModel(OwlViTPreTrainedModel):
    method __init__ (line 990) | def __init__(self, config: OwlViTVisionConfig):
    method get_input_embeddings (line 996) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 1001) | def forward(
  class OwlViTModel (line 1037) | class OwlViTModel(OwlViTPreTrainedModel):
    method __init__ (line 1040) | def __init__(self, config: OwlViTConfig):
    method get_text_features (line 1073) | def get_text_features(
    method get_image_features (line 1108) | def get_image_features(
    method forward (line 1154) | def forward(
  class OwlViTBoxPredictionHead (line 1251) | class OwlViTBoxPredictionHead(nn.Module):
    method __init__ (line 1252) | def __init__(self, config: OwlViTConfig):
    method forward (line 1261) | def forward(self, image_features: torch.Tensor) -> torch.FloatTensor:
  class OwlViTClassPredictionHead (line 1270) | class OwlViTClassPredictionHead(nn.Module):
    method __init__ (line 1271) | def __init__(self, config: OwlViTConfig):
    method forward (line 1282) | def forward(
  class OwlViTForObjectDetection (line 1319) | class OwlViTForObjectDetection(OwlViTPreTrainedModel):
    method __init__ (line 1322) | def __init__(self, config: OwlViTConfig):
    method normalize_grid_corner_coordinates (line 1332) | def normalize_grid_corner_coordinates(self, feature_map: torch.FloatTe...
    method compute_box_bias (line 1353) | def compute_box_bias(self, feature_map: torch.FloatTensor) -> torch.Fl...
    method box_predictor (line 1369) | def box_predictor(
    method class_predictor (line 1392) | def class_predictor(
    method image_text_embedder (line 1411) | def image_text_embedder(
    method image_embedder (line 1453) | def image_embedder(
    method embed_image_query (line 1485) | def embed_image_query(
    method image_guided_detection (line 1528) | def image_guided_detection(
    method forward (line 1623) | def forward(

FILE: transformers/models/owlvit/processing_owlvit.py
  class OwlViTProcessor (line 29) | class OwlViTProcessor(ProcessorMixin):
    method __init__ (line 45) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 62) | def __call__(self, text=None, images=None, query_images=None, padding=...
    method post_process (line 174) | def post_process(self, *args, **kwargs):
    method post_process_object_detection (line 181) | def post_process_object_detection(self, *args, **kwargs):
    method post_process_image_guided_detection (line 188) | def post_process_image_guided_detection(self, *args, **kwargs):
    method batch_decode (line 195) | def batch_decode(self, *args, **kwargs):
    method decode (line 202) | def decode(self, *args, **kwargs):
    method feature_extractor_class (line 210) | def feature_extractor_class(self):
    method feature_extractor (line 218) | def feature_extractor(self):

FILE: transformers/models/pegasus/configuration_pegasus.py
  class PegasusConfig (line 29) | class PegasusConfig(PretrainedConfig):
    method __init__ (line 104) | def __init__(
    method num_attention_heads (line 160) | def num_attention_heads(self) -> int:
    method hidden_size (line 164) | def hidden_size(self) -> int:

FILE: transformers/models/pegasus/convert_pegasus_tf_to_pytorch.py
  function rename_state_dict_key (line 48) | def rename_state_dict_key(k):
  function convert_pegasus (line 57) | def convert_pegasus(tf_weights: dict, cfg_updates: dict) -> PegasusForCo...
  function get_tf_weights_as_numpy (line 88) | def get_tf_weights_as_numpy(path="./ckpt/aeslc/model.ckpt-32000") -> Dict:
  function convert_pegasus_ckpt_to_pytorch (line 101) | def convert_pegasus_ckpt_to_pytorch(ckpt_path: str, save_dir: str):

FILE: transformers/models/pegasus/modeling_flax_pegasus.py
  function shift_tokens_right (line 213) | def shift_tokens_right(input_ids: np.array, pad_token_id: int, decoder_s...
  function create_sinusoidal_positions (line 226) | def create_sinusoidal_positions(n_pos, dim, dtype):
  class FlaxPegasusAttention (line 237) | class FlaxPegasusAttention(nn.Module):
    method setup (line 246) | def setup(self) -> None:
    method _split_heads (line 272) | def _split_heads(self, hidden_states):
    method _merge_heads (line 275) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 279) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 310) | def __call__(
  class FlaxPegasusEncoderLayer (line 405) | class FlaxPegasusEncoderLayer(nn.Module):
    method setup (line 409) | def setup(self) -> None:
    method __call__ (line 432) | def __call__(
  class FlaxPegasusEncoderLayerCollection (line 462) | class FlaxPegasusEncoderLayerCollection(nn.Module):
    method setup (line 466) | def setup(self):
    method __call__ (line 473) | def __call__(
  class FlaxPegasusDecoderLayer (line 517) | class FlaxPegasusDecoderLayer(nn.Module):
    method setup (line 521) | def setup(self) -> None:
    method __call__ (line 554) | def __call__(
  class FlaxPegasusDecoderLayerCollection (line 606) | class FlaxPegasusDecoderLayerCollection(nn.Module):
    method setup (line 610) | def setup(self):
    method __call__ (line 617) | def __call__(
  class FlaxPegasusEncoder (line 676) | class FlaxPegasusEncoder(nn.Module):
    method setup (line 681) | def setup(self):
    method __call__ (line 695) | def __call__(
  class FlaxPegasusDecoder (line 745) | class FlaxPegasusDecoder(nn.Module):
    method setup (line 750) | def setup(self):
    method __call__ (line 765) | def __call__(
  class FlaxPegasusModule (line 823) | class FlaxPegasusModule(nn.Module):
    method setup (line 827) | def setup(self):
    method _get_encoder_module (line 838) | def _get_encoder_module(self):
    method _get_decoder_module (line 841) | def _get_decoder_module(self):
    method __call__ (line 844) | def __call__(
  class FlaxPegasusPreTrainedModel (line 893) | class FlaxPegasusPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 898) | def __init__(
    method init_weights (line 910) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 944) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 987) | def encode(
    method decode (line 1050) | def decode(
    method __call__ (line 1163) | def __call__(
  class FlaxPegasusModel (line 1227) | class FlaxPegasusModel(FlaxPegasusPreTrainedModel):
  class FlaxPegasusForConditionalGenerationModule (line 1237) | class FlaxPegasusForConditionalGenerationModule(nn.Module):
    method setup (line 1242) | def setup(self):
    method _get_encoder_module (line 1252) | def _get_encoder_module(self):
    method _get_decoder_module (line 1255) | def _get_decoder_module(self):
    method __call__ (line 1258) | def __call__(
  class FlaxPegasusForConditionalGeneration (line 1312) | class FlaxPegasusForConditionalGeneration(FlaxPegasusPreTrainedModel):
    method decode (line 1318) | def decode(
    method prepare_inputs_for_generation (line 1453) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 1484) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/pegasus/modeling_pegasus.py
  function shift_tokens_right (line 60) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 77) | def _make_causal_mask(
  function _expand_mask (line 95) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class PegasusSinusoidalPositionalEmbedding (line 110) | class PegasusSinusoidalPositionalEmbedding(nn.Embedding):
    method __init__ (line 113) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method _init_weight (line 118) | def _init_weight(out: nn.Parameter) -> nn.Parameter:
    method forward (line 135) | def forward(self, input_ids_shape: torch.Size, past_key_values_length:...
  class PegasusAttention (line 145) | class PegasusAttention(nn.Module):
    method __init__ (line 148) | def __init__(
    method _shape (line 175) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 178) | def forward(
  class PegasusEncoderLayer (line 300) | class PegasusEncoderLayer(nn.Module):
    method __init__ (line 301) | def __init__(self, config: PegasusConfig):
    method forward (line 317) | def forward(
  class PegasusDecoderLayer (line 369) | class PegasusDecoderLayer(nn.Module):
    method __init__ (line 370) | def __init__(self, config: PegasusConfig):
    method forward (line 396) | def forward(
  class PegasusPreTrainedModel (line 486) | class PegasusPreTrainedModel(PreTrainedModel):
    method _init_weights (line 491) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 504) | def _set_gradient_checkpointing(self, module, value=False):
  class PegasusEncoder (line 638) | class PegasusEncoder(PegasusPreTrainedModel):
    method __init__ (line 648) | def __init__(self, config: PegasusConfig, embed_tokens: Optional[nn.Em...
    method resize_position_embeddings (line 676) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_position_embeddings (line 699) | def get_position_embeddings(self) -> nn.Embedding:
    method forward (line 705) | def forward(
  class PegasusDecoder (line 839) | class PegasusDecoder(PegasusPreTrainedModel):
    method __init__ (line 848) | def __init__(self, config: PegasusConfig, embed_tokens: Optional[nn.Em...
    method get_input_embeddings (line 873) | def get_input_embeddings(self):
    method set_input_embeddings (line 876) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 880) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method resize_position_embeddings (line 903) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_position_embeddings (line 926) | def get_position_embeddings(self) -> nn.Embedding:
    method forward (line 932) | def forward(
  class PegasusModel (line 1153) | class PegasusModel(PegasusPreTrainedModel):
    method __init__ (line 1156) | def __init__(self, config: PegasusConfig):
    method get_input_embeddings (line 1168) | def get_input_embeddings(self):
    method set_input_embeddings (line 1171) | def set_input_embeddings(self, value):
    method get_encoder (line 1176) | def get_encoder(self):
    method get_decoder (line 1179) | def get_decoder(self):
    method resize_position_embeddings (line 1182) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_position_embeddings (line 1199) | def get_position_embeddings(self) -> Tuple[nn.Embedding]:
    method forward (line 1207) | def forward(
  class PegasusForConditionalGeneration (line 1304) | class PegasusForConditionalGeneration(PegasusPreTrainedModel):
    method __init__ (line 1316) | def __init__(self, config: PegasusConfig):
    method get_encoder (line 1325) | def get_encoder(self):
    method get_decoder (line 1328) | def get_decoder(self):
    method resize_token_embeddings (line 1331) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_final_logits_bias (line 1336) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_output_embeddings (line 1345) | def get_output_embeddings(self):
    method set_output_embeddings (line 1348) | def set_output_embeddings(self, new_embeddings):
    method resize_position_embeddings (line 1351) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_position_embeddings (line 1368) | def get_position_embeddings(self) -> Tuple[nn.Embedding]:
    method forward (line 1377) | def forward(
    method prepare_inputs_for_generation (line 1456) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1484) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 1488) | def _reorder_cache(past_key_values, beam_idx):
  class PegasusDecoderWrapper (line 1499) | class PegasusDecoderWrapper(PegasusPreTrainedModel):
    method __init__ (line 1505) | def __init__(self, config):
    method forward (line 1509) | def forward(self, *args, **kwargs):
  class PegasusForCausalLM (line 1513) | class PegasusForCausalLM(PegasusPreTrainedModel):
    method __init__ (line 1516) | def __init__(self, config):
    method get_input_embeddings (line 1528) | def get_input_embeddings(self):
    method set_input_embeddings (line 1531) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1534) | def get_output_embeddings(self):
    method set_output_embeddings (line 1537) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 1540) | def set_decoder(self, decoder):
    method get_decoder (line 1543) | def get_decoder(self):
    method get_position_embeddings (line 1546) | def get_position_embeddings(self) -> nn.Embedding:
    method resize_position_embeddings (line 1552) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method forward (line 1570) | def forward(
    method prepare_inputs_for_generation (line 1714) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1732) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/pegasus/modeling_tf_pegasus.py
  function shift_tokens_right (line 65) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int, decoder_...
  function _make_causal_mask (line 90) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 108) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFPegasusSinusoidalPositionalEmbedding (line 122) | class TFPegasusSinusoidalPositionalEmbedding(tf.keras.layers.Layer):
    method __init__ (line 125) | def __init__(self, num_positions: int, embedding_dim: int, **kwargs):
    method build (line 134) | def build(self, input_shape: tf.TensorShape):
    method _init_weight (line 153) | def _init_weight(n_pos: int, dim: int):
    method call (line 170) | def call(
  class TFPegasusAttention (line 181) | class TFPegasusAttention(tf.keras.layers.Layer):
    method __init__ (line 184) | def __init__(
    method _shape (line 212) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 215) | def call(
  class TFPegasusEncoderLayer (line 335) | class TFPegasusEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 336) | def __init__(self, config: PegasusConfig, **kwargs):
    method call (line 350) | def call(
  class TFPegasusDecoderLayer (line 392) | class TFPegasusDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 393) | def __init__(self, config: PegasusConfig, **kwargs):
    method call (line 420) | def call(
  class TFPegasusPreTrainedModel (line 501) | class TFPegasusPreTrainedModel(TFPreTrainedModel):
  class TFPegasusEncoder (line 651) | class TFPegasusEncoder(tf.keras.layers.Layer):
    method __init__ (line 661) | def __init__(self, config: PegasusConfig, embed_tokens: Optional[tf.ke...
    method get_embed_tokens (line 679) | def get_embed_tokens(self):
    method set_embed_tokens (line 682) | def set_embed_tokens(self, embed_tokens):
    method call (line 686) | def call(
  class TFPegasusDecoder (line 817) | class TFPegasusDecoder(tf.keras.layers.Layer):
    method __init__ (line 827) | def __init__(self, config: PegasusConfig, embed_tokens: Optional[tf.ke...
    method get_embed_tokens (line 844) | def get_embed_tokens(self):
    method set_embed_tokens (line 847) | def set_embed_tokens(self, embed_tokens):
    method call (line 851) | def call(
  class TFPegasusMainLayer (line 1052) | class TFPegasusMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1055) | def __init__(self, config: PegasusConfig, **kwargs):
    method get_input_embeddings (line 1071) | def get_input_embeddings(self):
    method set_input_embeddings (line 1074) | def set_input_embeddings(self, new_embeddings):
    method call (line 1080) | def call(
  class TFPegasusModel (line 1166) | class TFPegasusModel(TFPegasusPreTrainedModel):
    method __init__ (line 1167) | def __init__(self, config: PegasusConfig, *inputs, **kwargs):
    method get_encoder (line 1172) | def get_encoder(self):
    method get_decoder (line 1175) | def get_decoder(self):
    method call (line 1185) | def call(
    method serving_output (line 1229) | def serving_output(self, output):
  class BiasLayer (line 1250) | class BiasLayer(tf.keras.layers.Layer):
    method __init__ (line 1256) | def __init__(self, shape, initializer, trainable, name, **kwargs):
    method call (line 1263) | def call(self, x):
  class TFPegasusForConditionalGeneration (line 1271) | class TFPegasusForConditionalGeneration(TFPegasusPreTrainedModel, TFCaus...
    method __init__ (line 1277) | def __init__(self, config, *inputs, **kwargs):
    method get_decoder (line 1286) | def get_decoder(self):
    method get_encoder (line 1289) | def get_encoder(self):
    method get_output_embeddings (line 1292) | def get_output_embeddings(self):
    method set_output_embeddings (line 1295) | def set_output_embeddings(self, value):
    method get_bias (line 1298) | def get_bias(self):
    method set_bias (line 1301) | def set_bias(self, value):
    method call (line 1313) | def call(
    method serving_output (line 1395) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 1415) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1453) | def prepare_decoder_input_ids_from_labels(self, labels: tf.Tensor):

FILE: transformers/models/pegasus/tokenization_pegasus.py
  class PegasusTokenizer (line 41) | class PegasusTokenizer(PreTrainedTokenizer):
    method __init__ (line 105) | def __init__(
    method vocab_size (line 186) | def vocab_size(self) -> int:
    method get_vocab (line 189) | def get_vocab(self) -> Dict[str, int]:
    method __getstate__ (line 194) | def __getstate__(self):
    method __setstate__ (line 199) | def __setstate__(self, d):
    method _tokenize (line 209) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 213) | def _convert_token_to_id(self, token: str) -> int:
    method _convert_id_to_token (line 222) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 232) | def convert_tokens_to_string(self, tokens):
    method num_special_tokens_to_add (line 246) | def num_special_tokens_to_add(self, pair=False):
    method _special_token_mask (line 250) | def _special_token_mask(self, seq):
    method get_special_tokens_mask (line 256) | def get_special_tokens_mask(
    method build_inputs_with_special_tokens (line 267) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method save_vocabulary (line 292) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/pegasus/tokenization_pegasus_fast.py
  class PegasusTokenizerFast (line 51) | class PegasusTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 100) | def __init__(
    method _special_token_mask (line 157) | def _special_token_mask(self, seq):
    method get_special_tokens_mask (line 169) | def get_special_tokens_mask(
    method build_inputs_with_special_tokens (line 180) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method save_vocabulary (line 201) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/pegasus_x/configuration_pegasus_x.py
  class PegasusXConfig (line 30) | class PegasusXConfig(PretrainedConfig):
    method __init__ (line 110) | def __init__(
    method num_attention_heads (line 174) | def num_attention_heads(self) -> int:
    method hidden_size (line 178) | def hidden_size(self) -> int:

FILE: transformers/models/pegasus_x/modeling_pegasus_x.py
  class DimensionInfo (line 60) | class DimensionInfo:
  function shift_tokens_right (line 78) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 95) | def _make_causal_mask(
  function _expand_mask (line 113) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class PegasusXSinusoidalPositionalEmbedding (line 127) | class PegasusXSinusoidalPositionalEmbedding(nn.Module):
    method __init__ (line 130) | def __init__(self, embed_dim, max_scale: int = 10000.0):
    method forward (line 136) | def forward(self, input_embeds: torch.Tensor, past_key_values_length: ...
  class PegasusXAttention (line 154) | class PegasusXAttention(nn.Module):
    method __init__ (line 157) | def __init__(
    method _shape (line 184) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 187) | def forward(
  class PegasusXGlobalLocalAttention (line 308) | class PegasusXGlobalLocalAttention(nn.Module):
    method __init__ (line 311) | def __init__(
    method _shape (line 339) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 342) | def forward(
    method compute_global_attention_representations (line 435) | def compute_global_attention_representations(
    method compute_local_attention_representations (line 478) | def compute_local_attention_representations(
  class PegasusXEncoderLayer (line 542) | class PegasusXEncoderLayer(nn.Module):
    method __init__ (line 543) | def __init__(self, stagger_blocks_this_layer: bool, config: PegasusXCo...
    method forward (line 563) | def forward(
    method pad_local_tokens (line 635) | def pad_local_tokens(cls, hidden_states, attention_mask, block_size):
    method unpad_local_tokens (line 651) | def unpad_local_tokens(cls, padded_hidden_states, block_size):
  class PegasusXDecoderLayer (line 657) | class PegasusXDecoderLayer(nn.Module):
    method __init__ (line 658) | def __init__(self, config: PegasusXConfig):
    method forward (line 686) | def forward(
  class PegasusXPreTrainedModel (line 769) | class PegasusXPreTrainedModel(PreTrainedModel):
    method _init_weights (line 774) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 783) | def _set_gradient_checkpointing(self, module, value=False):
  class PegasusXEncoder (line 901) | class PegasusXEncoder(PegasusXPreTrainedModel):
    method __init__ (line 911) | def __init__(self, config: PegasusXConfig, embed_tokens: Optional[nn.E...
    method resize_position_embeddings (line 942) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_position_embeddings (line 961) | def get_position_embeddings(self) -> nn.Embedding:
    method forward (line 967) | def forward(
  class PegasusXDecoder (line 1110) | class PegasusXDecoder(PegasusXPreTrainedModel):
    method __init__ (line 1119) | def __init__(self, config: PegasusXConfig, embed_tokens: Optional[nn.E...
    method get_input_embeddings (line 1139) | def get_input_embeddings(self):
    method set_input_embeddings (line 1142) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 1146) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method resize_position_embeddings (line 1169) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_position_embeddings (line 1188) | def get_position_embeddings(self) -> nn.Embedding:
    method forward (line 1194) | def forward(
  class PegasusXModel (line 1388) | class PegasusXModel(PegasusXPreTrainedModel):
    method __init__ (line 1391) | def __init__(self, config: PegasusXConfig):
    method get_input_embeddings (line 1403) | def get_input_embeddings(self):
    method set_input_embeddings (line 1406) | def set_input_embeddings(self, value):
    method get_encoder (line 1411) | def get_encoder(self):
    method get_decoder (line 1414) | def get_decoder(self):
    method resize_position_embeddings (line 1417) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_position_embeddings (line 1434) | def get_position_embeddings(self) -> Tuple[nn.Embedding]:
    method forward (line 1442) | def forward(
  class PegasusXForConditionalGeneration (line 1531) | class PegasusXForConditionalGeneration(PegasusXPreTrainedModel):
    method __init__ (line 1542) | def __init__(self, config: PegasusXConfig):
    method get_encoder (line 1550) | def get_encoder(self):
    method get_decoder (line 1553) | def get_decoder(self):
    method resize_token_embeddings (line 1556) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method get_output_embeddings (line 1560) | def get_output_embeddings(self):
    method set_output_embeddings (line 1563) | def set_output_embeddings(self, new_embeddings):
    method resize_position_embeddings (line 1566) | def resize_position_embeddings(self, new_num_position_embeddings: int):
    method get_position_embeddings (line 1583) | def get_position_embeddings(self) -> Tuple[nn.Embedding]:
    method forward (line 1592) | def forward(
    method prepare_inputs_for_generation (line 1665) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1687) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 1691) | def _reorder_cache(past_key_values, beam_idx):
  class PegasusXDecoderWrapper (line 1702) | class PegasusXDecoderWrapper(PegasusXPreTrainedModel):
    method __init__ (line 1708) | def __init__(self, config):
    method forward (line 1712) | def forward(self, *args, **kwargs):

FILE: transformers/models/perceiver/configuration_perceiver.py
  class PerceiverConfig (line 36) | class PerceiverConfig(PretrainedConfig):
    method __init__ (line 120) | def __init__(
  class PerceiverOnnxConfig (line 182) | class PerceiverOnnxConfig(OnnxConfig):
    method inputs (line 184) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 197) | def atol_for_validation(self) -> float:
    method generate_dummy_inputs (line 200) | def generate_dummy_inputs(

FILE: transformers/models/perceiver/convert_perceiver_haiku_to_pytorch.py
  function prepare_img (line 48) | def prepare_img():
  function rename_keys (line 55) | def rename_keys(state_dict, architecture):
  function convert_perceiver_checkpoint (line 264) | def convert_perceiver_checkpoint(pickle_file, pytorch_dump_folder_path, ...

FILE: transformers/models/perceiver/feature_extraction_perceiver.py
  class PerceiverFeatureExtractor (line 26) | class PerceiverFeatureExtractor(PerceiverImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/perceiver/image_processing_perceiver.py
  class PerceiverImageProcessor (line 44) | class PerceiverImageProcessor(BaseImageProcessor):
    method __init__ (line 83) | def __init__(
    method center_crop (line 114) | def center_crop(
    method resize (line 149) | def resize(
    method rescale (line 177) | def rescale(
    method normalize (line 197) | def normalize(
    method preprocess (line 220) | def preprocess(

FILE: transformers/models/perceiver/modeling_perceiver.py
  class PerceiverModelOutput (line 61) | class PerceiverModelOutput(ModelOutput):
  class PerceiverDecoderOutput (line 92) | class PerceiverDecoderOutput(ModelOutput):
  class PerceiverMaskedLMOutput (line 110) | class PerceiverMaskedLMOutput(ModelOutput):
  class PerceiverClassifierOutput (line 141) | class PerceiverClassifierOutput(ModelOutput):
  class PerceiverEmbeddings (line 172) | class PerceiverEmbeddings(nn.Module):
    method __init__ (line 175) | def __init__(self, config):
    method forward (line 179) | def forward(self, batch_size: int):
  class PerceiverSelfAttention (line 183) | class PerceiverSelfAttention(nn.Module):
    method __init__ (line 186) | def __init__(
    method transpose_for_scores (line 227) | def transpose_for_scores(self, x, channels_per_head):
    method forward (line 232) | def forward(
  class PerceiverSelfOutput (line 298) | class PerceiverSelfOutput(nn.Module):
    method __init__ (line 299) | def __init__(self, config, input_channels, output_channels):
    method forward (line 303) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class PerceiverAttention (line 308) | class PerceiverAttention(nn.Module):
    method __init__ (line 311) | def __init__(
    method prune_heads (line 359) | def prune_heads(self, heads):
    method forward (line 377) | def forward(
  class PerceiverMLP (line 408) | class PerceiverMLP(nn.Module):
    method __init__ (line 411) | def __init__(self, config, input_size, widening_factor):
    method forward (line 420) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class PerceiverLayer (line 427) | class PerceiverLayer(nn.Module):
    method __init__ (line 428) | def __init__(
    method forward (line 456) | def forward(
    method feed_forward_chunk (line 487) | def feed_forward_chunk(self, attention_output):
  class PerceiverEncoder (line 493) | class PerceiverEncoder(nn.Module):
    method __init__ (line 496) | def __init__(self, config, kv_dim=None):
    method forward (line 543) | def forward(
  class PerceiverPreTrainedModel (line 608) | class PerceiverPreTrainedModel(PreTrainedModel):
    method _init_weights (line 618) | def _init_weights(self, module):
  class PerceiverModel (line 716) | class PerceiverModel(PerceiverPreTrainedModel):
    method __init__ (line 717) | def __init__(
    method get_input_embeddings (line 738) | def get_input_embeddings(self):
    method set_input_embeddings (line 741) | def set_input_embeddings(self, value):
    method _prune_heads (line 744) | def _prune_heads(self, heads_to_prune):
    method forward (line 754) | def forward(
  class PerceiverForMaskedLM (line 954) | class PerceiverForMaskedLM(PerceiverPreTrainedModel):
    method __init__ (line 955) | def __init__(self, config: PerceiverConfig):
    method forward (line 988) | def forward(
  class PerceiverForSequenceClassification (line 1089) | class PerceiverForSequenceClassification(PerceiverPreTrainedModel):
    method __init__ (line 1090) | def __init__(self, config):
    method forward (line 1112) | def forward(
  class PerceiverForImageClassificationLearned (line 1214) | class PerceiverForImageClassificationLearned(PerceiverPreTrainedModel):
    method __init__ (line 1215) | def __init__(self, config):
    method forward (line 1247) | def forward(
  class PerceiverForImageClassificationFourier (line 1357) | class PerceiverForImageClassificationFourier(PerceiverPreTrainedModel):
    method __init__ (line 1358) | def __init__(self, config):
    method forward (line 1391) | def forward(
  class PerceiverForImageClassificationConvProcessing (line 1500) | class PerceiverForImageClassificationConvProcessing(PerceiverPreTrainedM...
    method __init__ (line 1501) | def __init__(self, config):
    method forward (line 1535) | def forward(
  class PerceiverForOpticalFlow (line 1644) | class PerceiverForOpticalFlow(PerceiverPreTrainedModel):
    method __init__ (line 1645) | def __init__(self, config):
    method forward (line 1696) | def forward(
  class PerceiverForMultimodalAutoencoding (line 1789) | class PerceiverForMultimodalAutoencoding(PerceiverPreTrainedModel):
    method __init__ (line 1790) | def __init__(self, config: PerceiverConfig):
    method forward (line 1907) | def forward(
  function build_position_encoding (line 1997) | def build_position_encoding(
  class PerceiverAbstractDecoder (line 2034) | class PerceiverAbstractDecoder(nn.Module, metaclass=abc.ABCMeta):
    method decoder_query (line 2038) | def decoder_query(self, inputs, modality_sizes=None, inputs_without_po...
    method num_query_channels (line 2043) | def num_query_channels(self):
    method forward (line 2047) | def forward(self, query, z, query_mask=None):
  class PerceiverProjectionDecoder (line 2051) | class PerceiverProjectionDecoder(PerceiverAbstractDecoder):
    method __init__ (line 2060) | def __init__(self, config):
    method decoder_query (line 2064) | def decoder_query(self, inputs, modality_sizes=None, inputs_without_po...
    method forward (line 2067) | def forward(
  class PerceiverBasicDecoder (line 2077) | class PerceiverBasicDecoder(PerceiverAbstractDecoder):
    method __init__ (line 2113) | def __init__(
    method num_query_channels (line 2171) | def num_query_channels(self) -> int:
    method decoder_query (line 2184) | def decoder_query(self, inputs, modality_sizes=None, inputs_without_po...
    method forward (line 2232) | def forward(
  class PerceiverClassificationDecoder (line 2263) | class PerceiverClassificationDecoder(PerceiverAbstractDecoder):
    method __init__ (line 2274) | def __init__(self, config, **decoder_kwargs):
    method num_query_channels (line 2286) | def num_query_channels(self) -> int:
    method decoder_query (line 2289) | def decoder_query(self, inputs, modality_sizes=None, inputs_without_po...
    method forward (line 2294) | def forward(
  class PerceiverOpticalFlowDecoder (line 2309) | class PerceiverOpticalFlowDecoder(PerceiverAbstractDecoder):
    method __init__ (line 2312) | def __init__(self, config, output_image_shape, output_num_channels=2, ...
    method num_query_channels (line 2321) | def num_query_channels(self) -> int:
    method decoder_query (line 2324) | def decoder_query(self, inputs, modality_sizes=None, inputs_without_po...
    method forward (line 2329) | def forward(
  class PerceiverBasicVideoAutoencodingDecoder (line 2344) | class PerceiverBasicVideoAutoencodingDecoder(PerceiverAbstractDecoder):
    method __init__ (line 2358) | def __init__(
    method num_query_channels (line 2376) | def num_query_channels(self) -> int:
    method decoder_query (line 2379) | def decoder_query(self, inputs, modality_sizes=None, inputs_without_po...
    method forward (line 2387) | def forward(
  function restructure (line 2397) | def restructure(modality_sizes: ModalitySizeType, inputs: torch.Tensor) ...
  class PerceiverMultimodalDecoder (line 2421) | class PerceiverMultimodalDecoder(PerceiverAbstractDecoder):
    method __init__ (line 2447) | def __init__(
    method num_query_channels (line 2479) | def num_query_channels(self) -> int:
    method decoder_query (line 2484) | def decoder_query(self, inputs, modality_sizes, inputs_without_pos=Non...
    method forward (line 2518) | def forward(
  function space_to_depth (line 2532) | def space_to_depth(frames: torch.Tensor, temporal_block_size: int = 1, s...
  class Conv2dSamePadding (line 2592) | class Conv2dSamePadding(nn.Conv2d):
    method __init__ (line 2598) | def __init__(self, *args, **kwargs):
    method forward (line 2604) | def forward(self, input):
  class Conv2DDownsample (line 2608) | class Conv2DDownsample(nn.Module):
    method __init__ (line 2611) | def __init__(
    method forward (line 2638) | def forward(self, inputs: torch.Tensor) -> torch.Tensor:
  function generate_fourier_features (line 2646) | def generate_fourier_features(pos, num_bands, max_resolution=(224, 224),...
  function build_linear_positions (line 2698) | def build_linear_positions(index_dims, output_range=(-1.0, 1.0)):
  class PerceiverAbstractPositionEncoding (line 2721) | class PerceiverAbstractPositionEncoding(nn.Module, metaclass=abc.ABCMeta):
    method num_dimensions (line 2726) | def num_dimensions(self) -> int:
    method output_size (line 2730) | def output_size(self, *args, **kwargs) -> int:
    method forward (line 2734) | def forward(self, batch_size, pos):
  class PerceiverTrainablePositionEncoding (line 2738) | class PerceiverTrainablePositionEncoding(PerceiverAbstractPositionEncodi...
    method __init__ (line 2741) | def __init__(self, index_dims, num_channels=128):
    method num_dimensions (line 2749) | def num_dimensions(self) -> int:
    method output_size (line 2754) | def output_size(self, *args, **kwargs) -> int:
    method forward (line 2757) | def forward(self, batch_size: int) -> torch.Tensor:
  function _check_or_build_spatial_positions (line 2765) | def _check_or_build_spatial_positions(pos, index_dims, batch_size):
  class PerceiverFourierPositionEncoding (line 2795) | class PerceiverFourierPositionEncoding(PerceiverAbstractPositionEncoding):
    method __init__ (line 2798) | def __init__(self, num_bands, max_resolution, concat_pos=True, sine_on...
    method num_dimensions (line 2806) | def num_dimensions(self) -> int:
    method output_size (line 2809) | def output_size(self):
    method forward (line 2820) | def forward(
  class AbstractPreprocessor (line 2839) | class AbstractPreprocessor(nn.Module):
    method num_channels (line 2841) | def num_channels(self) -> int:
  class PerceiverTextPreprocessor (line 2846) | class PerceiverTextPreprocessor(AbstractPreprocessor):
    method __init__ (line 2857) | def __init__(self, config: PerceiverConfig) -> None:
    method num_channels (line 2864) | def num_channels(self) -> int:
    method forward (line 2867) | def forward(self, inputs: torch.LongTensor, pos: Optional[torch.Tensor...
  class PerceiverEmbeddingDecoder (line 2877) | class PerceiverEmbeddingDecoder(nn.Module):
    method __init__ (line 2886) | def __init__(self, config: PerceiverConfig) -> None:
    method forward (line 2892) | def forward(self, hidden_states: torch.Tensor, embedding_layer: torch....
  class PerceiverMultimodalPostprocessor (line 2901) | class PerceiverMultimodalPostprocessor(nn.Module):
    method __init__ (line 2914) | def __init__(self, modalities: Mapping[str, PostprocessorType], input_...
    method forward (line 2919) | def forward(
  class PerceiverClassificationPostprocessor (line 2935) | class PerceiverClassificationPostprocessor(nn.Module):
    method __init__ (line 2946) | def __init__(self, config: PerceiverConfig, in_channels: int) -> None:
    method forward (line 2950) | def forward(self, inputs, pos: Optional[torch.Tensor] = None, modality...
  class PerceiverAudioPostprocessor (line 2955) | class PerceiverAudioPostprocessor(nn.Module):
    method __init__ (line 2968) | def __init__(self, config: PerceiverConfig, in_channels: int, postproc...
    method forward (line 2977) | def forward(self, inputs: torch.Tensor, pos: Optional[torch.Tensor] = ...
  class PerceiverProjectionPostprocessor (line 2982) | class PerceiverProjectionPostprocessor(nn.Module):
    method __init__ (line 2994) | def __init__(self, in_channels: int, out_channels: int) -> None:
    method forward (line 2998) | def forward(self, inputs: torch.Tensor, pos: Optional[torch.Tensor] = ...
  class PerceiverImagePreprocessor (line 3003) | class PerceiverImagePreprocessor(AbstractPreprocessor):
    method __init__ (line 3040) | def __init__(
    method num_channels (line 3115) | def num_channels(self) -> int:
    method _build_network_inputs (line 3147) | def _build_network_inputs(self, inputs: torch.Tensor, network_input_is...
    method forward (line 3182) | def forward(self, inputs: torch.Tensor, pos: Optional[torch.Tensor] = ...
  class PerceiverOneHotPreprocessor (line 3232) | class PerceiverOneHotPreprocessor(AbstractPreprocessor):
    method __init__ (line 3241) | def __init__(self, config: PerceiverConfig) -> None:
    method num_channels (line 3246) | def num_channels(self) -> int:
    method forward (line 3249) | def forward(self, inputs: torch.Tensor, pos: Optional[torch.Tensor] = ...
  class PerceiverAudioPreprocessor (line 3258) | class PerceiverAudioPreprocessor(AbstractPreprocessor):
    method __init__ (line 3281) | def __init__(
    method num_channels (line 3315) | def num_channels(self) -> int:
    method _build_network_inputs (line 3325) | def _build_network_inputs(self, inputs):
    method forward (line 3346) | def forward(self, inputs: torch.Tensor, pos: Optional[torch.Tensor] = ...
  class PerceiverMultimodalPreprocessor (line 3355) | class PerceiverMultimodalPreprocessor(AbstractPreprocessor):
    method __init__ (line 3372) | def __init__(
    method num_channels (line 3393) | def num_channels(self) -> int:
    method forward (line 3398) | def forward(

FILE: transformers/models/perceiver/tokenization_perceiver.py
  class PerceiverTokenizer (line 27) | class PerceiverTokenizer(PreTrainedTokenizer):
    method __init__ (line 60) | def __init__(
    method get_vocab (line 103) | def get_vocab(self) -> Dict[str, int]:
    method vocab_size (line 112) | def vocab_size(self):
    method get_special_tokens_mask (line 115) | def get_special_tokens_mask(
    method build_inputs_with_special_tokens (line 143) | def build_inputs_with_special_tokens(
    method _tokenize (line 167) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 172) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 184) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 194) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 213) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/phobert/tokenization_phobert.py
  function get_pairs (line 52) | def get_pairs(word):
  class PhobertTokenizer (line 68) | class PhobertTokenizer(PreTrainedTokenizer):
    method __init__ (line 121) | def __init__(
    method build_inputs_with_special_tokens (line 164) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 190) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 218) | def create_token_type_ids_from_sequences(
    method vocab_size (line 243) | def vocab_size(self):
    method get_vocab (line 246) | def get_vocab(self):
    method bpe (line 249) | def bpe(self, token):
    method _tokenize (line 293) | def _tokenize(self, text):
    method _convert_token_to_id (line 303) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 307) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 311) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 316) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method add_from_file (line 341) | def add_from_file(self, f):

FILE: transformers/models/pix2struct/configuration_pix2struct.py
  class Pix2StructTextConfig (line 34) | class Pix2StructTextConfig(PretrainedConfig):
    method __init__ (line 102) | def __init__(
    method from_pretrained (line 153) | def from_pretrained(
  class Pix2StructVisionConfig (line 171) | class Pix2StructVisionConfig(PretrainedConfig):
    method __init__ (line 240) | def __init__(
    method from_pretrained (line 286) | def from_pretrained(
  class Pix2StructConfig (line 304) | class Pix2StructConfig(PretrainedConfig):
    method __init__ (line 355) | def __init__(
    method from_text_vision_configs (line 392) | def from_text_vision_configs(
    method to_dict (line 405) | def to_dict(self):

FILE: transformers/models/pix2struct/convert_pix2struct_original_pytorch_to_hf.py
  function get_flax_param (line 34) | def get_flax_param(t5x_checkpoint_path):
  function rename_and_convert_flax_params (line 40) | def rename_and_convert_flax_params(flax_dict):
  function convert_pix2struct_original_pytorch_checkpoint_to_hf (line 105) | def convert_pix2struct_original_pytorch_checkpoint_to_hf(

FILE: transformers/models/pix2struct/image_processing_pix2struct.py
  function torch_extract_patches (line 52) | def torch_extract_patches(image_tensor, patch_height, patch_width):
  function render_text (line 79) | def render_text(
  function render_header (line 147) | def render_header(image: np.ndarray, header: str, **kwargs):
  class Pix2StructImageProcessor (line 187) | class Pix2StructImageProcessor(BaseImageProcessor):
    method __init__ (line 210) | def __init__(
    method extract_flattened_patches (line 226) | def extract_flattened_patches(self, image: np.ndarray, max_patches: in...
    method normalize (line 300) | def normalize(
    method preprocess (line 323) | def preprocess(

FILE: transformers/models/pix2struct/modeling_pix2struct.py
  class Pix2StructLayerNorm (line 76) | class Pix2StructLayerNorm(nn.Module):
    method __init__ (line 77) | def __init__(self, hidden_size, eps=1e-6):
    method forward (line 85) | def forward(self, hidden_states):
  class Pix2StructVisionEmbeddings (line 117) | class Pix2StructVisionEmbeddings(nn.Module):
    method __init__ (line 124) | def __init__(self, config: Pix2StructConfig) -> None:
    method forward (line 133) | def forward(self, flattened_patches: torch.Tensor) -> torch.Tensor:
  class Pix2StructVisionAttention (line 153) | class Pix2StructVisionAttention(nn.Module):
    method __init__ (line 154) | def __init__(self, config):
    method forward (line 170) | def forward(
  class Pix2StructVisionMlp (line 248) | class Pix2StructVisionMlp(nn.Module):
    method __init__ (line 249) | def __init__(self, config: Pix2StructVisionConfig):
    method forward (line 257) | def forward(self, hidden_states):
  class Pix2StructVisionLayer (line 277) | class Pix2StructVisionLayer(nn.Module):
    method __init__ (line 278) | def __init__(self, config: Pix2StructConfig) -> None:
    method forward (line 287) | def forward(
  class Pix2StructVisionEncoder (line 320) | class Pix2StructVisionEncoder(nn.Module):
    method __init__ (line 321) | def __init__(self, config: Pix2StructConfig) -> None:
    method forward (line 327) | def forward(
  class Pix2StructPreTrainedModel (line 379) | class Pix2StructPreTrainedModel(PreTrainedModel):
    method dummy_inputs (line 388) | def dummy_inputs(self):
    method _init_weights (line 398) | def _init_weights(self, module):
    method _shift_right (line 478) | def _shift_right(self, input_ids):
  class Pix2StructVisionModel (line 548) | class Pix2StructVisionModel(Pix2StructPreTrainedModel):
    method __init__ (line 554) | def __init__(self, config: Pix2StructConfig):
    method _set_gradient_checkpointing (line 566) | def _set_gradient_checkpointing(self, module: Pix2StructVisionEncoder,...
    method get_input_embeddings (line 570) | def get_input_embeddings(self):
    method _prune_heads (line 573) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
    method forward (line 583) | def forward(
  class Pix2StructTextDenseGatedActDense (line 662) | class Pix2StructTextDenseGatedActDense(nn.Module):
    method __init__ (line 663) | def __init__(self, config: Pix2StructTextConfig):
    method forward (line 671) | def forward(self, hidden_states):
  class Pix2StructTextLayerFF (line 691) | class Pix2StructTextLayerFF(nn.Module):
    method __init__ (line 692) | def __init__(self, config: Pix2StructTextConfig):
    method forward (line 700) | def forward(self, hidden_states):
  class Pix2StructTextAttention (line 707) | class Pix2StructTextAttention(nn.Module):
    method __init__ (line 708) | def __init__(self, config: Pix2StructTextConfig, has_relative_attentio...
    method _relative_position_bucket (line 732) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 780) | def compute_bias(self, query_length, key_length, device=None):
    method forward (line 797) | def forward(
  class Pix2StructTextLayerSelfAttention (line 927) | class Pix2StructTextLayerSelfAttention(nn.Module):
    method __init__ (line 928) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 934) | def forward(
  class Pix2StructTextLayerCrossAttention (line 960) | class Pix2StructTextLayerCrossAttention(nn.Module):
    method __init__ (line 961) | def __init__(self, config):
    method forward (line 967) | def forward(
  class Pix2StructTextBlock (line 996) | class Pix2StructTextBlock(nn.Module):
    method __init__ (line 997) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 1008) | def forward(
  class Pix2StructTextModel (line 1317) | class Pix2StructTextModel(Pix2StructPreTrainedModel):
    method _set_gradient_checkpointing (line 1322) | def _set_gradient_checkpointing(self, module, value=False):
    method __init__ (line 1326) | def __init__(self, config):
    method _reorder_cache (line 1343) | def _reorder_cache(self, past_key_values, beam_idx):
    method get_input_embeddings (line 1373) | def get_input_embeddings(self):
    method set_input_embeddings (line 1376) | def set_input_embeddings(self, new_embeddings):
    method get_output_embeddings (line 1379) | def get_output_embeddings(self):
    method set_output_embeddings (line 1382) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1387) | def forward(
  class Pix2StructForConditionalGeneration (line 1596) | class Pix2StructForConditionalGeneration(Pix2StructPreTrainedModel):
    method __init__ (line 1608) | def __init__(self, config: Pix2StructConfig):
    method get_input_embeddings (line 1619) | def get_input_embeddings(self):
    method set_input_embeddings (line 1622) | def set_input_embeddings(self, new_embeddings):
    method get_output_embeddings (line 1625) | def get_output_embeddings(self) -> nn.Module:
    method set_output_embeddings (line 1628) | def set_output_embeddings(self, new_embeddings):
    method resize_token_embeddings (line 1631) | def resize_token_embeddings(self, new_num_tokens: Optional[int] = None...
    method get_decoder (line 1639) | def get_decoder(self):
    method get_encoder (line 1642) | def get_encoder(self):
    method forward (line 1647) | def forward(
    method prepare_inputs_for_generation (line 1789) | def prepare_inputs_for_generation(

FILE: transformers/models/pix2struct/processing_pix2struct.py
  class Pix2StructProcessor (line 26) | class Pix2StructProcessor(ProcessorMixin):
    method __init__ (line 44) | def __init__(self, image_processor, tokenizer):
    method __call__ (line 48) | def __call__(
    method batch_decode (line 144) | def batch_decode(self, *args, **kwargs):
    method decode (line 151) | def decode(self, *args, **kwargs):
    method model_input_names (line 159) | def model_input_names(self):

FILE: transformers/models/plbart/configuration_plbart.py
  class PLBartConfig (line 32) | class PLBartConfig(PretrainedConfig):
    method __init__ (line 109) | def __init__(
  class PLBartOnnxConfig (line 167) | class PLBartOnnxConfig(OnnxConfigWithPast):
    method inputs (line 169) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 178) | def outputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/plbart/convert_plbart_original_checkpoint_to_torch.py
  function remove_ignore_keys_ (line 23) | def remove_ignore_keys_(state_dict):
  function make_linear_from_emb (line 36) | def make_linear_from_emb(emb):
  function convert_fairseq_plbart_checkpoint_from_disk (line 43) | def convert_fairseq_plbart_checkpoint_from_disk(

FILE: transformers/models/plbart/modeling_plbart.py
  function shift_tokens_right (line 61) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int):
  function _make_causal_mask (line 82) | def _make_causal_mask(
  function _expand_mask (line 100) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class PLBartLearnedPositionalEmbedding (line 115) | class PLBartLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 120) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 126) | def forward(self, input_ids: torch.Tensor, past_key_values_length: int...
  class PLBartAttention (line 138) | class PLBartAttention(nn.Module):
    method __init__ (line 141) | def __init__(
    method _shape (line 168) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 171) | def forward(
  class PLBartEncoderLayer (line 293) | class PLBartEncoderLayer(nn.Module):
    method __init__ (line 294) | def __init__(self, config: PLBartConfig):
    method forward (line 310) | def forward(
  class PLBartDecoderLayer (line 362) | class PLBartDecoderLayer(nn.Module):
    method __init__ (line 363) | def __init__(self, config: PLBartConfig):
    method forward (line 389) | def forward(
  class PLBartClassificationHead (line 480) | class PLBartClassificationHead(nn.Module):
    method __init__ (line 483) | def __init__(
    method forward (line 495) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class PLBartPreTrainedModel (line 504) | class PLBartPreTrainedModel(PreTrainedModel):
    method _init_weights (line 510) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 521) | def _set_gradient_checkpointing(self, module, value=False):
  class PLBartEncoder (line 665) | class PLBartEncoder(PLBartPreTrainedModel):
    method __init__ (line 675) | def __init__(self, config: PLBartConfig, embed_tokens: Optional[nn.Emb...
    method get_input_embeddings (line 702) | def get_input_embeddings(self):
    method set_input_embeddings (line 705) | def set_input_embeddings(self, value):
    method forward (line 708) | def forward(
  class PLBartDecoder (line 843) | class PLBartDecoder(PLBartPreTrainedModel):
    method __init__ (line 852) | def __init__(self, config: PLBartConfig, embed_tokens: Optional[nn.Emb...
    method get_input_embeddings (line 876) | def get_input_embeddings(self):
    method set_input_embeddings (line 879) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 882) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 905) | def forward(
  class PLBartModel (line 1129) | class PLBartModel(PLBartPreTrainedModel):
    method __init__ (line 1132) | def __init__(self, config: PLBartConfig):
    method get_input_embeddings (line 1143) | def get_input_embeddings(self):
    method set_input_embeddings (line 1146) | def set_input_embeddings(self, value):
    method get_encoder (line 1151) | def get_encoder(self):
    method get_decoder (line 1154) | def get_decoder(self):
    method forward (line 1163) | def forward(
  class PLBartForConditionalGeneration (line 1246) | class PLBartForConditionalGeneration(PLBartPreTrainedModel):
    method __init__ (line 1257) | def __init__(self, config: PLBartConfig):
    method get_encoder (line 1265) | def get_encoder(self):
    method get_decoder (line 1268) | def get_decoder(self):
    method resize_token_embeddings (line 1271) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method _resize_final_logits_bias (line 1276) | def _resize_final_logits_bias(self, new_num_tokens: int) -> None:
    method get_output_embeddings (line 1285) | def get_output_embeddings(self):
    method set_output_embeddings (line 1288) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1294) | def forward(
    method prepare_inputs_for_generation (line 1369) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1397) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 1401) | def _reorder_cache(past_key_values, beam_idx):
  class PLBartForSequenceClassification (line 1418) | class PLBartForSequenceClassification(PLBartPreTrainedModel):
    method __init__ (line 1421) | def __init__(self, config: PLBartConfig, **kwargs):
    method forward (line 1441) | def forward(
  class PLBartDecoderWrapper (line 1541) | class PLBartDecoderWrapper(PLBartPreTrainedModel):
    method __init__ (line 1547) | def __init__(self, config):
    method forward (line 1551) | def forward(self, *args, **kwargs):
  class PLBartForCausalLM (line 1556) | class PLBartForCausalLM(PLBartPreTrainedModel):
    method __init__ (line 1559) | def __init__(self, config):
    method get_input_embeddings (line 1571) | def get_input_embeddings(self):
    method set_input_embeddings (line 1574) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1577) | def get_output_embeddings(self):
    method set_output_embeddings (line 1580) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 1583) | def set_decoder(self, decoder):
    method get_decoder (line 1586) | def get_decoder(self):
    method forward (line 1590) | def forward(
    method prepare_inputs_for_generation (line 1734) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1752) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/plbart/tokenization_plbart.py
  class PLBartTokenizer (line 106) | class PLBartTokenizer(PreTrainedTokenizer):
    method __init__ (line 176) | def __init__(
    method __getstate__ (line 268) | def __getstate__(self):
    method __setstate__ (line 274) | def __setstate__(self, d):
    method vocab_size (line 285) | def vocab_size(self):
    method src_lang (line 294) | def src_lang(self) -> str:
    method src_lang (line 298) | def src_lang(self, new_src_lang: str) -> None:
    method get_special_tokens_mask (line 303) | def get_special_tokens_mask(
    method build_inputs_with_special_tokens (line 333) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 360) | def create_token_type_ids_from_sequences(
    method _build_translation_inputs (line 384) | def _build_translation_inputs(
    method get_vocab (line 397) | def get_vocab(self):
    method _tokenize (line 402) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 405) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 414) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 420) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 425) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method prepare_seq2seq_batch (line 442) | def prepare_seq2seq_batch(
    method _switch_to_input_mode (line 454) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 457) | def _switch_to_target_mode(self):
    method set_src_lang_special_tokens (line 460) | def set_src_lang_special_tokens(self, src_lang) -> None:
    method set_tgt_lang_special_tokens (line 470) | def set_tgt_lang_special_tokens(self, lang: str) -> None:
    method _convert_lang_code_special_format (line 481) | def _convert_lang_code_special_format(self, lang: str) -> str:

FILE: transformers/models/poolformer/configuration_poolformer.py
  class PoolFormerConfig (line 34) | class PoolFormerConfig(PretrainedConfig):
    method __init__ (line 96) | def __init__(
  class PoolFormerOnnxConfig (line 135) | class PoolFormerOnnxConfig(OnnxConfig):
    method inputs (line 139) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 147) | def atol_for_validation(self) -> float:

FILE: transformers/models/poolformer/convert_poolformer_original_to_pytorch.py
  function replace_key_with_offset (line 35) | def replace_key_with_offset(key, offset, original_name, new_name):
  function rename_keys (line 49) | def rename_keys(state_dict):
  function prepare_img (line 85) | def prepare_img():
  function convert_poolformer_checkpoint (line 93) | def convert_poolformer_checkpoint(model_name, checkpoint_path, pytorch_d...

FILE: transformers/models/poolformer/feature_extraction_poolformer.py
  class PoolFormerFeatureExtractor (line 26) | class PoolFormerFeatureExtractor(PoolFormerImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/poolformer/image_processing_poolformer.py
  class PoolFormerImageProcessor (line 50) | class PoolFormerImageProcessor(BaseImageProcessor):
    method __init__ (line 103) | def __init__(
    method resize (line 136) | def resize(
    method center_crop (line 198) | def center_crop(
    method rescale (line 222) | def rescale(
    method normalize (line 242) | def normalize(
    method preprocess (line 265) | def preprocess(

FILE: transformers/models/poolformer/modeling_poolformer.py
  function drop_path (line 53) | def drop_path(input, drop_prob: float = 0.0, training: bool = False):
  class PoolFormerDropPath (line 74) | class PoolFormerDropPath(nn.Module):
    method __init__ (line 77) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 81) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 84) | def extra_repr(self) -> str:
  class PoolFormerEmbeddings (line 88) | class PoolFormerEmbeddings(nn.Module):
    method __init__ (line 93) | def __init__(self, hidden_size, num_channels, patch_size, stride, padd...
    method forward (line 102) | def forward(self, pixel_values):
  class PoolFormerGroupNorm (line 108) | class PoolFormerGroupNorm(nn.GroupNorm):
    method __init__ (line 113) | def __init__(self, num_channels, **kwargs):
  class PoolFormerPooling (line 117) | class PoolFormerPooling(nn.Module):
    method __init__ (line 118) | def __init__(self, pool_size):
    method forward (line 122) | def forward(self, hidden_states):
  class PoolFormerOutput (line 126) | class PoolFormerOutput(nn.Module):
    method __init__ (line 127) | def __init__(self, config, dropout_prob, hidden_size, intermediate_size):
    method forward (line 137) | def forward(self, hidden_states):
  class PoolFormerLayer (line 147) | class PoolFormerLayer(nn.Module):
    method __init__ (line 150) | def __init__(self, config, num_channels, pool_size, hidden_size, inter...
    method forward (line 168) | def forward(self, hidden_states):
  class PoolFormerEncoder (line 198) | class PoolFormerEncoder(nn.Module):
    method __init__ (line 199) | def __init__(self, config):
    method forward (line 242) | def forward(self, pixel_values, output_hidden_states=False, return_dic...
  class PoolFormerPreTrainedModel (line 264) | class PoolFormerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 275) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 285) | def _set_gradient_checkpointing(self, module, value=False):
  class PoolFormerModel (line 313) | class PoolFormerModel(PoolFormerPreTrainedModel):
    method __init__ (line 314) | def __init__(self, config):
    method get_input_embeddings (line 323) | def get_input_embeddings(self):
    method forward (line 334) | def forward(
  class PoolFormerFinalPooler (line 364) | class PoolFormerFinalPooler(nn.Module):
    method __init__ (line 365) | def __init__(self, config):
    method forward (line 369) | def forward(self, hidden_states):
  class PoolFormerForImageClassification (line 380) | class PoolFormerForImageClassification(PoolFormerPreTrainedModel):
    method __init__ (line 381) | def __init__(self, config):
    method forward (line 403) | def forward(

FILE: transformers/models/prophetnet/configuration_prophetnet.py
  class ProphetNetConfig (line 32) | class ProphetNetConfig(PretrainedConfig):
    method __init__ (line 107) | def __init__(
    method num_hidden_layers (line 174) | def num_hidden_layers(self) -> int:
    method num_hidden_layers (line 178) | def num_hidden_layers(self, value):

FILE: transformers/models/prophetnet/convert_prophetnet_original_pytorch_checkpoint_to_pytorch.py
  function convert_prophetnet_checkpoint_to_pytorch (line 38) | def convert_prophetnet_checkpoint_to_pytorch(prophetnet_checkpoint_path:...

FILE: transformers/models/prophetnet/modeling_prophetnet.py
  function softmax (line 178) | def softmax(hidden_state, dim, onnx_trace=False):
  function ngram_attention_bias (line 185) | def ngram_attention_bias(sequence_length, ngram, device, dtype):
  function compute_relative_buckets (line 202) | def compute_relative_buckets(num_buckets, max_distance, relative_positio...
  function compute_all_stream_relative_buckets (line 229) | def compute_all_stream_relative_buckets(num_buckets, max_distance, posit...
  class ProphetNetSeq2SeqLMOutput (line 253) | class ProphetNetSeq2SeqLMOutput(ModelOutput):
    method decoder_cross_attentions (line 328) | def decoder_cross_attentions(self):
  class ProphetNetSeq2SeqModelOutput (line 338) | class ProphetNetSeq2SeqModelOutput(ModelOutput):
    method decoder_cross_attentions (line 414) | def decoder_cross_attentions(self):
  class ProphetNetDecoderModelOutput (line 424) | class ProphetNetDecoderModelOutput(ModelOutput):
  class ProphetNetDecoderLMOutput (line 484) | class ProphetNetDecoderLMOutput(ModelOutput):
  class ProphetNetPreTrainedModel (line 545) | class ProphetNetPreTrainedModel(PreTrainedModel):
    method _init_weights (line 550) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 560) | def _set_gradient_checkpointing(self, module, value=False):
    method _shift_right (line 564) | def _shift_right(self, input_ids):
  class ProphetNetPositionalEmbeddings (line 587) | class ProphetNetPositionalEmbeddings(nn.Embedding):
    method __init__ (line 594) | def __init__(self, config: ProphetNetConfig) -> None:
    method forward (line 598) | def forward(self, inputs_shape, device, attention_mask=None, past_key_...
    method _forward (line 626) | def _forward(self, position_ids):
  class ProphetNetAttention (line 630) | class ProphetNetAttention(nn.Module):
    method __init__ (line 633) | def __init__(
    method _shape (line 657) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 660) | def forward(
  class ProphetNetFeedForward (line 759) | class ProphetNetFeedForward(nn.Module):
    method __init__ (line 764) | def __init__(self, config: ProphetNetConfig, ffn_dim: int):
    method forward (line 772) | def forward(self, hidden_states):
  class ProphetNetNgramSelfAttention (line 782) | class ProphetNetNgramSelfAttention(nn.Module):
    method __init__ (line 783) | def __init__(self, config: ProphetNetConfig):
    method _shape (line 812) | def _shape(self, tensor, seq_len, batch_size):
    method prepare_for_onnx_export_ (line 815) | def prepare_for_onnx_export_(self):
    method forward (line 818) | def forward(
    method get_main_relative_pos_embeddings (line 996) | def get_main_relative_pos_embeddings(
    method get_predict_relative_pos_embeddings (line 1041) | def get_predict_relative_pos_embeddings(
  class ProphetNetEncoderLayer (line 1101) | class ProphetNetEncoderLayer(nn.Module):
    method __init__ (line 1106) | def __init__(self, config: ProphetNetConfig):
    method forward (line 1116) | def forward(
  class ProphetNetDecoderLayer (line 1144) | class ProphetNetDecoderLayer(nn.Module):
    method __init__ (line 1149) | def __init__(self, config: ProphetNetConfig):
    method forward (line 1164) | def forward(
  class ProphetNetEncoder (line 1232) | class ProphetNetEncoder(ProphetNetPreTrainedModel):
    method __init__ (line 1239) | def __init__(self, config: ProphetNetConfig, word_embeddings: nn.Embed...
    method get_input_embeddings (line 1256) | def get_input_embeddings(self):
    method set_input_embeddings (line 1259) | def set_input_embeddings(self, value):
    method forward (line 1264) | def forward(
  class ProphetNetDecoder (line 1372) | class ProphetNetDecoder(ProphetNetPreTrainedModel):
    method __init__ (line 1379) | def __init__(self, config: ProphetNetConfig, word_embeddings: Optional...
    method get_input_embeddings (line 1403) | def get_input_embeddings(self):
    method set_input_embeddings (line 1406) | def set_input_embeddings(self, value):
    method forward (line 1411) | def forward(
    method compute_buffered_relative_buckets (line 1660) | def compute_buffered_relative_buckets(self, position_ids):
    method prepare_attention_mask (line 1682) | def prepare_attention_mask(self, hidden_states, attention_mask):
    method prepare_predict_attention_mask (line 1706) | def prepare_predict_attention_mask(self, hidden_states, attention_mask):
  class ProphetNetModel (line 1746) | class ProphetNetModel(ProphetNetPreTrainedModel):
    method __init__ (line 1749) | def __init__(self, config: ProphetNetConfig):
    method get_input_embeddings (line 1766) | def get_input_embeddings(self):
    method set_input_embeddings (line 1769) | def set_input_embeddings(self, value):
    method get_encoder (line 1774) | def get_encoder(self):
    method get_decoder (line 1777) | def get_decoder(self):
    method forward (line 1782) | def forward(
  class ProphetNetForConditionalGeneration (line 1875) | class ProphetNetForConditionalGeneration(ProphetNetPreTrainedModel):
    method __init__ (line 1882) | def __init__(self, config: ProphetNetConfig):
    method get_output_embeddings (line 1893) | def get_output_embeddings(self):
    method set_output_embeddings (line 1896) | def set_output_embeddings(self, new_embeddings):
    method get_input_embeddings (line 1899) | def get_input_embeddings(self):
    method forward (line 1904) | def forward(
    method _compute_loss (line 2008) | def _compute_loss(self, logits, labels, ignore_index=-100):
    method prepare_inputs_for_generation (line 2036) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 2065) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 2070) | def _reorder_cache(past_key_values, beam_idx):
    method get_encoder (line 2079) | def get_encoder(self):
    method get_decoder (line 2082) | def get_decoder(self):
  class ProphetNetForCausalLM (line 2091) | class ProphetNetForCausalLM(ProphetNetPreTrainedModel):
    method __init__ (line 2094) | def __init__(self, config: ProphetNetConfig):
    method get_input_embeddings (line 2110) | def get_input_embeddings(self):
    method set_input_embeddings (line 2113) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 2116) | def get_output_embeddings(self):
    method set_output_embeddings (line 2119) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 2122) | def set_decoder(self, decoder):
    method get_decoder (line 2125) | def get_decoder(self):
    method forward (line 2130) | def forward(
    method _compute_loss (line 2262) | def _compute_loss(self, logits, labels, ignore_index=-100):
    method prepare_inputs_for_generation (line 2290) | def prepare_inputs_for_generation(
    method _reorder_cache (line 2316) | def _reorder_cache(past_key_values, beam_idx):
  class ProphetNetDecoderWrapper (line 2323) | class ProphetNetDecoderWrapper(ProphetNetPreTrainedModel):
    method __init__ (line 2329) | def __init__(self, config: ProphetNetConfig):
    method forward (line 2333) | def forward(self, *args, **kwargs):

FILE: transformers/models/prophetnet/tokenization_prophetnet.py
  function whitespace_tokenize (line 47) | def whitespace_tokenize(text):
  class BasicTokenizer (line 57) | class BasicTokenizer(object):
    method __init__ (line 77) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 85) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 122) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 133) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 155) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 168) | def _is_chinese_char(self, cp):
    method _clean_text (line 192) | def _clean_text(self, text):
  class WordpieceTokenizer (line 207) | class WordpieceTokenizer(object):
    method __init__ (line 210) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 215) | def tokenize(self, text):
  function load_vocab (line 264) | def load_vocab(vocab_file):
  class ProphetNetTokenizer (line 275) | class ProphetNetTokenizer(PreTrainedTokenizer):
    method __init__ (line 330) | def __init__(
    method vocab_size (line 378) | def vocab_size(self):
    method get_vocab (line 381) | def get_vocab(self):
    method _tokenize (line 384) | def _tokenize(self, text):
    method _convert_token_to_id (line 397) | def _convert_token_to_id(self, token: str):
    method _convert_id_to_token (line 401) | def _convert_id_to_token(self, index: int):
    method convert_tokens_to_string (line 405) | def convert_tokens_to_string(self, tokens: str):
    method get_special_tokens_mask (line 410) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 440) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 468) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 488) | def build_inputs_with_special_tokens(

FILE: transformers/models/qdqbert/configuration_qdqbert.py
  class QDQBertConfig (line 29) | class QDQBertConfig(PretrainedConfig):
    method __init__ (line 90) | def __init__(

FILE: transformers/models/qdqbert/modeling_qdqbert.py
  function load_tf_weights_in_qdqbert (line 78) | def load_tf_weights_in_qdqbert(model, tf_checkpoint_path):
  class QDQBertEmbeddings (line 152) | class QDQBertEmbeddings(nn.Module):
    method __init__ (line 155) | def __init__(self, config):
    method forward (line 172) | def forward(
  class QDQBertSelfAttention (line 214) | class QDQBertSelfAttention(nn.Module):
    method __init__ (line 215) | def __init__(self, config):
    method transpose_for_scores (line 244) | def transpose_for_scores(self, x):
    method forward (line 249) | def forward(
  class QDQBertSelfOutput (line 348) | class QDQBertSelfOutput(nn.Module):
    method __init__ (line 349) | def __init__(self, config):
    method forward (line 361) | def forward(self, hidden_states, input_tensor):
  class QDQBertAttention (line 372) | class QDQBertAttention(nn.Module):
    method __init__ (line 373) | def __init__(self, config):
    method prune_heads (line 379) | def prune_heads(self, heads):
    method forward (line 397) | def forward(
  class QDQBertIntermediate (line 421) | class QDQBertIntermediate(nn.Module):
    method __init__ (line 422) | def __init__(self, config):
    method forward (line 431) | def forward(self, hidden_states):
  class QDQBertOutput (line 437) | class QDQBertOutput(nn.Module):
    method __init__ (line 438) | def __init__(self, config):
    method forward (line 449) | def forward(self, hidden_states, input_tensor):
  class QDQBertLayer (line 460) | class QDQBertLayer(nn.Module):
    method __init__ (line 461) | def __init__(self, config):
    method forward (line 474) | def forward(
    method feed_forward_chunk (line 537) | def feed_forward_chunk(self, attention_output):
  class QDQBertEncoder (line 544) | class QDQBertEncoder(nn.Module):
    method __init__ (line 545) | def __init__(self, config):
    method forward (line 551) | def forward(
  class QDQBertPooler (line 641) | class QDQBertPooler(nn.Module):
    method __init__ (line 642) | def __init__(self, config):
    method forward (line 647) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class QDQBertPredictionHeadTransform (line 657) | class QDQBertPredictionHeadTransform(nn.Module):
    method __init__ (line 658) | def __init__(self, config):
    method forward (line 667) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class QDQBertLMPredictionHead (line 675) | class QDQBertLMPredictionHead(nn.Module):
    method __init__ (line 676) | def __init__(self, config):
    method forward (line 689) | def forward(self, hidden_states):
  class QDQBertOnlyMLMHead (line 696) | class QDQBertOnlyMLMHead(nn.Module):
    method __init__ (line 697) | def __init__(self, config):
    method forward (line 701) | def forward(self, sequence_output):
  class QDQBertOnlyNSPHead (line 707) | class QDQBertOnlyNSPHead(nn.Module):
    method __init__ (line 708) | def __init__(self, config):
    method forward (line 712) | def forward(self, pooled_output):
  class QDQBertPreTrainingHeads (line 718) | class QDQBertPreTrainingHeads(nn.Module):
    method __init__ (line 719) | def __init__(self, config):
    method forward (line 724) | def forward(self, sequence_output, pooled_output):
  class QDQBertPreTrainedModel (line 731) | class QDQBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 743) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 759) | def _set_gradient_checkpointing(self, module, value=False):
  class QDQBertModel (line 834) | class QDQBertModel(QDQBertPreTrainedModel):
    method __init__ (line 847) | def __init__(self, config, add_pooling_layer: bool = True):
    method get_input_embeddings (line 860) | def get_input_embeddings(self):
    method set_input_embeddings (line 863) | def set_input_embeddings(self, value):
    method _prune_heads (line 866) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]):
    method forward (line 880) | def forward(
  class QDQBertLMHeadModel (line 1014) | class QDQBertLMHeadModel(QDQBertPreTrainedModel):
    method __init__ (line 1018) | def __init__(self, config):
    method get_output_embeddings (line 1030) | def get_output_embeddings(self):
    method set_output_embeddings (line 1033) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1038) | def forward(
    method prepare_inputs_for_generation (line 1141) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1159) | def _reorder_cache(self, past_key_values, beam_idx):
  class QDQBertForMaskedLM (line 1167) | class QDQBertForMaskedLM(QDQBertPreTrainedModel):
    method __init__ (line 1171) | def __init__(self, config):
    method get_output_embeddings (line 1186) | def get_output_embeddings(self):
    method set_output_embeddings (line 1189) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1198) | def forward(
    method prepare_inputs_for_generation (line 1255) | def prepare_inputs_for_generation(
  class QDQBertForNextSentencePrediction (line 1278) | class QDQBertForNextSentencePrediction(QDQBertPreTrainedModel):
    method __init__ (line 1279) | def __init__(self, config):
    method forward (line 1290) | def forward(
  class QDQBertForSequenceClassification (line 1382) | class QDQBertForSequenceClassification(QDQBertPreTrainedModel):
    method __init__ (line 1383) | def __init__(self, config):
    method forward (line 1400) | def forward(
  class QDQBertForMultipleChoice (line 1479) | class QDQBertForMultipleChoice(QDQBertPreTrainedModel):
    method __init__ (line 1480) | def __init__(self, config):
    method forward (line 1496) | def forward(
  class QDQBertForTokenClassification (line 1570) | class QDQBertForTokenClassification(QDQBertPreTrainedModel):
    method __init__ (line 1573) | def __init__(self, config):
    method forward (line 1590) | def forward(
  class QDQBertForQuestionAnswering (line 1650) | class QDQBertForQuestionAnswering(QDQBertPreTrainedModel):
    method __init__ (line 1653) | def __init__(self, config):
    method forward (line 1669) | def forward(

FILE: transformers/models/rag/configuration_rag.py
  class RagConfig (line 82) | class RagConfig(PretrainedConfig):
    method __init__ (line 86) | def __init__(
    method from_question_encoder_generator_configs (line 171) | def from_question_encoder_generator_configs(
    method to_dict (line 183) | def to_dict(self):

FILE: transformers/models/rag/modeling_rag.py
  class RetrievAugLMMarginOutput (line 39) | class RetrievAugLMMarginOutput(ModelOutput):
  class RetrievAugLMOutput (line 134) | class RetrievAugLMOutput(ModelOutput):
  class RagPreTrainedModel (line 223) | class RagPreTrainedModel(PreTrainedModel):
    method from_pretrained (line 237) | def from_pretrained(cls, *args, **kwargs):
    method from_pretrained_question_encoder_generator (line 244) | def from_pretrained_question_encoder_generator(
  class RagModel (line 494) | class RagModel(RagPreTrainedModel):
    method __init__ (line 495) | def __init__(
    method forward (line 539) | def forward(
  class RagSequenceForGeneration (line 740) | class RagSequenceForGeneration(RagPreTrainedModel):
    method __init__ (line 741) | def __init__(
    method set_retriever (line 762) | def set_retriever(self, retriever: RagRetriever):
    method set_context_encoder_for_training (line 765) | def set_context_encoder_for_training(self, ctx_encoder: PreTrainedModel):
    method forward (line 771) | def forward(
    method retriever (line 899) | def retriever(self):
    method generator (line 903) | def generator(self):
    method question_encoder (line 907) | def question_encoder(self):
    method generate (line 911) | def generate(
    method get_nll (line 1061) | def get_nll(
    method _cat_and_pad (line 1121) | def _cat_and_pad(tensors, pad_token_id):
  class RagTokenForGeneration (line 1138) | class RagTokenForGeneration(RagPreTrainedModel):
    method __init__ (line 1139) | def __init__(
    method set_retriever (line 1161) | def set_retriever(self, retriever: RagRetriever):
    method set_context_encoder_for_training (line 1164) | def set_context_encoder_for_training(self, ctx_encoder: PreTrainedModel):
    method prepare_inputs_for_generation (line 1168) | def prepare_inputs_for_generation(
    method retriever (line 1196) | def retriever(self):
    method generator (line 1200) | def generator(self):
    method question_encoder (line 1204) | def question_encoder(self):
    method _reorder_cache (line 1208) | def _reorder_cache(past_key_values, beam_idx):
    method marginalize (line 1225) | def marginalize(self, seq_logits, doc_scores, n_docs=None):
    method forward (line 1238) | def forward(
    method generate (line 1378) | def generate(
    method get_input_embeddings (line 1582) | def get_input_embeddings(self):
    method get_output_embeddings (line 1585) | def get_output_embeddings(self):
    method set_output_embeddings (line 1588) | def set_output_embeddings(self, new_embeddings):
    method shift_tokens_right (line 1591) | def shift_tokens_right(self, input_ids, start_token_id=None):
    method get_nll (line 1600) | def get_nll(self, seq_logits, doc_scores, target, reduce_loss=False, e...

FILE: transformers/models/rag/modeling_tf_rag.py
  class TFRetrievAugLMMarginOutput (line 48) | class TFRetrievAugLMMarginOutput(ModelOutput):
  class TFRetrievAugLMOutput (line 136) | class TFRetrievAugLMOutput(ModelOutput):
  class TFRagPreTrainedModel (line 218) | class TFRagPreTrainedModel(TFPreTrainedModel):
    method from_pretrained_question_encoder_generator (line 232) | def from_pretrained_question_encoder_generator(
  class TFRagModel (line 496) | class TFRagModel(TFRagPreTrainedModel):
    method __init__ (line 499) | def __init__(
    method set_retriever (line 543) | def set_retriever(self, retriever: RagRetriever):
    method call (line 549) | def call(
  class TFRagTokenForGeneration (line 729) | class TFRagTokenForGeneration(TFRagPreTrainedModel, TFCausalLanguageMode...
    method __init__ (line 732) | def __init__(
    method set_retriever (line 761) | def set_retriever(self, retriever: RagRetriever):
    method prepare_inputs_for_generation (line 765) | def prepare_inputs_for_generation(
    method retriever (line 793) | def retriever(self):
    method generator (line 797) | def generator(self):
    method question_encoder (line 801) | def question_encoder(self):
    method _gather_beams (line 805) | def _gather_beams(nested, beam_indices, batch_axis=0):
    method marginalize (line 830) | def marginalize(self, seq_logits, doc_scores, n_docs=None):
    method call (line 845) | def call(
    method generate (line 997) | def generate(
    method get_input_embeddings (line 1210) | def get_input_embeddings(self):
    method get_output_embeddings (line 1213) | def get_output_embeddings(self):
    method shift_tokens_right (line 1217) | def shift_tokens_right(self, input_ids, start_token_id=None):
    method get_nll (line 1250) | def get_nll(self, seq_logits, doc_scores, target, reduce_loss=False, e...
    method hf_compute_loss (line 1264) | def hf_compute_loss(self, labels, y_pred, smooth_epsilon=0.0, from_log...
  class TFRagSequenceForGeneration (line 1301) | class TFRagSequenceForGeneration(TFRagPreTrainedModel, TFCausalLanguageM...
    method __init__ (line 1304) | def __init__(
    method set_retriever (line 1333) | def set_retriever(self, retriever: RagRetriever):
    method retriever (line 1337) | def retriever(self):
    method generator (line 1341) | def generator(self):
    method question_encoder (line 1345) | def question_encoder(self):
    method call (line 1351) | def call(
    method get_nll (line 1499) | def get_nll(
    method generate (line 1583) | def generate(
    method _cat_and_pad (line 1726) | def _cat_and_pad(tensors, pad_token_id):

FILE: transformers/models/rag/retrieval_rag.py
  class Index (line 44) | class Index:
    method get_doc_dicts (line 49) | def get_doc_dicts(self, doc_ids: np.ndarray) -> List[dict]:
    method get_top_docs (line 59) | def get_top_docs(self, question_hidden_states: np.ndarray, n_docs=5) -...
    method is_initialized (line 75) | def is_initialized(self):
    method init_index (line 81) | def init_index(self):
  class LegacyIndex (line 90) | class LegacyIndex(Index):
    method __init__ (line 105) | def __init__(self, vector_size, index_path):
    method _resolve_path (line 113) | def _resolve_path(self, index_path, filename):
    method _load_passages (line 131) | def _load_passages(self):
    method _deserialize_index (line 138) | def _deserialize_index(self):
    method is_initialized (line 149) | def is_initialized(self):
    method init_index (line 152) | def init_index(self):
    method get_doc_dicts (line 160) | def get_doc_dicts(self, doc_ids: np.array):
    method get_top_docs (line 174) | def get_top_docs(self, question_hidden_states: np.ndarray, n_docs=5) -...
  class HFIndexBase (line 183) | class HFIndexBase(Index):
    method __init__ (line 184) | def __init__(self, vector_size, dataset, index_initialized=False):
    method _check_dataset_format (line 191) | def _check_dataset_format(self, with_index: bool):
    method init_index (line 206) | def init_index(self):
    method is_initialized (line 209) | def is_initialized(self):
    method get_doc_dicts (line 212) | def get_doc_dicts(self, doc_ids: np.ndarray) -> List[dict]:
    method get_top_docs (line 215) | def get_top_docs(self, question_hidden_states: np.ndarray, n_docs=5) -...
  class CanonicalHFIndex (line 225) | class CanonicalHFIndex(HFIndexBase):
    method __init__ (line 247) | def __init__(
    method init_index (line 269) | def init_index(self):
  class CustomHFIndex (line 287) | class CustomHFIndex(HFIndexBase):
    method __init__ (line 301) | def __init__(self, vector_size: int, dataset, index_path=None):
    method load_from_disk (line 306) | def load_from_disk(cls, vector_size, dataset_path, index_path):
    method init_index (line 316) | def init_index(self):
  class RagRetriever (line 323) | class RagRetriever:
    method __init__ (line 377) | def __init__(self, config, question_encoder_tokenizer, generator_token...
    method _build_index (line 396) | def _build_index(config):
    method from_pretrained (line 419) | def from_pretrained(cls, retriever_name_or_path, indexed_dataset=None,...
    method save_pretrained (line 437) | def save_pretrained(self, save_directory):
    method init_retrieval (line 457) | def init_retrieval(self):
    method postprocess_docs (line 465) | def postprocess_docs(self, docs, input_strings, prefix, n_docs, return...
    method _chunk_tensor (line 517) | def _chunk_tensor(self, t: Iterable, chunk_size: int) -> List[Iterable]:
    method _main_retrieve (line 520) | def _main_retrieve(self, question_hidden_states: np.ndarray, n_docs: i...
    method retrieve (line 537) | def retrieve(self, question_hidden_states: np.ndarray, n_docs: int) ->...
    method set_ctx_encoder_tokenizer (line 559) | def set_ctx_encoder_tokenizer(self, ctx_encoder_tokenizer: PreTrainedT...
    method __call__ (line 564) | def __call__(

FILE: transformers/models/rag/tokenization_rag.py
  class RagTokenizer (line 28) | class RagTokenizer:
    method __init__ (line 29) | def __init__(self, question_encoder, generator):
    method save_pretrained (line 34) | def save_pretrained(self, save_directory):
    method from_pretrained (line 44) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method __call__ (line 61) | def __call__(self, *args, **kwargs):
    method batch_decode (line 64) | def batch_decode(self, *args, **kwargs):
    method decode (line 67) | def decode(self, *args, **kwargs):
    method _switch_to_input_mode (line 70) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 73) | def _switch_to_target_mode(self):
    method prepare_seq2seq_batch (line 76) | def prepare_seq2seq_batch(

FILE: transformers/models/realm/configuration_realm.py
  class RealmConfig (line 44) | class RealmConfig(PretrainedConfig):
    method __init__ (line 130) | def __init__(

FILE: transformers/models/realm/modeling_realm.py
  function load_tf_weights_in_realm (line 58) | def load_tf_weights_in_realm(model, config, tf_checkpoint_path):
  class RealmEmbeddings (line 166) | class RealmEmbeddings(nn.Module):
    method __init__ (line 169) | def __init__(self, config):
    method forward (line 186) | def forward(
  class RealmSelfAttention (line 229) | class RealmSelfAttention(nn.Module):
    method __init__ (line 230) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 256) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 261) | def forward(
  class RealmSelfOutput (line 364) | class RealmSelfOutput(nn.Module):
    method __init__ (line 365) | def __init__(self, config):
    method forward (line 371) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RealmAttention (line 379) | class RealmAttention(nn.Module):
    method __init__ (line 380) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 386) | def prune_heads(self, heads):
    method forward (line 404) | def forward(
  class RealmIntermediate (line 429) | class RealmIntermediate(nn.Module):
    method __init__ (line 430) | def __init__(self, config):
    method forward (line 438) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RealmOutput (line 445) | class RealmOutput(nn.Module):
    method __init__ (line 446) | def __init__(self, config):
    method forward (line 452) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RealmLayer (line 460) | class RealmLayer(nn.Module):
    method __init__ (line 461) | def __init__(self, config):
    method forward (line 475) | def forward(
    method feed_forward_chunk (line 540) | def feed_forward_chunk(self, attention_output):
  class RealmEncoder (line 547) | class RealmEncoder(nn.Module):
    method __init__ (line 548) | def __init__(self, config):
    method forward (line 554) | def forward(
  class RealmPooler (line 646) | class RealmPooler(nn.Module):
    method __init__ (line 647) | def __init__(self, config):
    method forward (line 652) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RealmEmbedderOutput (line 662) | class RealmEmbedderOutput(ModelOutput):
  class RealmScorerOutput (line 689) | class RealmScorerOutput(ModelOutput):
  class RealmReaderOutput (line 708) | class RealmReaderOutput(ModelOutput):
  class RealmForOpenQAOutput (line 758) | class RealmForOpenQAOutput(ModelOutput):
  class RealmPredictionHeadTransform (line 774) | class RealmPredictionHeadTransform(nn.Module):
    method __init__ (line 775) | def __init__(self, config):
    method forward (line 784) | def forward(self, hidden_states):
  class RealmLMPredictionHead (line 791) | class RealmLMPredictionHead(nn.Module):
    method __init__ (line 792) | def __init__(self, config):
    method forward (line 805) | def forward(self, hidden_states):
  class RealmOnlyMLMHead (line 811) | class RealmOnlyMLMHead(nn.Module):
    method __init__ (line 812) | def __init__(self, config):
    method forward (line 816) | def forward(self, sequence_output):
  class RealmScorerProjection (line 821) | class RealmScorerProjection(nn.Module):
    method __init__ (line 822) | def __init__(self, config):
    method forward (line 828) | def forward(self, hidden_states):
  class RealmReaderProjection (line 834) | class RealmReaderProjection(nn.Module):
    method __init__ (line 835) | def __init__(self, config):
    method forward (line 843) | def forward(self, hidden_states, block_mask):
  class RealmPreTrainedModel (line 962) | class RealmPreTrainedModel(PreTrainedModel):
    method _init_weights (line 973) | def _init_weights(self, module):
    method _flatten_inputs (line 989) | def _flatten_inputs(self, *inputs):
  class RealmBertModel (line 1003) | class RealmBertModel(RealmPreTrainedModel):
    method __init__ (line 1008) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 1021) | def get_input_embeddings(self):
    method set_input_embeddings (line 1024) | def set_input_embeddings(self, value):
    method _prune_heads (line 1027) | def _prune_heads(self, heads_to_prune):
    method forward (line 1035) | def forward(
  class RealmEmbedder (line 1149) | class RealmEmbedder(RealmPreTrainedModel):
    method __init__ (line 1152) | def __init__(self, config):
    method get_input_embeddings (line 1159) | def get_input_embeddings(self):
    method set_input_embeddings (line 1162) | def set_input_embeddings(self, value):
    method forward (line 1167) | def forward(
  class RealmScorer (line 1231) | class RealmScorer(RealmPreTrainedModel):
    method __init__ (line 1238) | def __init__(self, config, query_embedder=None):
    method forward (line 1249) | def forward(
  class RealmKnowledgeAugEncoder (line 1379) | class RealmKnowledgeAugEncoder(RealmPreTrainedModel):
    method __init__ (line 1382) | def __init__(self, config):
    method get_input_embeddings (line 1388) | def get_input_embeddings(self):
    method set_input_embeddings (line 1391) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 1394) | def get_output_embeddings(self):
    method set_output_embeddings (line 1397) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1404) | def forward(
  class RealmReader (line 1529) | class RealmReader(RealmPreTrainedModel):
    method __init__ (line 1532) | def __init__(self, config):
    method forward (line 1544) | def forward(
  class RealmForOpenQA (line 1735) | class RealmForOpenQA(RealmPreTrainedModel):
    method __init__ (line 1736) | def __init__(self, config, retriever=None):
    method searcher_beam_size (line 1753) | def searcher_beam_size(self):
    method block_embedding_to (line 1758) | def block_embedding_to(self, device):
    method forward (line 1770) | def forward(

FILE: transformers/models/realm/retrieval_realm.py
  function convert_tfrecord_to_np (line 33) | def convert_tfrecord_to_np(block_records_path: str, num_block_records: i...
  class ScaNNSearcher (line 43) | class ScaNNSearcher:
    method __init__ (line 46) | def __init__(
    method search_batched (line 67) | def search_batched(self, question_projection):
  class RealmRetriever (line 72) | class RealmRetriever:
    method __init__ (line 83) | def __init__(self, block_records, tokenizer):
    method __call__ (line 88) | def __call__(self, retrieved_block_ids, question_input_ids, answer_ids...
    method from_pretrained (line 110) | def from_pretrained(cls, pretrained_model_name_or_path: Optional[Union...
    method save_pretrained (line 123) | def save_pretrained(self, save_directory):
    method block_has_answer (line 129) | def block_has_answer(self, concat_inputs, answer_ids):

FILE: transformers/models/realm/tokenization_realm.py
  function load_vocab (line 75) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 86) | def whitespace_tokenize(text):
  class RealmTokenizer (line 95) | class RealmTokenizer(PreTrainedTokenizer):
    method __init__ (line 145) | def __init__(
    method do_lower_case (line 192) | def do_lower_case(self):
    method vocab_size (line 196) | def vocab_size(self):
    method get_vocab (line 199) | def get_vocab(self):
    method _tokenize (line 202) | def _tokenize(self, text):
    method _convert_token_to_id (line 215) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 219) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 223) | def convert_tokens_to_string(self, tokens):
    method batch_encode_candidates (line 228) | def batch_encode_candidates(self, text, **kwargs):
    method build_inputs_with_special_tokens (line 301) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 326) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 354) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 383) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 404) | class BasicTokenizer(object):
    method __init__ (line 424) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 432) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 469) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 480) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 502) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 515) | def _is_chinese_char(self, cp):
    method _clean_text (line 539) | def _clean_text(self, text):
  class WordpieceTokenizer (line 553) | class WordpieceTokenizer(object):
    method __init__ (line 556) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 561) | def tokenize(self, text):

FILE: transformers/models/realm/tokenization_realm_fast.py
  class RealmTokenizerFast (line 102) | class RealmTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 151) | def __init__(
    method batch_encode_candidates (line 193) | def batch_encode_candidates(self, text, **kwargs):
    method build_inputs_with_special_tokens (line 266) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 290) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 319) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/reformer/configuration_reformer.py
  class ReformerConfig (line 32) | class ReformerConfig(PretrainedConfig):
    method __init__ (line 165) | def __init__(

FILE: transformers/models/reformer/convert_reformer_trax_checkpoint_to_pytorch.py
  function set_param (line 32) | def set_param(torch_layer, weight, bias=None):
  function set_layer_weights_in_torch_lsh (line 41) | def set_layer_weights_in_torch_lsh(weights, torch_layer, hidden_size):
  function set_layer_weights_in_torch_local (line 61) | def set_layer_weights_in_torch_local(weights, torch_layer, hidden_size):
  function set_block_weights_in_torch (line 86) | def set_block_weights_in_torch(weights, torch_block, hidden_size):
  function set_model_weights_in_torch (line 139) | def set_model_weights_in_torch(weights, torch_model, hidden_size):
  function convert_trax_checkpoint_to_pytorch (line 186) | def convert_trax_checkpoint_to_pytorch(trax_model_pkl_path, config_file,...

FILE: transformers/models/reformer/modeling_reformer.py
  function _stable_argsort (line 74) | def _stable_argsort(vector, dim):
  function _get_least_common_mult_chunk_len (line 83) | def _get_least_common_mult_chunk_len(config):
  function _get_min_chunk_len (line 99) | def _get_min_chunk_len(config):
  class AxialPositionEmbeddings (line 115) | class AxialPositionEmbeddings(nn.Module):
    method __init__ (line 120) | def __init__(self, config):
    method forward (line 145) | def forward(self, position_ids):
  class PositionEmbeddings (line 211) | class PositionEmbeddings(nn.Module):
    method __init__ (line 214) | def __init__(self, config):
    method forward (line 219) | def forward(self, position_ids):
  class ReformerEmbeddings (line 225) | class ReformerEmbeddings(nn.Module):
    method __init__ (line 228) | def __init__(self, config):
    method forward (line 238) | def forward(self, input_ids=None, position_ids=None, inputs_embeds=Non...
  class EfficientAttentionMixin (line 271) | class EfficientAttentionMixin:
    method _look_adjacent (line 276) | def _look_adjacent(self, vectors, num_chunks_before, num_chunks_after):
    method _split_hidden_size_dim (line 299) | def _split_hidden_size_dim(self, x, num_attn_heads, attn_head_size):
    method _merge_hidden_size_dims (line 307) | def _merge_hidden_size_dims(self, x, num_attn_heads, attn_head_size):
    method _split_seq_length_dim_to (line 314) | def _split_seq_length_dim_to(self, vectors, dim_factor_1, dim_factor_2...
  class LSHSelfAttention (line 329) | class LSHSelfAttention(nn.Module, EfficientAttentionMixin):
    method __init__ (line 330) | def __init__(self, config):
    method forward (line 360) | def forward(
    method _query_per_attn_head (line 597) | def _query_per_attn_head(self, hidden_states):
    method _value_per_attn_head (line 605) | def _value_per_attn_head(self, hidden_states):
    method _hash_vectors (line 613) | def _hash_vectors(self, vectors, num_hashes, attention_mask, increase_...
    method _get_sorted_bucket_idx_and_undo_sorted_bucket_idx (line 687) | def _get_sorted_bucket_idx_and_undo_sorted_bucket_idx(self, sequence_l...
    method _set_num_buckets (line 706) | def _set_num_buckets(self, sequence_length):
    method _attend (line 726) | def _attend(
    method _compute_attn_mask (line 838) | def _compute_attn_mask(
    method _get_relevant_hid_states_and_buckets (line 866) | def _get_relevant_hid_states_and_buckets(
    method _expand_to_indices_in_relevant_chunk (line 952) | def _expand_to_indices_in_relevant_chunk(self, indices, sequence_length):
    method _len_and_dim_norm (line 972) | def _len_and_dim_norm(self, vectors, sqrt_num):
    method _len_norm (line 980) | def _len_norm(self, x, epsilon=1e-6):
    method _gather_by_expansion (line 988) | def _gather_by_expansion(self, vectors, idxs, num_hashes):
  class ReverseSort (line 997) | class ReverseSort(Function):
    method forward (line 1004) | def forward(ctx, out_vectors, logits, sorted_bucket_idx, undo_sorted_b...
    method backward (line 1016) | def backward(ctx, grad_out_vectors, grad_logits):
  class LocalSelfAttention (line 1029) | class LocalSelfAttention(nn.Module, EfficientAttentionMixin):
    method __init__ (line 1030) | def __init__(self, config):
    method forward (line 1055) | def forward(
    method _compute_attn_mask (line 1222) | def _compute_attn_mask(
    method _retrieve_relevant_hidden_states (line 1248) | def _retrieve_relevant_hidden_states(previous_hidden_states, chunk_len...
  class ReformerSelfOutput (line 1253) | class ReformerSelfOutput(nn.Module):
    method __init__ (line 1254) | def __init__(self, config):
    method forward (line 1261) | def forward(self, hidden_states):
  class ReformerAttention (line 1267) | class ReformerAttention(nn.Module):
    method __init__ (line 1268) | def __init__(self, config, layer_id=0):
    method forward (line 1292) | def forward(
  class ReformerFeedForwardDense (line 1359) | class ReformerFeedForwardDense(nn.Module):
    method __init__ (line 1360) | def __init__(self, config):
    method forward (line 1371) | def forward(self, hidden_states):
  class ReformerFeedForwardOutput (line 1378) | class ReformerFeedForwardOutput(nn.Module):
    method __init__ (line 1379) | def __init__(self, config):
    method forward (line 1385) | def forward(self, hidden_states):
  class ChunkReformerFeedForward (line 1391) | class ChunkReformerFeedForward(nn.Module):
    method __init__ (line 1392) | def __init__(self, config):
    method forward (line 1401) | def forward(self, attention_output):
    method forward_chunk (line 1409) | def forward_chunk(self, hidden_states):
  class ReformerLayer (line 1415) | class ReformerLayer(nn.Module):
    method __init__ (line 1416) | def __init__(self, config, layer_id=0):
    method _init_attention_seed (line 1426) | def _init_attention_seed(self):
    method _init_feed_forward_seed (line 1444) | def _init_feed_forward_seed(self):
    method forward (line 1461) | def forward(
    method backward_pass (line 1514) | def backward_pass(
  class _ReversibleFunction (line 1583) | class _ReversibleFunction(Function):
    method forward (line 1591) | def forward(
    method backward (line 1649) | def backward(ctx, grad_hidden_states):
  class ReformerEncoder (line 1695) | class ReformerEncoder(nn.Module):
    method __init__ (line 1696) | def __init__(self, config):
    method forward (line 1705) | def forward(
  class ReformerOnlyLMHead (line 1756) | class ReformerOnlyLMHead(nn.Module):
    method __init__ (line 1757) | def __init__(self, config):
    method forward (line 1767) | def forward(self, hidden_states):
    method forward_chunk (line 1770) | def forward_chunk(self, hidden_states):
    method _tie_weights (line 1774) | def _tie_weights(self):
  class ReformerPreTrainedModel (line 1779) | class ReformerPreTrainedModel(PreTrainedModel):
    method dummy_inputs (line 1789) | def dummy_inputs(self):
    method _init_weights (line 1798) | def _init_weights(self, module):
  class ReformerModelOutput (line 1819) | class ReformerModelOutput(ModelOutput):
  class ReformerModelWithLMHeadOutput (line 1856) | class ReformerModelWithLMHeadOutput(ModelOutput):
  class ReformerModel (line 1976) | class ReformerModel(ReformerPreTrainedModel):
    method __init__ (line 1977) | def __init__(self, config):
    method get_input_embeddings (line 1990) | def get_input_embeddings(self):
    method set_input_embeddings (line 1993) | def set_input_embeddings(self, value):
    method _prune_heads (line 1996) | def _prune_heads(self, heads_to_prune):
    method forward (line 2010) | def forward(
    method _pad_to_mult_of_chunk_length (line 2130) | def _pad_to_mult_of_chunk_length(
  class ReformerModelWithLMHead (line 2187) | class ReformerModelWithLMHead(ReformerPreTrainedModel):
    method __init__ (line 2190) | def __init__(self, config):
    method get_output_embeddings (line 2208) | def get_output_embeddings(self):
    method set_output_embeddings (line 2211) | def set_output_embeddings(self, new_embeddings):
    method forward (line 2220) | def forward(
    method prepare_inputs_for_generation (line 2281) | def prepare_inputs_for_generation(
    method _reorder_cache (line 2297) | def _reorder_cache(self, past_key_values, beam_idx):
  class ReformerForMaskedLM (line 2313) | class ReformerForMaskedLM(ReformerPreTrainedModel):
    method __init__ (line 2314) | def __init__(self, config):
    method get_output_embeddings (line 2326) | def get_output_embeddings(self):
    method set_output_embeddings (line 2329) | def set_output_embeddings(self, new_embeddings):
    method forward (line 2334) | def forward(
  class ReformerForSequenceClassification (line 2441) | class ReformerForSequenceClassification(ReformerPreTrainedModel):
    method __init__ (line 2442) | def __init__(self, config):
    method forward (line 2457) | def forward(
  class ReformerClassificationHead (line 2559) | class ReformerClassificationHead(nn.Module):
    method __init__ (line 2562) | def __init__(self, config):
    method forward (line 2571) | def forward(self, hidden_states, **kwargs):
  class ReformerForQuestionAnswering (line 2588) | class ReformerForQuestionAnswering(ReformerPreTrainedModel):
    method __init__ (line 2589) | def __init__(self, config):
    method forward (line 2606) | def forward(

FILE: transformers/models/reformer/tokenization_reformer.py
  class ReformerTokenizer (line 48) | class ReformerTokenizer(PreTrainedTokenizer):
    method __init__ (line 98) | def __init__(
    method vocab_size (line 122) | def vocab_size(self):
    method get_vocab (line 125) | def get_vocab(self) -> Dict[str, int]:
    method __getstate__ (line 130) | def __getstate__(self):
    method __setstate__ (line 135) | def __setstate__(self, d):
    method _tokenize (line 145) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 149) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 153) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 159) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 173) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/reformer/tokenization_reformer_fast.py
  class ReformerTokenizerFast (line 57) | class ReformerTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 94) | def __init__(
    method save_vocabulary (line 115) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/regnet/configuration_regnet.py
  class RegNetConfig (line 28) | class RegNetConfig(PretrainedConfig):
    method __init__ (line 72) | def __init__(

FILE: transformers/models/regnet/convert_regnet_seer_10b_to_pytorch.py
  class Tracker (line 47) | class Tracker:
    method _forward_hook (line 53) | def _forward_hook(self, m, inputs: Tensor, outputs: Tensor, name: str):
    method __call__ (line 59) | def __call__(self, x: Tensor):
    method parametrized (line 67) | def parametrized(self):
  class FakeRegNetVisslWrapper (line 72) | class FakeRegNetVisslWrapper(nn.Module):
    method __init__ (line 77) | def __init__(self, model: nn.Module):
    method forward (line 91) | def forward(self, x: Tensor):
  class FakeRegNetParams (line 99) | class FakeRegNetParams(RegNetParams):
    method get_expanded_params (line 105) | def get_expanded_params(self):
  function get_from_to_our_keys (line 109) | def get_from_to_our_keys(model_name: str) -> Dict[str, str]:
  function convert_weights_and_push (line 162) | def convert_weights_and_push(save_directory: Path, model_name: str = Non...

FILE: transformers/models/regnet/convert_regnet_to_pytorch.py
  class Tracker (line 42) | class Tracker:
    method _forward_hook (line 47) | def _forward_hook(self, m, inputs: Tensor, outputs: Tensor):
    method __call__ (line 52) | def __call__(self, x: Tensor):
    method parametrized (line 60) | def parametrized(self):
  class ModuleTransfer (line 66) | class ModuleTransfer:
    method __call__ (line 74) | def __call__(self, x: Tensor):
  class FakeRegNetVisslWrapper (line 97) | class FakeRegNetVisslWrapper(nn.Module):
    method __init__ (line 102) | def __init__(self, model: nn.Module):
    method forward (line 116) | def forward(self, x: Tensor):
  class NameToFromModelFuncMap (line 124) | class NameToFromModelFuncMap(dict):
    method convert_name_to_timm (line 129) | def convert_name_to_timm(self, x: str) -> str:
    method __getitem__ (line 133) | def __getitem__(self, x: str) -> Callable[[], Tuple[nn.Module, Dict]]:
  class NameToOurModelFuncMap (line 145) | class NameToOurModelFuncMap(dict):
    method __getitem__ (line 150) | def __getitem__(self, x: str) -> Callable[[], nn.Module]:
  function manually_copy_vissl_head (line 158) | def manually_copy_vissl_head(from_state_dict, to_state_dict, keys: List[...
  function convert_weight_and_push (line 165) | def convert_weight_and_push(
  function convert_weights_and_push (line 222) | def convert_weights_and_push(save_directory: Path, model_name: str = Non...

FILE: transformers/models/regnet/modeling_flax_regnet.py
  class Identity (line 94) | class Identity(nn.Module):
    method __call__ (line 98) | def __call__(self, x, **kwargs):
  class FlaxRegNetConvLayer (line 102) | class FlaxRegNetConvLayer(nn.Module):
    method setup (line 110) | def setup(self):
    method __call__ (line 124) | def __call__(self, hidden_state: jnp.ndarray, deterministic: bool = Tr...
  class FlaxRegNetEmbeddings (line 131) | class FlaxRegNetEmbeddings(nn.Module):
    method setup (line 135) | def setup(self):
    method __call__ (line 144) | def __call__(self, pixel_values: jnp.ndarray, deterministic: bool = Tr...
  class FlaxRegNetShortCut (line 155) | class FlaxRegNetShortCut(nn.Module):
    method setup (line 165) | def setup(self):
    method __call__ (line 176) | def __call__(self, x: jnp.ndarray, deterministic: bool = True) -> jnp....
  class FlaxRegNetSELayerCollection (line 182) | class FlaxRegNetSELayerCollection(nn.Module):
    method setup (line 187) | def setup(self):
    method __call__ (line 203) | def __call__(self, hidden_state: jnp.ndarray) -> jnp.ndarray:
  class FlaxRegNetSELayer (line 212) | class FlaxRegNetSELayer(nn.Module):
    method setup (line 221) | def setup(self):
    method __call__ (line 225) | def __call__(self, hidden_state: jnp.ndarray) -> jnp.ndarray:
  class FlaxRegNetXLayerCollection (line 236) | class FlaxRegNetXLayerCollection(nn.Module):
    method setup (line 242) | def setup(self):
    method __call__ (line 270) | def __call__(self, hidden_state: jnp.ndarray, deterministic: bool = Tr...
  class FlaxRegNetXLayer (line 276) | class FlaxRegNetXLayer(nn.Module):
    method setup (line 287) | def setup(self):
    method __call__ (line 307) | def __call__(self, hidden_state: jnp.ndarray, deterministic: bool = Tr...
  class FlaxRegNetYLayerCollection (line 316) | class FlaxRegNetYLayerCollection(nn.Module):
    method setup (line 323) | def setup(self):
    method __call__ (line 357) | def __call__(self, hidden_state: jnp.ndarray) -> jnp.ndarray:
  class FlaxRegNetYLayer (line 363) | class FlaxRegNetYLayer(nn.Module):
    method setup (line 374) | def setup(self):
    method __call__ (line 395) | def __call__(self, hidden_state: jnp.ndarray, deterministic: bool = Tr...
  class FlaxRegNetStageLayersCollection (line 404) | class FlaxRegNetStageLayersCollection(nn.Module):
    method setup (line 416) | def setup(self):
    method __call__ (line 444) | def __call__(self, x: jnp.ndarray, deterministic: bool = True) -> jnp....
  class FlaxRegNetStage (line 452) | class FlaxRegNetStage(nn.Module):
    method setup (line 464) | def setup(self):
    method __call__ (line 474) | def __call__(self, x: jnp.ndarray, deterministic: bool = True) -> jnp....
  class FlaxRegNetStageCollection (line 479) | class FlaxRegNetStageCollection(nn.Module):
    method setup (line 483) | def setup(self):
    method __call__ (line 504) | def __call__(
  class FlaxRegNetEncoder (line 522) | class FlaxRegNetEncoder(nn.Module):
    method setup (line 526) | def setup(self):
    method __call__ (line 529) | def __call__(
  class FlaxRegNetPreTrainedModel (line 553) | class FlaxRegNetPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 564) | def __init__(
    method init_weights (line 578) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 597) | def __call__(
  class FlaxRegNetModule (line 630) | class FlaxRegNetModule(nn.Module):
    method setup (line 634) | def setup(self):
    method __call__ (line 644) | def __call__(
  class FlaxRegNetModel (line 689) | class FlaxRegNetModel(FlaxRegNetPreTrainedModel):
  class FlaxRegNetClassifierCollection (line 724) | class FlaxRegNetClassifierCollection(nn.Module):
    method setup (line 728) | def setup(self):
    method __call__ (line 731) | def __call__(self, x: jnp.ndarray) -> jnp.ndarray:
  class FlaxRegNetForImageClassificationModule (line 736) | class FlaxRegNetForImageClassificationModule(nn.Module):
    method setup (line 740) | def setup(self):
    method __call__ (line 748) | def __call__(
  class FlaxRegNetForImageClassification (line 782) | class FlaxRegNetForImageClassification(FlaxRegNetPreTrainedModel):

FILE: transformers/models/regnet/modeling_regnet.py
  class RegNetConvLayer (line 55) | class RegNetConvLayer(nn.Module):
    method __init__ (line 56) | def __init__(
    method forward (line 78) | def forward(self, hidden_state):
  class RegNetEmbeddings (line 85) | class RegNetEmbeddings(nn.Module):
    method __init__ (line 90) | def __init__(self, config: RegNetConfig):
    method forward (line 97) | def forward(self, pixel_values):
  class RegNetShortCut (line 108) | class RegNetShortCut(nn.Module):
    method __init__ (line 114) | def __init__(self, in_channels: int, out_channels: int, stride: int = 2):
    method forward (line 119) | def forward(self, input: Tensor) -> Tensor:
  class RegNetSELayer (line 125) | class RegNetSELayer(nn.Module):
    method __init__ (line 130) | def __init__(self, in_channels: int, reduced_channels: int):
    method forward (line 141) | def forward(self, hidden_state):
  class RegNetXLayer (line 149) | class RegNetXLayer(nn.Module):
    method __init__ (line 154) | def __init__(self, config: RegNetConfig, in_channels: int, out_channel...
    method forward (line 168) | def forward(self, hidden_state):
  class RegNetYLayer (line 177) | class RegNetYLayer(nn.Module):
    method __init__ (line 182) | def __init__(self, config: RegNetConfig, in_channels: int, out_channel...
    method forward (line 197) | def forward(self, hidden_state):
  class RegNetStage (line 206) | class RegNetStage(nn.Module):
    method __init__ (line 211) | def __init__(
    method forward (line 234) | def forward(self, hidden_state):
  class RegNetEncoder (line 239) | class RegNetEncoder(nn.Module):
    method __init__ (line 240) | def __init__(self, config: RegNetConfig):
    method forward (line 257) | def forward(
  class RegNetPreTrainedModel (line 277) | class RegNetPreTrainedModel(PreTrainedModel):
    method _init_weights (line 289) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 296) | def _set_gradient_checkpointing(self, module, value=False):
  class RegNetModel (line 331) | class RegNetModel(RegNetPreTrainedModel):
    method __init__ (line 332) | def __init__(self, config):
    method forward (line 349) | def forward(
  class RegNetForImageClassification (line 385) | class RegNetForImageClassification(RegNetPreTrainedModel):
    method __init__ (line 386) | def __init__(self, config):
    method forward (line 405) | def forward(

FILE: transformers/models/regnet/modeling_tf_regnet.py
  class TFRegNetConvLayer (line 53) | class TFRegNetConvLayer(tf.keras.layers.Layer):
    method __init__ (line 54) | def __init__(
    method call (line 79) | def call(self, hidden_state):
  class TFRegNetEmbeddings (line 86) | class TFRegNetEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 91) | def __init__(self, config: RegNetConfig, **kwargs):
    method call (line 102) | def call(self, pixel_values):
  class TFRegNetShortCut (line 117) | class TFRegNetShortCut(tf.keras.layers.Layer):
    method __init__ (line 123) | def __init__(self, out_channels: int, stride: int = 2, **kwargs):
    method call (line 130) | def call(self, inputs: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFRegNetSELayer (line 134) | class TFRegNetSELayer(tf.keras.layers.Layer):
    method __init__ (line 139) | def __init__(self, in_channels: int, reduced_channels: int, **kwargs):
    method call (line 147) | def call(self, hidden_state):
  class TFRegNetXLayer (line 156) | class TFRegNetXLayer(tf.keras.layers.Layer):
    method __init__ (line 161) | def __init__(self, config: RegNetConfig, in_channels: int, out_channel...
    method call (line 180) | def call(self, hidden_state):
  class TFRegNetYLayer (line 190) | class TFRegNetYLayer(tf.keras.layers.Layer):
    method __init__ (line 195) | def __init__(self, config: RegNetConfig, in_channels: int, out_channel...
    method call (line 214) | def call(self, hidden_state):
  class TFRegNetStage (line 224) | class TFRegNetStage(tf.keras.layers.Layer):
    method __init__ (line 229) | def __init__(
    method call (line 241) | def call(self, hidden_state):
  class TFRegNetEncoder (line 247) | class TFRegNetEncoder(tf.keras.layers.Layer):
    method __init__ (line 248) | def __init__(self, config: RegNetConfig, **kwargs):
    method call (line 266) | def call(
  class TFRegNetMainLayer (line 287) | class TFRegNetMainLayer(tf.keras.layers.Layer):
    method __init__ (line 290) | def __init__(self, config, **kwargs):
    method call (line 298) | def call(
  class TFRegNetPreTrainedModel (line 337) | class TFRegNetPreTrainedModel(TFPreTrainedModel):
    method input_signature (line 348) | def input_signature(self):
  class TFRegNetModel (line 380) | class TFRegNetModel(TFRegNetPreTrainedModel):
    method __init__ (line 381) | def __init__(self, config: RegNetConfig, *inputs, **kwargs):
    method call (line 394) | def call(
  class TFRegNetForImageClassification (line 429) | class TFRegNetForImageClassification(TFRegNetPreTrainedModel, TFSequence...
    method __init__ (line 430) | def __init__(self, config: RegNetConfig, *inputs, **kwargs):
    method call (line 448) | def call(

FILE: transformers/models/rembert/configuration_rembert.py
  class RemBertConfig (line 32) | class RemBertConfig(PretrainedConfig):
    method __init__ (line 101) | def __init__(
  class RemBertOnnxConfig (line 145) | class RemBertOnnxConfig(OnnxConfig):
    method inputs (line 147) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 161) | def atol_for_validation(self) -> float:

FILE: transformers/models/rembert/convert_rembert_tf_checkpoint_to_pytorch.py
  function convert_rembert_tf_checkpoint_to_pytorch (line 29) | def convert_rembert_tf_checkpoint_to_pytorch(tf_checkpoint_path, bert_co...

FILE: transformers/models/rembert/modeling_rembert.py
  function load_tf_weights_in_rembert (line 61) | def load_tf_weights_in_rembert(model, config, tf_checkpoint_path):
  class RemBertEmbeddings (line 144) | class RemBertEmbeddings(nn.Module):
    method __init__ (line 147) | def __init__(self, config):
    method forward (line 163) | def forward(
  class RemBertPooler (line 197) | class RemBertPooler(nn.Module):
    method __init__ (line 198) | def __init__(self, config):
    method forward (line 203) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RemBertSelfAttention (line 212) | class RemBertSelfAttention(nn.Module):
    method __init__ (line 213) | def __init__(self, config):
    method transpose_for_scores (line 233) | def transpose_for_scores(self, x):
    method forward (line 238) | def forward(
  class RemBertSelfOutput (line 318) | class RemBertSelfOutput(nn.Module):
    method __init__ (line 319) | def __init__(self, config):
    method forward (line 325) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RemBertAttention (line 332) | class RemBertAttention(nn.Module):
    method __init__ (line 333) | def __init__(self, config):
    method prune_heads (line 340) | def prune_heads(self, heads):
    method forward (line 359) | def forward(
  class RemBertIntermediate (line 384) | class RemBertIntermediate(nn.Module):
    method __init__ (line 385) | def __init__(self, config):
    method forward (line 393) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RemBertOutput (line 400) | class RemBertOutput(nn.Module):
    method __init__ (line 401) | def __init__(self, config):
    method forward (line 407) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RemBertLayer (line 414) | class RemBertLayer(nn.Module):
    method __init__ (line 415) | def __init__(self, config):
    method forward (line 430) | def forward(
    method feed_forward_chunk (line 496) | def feed_forward_chunk(self, attention_output):
  class RemBertEncoder (line 502) | class RemBertEncoder(nn.Module):
    method __init__ (line 503) | def __init__(self, config):
    method forward (line 511) | def forward(
  class RemBertPredictionHeadTransform (line 603) | class RemBertPredictionHeadTransform(nn.Module):
    method __init__ (line 604) | def __init__(self, config):
    method forward (line 613) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RemBertLMPredictionHead (line 620) | class RemBertLMPredictionHead(nn.Module):
    method __init__ (line 621) | def __init__(self, config):
    method forward (line 628) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RemBertOnlyMLMHead (line 637) | class RemBertOnlyMLMHead(nn.Module):
    method __init__ (line 638) | def __init__(self, config):
    method forward (line 642) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class RemBertPreTrainedModel (line 647) | class RemBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 659) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 675) | def _set_gradient_checkpointing(self, module, value=False):
  class RemBertModel (line 745) | class RemBertModel(RemBertPreTrainedModel):
    method __init__ (line 758) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 770) | def get_input_embeddings(self):
    method set_input_embeddings (line 773) | def set_input_embeddings(self, value):
    method _prune_heads (line 776) | def _prune_heads(self, heads_to_prune):
    method forward (line 790) | def forward(
  class RemBertForMaskedLM (line 914) | class RemBertForMaskedLM(RemBertPreTrainedModel):
    method __init__ (line 915) | def __init__(self, config):
    method get_output_embeddings (line 930) | def get_output_embeddings(self):
    method set_output_embeddings (line 933) | def set_output_embeddings(self, new_embeddings):
    method forward (line 942) | def forward(
    method prepare_inputs_for_generation (line 998) | def prepare_inputs_for_generation(self, input_ids, attention_mask=None...
  class RemBertForCausalLM (line 1016) | class RemBertForCausalLM(RemBertPreTrainedModel):
    method __init__ (line 1019) | def __init__(self, config):
    method get_output_embeddings (line 1031) | def get_output_embeddings(self):
    method set_output_embeddings (line 1034) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1039) | def forward(
    method prepare_inputs_for_generation (line 1139) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1152) | def _reorder_cache(self, past_key_values, beam_idx):
  class RemBertForSequenceClassification (line 1168) | class RemBertForSequenceClassification(RemBertPreTrainedModel):
    method __init__ (line 1169) | def __init__(self, config):
    method forward (line 1185) | def forward(
  class RemBertForMultipleChoice (line 1264) | class RemBertForMultipleChoice(RemBertPreTrainedModel):
    method __init__ (line 1265) | def __init__(self, config):
    method forward (line 1281) | def forward(
  class RemBertForTokenClassification (line 1355) | class RemBertForTokenClassification(RemBertPreTrainedModel):
    method __init__ (line 1356) | def __init__(self, config):
    method forward (line 1373) | def forward(
  class RemBertForQuestionAnswering (line 1433) | class RemBertForQuestionAnswering(RemBertPreTrainedModel):
    method __init__ (line 1434) | def __init__(self, config):
    method forward (line 1451) | def forward(

FILE: transformers/models/rembert/modeling_tf_rembert.py
  class TFRemBertEmbeddings (line 70) | class TFRemBertEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 73) | def __init__(self, config: RemBertConfig, **kwargs):
    method build (line 83) | def build(self, input_shape: tf.TensorShape):
    method call (line 107) | def call(
  class TFRemBertSelfAttention (line 148) | class TFRemBertSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 149) | def __init__(self, config: RemBertConfig, **kwargs):
    method transpose_for_scores (line 176) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 183) | def call(
  class TFRemBertSelfOutput (line 266) | class TFRemBertSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 267) | def __init__(self, config: RemBertConfig, **kwargs):
    method call (line 276) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFRemBertAttention (line 285) | class TFRemBertAttention(tf.keras.layers.Layer):
    method __init__ (line 286) | def __init__(self, config: RemBertConfig, **kwargs):
    method prune_heads (line 292) | def prune_heads(self, heads):
    method call (line 295) | def call(
  class TFRemBertIntermediate (line 326) | class TFRemBertIntermediate(tf.keras.layers.Layer):
    method __init__ (line 327) | def __init__(self, config: RemBertConfig, **kwargs):
    method call (line 339) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFRemBertOutput (line 347) | class TFRemBertOutput(tf.keras.layers.Layer):
    method __init__ (line 348) | def __init__(self, config: RemBertConfig, **kwargs):
    method call (line 357) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFRemBertLayer (line 366) | class TFRemBertLayer(tf.keras.layers.Layer):
    method __init__ (line 367) | def __init__(self, config: RemBertConfig, **kwargs):
    method call (line 380) | def call(
  class TFRemBertEncoder (line 452) | class TFRemBertEncoder(tf.keras.layers.Layer):
    method __init__ (line 453) | def __init__(self, config: RemBertConfig, **kwargs):
    method call (line 464) | def call(
  class TFRemBertPooler (line 529) | class TFRemBertPooler(tf.keras.layers.Layer):
    method __init__ (line 530) | def __init__(self, config: RemBertConfig, **kwargs):
    method call (line 540) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFRemBertLMPredictionHead (line 549) | class TFRemBertLMPredictionHead(tf.keras.layers.Layer):
    method __init__ (line 550) | def __init__(self, config: RemBertConfig, input_embeddings: tf.keras.l...
    method build (line 565) | def build(self, input_shape: tf.TensorShape):
    method get_output_embeddings (line 577) | def get_output_embeddings(self) -> tf.keras.layers.Layer:
    method set_output_embeddings (line 580) | def set_output_embeddings(self, value):
    method get_bias (line 584) | def get_bias(self) -> Dict[str, tf.Variable]:
    method set_bias (line 587) | def set_bias(self, value: tf.Variable):
    method call (line 591) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFRemBertMLMHead (line 604) | class TFRemBertMLMHead(tf.keras.layers.Layer):
    method __init__ (line 605) | def __init__(self, config: RemBertConfig, input_embeddings: tf.keras.l...
    method call (line 610) | def call(self, sequence_output: tf.Tensor) -> tf.Tensor:
  class TFRemBertMainLayer (line 617) | class TFRemBertMainLayer(tf.keras.layers.Layer):
    method __init__ (line 620) | def __init__(self, config: RemBertConfig, add_pooling_layer: bool = Tr...
    method get_input_embeddings (line 630) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 633) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 637) | def _prune_heads(self, heads_to_prune):
    method call (line 646) | def call(
  class TFRemBertPreTrainedModel (line 804) | class TFRemBertPreTrainedModel(TFPreTrainedModel):
  class TFRemBertModel (line 916) | class TFRemBertModel(TFRemBertPreTrainedModel):
    method __init__ (line 917) | def __init__(self, config: RemBertConfig, *inputs, **kwargs):
    method call (line 929) | def call(
  class TFRemBertForMaskedLM (line 987) | class TFRemBertForMaskedLM(TFRemBertPreTrainedModel, TFMaskedLanguageMod...
    method __init__ (line 988) | def __init__(self, config: RemBertConfig, *inputs, **kwargs):
    method get_lm_head (line 1000) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method call (line 1010) | def call(
  class TFRemBertForCausalLM (line 1061) | class TFRemBertForCausalLM(TFRemBertPreTrainedModel, TFCausalLanguageMod...
    method __init__ (line 1062) | def __init__(self, config: RemBertConfig, *inputs, **kwargs):
    method get_lm_head (line 1071) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method prepare_inputs_for_generation (line 1075) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method call (line 1093) | def call(
  class TFRemBertForSequenceClassification (line 1180) | class TFRemBertForSequenceClassification(TFRemBertPreTrainedModel, TFSeq...
    method __init__ (line 1181) | def __init__(self, config: RemBertConfig, *inputs, **kwargs):
    method call (line 1201) | def call(
  class TFRemBertForMultipleChoice (line 1257) | class TFRemBertForMultipleChoice(TFRemBertPreTrainedModel, TFMultipleCho...
    method __init__ (line 1258) | def __init__(self, config: RemBertConfig, *inputs, **kwargs):
    method call (line 1274) | def call(
  class TFRemBertForTokenClassification (line 1353) | class TFRemBertForTokenClassification(TFRemBertPreTrainedModel, TFTokenC...
    method __init__ (line 1354) | def __init__(self, config: RemBertConfig, *inputs, **kwargs):
    method call (line 1372) | def call(
  class TFRemBertForQuestionAnswering (line 1426) | class TFRemBertForQuestionAnswering(TFRemBertPreTrainedModel, TFQuestion...
    method __init__ (line 1427) | def __init__(self, config: RemBertConfig, *inputs, **kwargs):
    method call (line 1444) | def call(

FILE: transformers/models/rembert/tokenization_rembert.py
  class RemBertTokenizer (line 43) | class RemBertTokenizer(PreTrainedTokenizer):
    method __init__ (line 99) | def __init__(
    method vocab_size (line 137) | def vocab_size(self):
    method get_vocab (line 140) | def get_vocab(self):
    method __getstate__ (line 145) | def __getstate__(self):
    method __setstate__ (line 150) | def __setstate__(self, d):
    method _tokenize (line 155) | def _tokenize(self, text, sample=False):
    method _convert_token_to_id (line 160) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 164) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 168) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 172) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 197) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 228) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 258) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/rembert/tokenization_rembert_fast.py
  class RemBertTokenizerFast (line 51) | class RemBertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 103) | def __init__(
    method build_inputs_with_special_tokens (line 144) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 169) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 200) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 230) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/resnet/configuration_resnet.py
  class ResNetConfig (line 35) | class ResNetConfig(BackboneConfigMixin, PretrainedConfig):
    method __init__ (line 88) | def __init__(
  class ResNetOnnxConfig (line 117) | class ResNetOnnxConfig(OnnxConfig):
    method inputs (line 121) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 129) | def atol_for_validation(self) -> float:

FILE: transformers/models/resnet/convert_resnet_to_pytorch.py
  class Tracker (line 40) | class Tracker:
    method _forward_hook (line 45) | def _forward_hook(self, m, inputs: Tensor, outputs: Tensor):
    method __call__ (line 50) | def __call__(self, x: Tensor):
    method parametrized (line 58) | def parametrized(self):
  class ModuleTransfer (line 64) | class ModuleTransfer:
    method __call__ (line 71) | def __call__(self, x: Tensor):
  function convert_weight_and_push (line 94) | def convert_weight_and_push(name: str, config: ResNetConfig, save_direct...
  function convert_weights_and_push (line 126) | def convert_weights_and_push(save_directory: Path, model_name: str = Non...

FILE: transformers/models/resnet/modeling_flax_resnet.py
  class Identity (line 88) | class Identity(nn.Module):
    method __call__ (line 92) | def __call__(self, x, **kwargs):
  class FlaxResNetConvLayer (line 96) | class FlaxResNetConvLayer(nn.Module):
    method setup (line 103) | def setup(self):
    method __call__ (line 116) | def __call__(self, x: jnp.ndarray, deterministic: bool = True) -> jnp....
  class FlaxResNetEmbeddings (line 123) | class FlaxResNetEmbeddings(nn.Module):
    method setup (line 131) | def setup(self):
    method __call__ (line 142) | def __call__(self, pixel_values: jnp.ndarray, deterministic: bool = Tr...
  class FlaxResNetShortCut (line 153) | class FlaxResNetShortCut(nn.Module):
    method setup (line 163) | def setup(self):
    method __call__ (line 174) | def __call__(self, x: jnp.ndarray, deterministic: bool = True) -> jnp....
  class FlaxResNetBasicLayerCollection (line 180) | class FlaxResNetBasicLayerCollection(nn.Module):
    method setup (line 185) | def setup(self):
    method __call__ (line 191) | def __call__(self, hidden_state: jnp.ndarray, deterministic: bool = Tr...
  class FlaxResNetBasicLayer (line 197) | class FlaxResNetBasicLayer(nn.Module):
    method setup (line 208) | def setup(self):
    method __call__ (line 223) | def __call__(self, hidden_state, deterministic: bool = True):
  class FlaxResNetBottleNeckLayerCollection (line 235) | class FlaxResNetBottleNeckLayerCollection(nn.Module):
    method setup (line 242) | def setup(self):
    method __call__ (line 251) | def __call__(self, hidden_state: jnp.ndarray, deterministic: bool = Tr...
  class FlaxResNetBottleNeckLayer (line 257) | class FlaxResNetBottleNeckLayer(nn.Module):
    method setup (line 271) | def setup(self):
    method __call__ (line 289) | def __call__(self, hidden_state: jnp.ndarray, deterministic: bool = Tr...
  class FlaxResNetStageLayersCollection (line 300) | class FlaxResNetStageLayersCollection(nn.Module):
    method setup (line 312) | def setup(self):
    method __call__ (line 340) | def __call__(self, x: jnp.ndarray, deterministic: bool = True) -> jnp....
  class FlaxResNetStage (line 347) | class FlaxResNetStage(nn.Module):
    method setup (line 359) | def setup(self):
    method __call__ (line 369) | def __call__(self, x: jnp.ndarray, deterministic: bool = True) -> jnp....
  class FlaxResNetStageCollection (line 373) | class FlaxResNetStageCollection(nn.Module):
    method setup (line 377) | def setup(self):
    method __call__ (line 398) | def __call__(
  class FlaxResNetEncoder (line 415) | class FlaxResNetEncoder(nn.Module):
    method setup (line 419) | def setup(self):
    method __call__ (line 422) | def __call__(
  class FlaxResNetPreTrainedModel (line 445) | class FlaxResNetPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 456) | def __init__(
    method init_weights (line 470) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 489) | def __call__(
  class FlaxResNetModule (line 521) | class FlaxResNetModule(nn.Module):
    method setup (line 525) | def setup(self):
    method __call__ (line 535) | def __call__(
  class FlaxResNetModel (line 580) | class FlaxResNetModel(FlaxResNetPreTrainedModel):
  class FlaxResNetClassifierCollection (line 610) | class FlaxResNetClassifierCollection(nn.Module):
    method setup (line 614) | def setup(self):
    method __call__ (line 617) | def __call__(self, x: jnp.ndarray) -> jnp.ndarray:
  class FlaxResNetForImageClassificationModule (line 621) | class FlaxResNetForImageClassificationModule(nn.Module):
    method setup (line 625) | def setup(self):
    method __call__ (line 633) | def __call__(
  class FlaxResNetForImageClassification (line 667) | class FlaxResNetForImageClassification(FlaxResNetPreTrainedModel):

FILE: transformers/models/resnet/modeling_resnet.py
  class ResNetConvLayer (line 62) | class ResNetConvLayer(nn.Module):
    method __init__ (line 63) | def __init__(
    method forward (line 73) | def forward(self, input: Tensor) -> Tensor:
  class ResNetEmbeddings (line 80) | class ResNetEmbeddings(nn.Module):
    method __init__ (line 85) | def __init__(self, config: ResNetConfig):
    method forward (line 93) | def forward(self, pixel_values: Tensor) -> Tensor:
  class ResNetShortCut (line 104) | class ResNetShortCut(nn.Module):
    method __init__ (line 110) | def __init__(self, in_channels: int, out_channels: int, stride: int = 2):
    method forward (line 115) | def forward(self, input: Tensor) -> Tensor:
  class ResNetBasicLayer (line 121) | class ResNetBasicLayer(nn.Module):
    method __init__ (line 126) | def __init__(self, in_channels: int, out_channels: int, stride: int = ...
    method forward (line 138) | def forward(self, hidden_state):
  class ResNetBottleNeckLayer (line 147) | class ResNetBottleNeckLayer(nn.Module):
    method __init__ (line 155) | def __init__(
    method forward (line 171) | def forward(self, hidden_state):
  class ResNetStage (line 180) | class ResNetStage(nn.Module):
    method __init__ (line 185) | def __init__(
    method forward (line 203) | def forward(self, input: Tensor) -> Tensor:
  class ResNetEncoder (line 210) | class ResNetEncoder(nn.Module):
    method __init__ (line 211) | def __init__(self, config: ResNetConfig):
    method forward (line 228) | def forward(
  class ResNetPreTrainedModel (line 251) | class ResNetPreTrainedModel(PreTrainedModel):
    method _init_weights (line 262) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 269) | def _set_gradient_checkpointing(self, module, value=False):
  class ResNetModel (line 303) | class ResNetModel(ResNetPreTrainedModel):
    method __init__ (line 304) | def __init__(self, config):
    method forward (line 321) | def forward(
  class ResNetForImageClassification (line 356) | class ResNetForImageClassification(ResNetPreTrainedModel):
    method __init__ (line 357) | def __init__(self, config):
    method forward (line 376) | def forward(
  class ResNetBackbone (line 432) | class ResNetBackbone(ResNetPreTrainedModel, BackboneMixin):
    method __init__ (line 433) | def __init__(self, config):
    method forward (line 446) | def forward(

FILE: transformers/models/resnet/modeling_tf_resnet.py
  class TFResNetConvLayer (line 52) | class TFResNetConvLayer(tf.keras.layers.Layer):
    method __init__ (line 53) | def __init__(
    method convolution (line 65) | def convolution(self, hidden_state: tf.Tensor) -> tf.Tensor:
    method call (line 72) | def call(self, hidden_state: tf.Tensor, training: bool = False) -> tf....
  class TFResNetEmbeddings (line 79) | class TFResNetEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 84) | def __init__(self, config: ResNetConfig, **kwargs) -> None:
    method call (line 96) | def call(self, pixel_values: tf.Tensor, training: bool = False) -> tf....
  class TFResNetShortCut (line 109) | class TFResNetShortCut(tf.keras.layers.Layer):
    method __init__ (line 115) | def __init__(self, out_channels: int, stride: int = 2, **kwargs) -> None:
    method call (line 123) | def call(self, x: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFResNetBasicLayer (line 130) | class TFResNetBasicLayer(tf.keras.layers.Layer):
    method __init__ (line 135) | def __init__(
    method call (line 149) | def call(self, hidden_state: tf.Tensor, training: bool = False) -> tf....
  class TFResNetBottleNeckLayer (line 159) | class TFResNetBottleNeckLayer(tf.keras.layers.Layer):
    method __init__ (line 167) | def __init__(
    method call (line 189) | def call(self, hidden_state: tf.Tensor, training: bool = False) -> tf....
  class TFResNetStage (line 200) | class TFResNetStage(tf.keras.layers.Layer):
    method __init__ (line 205) | def __init__(
    method call (line 219) | def call(self, hidden_state: tf.Tensor, training: bool = False) -> tf....
  class TFResNetEncoder (line 225) | class TFResNetEncoder(tf.keras.layers.Layer):
    method __init__ (line 226) | def __init__(self, config: ResNetConfig, **kwargs) -> None:
    method call (line 244) | def call(
  class TFResNetPreTrainedModel (line 268) | class TFResNetPreTrainedModel(TFPreTrainedModel):
    method input_signature (line 279) | def input_signature(self):
  class TFResNetMainLayer (line 311) | class TFResNetMainLayer(tf.keras.layers.Layer):
    method __init__ (line 314) | def __init__(self, config: ResNetConfig, **kwargs) -> None:
    method call (line 322) | def call(
  class TFResNetModel (line 372) | class TFResNetModel(TFResNetPreTrainedModel):
    method __init__ (line 373) | def __init__(self, config: ResNetConfig, **kwargs) -> None:
    method call (line 386) | def call(
  class TFResNetForImageClassification (line 414) | class TFResNetForImageClassification(TFResNetPreTrainedModel, TFSequence...
    method __init__ (line 415) | def __init__(self, config: ResNetConfig, **kwargs) -> None:
    method classifier (line 426) | def classifier(self, x: tf.Tensor) -> tf.Tensor:
    method call (line 439) | def call(

FILE: transformers/models/retribert/configuration_retribert.py
  class RetriBertConfig (line 31) | class RetriBertConfig(PretrainedConfig):
    method __init__ (line 77) | def __init__(

FILE: transformers/models/retribert/modeling_retribert.py
  class RetriBertPreTrainedModel (line 42) | class RetriBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 52) | def _init_weights(self, module):
  class RetriBertModel (line 88) | class RetriBertModel(RetriBertPreTrainedModel):
    method __init__ (line 89) | def __init__(self, config: RetriBertConfig) -> None:
    method embed_sentences_checkpointed (line 104) | def embed_sentences_checkpointed(
    method embed_questions (line 148) | def embed_questions(
    method embed_answers (line 162) | def embed_answers(
    method forward (line 176) | def forward(

FILE: transformers/models/retribert/tokenization_retribert.py
  function load_vocab (line 49) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 61) | def whitespace_tokenize(text):
  class RetriBertTokenizer (line 70) | class RetriBertTokenizer(PreTrainedTokenizer):
    method __init__ (line 120) | def __init__(
    method do_lower_case (line 168) | def do_lower_case(self):
    method vocab_size (line 173) | def vocab_size(self):
    method get_vocab (line 177) | def get_vocab(self):
    method _tokenize (line 181) | def _tokenize(self, text):
    method _convert_token_to_id (line 195) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 200) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 205) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 211) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 237) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 266) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 296) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 318) | class BasicTokenizer(object):
    method __init__ (line 338) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 346) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 383) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 394) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 416) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 429) | def _is_chinese_char(self, cp):
    method _clean_text (line 453) | def _clean_text(self, text):
  class WordpieceTokenizer (line 468) | class WordpieceTokenizer(object):
    method __init__ (line 471) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 476) | def tokenize(self, text):

FILE: transformers/models/retribert/tokenization_retribert_fast.py
  class RetriBertTokenizerFast (line 54) | class RetriBertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 105) | def __init__(
    method build_inputs_with_special_tokens (line 148) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 173) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 203) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/roberta/configuration_roberta.py
  class RobertaConfig (line 37) | class RobertaConfig(PretrainedConfig):
    method __init__ (line 106) | def __init__(
  class RobertaOnnxConfig (line 147) | class RobertaOnnxConfig(OnnxConfig):
    method inputs (line 149) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/roberta/convert_roberta_original_pytorch_checkpoint_to_pytorch.py
  function convert_roberta_checkpoint_to_pytorch (line 48) | def convert_roberta_checkpoint_to_pytorch(

FILE: transformers/models/roberta/modeling_flax_roberta.py
  function create_position_ids_from_input_ids (line 52) | def create_position_ids_from_input_ids(input_ids, padding_idx):
  class FlaxRobertaEmbeddings (line 137) | class FlaxRobertaEmbeddings(nn.Module):
    method setup (line 143) | def setup(self):
    method __call__ (line 165) | def __call__(self, input_ids, token_type_ids, position_ids, attention_...
  class FlaxRobertaSelfAttention (line 181) | class FlaxRobertaSelfAttention(nn.Module):
    method setup (line 186) | def setup(self):
    method _split_heads (line 215) | def _split_heads(self, hidden_states):
    method _merge_heads (line 218) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 223) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 254) | def __call__(
  class FlaxRobertaSelfOutput (line 353) | class FlaxRobertaSelfOutput(nn.Module):
    method setup (line 357) | def setup(self):
    method __call__ (line 366) | def __call__(self, hidden_states, input_tensor, deterministic: bool = ...
  class FlaxRobertaAttention (line 374) | class FlaxRobertaAttention(nn.Module):
    method setup (line 379) | def setup(self):
    method __call__ (line 383) | def __call__(
  class FlaxRobertaIntermediate (line 417) | class FlaxRobertaIntermediate(nn.Module):
    method setup (line 421) | def setup(self):
    method __call__ (line 429) | def __call__(self, hidden_states):
  class FlaxRobertaOutput (line 436) | class FlaxRobertaOutput(nn.Module):
    method setup (line 440) | def setup(self):
    method __call__ (line 449) | def __call__(self, hidden_states, attention_output, deterministic: boo...
  class FlaxRobertaLayer (line 457) | class FlaxRobertaLayer(nn.Module):
    method setup (line 461) | def setup(self):
    method __call__ (line 468) | def __call__(
  class FlaxRobertaLayerCollection (line 515) | class FlaxRobertaLayerCollection(nn.Module):
    method setup (line 520) | def setup(self):
    method __call__ (line 533) | def __call__(
  class FlaxRobertaEncoder (line 598) | class FlaxRobertaEncoder(nn.Module):
    method setup (line 603) | def setup(self):
    method __call__ (line 610) | def __call__(
  class FlaxRobertaPooler (line 638) | class FlaxRobertaPooler(nn.Module):
    method setup (line 642) | def setup(self):
    method __call__ (line 649) | def __call__(self, hidden_states):
  class FlaxRobertaLMHead (line 655) | class FlaxRobertaLMHead(nn.Module):
    method setup (line 660) | def setup(self):
    method __call__ (line 675) | def __call__(self, hidden_states, shared_embedding=None):
  class FlaxRobertaClassificationHead (line 690) | class FlaxRobertaClassificationHead(nn.Module):
    method setup (line 694) | def setup(self):
    method __call__ (line 712) | def __call__(self, hidden_states, deterministic=True):
  class FlaxRobertaPreTrainedModel (line 722) | class FlaxRobertaPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 733) | def __init__(
    method enable_gradient_checkpointing (line 747) | def enable_gradient_checkpointing(self):
    method init_weights (line 754) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 797) | def init_cache(self, batch_size, max_length):
    method __call__ (line 817) | def __call__(
  class FlaxRobertaModule (line 915) | class FlaxRobertaModule(nn.Module):
    method setup (line 921) | def setup(self):
    method __call__ (line 930) | def __call__(
  class FlaxRobertaModel (line 990) | class FlaxRobertaModel(FlaxRobertaPreTrainedModel):
  class FlaxRobertaForMaskedLMModule (line 997) | class FlaxRobertaForMaskedLMModule(nn.Module):
    method setup (line 1002) | def setup(self):
    method __call__ (line 1011) | def __call__(
  class FlaxRobertaForMaskedLM (line 1056) | class FlaxRobertaForMaskedLM(FlaxRobertaPreTrainedModel):
  class FlaxRobertaForSequenceClassificationModule (line 1069) | class FlaxRobertaForSequenceClassificationModule(nn.Module):
    method setup (line 1074) | def setup(self):
    method __call__ (line 1083) | def __call__(
  class FlaxRobertaForSequenceClassification (line 1128) | class FlaxRobertaForSequenceClassification(FlaxRobertaPreTrainedModel):
  class FlaxRobertaForMultipleChoiceModule (line 1141) | class FlaxRobertaForMultipleChoiceModule(nn.Module):
    method setup (line 1146) | def setup(self):
    method __call__ (line 1155) | def __call__(
  class FlaxRobertaForMultipleChoice (line 1209) | class FlaxRobertaForMultipleChoice(FlaxRobertaPreTrainedModel):
  class FlaxRobertaForTokenClassificationModule (line 1225) | class FlaxRobertaForTokenClassificationModule(nn.Module):
    method setup (line 1230) | def setup(self):
    method __call__ (line 1245) | def __call__(
  class FlaxRobertaForTokenClassification (line 1291) | class FlaxRobertaForTokenClassification(FlaxRobertaPreTrainedModel):
  class FlaxRobertaForQuestionAnsweringModule (line 1304) | class FlaxRobertaForQuestionAnsweringModule(nn.Module):
    method setup (line 1309) | def setup(self):
    method __call__ (line 1318) | def __call__(
  class FlaxRobertaForQuestionAnswering (line 1368) | class FlaxRobertaForQuestionAnswering(FlaxRobertaPreTrainedModel):
  class FlaxRobertaForCausalLMModule (line 1380) | class FlaxRobertaForCausalLMModule(nn.Module):
    method setup (line 1385) | def setup(self):
    method __call__ (line 1394) | def __call__(
  class FlaxRobertaForCausalLM (line 1452) | class FlaxRobertaForCausalLM(FlaxRobertaPreTrainedModel):
    method prepare_inputs_for_generation (line 1455) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 1476) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/roberta/modeling_roberta.py
  class RobertaEmbeddings (line 65) | class RobertaEmbeddings(nn.Module):
    method __init__ (line 71) | def __init__(self, config):
    method forward (line 94) | def forward(
    method create_position_ids_from_inputs_embeds (line 134) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class RobertaSelfAttention (line 153) | class RobertaSelfAttention(nn.Module):
    method __init__ (line 154) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 180) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 185) | def forward(
  class RobertaSelfOutput (line 288) | class RobertaSelfOutput(nn.Module):
    method __init__ (line 289) | def __init__(self, config):
    method forward (line 295) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RobertaAttention (line 303) | class RobertaAttention(nn.Module):
    method __init__ (line 304) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 310) | def prune_heads(self, heads):
    method forward (line 328) | def forward(
  class RobertaIntermediate (line 353) | class RobertaIntermediate(nn.Module):
    method __init__ (line 354) | def __init__(self, config):
    method forward (line 362) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RobertaOutput (line 369) | class RobertaOutput(nn.Module):
    method __init__ (line 370) | def __init__(self, config):
    method forward (line 376) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RobertaLayer (line 384) | class RobertaLayer(nn.Module):
    method __init__ (line 385) | def __init__(self, config):
    method forward (line 399) | def forward(
    method feed_forward_chunk (line 464) | def feed_forward_chunk(self, attention_output):
  class RobertaEncoder (line 471) | class RobertaEncoder(nn.Module):
    method __init__ (line 472) | def __init__(self, config):
    method forward (line 478) | def forward(
  class RobertaPooler (line 570) | class RobertaPooler(nn.Module):
    method __init__ (line 571) | def __init__(self, config):
    method forward (line 576) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RobertaPreTrainedModel (line 585) | class RobertaPreTrainedModel(PreTrainedModel):
    method _init_weights (line 597) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 613) | def _set_gradient_checkpointing(self, module, value=False):
    method update_keys_to_ignore (line 617) | def update_keys_to_ignore(self, config, del_keys_to_ignore):
  class RobertaModel (line 698) | class RobertaModel(RobertaPreTrainedModel):
    method __init__ (line 717) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 729) | def get_input_embeddings(self):
    method set_input_embeddings (line 732) | def set_input_embeddings(self, value):
    method _prune_heads (line 735) | def _prune_heads(self, heads_to_prune):
    method forward (line 750) | def forward(
  class RobertaForCausalLM (line 883) | class RobertaForCausalLM(RobertaPreTrainedModel):
    method __init__ (line 888) | def __init__(self, config):
    method get_output_embeddings (line 903) | def get_output_embeddings(self):
    method set_output_embeddings (line 906) | def set_output_embeddings(self, new_embeddings):
    method forward (line 911) | def forward(
    method prepare_inputs_for_generation (line 1017) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1029) | def _reorder_cache(self, past_key_values, beam_idx):
  class RobertaForMaskedLM (line 1037) | class RobertaForMaskedLM(RobertaPreTrainedModel):
    method __init__ (line 1042) | def __init__(self, config):
    method get_output_embeddings (line 1060) | def get_output_embeddings(self):
    method set_output_embeddings (line 1063) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1075) | def forward(
  class RobertaLMHead (line 1135) | class RobertaLMHead(nn.Module):
    method __init__ (line 1138) | def __init__(self, config):
    method forward (line 1147) | def forward(self, features, **kwargs):
    method _tie_weights (line 1157) | def _tie_weights(self):
  class RobertaForSequenceClassification (line 1173) | class RobertaForSequenceClassification(RobertaPreTrainedModel):
    method __init__ (line 1176) | def __init__(self, config):
    method forward (line 1195) | def forward(
  class RobertaForMultipleChoice (line 1274) | class RobertaForMultipleChoice(RobertaPreTrainedModel):
    method __init__ (line 1277) | def __init__(self, config):
    method forward (line 1293) | def forward(
  class RobertaForTokenClassification (line 1368) | class RobertaForTokenClassification(RobertaPreTrainedModel):
    method __init__ (line 1372) | def __init__(self, config):
    method forward (line 1394) | def forward(
  class RobertaClassificationHead (line 1449) | class RobertaClassificationHead(nn.Module):
    method __init__ (line 1452) | def __init__(self, config):
    method forward (line 1461) | def forward(self, features, **kwargs):
  class RobertaForQuestionAnswering (line 1478) | class RobertaForQuestionAnswering(RobertaPreTrainedModel):
    method __init__ (line 1482) | def __init__(self, config):
    method forward (line 1500) | def forward(
  function create_position_ids_from_input_ids (line 1575) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/roberta/modeling_tf_roberta.py
  class TFRobertaEmbeddings (line 76) | class TFRobertaEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 81) | def __init__(self, config, **kwargs):
    method build (line 92) | def build(self, input_shape: tf.TensorShape):
    method create_position_ids_from_input_ids (line 116) | def create_position_ids_from_input_ids(self, input_ids, past_key_value...
    method call (line 130) | def call(
  class TFRobertaPooler (line 177) | class TFRobertaPooler(tf.keras.layers.Layer):
    method __init__ (line 178) | def __init__(self, config: RobertaConfig, **kwargs):
    method call (line 188) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFRobertaSelfAttention (line 198) | class TFRobertaSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 199) | def __init__(self, config: RobertaConfig, **kwargs):
    method transpose_for_scores (line 226) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 233) | def call(
  class TFRobertaSelfOutput (line 316) | class TFRobertaSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 317) | def __init__(self, config: RobertaConfig, **kwargs):
    method call (line 326) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFRobertaAttention (line 335) | class TFRobertaAttention(tf.keras.layers.Layer):
    method __init__ (line 336) | def __init__(self, config: RobertaConfig, **kwargs):
    method prune_heads (line 342) | def prune_heads(self, heads):
    method call (line 345) | def call(
  class TFRobertaIntermediate (line 376) | class TFRobertaIntermediate(tf.keras.layers.Layer):
    method __init__ (line 377) | def __init__(self, config: RobertaConfig, **kwargs):
    method call (line 389) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFRobertaOutput (line 397) | class TFRobertaOutput(tf.keras.layers.Layer):
    method __init__ (line 398) | def __init__(self, config: RobertaConfig, **kwargs):
    method call (line 407) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFRobertaLayer (line 416) | class TFRobertaLayer(tf.keras.layers.Layer):
    method __init__ (line 417) | def __init__(self, config: RobertaConfig, **kwargs):
    method call (line 430) | def call(
  class TFRobertaEncoder (line 503) | class TFRobertaEncoder(tf.keras.layers.Layer):
    method __init__ (line 504) | def __init__(self, config: RobertaConfig, **kwargs):
    method call (line 509) | def call(
  class TFRobertaMainLayer (line 573) | class TFRobertaMainLayer(tf.keras.layers.Layer):
    method __init__ (line 576) | def __init__(self, config, add_pooling_layer=True, **kwargs):
    method get_input_embeddings (line 593) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 597) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 602) | def _prune_heads(self, heads_to_prune):
    method call (line 611) | def call(
  class TFRobertaPreTrainedModel (line 769) | class TFRobertaPreTrainedModel(TFPreTrainedModel):
  class TFRobertaModel (line 881) | class TFRobertaModel(TFRobertaPreTrainedModel):
    method __init__ (line 882) | def __init__(self, config, *inputs, **kwargs):
    method call (line 893) | def call(
  class TFRobertaLMHead (line 950) | class TFRobertaLMHead(tf.keras.layers.Layer):
    method __init__ (line 953) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 968) | def build(self, input_shape):
    method get_output_embeddings (line 973) | def get_output_embeddings(self):
    method set_output_embeddings (line 976) | def set_output_embeddings(self, value):
    method get_bias (line 980) | def get_bias(self):
    method set_bias (line 983) | def set_bias(self, value):
    method call (line 987) | def call(self, hidden_states):
  class TFRobertaForMaskedLM (line 1003) | class TFRobertaForMaskedLM(TFRobertaPreTrainedModel, TFMaskedLanguageMod...
    method __init__ (line 1007) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 1013) | def get_lm_head(self):
    method get_prefix_bias_name (line 1016) | def get_prefix_bias_name(self):
    method call (line 1030) | def call(
  class TFRobertaForCausalLM (line 1080) | class TFRobertaForCausalLM(TFRobertaPreTrainedModel, TFCausalLanguageMod...
    method __init__ (line 1084) | def __init__(self, config: RobertaConfig, *inputs, **kwargs):
    method get_lm_head (line 1093) | def get_lm_head(self):
    method get_prefix_bias_name (line 1096) | def get_prefix_bias_name(self):
    method prepare_inputs_for_generation (line 1101) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method call (line 1120) | def call(
  class TFRobertaClassificationHead (line 1202) | class TFRobertaClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 1205) | def __init__(self, config, **kwargs):
    method call (line 1221) | def call(self, features, training=False):
  class TFRobertaForSequenceClassification (line 1237) | class TFRobertaForSequenceClassification(TFRobertaPreTrainedModel, TFSeq...
    method __init__ (line 1241) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1257) | def call(
  class TFRobertaForMultipleChoice (line 1313) | class TFRobertaForMultipleChoice(TFRobertaPreTrainedModel, TFMultipleCho...
    method __init__ (line 1318) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1334) | def call(
  class TFRobertaForTokenClassification (line 1403) | class TFRobertaForTokenClassification(TFRobertaPreTrainedModel, TFTokenC...
    method __init__ (line 1408) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1430) | def call(
  class TFRobertaForQuestionAnswering (line 1486) | class TFRobertaForQuestionAnswering(TFRobertaPreTrainedModel, TFQuestion...
    method __init__ (line 1490) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1508) | def call(

FILE: transformers/models/roberta/tokenization_roberta.py
  function bytes_to_unicode (line 69) | def bytes_to_unicode():
  function get_pairs (line 93) | def get_pairs(word):
  class RobertaTokenizer (line 107) | class RobertaTokenizer(PreTrainedTokenizer):
    method __init__ (line 190) | def __init__(
    method vocab_size (line 245) | def vocab_size(self):
    method get_vocab (line 248) | def get_vocab(self):
    method bpe (line 251) | def bpe(self, token):
    method _tokenize (line 293) | def _tokenize(self, text):
    method _convert_token_to_id (line 303) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 307) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 311) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 317) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 346) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 371) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 398) | def create_token_type_ids_from_sequences(
    method prepare_for_tokenization (line 421) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...

FILE: transformers/models/roberta/tokenization_roberta_fast.py
  class RobertaTokenizerFast (line 76) | class RobertaTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 163) | def __init__(
    method mask_token (line 232) | def mask_token(self) -> str:
    method mask_token (line 247) | def mask_token(self, value):
    method _batch_encode_plus (line 258) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 267) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method save_vocabulary (line 277) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 281) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 288) | def create_token_type_ids_from_sequences(

FILE: transformers/models/roberta_prelayernorm/configuration_roberta_prelayernorm.py
  class RobertaPreLayerNormConfig (line 35) | class RobertaPreLayerNormConfig(PretrainedConfig):
    method __init__ (line 107) | def __init__(
  class RobertaPreLayerNormOnnxConfig (line 149) | class RobertaPreLayerNormOnnxConfig(OnnxConfig):
    method inputs (line 151) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/roberta_prelayernorm/convert_roberta_prelayernorm_original_pytorch_checkpoint_to_pytorch.py
  function convert_roberta_prelayernorm_checkpoint_to_pytorch (line 31) | def convert_roberta_prelayernorm_checkpoint_to_pytorch(checkpoint_repo: ...

FILE: transformers/models/roberta_prelayernorm/modeling_flax_roberta_prelayernorm.py
  function create_position_ids_from_input_ids (line 54) | def create_position_ids_from_input_ids(input_ids, padding_idx):
  class FlaxRobertaPreLayerNormEmbeddings (line 139) | class FlaxRobertaPreLayerNormEmbeddings(nn.Module):
    method setup (line 145) | def setup(self):
    method __call__ (line 167) | def __call__(self, input_ids, token_type_ids, position_ids, attention_...
  class FlaxRobertaPreLayerNormSelfAttention (line 183) | class FlaxRobertaPreLayerNormSelfAttention(nn.Module):
    method setup (line 188) | def setup(self):
    method _split_heads (line 217) | def _split_heads(self, hidden_states):
    method _merge_heads (line 220) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 225) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 256) | def __call__(
  class FlaxRobertaPreLayerNormSelfOutput (line 354) | class FlaxRobertaPreLayerNormSelfOutput(nn.Module):
    method setup (line 358) | def setup(self):
    method __call__ (line 366) | def __call__(self, hidden_states, input_tensor, deterministic: bool = ...
  class FlaxRobertaPreLayerNormAttention (line 373) | class FlaxRobertaPreLayerNormAttention(nn.Module):
    method setup (line 378) | def setup(self):
    method __call__ (line 383) | def __call__(
  class FlaxRobertaPreLayerNormIntermediate (line 417) | class FlaxRobertaPreLayerNormIntermediate(nn.Module):
    method setup (line 421) | def setup(self):
    method __call__ (line 430) | def __call__(self, hidden_states):
  class FlaxRobertaPreLayerNormOutput (line 437) | class FlaxRobertaPreLayerNormOutput(nn.Module):
    method setup (line 441) | def setup(self):
    method __call__ (line 449) | def __call__(self, hidden_states, attention_output, deterministic: boo...
  class FlaxRobertaPreLayerNormLayer (line 457) | class FlaxRobertaPreLayerNormLayer(nn.Module):
    method setup (line 461) | def setup(self):
    method __call__ (line 468) | def __call__(
  class FlaxRobertaPreLayerNormLayerCollection (line 515) | class FlaxRobertaPreLayerNormLayerCollection(nn.Module):
    method setup (line 520) | def setup(self):
    method __call__ (line 533) | def __call__(
  class FlaxRobertaPreLayerNormEncoder (line 598) | class FlaxRobertaPreLayerNormEncoder(nn.Module):
    method setup (line 603) | def setup(self):
    method __call__ (line 610) | def __call__(
  class FlaxRobertaPreLayerNormPooler (line 638) | class FlaxRobertaPreLayerNormPooler(nn.Module):
    method setup (line 642) | def setup(self):
    method __call__ (line 649) | def __call__(self, hidden_states):
  class FlaxRobertaPreLayerNormLMHead (line 656) | class FlaxRobertaPreLayerNormLMHead(nn.Module):
    method setup (line 661) | def setup(self):
    method __call__ (line 676) | def __call__(self, hidden_states, shared_embedding=None):
  class FlaxRobertaPreLayerNormClassificationHead (line 692) | class FlaxRobertaPreLayerNormClassificationHead(nn.Module):
    method setup (line 696) | def setup(self):
    method __call__ (line 714) | def __call__(self, hidden_states, deterministic=True):
  class FlaxRobertaPreLayerNormPreTrainedModel (line 725) | class FlaxRobertaPreLayerNormPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 736) | def __init__(
    method enable_gradient_checkpointing (line 750) | def enable_gradient_checkpointing(self):
    method init_weights (line 757) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 800) | def init_cache(self, batch_size, max_length):
    method __call__ (line 820) | def __call__(
  class FlaxRobertaPreLayerNormModule (line 917) | class FlaxRobertaPreLayerNormModule(nn.Module):
    method setup (line 923) | def setup(self):
    method __call__ (line 933) | def __call__(
  class FlaxRobertaPreLayerNormModel (line 995) | class FlaxRobertaPreLayerNormModel(FlaxRobertaPreLayerNormPreTrainedModel):
  class FlaxRobertaPreLayerNormForMaskedLMModule (line 1008) | class FlaxRobertaPreLayerNormForMaskedLMModule(nn.Module):
    method setup (line 1013) | def setup(self):
    method __call__ (line 1022) | def __call__(
  class FlaxRobertaPreLayerNormForMaskedLM (line 1072) | class FlaxRobertaPreLayerNormForMaskedLM(FlaxRobertaPreLayerNormPreTrain...
  class FlaxRobertaPreLayerNormForSequenceClassificationModule (line 1086) | class FlaxRobertaPreLayerNormForSequenceClassificationModule(nn.Module):
    method setup (line 1091) | def setup(self):
    method __call__ (line 1100) | def __call__(
  class FlaxRobertaPreLayerNormForSequenceClassification (line 1146) | class FlaxRobertaPreLayerNormForSequenceClassification(FlaxRobertaPreLay...
  class FlaxRobertaPreLayerNormForMultipleChoiceModule (line 1159) | class FlaxRobertaPreLayerNormForMultipleChoiceModule(nn.Module):
    method setup (line 1164) | def setup(self):
    method __call__ (line 1173) | def __call__(
  class FlaxRobertaPreLayerNormForMultipleChoice (line 1228) | class FlaxRobertaPreLayerNormForMultipleChoice(FlaxRobertaPreLayerNormPr...
  class FlaxRobertaPreLayerNormForTokenClassificationModule (line 1245) | class FlaxRobertaPreLayerNormForTokenClassificationModule(nn.Module):
    method setup (line 1250) | def setup(self):
    method __call__ (line 1265) | def __call__(
  class FlaxRobertaPreLayerNormForTokenClassification (line 1312) | class FlaxRobertaPreLayerNormForTokenClassification(FlaxRobertaPreLayerN...
  class FlaxRobertaPreLayerNormForQuestionAnsweringModule (line 1325) | class FlaxRobertaPreLayerNormForQuestionAnsweringModule(nn.Module):
    method setup (line 1330) | def setup(self):
    method __call__ (line 1339) | def __call__(
  class FlaxRobertaPreLayerNormForQuestionAnswering (line 1390) | class FlaxRobertaPreLayerNormForQuestionAnswering(FlaxRobertaPreLayerNor...
  class FlaxRobertaPreLayerNormForCausalLMModule (line 1403) | class FlaxRobertaPreLayerNormForCausalLMModule(nn.Module):
    method setup (line 1408) | def setup(self):
    method __call__ (line 1417) | def __call__(
  class FlaxRobertaPreLayerNormForCausalLM (line 1478) | class FlaxRobertaPreLayerNormForCausalLM(FlaxRobertaPreLayerNormPreTrain...
    method prepare_inputs_for_generation (line 1481) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 1502) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/roberta_prelayernorm/modeling_roberta_prelayernorm.py
  class RobertaPreLayerNormEmbeddings (line 68) | class RobertaPreLayerNormEmbeddings(nn.Module):
    method __init__ (line 74) | def __init__(self, config):
    method forward (line 97) | def forward(
    method create_position_ids_from_inputs_embeds (line 137) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class RobertaPreLayerNormSelfAttention (line 156) | class RobertaPreLayerNormSelfAttention(nn.Module):
    method __init__ (line 157) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 183) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 188) | def forward(
  class RobertaPreLayerNormSelfOutput (line 290) | class RobertaPreLayerNormSelfOutput(nn.Module):
    method __init__ (line 291) | def __init__(self, config):
    method forward (line 296) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RobertaPreLayerNormAttention (line 303) | class RobertaPreLayerNormAttention(nn.Module):
    method __init__ (line 304) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 312) | def prune_heads(self, heads):
    method forward (line 330) | def forward(
  class RobertaPreLayerNormIntermediate (line 355) | class RobertaPreLayerNormIntermediate(nn.Module):
    method __init__ (line 356) | def __init__(self, config):
    method forward (line 365) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RobertaPreLayerNormOutput (line 372) | class RobertaPreLayerNormOutput(nn.Module):
    method __init__ (line 373) | def __init__(self, config):
    method forward (line 378) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RobertaPreLayerNormLayer (line 386) | class RobertaPreLayerNormLayer(nn.Module):
    method __init__ (line 387) | def __init__(self, config):
    method forward (line 401) | def forward(
    method feed_forward_chunk (line 466) | def feed_forward_chunk(self, attention_output):
  class RobertaPreLayerNormEncoder (line 473) | class RobertaPreLayerNormEncoder(nn.Module):
    method __init__ (line 474) | def __init__(self, config):
    method forward (line 480) | def forward(
  class RobertaPreLayerNormPooler (line 572) | class RobertaPreLayerNormPooler(nn.Module):
    method __init__ (line 573) | def __init__(self, config):
    method forward (line 578) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RobertaPreLayerNormPreTrainedModel (line 588) | class RobertaPreLayerNormPreTrainedModel(PreTrainedModel):
    method _init_weights (line 600) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 616) | def _set_gradient_checkpointing(self, module, value=False):
    method update_keys_to_ignore (line 620) | def update_keys_to_ignore(self, config, del_keys_to_ignore):
  class RobertaPreLayerNormModel (line 701) | class RobertaPreLayerNormModel(RobertaPreLayerNormPreTrainedModel):
    method __init__ (line 719) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 732) | def get_input_embeddings(self):
    method set_input_embeddings (line 735) | def set_input_embeddings(self, value):
    method _prune_heads (line 738) | def _prune_heads(self, heads_to_prune):
    method forward (line 752) | def forward(
  class RobertaPreLayerNormForCausalLM (line 888) | class RobertaPreLayerNormForCausalLM(RobertaPreLayerNormPreTrainedModel):
    method __init__ (line 893) | def __init__(self, config):
    method get_output_embeddings (line 910) | def get_output_embeddings(self):
    method set_output_embeddings (line 913) | def set_output_embeddings(self, new_embeddings):
    method forward (line 918) | def forward(
    method prepare_inputs_for_generation (line 1024) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1036) | def _reorder_cache(self, past_key_values, beam_idx):
  class RobertaPreLayerNormForMaskedLM (line 1046) | class RobertaPreLayerNormForMaskedLM(RobertaPreLayerNormPreTrainedModel):
    method __init__ (line 1052) | def __init__(self, config):
    method get_output_embeddings (line 1070) | def get_output_embeddings(self):
    method set_output_embeddings (line 1073) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1086) | def forward(
  class RobertaPreLayerNormLMHead (line 1147) | class RobertaPreLayerNormLMHead(nn.Module):
    method __init__ (line 1150) | def __init__(self, config):
    method forward (line 1159) | def forward(self, features, **kwargs):
    method _tie_weights (line 1169) | def _tie_weights(self):
  class RobertaPreLayerNormForSequenceClassification (line 1185) | class RobertaPreLayerNormForSequenceClassification(RobertaPreLayerNormPr...
    method __init__ (line 1188) | def __init__(self, config):
    method forward (line 1206) | def forward(
  class RobertaPreLayerNormForMultipleChoice (line 1286) | class RobertaPreLayerNormForMultipleChoice(RobertaPreLayerNormPreTrained...
    method __init__ (line 1289) | def __init__(self, config):
    method forward (line 1307) | def forward(
  class RobertaPreLayerNormForTokenClassification (line 1382) | class RobertaPreLayerNormForTokenClassification(RobertaPreLayerNormPreTr...
    method __init__ (line 1386) | def __init__(self, config):
    method forward (line 1407) | def forward(
  class RobertaPreLayerNormClassificationHead (line 1463) | class RobertaPreLayerNormClassificationHead(nn.Module):
    method __init__ (line 1466) | def __init__(self, config):
    method forward (line 1475) | def forward(self, features, **kwargs):
  class RobertaPreLayerNormForQuestionAnswering (line 1492) | class RobertaPreLayerNormForQuestionAnswering(RobertaPreLayerNormPreTrai...
    method __init__ (line 1496) | def __init__(self, config):
    method forward (line 1513) | def forward(
  function create_position_ids_from_input_ids (line 1588) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/roberta_prelayernorm/modeling_tf_roberta_prelayernorm.py
  class TFRobertaPreLayerNormEmbeddings (line 81) | class TFRobertaPreLayerNormEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 86) | def __init__(self, config, **kwargs):
    method build (line 97) | def build(self, input_shape: tf.TensorShape):
    method create_position_ids_from_input_ids (line 121) | def create_position_ids_from_input_ids(self, input_ids, past_key_value...
    method call (line 135) | def call(
  class TFRobertaPreLayerNormPooler (line 182) | class TFRobertaPreLayerNormPooler(tf.keras.layers.Layer):
    method __init__ (line 183) | def __init__(self, config: RobertaPreLayerNormConfig, **kwargs):
    method call (line 193) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFRobertaPreLayerNormSelfAttention (line 203) | class TFRobertaPreLayerNormSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 204) | def __init__(self, config: RobertaPreLayerNormConfig, **kwargs):
    method transpose_for_scores (line 231) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 238) | def call(
  class TFRobertaPreLayerNormSelfOutput (line 320) | class TFRobertaPreLayerNormSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 321) | def __init__(self, config: RobertaPreLayerNormConfig, **kwargs):
    method call (line 329) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFRobertaPreLayerNormAttention (line 337) | class TFRobertaPreLayerNormAttention(tf.keras.layers.Layer):
    method __init__ (line 338) | def __init__(self, config: RobertaPreLayerNormConfig, **kwargs):
    method prune_heads (line 346) | def prune_heads(self, heads):
    method call (line 349) | def call(
  class TFRobertaPreLayerNormIntermediate (line 380) | class TFRobertaPreLayerNormIntermediate(tf.keras.layers.Layer):
    method __init__ (line 381) | def __init__(self, config: RobertaPreLayerNormConfig, **kwargs):
    method call (line 394) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFRobertaPreLayerNormOutput (line 402) | class TFRobertaPreLayerNormOutput(tf.keras.layers.Layer):
    method __init__ (line 403) | def __init__(self, config: RobertaPreLayerNormConfig, **kwargs):
    method call (line 411) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFRobertaPreLayerNormLayer (line 420) | class TFRobertaPreLayerNormLayer(tf.keras.layers.Layer):
    method __init__ (line 421) | def __init__(self, config: RobertaPreLayerNormConfig, **kwargs):
    method call (line 434) | def call(
  class TFRobertaPreLayerNormEncoder (line 507) | class TFRobertaPreLayerNormEncoder(tf.keras.layers.Layer):
    method __init__ (line 508) | def __init__(self, config: RobertaPreLayerNormConfig, **kwargs):
    method call (line 513) | def call(
  class TFRobertaPreLayerNormMainLayer (line 577) | class TFRobertaPreLayerNormMainLayer(tf.keras.layers.Layer):
    method __init__ (line 580) | def __init__(self, config, add_pooling_layer=True, **kwargs):
    method get_input_embeddings (line 597) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 600) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 604) | def _prune_heads(self, heads_to_prune):
    method call (line 612) | def call(
  class TFRobertaPreLayerNormPreTrainedModel (line 770) | class TFRobertaPreLayerNormPreTrainedModel(TFPreTrainedModel):
  class TFRobertaPreLayerNormModel (line 883) | class TFRobertaPreLayerNormModel(TFRobertaPreLayerNormPreTrainedModel):
    method __init__ (line 884) | def __init__(self, config, *inputs, **kwargs):
    method call (line 895) | def call(
  class TFRobertaPreLayerNormLMHead (line 953) | class TFRobertaPreLayerNormLMHead(tf.keras.layers.Layer):
    method __init__ (line 956) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 971) | def build(self, input_shape):
    method get_output_embeddings (line 976) | def get_output_embeddings(self):
    method set_output_embeddings (line 979) | def set_output_embeddings(self, value):
    method get_bias (line 983) | def get_bias(self):
    method set_bias (line 986) | def set_bias(self, value):
    method call (line 990) | def call(self, hidden_states):
  class TFRobertaPreLayerNormForMaskedLM (line 1008) | class TFRobertaPreLayerNormForMaskedLM(TFRobertaPreLayerNormPreTrainedMo...
    method __init__ (line 1013) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 1021) | def get_lm_head(self):
    method get_prefix_bias_name (line 1024) | def get_prefix_bias_name(self):
    method call (line 1039) | def call(
  class TFRobertaPreLayerNormForCausalLM (line 1090) | class TFRobertaPreLayerNormForCausalLM(TFRobertaPreLayerNormPreTrainedMo...
    method __init__ (line 1094) | def __init__(self, config: RobertaPreLayerNormConfig, *inputs, **kwargs):
    method get_lm_head (line 1109) | def get_lm_head(self):
    method get_prefix_bias_name (line 1112) | def get_prefix_bias_name(self):
    method prepare_inputs_for_generation (line 1117) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method call (line 1136) | def call(
  class TFRobertaPreLayerNormClassificationHead (line 1219) | class TFRobertaPreLayerNormClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 1222) | def __init__(self, config, **kwargs):
    method call (line 1238) | def call(self, features, training=False):
  class TFRobertaPreLayerNormForSequenceClassification (line 1254) | class TFRobertaPreLayerNormForSequenceClassification(
    method __init__ (line 1260) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1277) | def call(
  class TFRobertaPreLayerNormForMultipleChoice (line 1334) | class TFRobertaPreLayerNormForMultipleChoice(TFRobertaPreLayerNormPreTra...
    method __init__ (line 1339) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1357) | def call(
  class TFRobertaPreLayerNormForTokenClassification (line 1426) | class TFRobertaPreLayerNormForTokenClassification(TFRobertaPreLayerNormP...
    method __init__ (line 1431) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1454) | def call(
  class TFRobertaPreLayerNormForQuestionAnswering (line 1510) | class TFRobertaPreLayerNormForQuestionAnswering(TFRobertaPreLayerNormPre...
    method __init__ (line 1514) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1533) | def call(

FILE: transformers/models/roc_bert/configuration_roc_bert.py
  class RoCBertConfig (line 28) | class RoCBertConfig(PretrainedConfig):
    method __init__ (line 114) | def __init__(

FILE: transformers/models/roc_bert/modeling_roc_bert.py
  function load_tf_weights_in_roc_bert (line 84) | def load_tf_weights_in_roc_bert(model, config, tf_checkpoint_path):
  class RoCBertEmbeddings (line 157) | class RoCBertEmbeddings(nn.Module):
    method __init__ (line 160) | def __init__(self, config):
    method forward (line 201) | def forward(
  class RoCBertSelfAttention (line 287) | class RoCBertSelfAttention(nn.Module):
    method __init__ (line 288) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 314) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 319) | def forward(
  class RoCBertSelfOutput (line 422) | class RoCBertSelfOutput(nn.Module):
    method __init__ (line 423) | def __init__(self, config):
    method forward (line 429) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RoCBertAttention (line 437) | class RoCBertAttention(nn.Module):
    method __init__ (line 438) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 444) | def prune_heads(self, heads):
    method forward (line 462) | def forward(
  class RoCBertIntermediate (line 487) | class RoCBertIntermediate(nn.Module):
    method __init__ (line 488) | def __init__(self, config):
    method forward (line 496) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RoCBertOutput (line 503) | class RoCBertOutput(nn.Module):
    method __init__ (line 504) | def __init__(self, config):
    method forward (line 510) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RoCBertLayer (line 518) | class RoCBertLayer(nn.Module):
    method __init__ (line 519) | def __init__(self, config):
    method forward (line 533) | def forward(
    method feed_forward_chunk (line 598) | def feed_forward_chunk(self, attention_output):
  class RoCBertEncoder (line 605) | class RoCBertEncoder(nn.Module):
    method __init__ (line 606) | def __init__(self, config):
    method forward (line 612) | def forward(
  class RoCBertPooler (line 704) | class RoCBertPooler(nn.Module):
    method __init__ (line 705) | def __init__(self, config):
    method forward (line 710) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RoCBertPredictionHeadTransform (line 720) | class RoCBertPredictionHeadTransform(nn.Module):
    method __init__ (line 721) | def __init__(self, config):
    method forward (line 730) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RoCBertLMPredictionHead (line 738) | class RoCBertLMPredictionHead(nn.Module):
    method __init__ (line 739) | def __init__(self, config):
    method forward (line 752) | def forward(self, hidden_states):
  class RoCBertOnlyMLMHead (line 759) | class RoCBertOnlyMLMHead(nn.Module):
    method __init__ (line 760) | def __init__(self, config):
    method forward (line 764) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class RoCBertPreTrainedModel (line 770) | class RoCBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 782) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 798) | def _set_gradient_checkpointing(self, module, value=False):
  class RoCBertModel (line 882) | class RoCBertModel(RoCBertPreTrainedModel):
    method __init__ (line 896) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 909) | def get_input_embeddings(self):
    method set_input_embeddings (line 913) | def set_input_embeddings(self, value):
    method get_pronunciation_embeddings (line 916) | def get_pronunciation_embeddings(self):
    method set_pronunciation_embeddings (line 919) | def set_pronunciation_embeddings(self, value):
    method get_shape_embeddings (line 922) | def get_shape_embeddings(self):
    method set_shape_embeddings (line 925) | def set_shape_embeddings(self, value):
    method _prune_heads (line 929) | def _prune_heads(self, heads_to_prune):
    method forward (line 944) | def forward(
  class RoCBertForPreTraining (line 1083) | class RoCBertForPreTraining(RoCBertPreTrainedModel):
    method __init__ (line 1086) | def __init__(self, config):
    method get_output_embeddings (line 1096) | def get_output_embeddings(self):
    method set_output_embeddings (line 1100) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1105) | def forward(
  class RoCBertForMaskedLM (line 1268) | class RoCBertForMaskedLM(RoCBertPreTrainedModel):
    method __init__ (line 1273) | def __init__(self, config):
    method get_output_embeddings (line 1289) | def get_output_embeddings(self):
    method set_output_embeddings (line 1293) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1297) | def forward(
    method prepare_inputs_for_generation (line 1378) | def prepare_inputs_for_generation(
  class RoCBertForCausalLM (line 1409) | class RoCBertForCausalLM(RoCBertPreTrainedModel):
    method __init__ (line 1414) | def __init__(self, config):
    method get_output_embeddings (line 1427) | def get_output_embeddings(self):
    method set_output_embeddings (line 1431) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1436) | def forward(
    method prepare_inputs_for_generation (line 1548) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1580) | def _reorder_cache(self, past_key_values, beam_idx):
  class RoCBertForSequenceClassification (line 1592) | class RoCBertForSequenceClassification(RoCBertPreTrainedModel):
    method __init__ (line 1594) | def __init__(self, config):
    method forward (line 1617) | def forward(
  class RoCBertForMultipleChoice (line 1698) | class RoCBertForMultipleChoice(RoCBertPreTrainedModel):
    method __init__ (line 1700) | def __init__(self, config):
    method forward (line 1721) | def forward(
  class RoCBertForTokenClassification (line 1803) | class RoCBertForTokenClassification(RoCBertPreTrainedModel):
    method __init__ (line 1807) | def __init__(self, config):
    method forward (line 1829) | def forward(
  class RoCBertForQuestionAnswering (line 1891) | class RoCBertForQuestionAnswering(RoCBertPreTrainedModel):
    method __init__ (line 1895) | def __init__(self, config):
    method forward (line 1915) | def forward(

FILE: transformers/models/roc_bert/tokenization_roc_bert.py
  function load_vocab (line 74) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 86) | def whitespace_tokenize(text):
  class RoCBertTokenizer (line 95) | class RoCBertTokenizer(PreTrainedTokenizer):
    method __init__ (line 142) | def __init__(
    method do_lower_case (line 201) | def do_lower_case(self):
    method vocab_size (line 205) | def vocab_size(self):
    method get_vocab (line 209) | def get_vocab(self):
    method _tokenize (line 213) | def _tokenize(self, text):
    method _encode_plus (line 226) | def _encode_plus(
    method prepare_for_model (line 321) | def prepare_for_model(
    method _pad (line 501) | def _pad(
    method _batch_encode_plus (line 561) | def _batch_encode_plus(
    method _batch_prepare_for_model (line 666) | def _batch_prepare_for_model(
    method _convert_token_to_id (line 741) | def _convert_token_to_id(self, token):
    method _convert_token_to_shape_id (line 745) | def _convert_token_to_shape_id(self, token):
    method convert_tokens_to_shape_ids (line 749) | def convert_tokens_to_shape_ids(self, tokens: Union[str, List[str]]) -...
    method _convert_token_to_pronunciation_id (line 758) | def _convert_token_to_pronunciation_id(self, token):
    method convert_tokens_to_pronunciation_ids (line 762) | def convert_tokens_to_pronunciation_ids(self, tokens: Union[str, List[...
    method _convert_id_to_token (line 772) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 777) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 782) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 812) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 841) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 870) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class RoCBertBasicTokenizer (line 916) | class RoCBertBasicTokenizer(object):
    method __init__ (line 936) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 944) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 981) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 992) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 1014) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 1027) | def _is_chinese_char(self, cp):
    method _clean_text (line 1051) | def _clean_text(self, text):
  class RoCBertWordpieceTokenizer (line 1066) | class RoCBertWordpieceTokenizer(object):
    method __init__ (line 1069) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 1074) | def tokenize(self, text):

FILE: transformers/models/roformer/configuration_roformer.py
  class RoFormerConfig (line 46) | class RoFormerConfig(PretrainedConfig):
    method __init__ (line 111) | def __init__(
  class RoFormerOnnxConfig (line 150) | class RoFormerOnnxConfig(OnnxConfig):
    method inputs (line 152) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/roformer/convert_roformer_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 29) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, bert_config_fil...

FILE: transformers/models/roformer/modeling_flax_roformer.py
  function create_sinusoidal_positions (line 130) | def create_sinusoidal_positions(n_pos, dim):
  class FlaxRoFormerEmbeddings (line 140) | class FlaxRoFormerEmbeddings(nn.Module):
    method setup (line 146) | def setup(self):
    method __call__ (line 160) | def __call__(self, input_ids, token_type_ids, attention_mask, determin...
  class FlaxRoFormerSelfAttention (line 174) | class FlaxRoFormerSelfAttention(nn.Module):
    method setup (line 178) | def setup(self) -> None:
    method __call__ (line 203) | def __call__(
    method apply_rotary_position_embeddings (line 273) | def apply_rotary_position_embeddings(sinusoidal_pos, query_layer, key_...
  class FlaxRoFormerSelfOutput (line 293) | class FlaxRoFormerSelfOutput(nn.Module):
    method setup (line 297) | def setup(self):
    method __call__ (line 306) | def __call__(self, hidden_states, input_tensor, deterministic: bool = ...
  class FlaxRoFormerAttention (line 313) | class FlaxRoFormerAttention(nn.Module):
    method setup (line 317) | def setup(self):
    method __call__ (line 321) | def __call__(
  class FlaxRoFormerIntermediate (line 353) | class FlaxRoFormerIntermediate(nn.Module):
    method setup (line 357) | def setup(self):
    method __call__ (line 365) | def __call__(self, hidden_states):
  class FlaxRoFormerOutput (line 372) | class FlaxRoFormerOutput(nn.Module):
    method setup (line 376) | def setup(self):
    method __call__ (line 385) | def __call__(self, hidden_states, attention_output, deterministic: boo...
  class FlaxRoFormerLayer (line 392) | class FlaxRoFormerLayer(nn.Module):
    method setup (line 396) | def setup(self):
    method __call__ (line 401) | def __call__(
  class FlaxRoFormerLayerCollection (line 430) | class FlaxRoFormerLayerCollection(nn.Module):
    method setup (line 434) | def setup(self):
    method __call__ (line 439) | def __call__(
  class FlaxRoFormerEncoder (line 492) | class FlaxRoFormerEncoder(nn.Module):
    method setup (line 496) | def setup(self):
    method __call__ (line 502) | def __call__(
  class FlaxRoFormerPredictionHeadTransform (line 527) | class FlaxRoFormerPredictionHeadTransform(nn.Module):
    method setup (line 531) | def setup(self):
    method __call__ (line 536) | def __call__(self, hidden_states):
  class FlaxRoFormerLMPredictionHead (line 543) | class FlaxRoFormerLMPredictionHead(nn.Module):
    method setup (line 548) | def setup(self):
    method __call__ (line 553) | def __call__(self, hidden_states, shared_embedding=None):
  class FlaxRoFormerOnlyMLMHead (line 567) | class FlaxRoFormerOnlyMLMHead(nn.Module):
    method setup (line 571) | def setup(self):
    method __call__ (line 574) | def __call__(self, hidden_states, shared_embedding=None):
  class FlaxRoFormerClassificationHead (line 579) | class FlaxRoFormerClassificationHead(nn.Module):
    method setup (line 583) | def setup(self):
    method __call__ (line 597) | def __call__(self, hidden_states, deterministic=True):
  class FlaxRoFormerPreTrainedModel (line 607) | class FlaxRoFormerPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 617) | def __init__(
    method init_weights (line 629) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 654) | def __call__(
  class FlaxRoFormerModule (line 702) | class FlaxRoFormerModule(nn.Module):
    method setup (line 706) | def setup(self):
    method __call__ (line 710) | def __call__(
  class FlaxRoFormerModel (line 747) | class FlaxRoFormerModel(FlaxRoFormerPreTrainedModel):
  class FlaxRoFormerForMaskedLMModule (line 754) | class FlaxRoFormerForMaskedLMModule(nn.Module):
    method setup (line 758) | def setup(self):
    method __call__ (line 762) | def __call__(
  class FlaxRoFormerForMaskedLM (line 805) | class FlaxRoFormerForMaskedLM(FlaxRoFormerPreTrainedModel):
  class FlaxRoFormerForSequenceClassificationModule (line 818) | class FlaxRoFormerForSequenceClassificationModule(nn.Module):
    method setup (line 822) | def setup(self):
    method __call__ (line 826) | def __call__(
  class FlaxRoFormerForSequenceClassification (line 869) | class FlaxRoFormerForSequenceClassification(FlaxRoFormerPreTrainedModel):
  class FlaxRoFormerForMultipleChoiceModule (line 881) | class FlaxRoFormerForMultipleChoiceModule(nn.Module):
    method setup (line 885) | def setup(self):
    method __call__ (line 890) | def __call__(
  class FlaxRoFormerForMultipleChoice (line 944) | class FlaxRoFormerForMultipleChoice(FlaxRoFormerPreTrainedModel):
  class FlaxRoFormerForTokenClassificationModule (line 959) | class FlaxRoFormerForTokenClassificationModule(nn.Module):
    method setup (line 963) | def setup(self):
    method __call__ (line 968) | def __call__(
  class FlaxRoFormerForTokenClassification (line 1012) | class FlaxRoFormerForTokenClassification(FlaxRoFormerPreTrainedModel):
  class FlaxRoFormerForQuestionAnsweringModule (line 1024) | class FlaxRoFormerForQuestionAnsweringModule(nn.Module):
    method setup (line 1028) | def setup(self):
    method __call__ (line 1032) | def __call__(
  class FlaxRoFormerForQuestionAnswering (line 1080) | class FlaxRoFormerForQuestionAnswering(FlaxRoFormerPreTrainedModel):

FILE: transformers/models/roformer/modeling_roformer.py
  class RoFormerSinusoidalPositionalEmbedding (line 67) | class RoFormerSinusoidalPositionalEmbedding(nn.Embedding):
    method __init__ (line 70) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method _init_weight (line 75) | def _init_weight(out: nn.Parameter) -> nn.Parameter:
    method forward (line 92) | def forward(self, input_ids_shape: torch.Size, past_key_values_length:...
  function load_tf_weights_in_roformer (line 101) | def load_tf_weights_in_roformer(model, config, tf_checkpoint_path):
  class RoFormerEmbeddings (line 174) | class RoFormerEmbeddings(nn.Module):
    method __init__ (line 177) | def __init__(self, config):
    method forward (line 187) | def forward(self, input_ids=None, token_type_ids=None, inputs_embeds=N...
  class RoFormerSelfAttention (line 208) | class RoFormerSelfAttention(nn.Module):
    method __init__ (line 209) | def __init__(self, config):
    method transpose_for_scores (line 230) | def transpose_for_scores(self, x):
    method forward (line 235) | def forward(
    method apply_rotary_position_embeddings (line 319) | def apply_rotary_position_embeddings(sinusoidal_pos, query_layer, key_...
  class RoFormerSelfOutput (line 347) | class RoFormerSelfOutput(nn.Module):
    method __init__ (line 348) | def __init__(self, config):
    method forward (line 354) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RoFormerAttention (line 361) | class RoFormerAttention(nn.Module):
    method __init__ (line 362) | def __init__(self, config):
    method prune_heads (line 369) | def prune_heads(self, heads):
    method forward (line 388) | def forward(
  class RoFormerIntermediate (line 415) | class RoFormerIntermediate(nn.Module):
    method __init__ (line 416) | def __init__(self, config):
    method forward (line 424) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class RoFormerOutput (line 431) | class RoFormerOutput(nn.Module):
    method __init__ (line 432) | def __init__(self, config):
    method forward (line 438) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class RoFormerLayer (line 445) | class RoFormerLayer(nn.Module):
    method __init__ (line 446) | def __init__(self, config):
    method forward (line 460) | def forward(
    method feed_forward_chunk (line 528) | def feed_forward_chunk(self, attention_output):
  class RoFormerEncoder (line 534) | class RoFormerEncoder(nn.Module):
    method __init__ (line 535) | def __init__(self, config):
    method forward (line 544) | def forward(
  class RoFormerPredictionHeadTransform (line 641) | class RoFormerPredictionHeadTransform(nn.Module):
    method __init__ (line 642) | def __init__(self, config):
    method forward (line 651) | def forward(self, hidden_states):
  class RoFormerLMPredictionHead (line 658) | class RoFormerLMPredictionHead(nn.Module):
    method __init__ (line 659) | def __init__(self, config):
    method forward (line 672) | def forward(self, hidden_states):
  class RoFormerOnlyMLMHead (line 679) | class RoFormerOnlyMLMHead(nn.Module):
    method __init__ (line 680) | def __init__(self, config):
    method forward (line 684) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class RoFormerPreTrainedModel (line 689) | class RoFormerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 705) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 723) | def _set_gradient_checkpointing(self, module, value=False):
  class RoFormerModel (line 788) | class RoFormerModel(RoFormerPreTrainedModel):
    method __init__ (line 801) | def __init__(self, config):
    method get_input_embeddings (line 814) | def get_input_embeddings(self):
    method set_input_embeddings (line 817) | def set_input_embeddings(self, value):
    method _prune_heads (line 820) | def _prune_heads(self, heads_to_prune):
    method forward (line 834) | def forward(
  class RoFormerForMaskedLM (line 954) | class RoFormerForMaskedLM(RoFormerPreTrainedModel):
    method __init__ (line 957) | def __init__(self, config):
    method get_output_embeddings (line 972) | def get_output_embeddings(self):
    method set_output_embeddings (line 975) | def set_output_embeddings(self, new_embeddings):
    method forward (line 984) | def forward(
    method prepare_inputs_for_generation (line 1038) | def prepare_inputs_for_generation(self, input_ids, attention_mask=None...
  class RoFormerForCausalLM (line 1056) | class RoFormerForCausalLM(RoFormerPreTrainedModel):
    method __init__ (line 1059) | def __init__(self, config):
    method get_output_embeddings (line 1071) | def get_output_embeddings(self):
    method set_output_embeddings (line 1074) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1079) | def forward(
    method prepare_inputs_for_generation (line 1178) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1191) | def _reorder_cache(self, past_key_values, beam_idx):
  class RoFormerClassificationHead (line 1200) | class RoFormerClassificationHead(nn.Module):
    method __init__ (line 1203) | def __init__(self, config):
    method forward (line 1211) | def forward(self, features, **kwargs):
  class RoFormerForSequenceClassification (line 1228) | class RoFormerForSequenceClassification(RoFormerPreTrainedModel):
    method __init__ (line 1229) | def __init__(self, config):
    method forward (line 1244) | def forward(
  class RoFormerForMultipleChoice (line 1319) | class RoFormerForMultipleChoice(RoFormerPreTrainedModel):
    method __init__ (line 1320) | def __init__(self, config):
    method forward (line 1338) | def forward(
  class RoFormerForTokenClassification (line 1410) | class RoFormerForTokenClassification(RoFormerPreTrainedModel):
    method __init__ (line 1411) | def __init__(self, config):
    method forward (line 1428) | def forward(
  class RoFormerForQuestionAnswering (line 1486) | class RoFormerForQuestionAnswering(RoFormerPreTrainedModel):
    method __init__ (line 1487) | def __init__(self, config):
    method forward (line 1505) | def forward(

FILE: transformers/models/roformer/modeling_tf_roformer.py
  class TFRoFormerSinusoidalPositionalEmbedding (line 77) | class TFRoFormerSinusoidalPositionalEmbedding(tf.keras.layers.Layer):
    method __init__ (line 80) | def __init__(self, num_positions: int, embedding_dim: int, **kwargs):
    method build (line 89) | def build(self, input_shape: tf.TensorShape):
    method _init_weight (line 108) | def _init_weight(n_pos: int, dim: int):
    method call (line 125) | def call(self, input_shape: tf.TensorShape, past_key_values_length: in...
  class TFRoFormerEmbeddings (line 133) | class TFRoFormerEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 136) | def __init__(self, config: RoFormerConfig, **kwargs):
    method build (line 145) | def build(self, input_shape: tf.TensorShape):
    method call (line 162) | def call(
  class TFRoFormerSelfAttention (line 195) | class TFRoFormerSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 196) | def __init__(self, config: RoFormerConfig, **kwargs):
    method transpose_for_scores (line 222) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 229) | def call(
    method apply_rotary_position_embeddings (line 285) | def apply_rotary_position_embeddings(sinusoidal_pos, query_layer, key_...
  class TFRoFormerSelfOutput (line 312) | class TFRoFormerSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 313) | def __init__(self, config: RoFormerConfig, **kwargs):
    method call (line 322) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFRoFormerAttention (line 330) | class TFRoFormerAttention(tf.keras.layers.Layer):
    method __init__ (line 331) | def __init__(self, config: RoFormerConfig, **kwargs):
    method prune_heads (line 337) | def prune_heads(self, heads):
    method call (line 340) | def call(
  class TFRoFormerIntermediate (line 366) | class TFRoFormerIntermediate(tf.keras.layers.Layer):
    method __init__ (line 367) | def __init__(self, config: RoFormerConfig, **kwargs):
    method call (line 379) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFRoFormerOutput (line 387) | class TFRoFormerOutput(tf.keras.layers.Layer):
    method __init__ (line 388) | def __init__(self, config: RoFormerConfig, **kwargs):
    method call (line 397) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFRoFormerLayer (line 405) | class TFRoFormerLayer(tf.keras.layers.Layer):
    method __init__ (line 406) | def __init__(self, config: RoFormerConfig, **kwargs):
    method call (line 413) | def call(
  class TFRoFormerEncoder (line 440) | class TFRoFormerEncoder(tf.keras.layers.Layer):
    method __init__ (line 441) | def __init__(self, config: RoFormerConfig, **kwargs):
    method call (line 450) | def call(
  class TFRoFormerPredictionHeadTransform (line 495) | class TFRoFormerPredictionHeadTransform(tf.keras.layers.Layer):
    method __init__ (line 496) | def __init__(self, config: RoFormerConfig, **kwargs):
    method call (line 512) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFRoFormerLMPredictionHead (line 520) | class TFRoFormerLMPredictionHead(tf.keras.layers.Layer):
    method __init__ (line 521) | def __init__(self, config: RoFormerConfig, input_embeddings: tf.keras....
    method build (line 533) | def build(self, input_shape: tf.TensorShape):
    method get_output_embeddings (line 538) | def get_output_embeddings(self) -> tf.keras.layers.Layer:
    method set_output_embeddings (line 541) | def set_output_embeddings(self, value: tf.Variable):
    method get_bias (line 545) | def get_bias(self) -> Dict[str, tf.Variable]:
    method set_bias (line 548) | def set_bias(self, value: tf.Variable):
    method call (line 552) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFRoFormerMLMHead (line 564) | class TFRoFormerMLMHead(tf.keras.layers.Layer):
    method __init__ (line 565) | def __init__(self, config: RoFormerConfig, input_embeddings: tf.keras....
    method call (line 570) | def call(self, sequence_output: tf.Tensor) -> tf.Tensor:
  class TFRoFormerMainLayer (line 577) | class TFRoFormerMainLayer(tf.keras.layers.Layer):
    method __init__ (line 580) | def __init__(self, config: RoFormerConfig, add_pooling_layer: bool = T...
    method get_input_embeddings (line 591) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 594) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 598) | def _prune_heads(self, heads_to_prune):
    method call (line 606) | def call(
  class TFRoFormerPreTrainedModel (line 691) | class TFRoFormerPreTrainedModel(TFPreTrainedModel):
  class TFRoFormerModel (line 798) | class TFRoFormerModel(TFRoFormerPreTrainedModel):
    method __init__ (line 799) | def __init__(self, config: RoFormerConfig, *inputs, **kwargs):
    method call (line 811) | def call(
  class TFRoFormerForMaskedLM (line 839) | class TFRoFormerForMaskedLM(TFRoFormerPreTrainedModel, TFMaskedLanguageM...
    method __init__ (line 840) | def __init__(self, config: RoFormerConfig, *inputs, **kwargs):
    method get_lm_head (line 852) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method call (line 862) | def call(
  class TFRoFormerForCausalLM (line 911) | class TFRoFormerForCausalLM(TFRoFormerPreTrainedModel, TFCausalLanguageM...
    method __init__ (line 912) | def __init__(self, config: RoFormerConfig, *inputs, **kwargs):
    method get_lm_head (line 921) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method call (line 930) | def call(
  class TFRoFormerClassificationHead (line 981) | class TFRoFormerClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 984) | def __init__(self, config: RoFormerConfig, *inputs, **kwargs):
    method call (line 1000) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFRoFormerForSequenceClassification (line 1017) | class TFRoFormerForSequenceClassification(TFRoFormerPreTrainedModel, TFS...
    method __init__ (line 1018) | def __init__(self, config: RoFormerConfig, *inputs, **kwargs):
    method call (line 1033) | def call(
  class TFRoFormerForMultipleChoice (line 1086) | class TFRoFormerForMultipleChoice(TFRoFormerPreTrainedModel, TFMultipleC...
    method __init__ (line 1087) | def __init__(self, config: RoFormerConfig, *inputs, **kwargs):
    method call (line 1105) | def call(
  class TFRoFormerForTokenClassification (line 1178) | class TFRoFormerForTokenClassification(TFRoFormerPreTrainedModel, TFToke...
    method __init__ (line 1179) | def __init__(self, config: RoFormerConfig, *inputs, **kwargs):
    method call (line 1197) | def call(
  class TFRoFormerForQuestionAnswering (line 1249) | class TFRoFormerForQuestionAnswering(TFRoFormerPreTrainedModel, TFQuesti...
    method __init__ (line 1250) | def __init__(self, config: RoFormerConfig, *inputs, **kwargs):
    method call (line 1267) | def call(

FILE: transformers/models/roformer/tokenization_roformer.py
  function load_vocab (line 70) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 82) | def whitespace_tokenize(text):
  class BasicTokenizer (line 92) | class BasicTokenizer(object):
    method __init__ (line 112) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 120) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 157) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 168) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 190) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 203) | def _is_chinese_char(self, cp):
    method _clean_text (line 227) | def _clean_text(self, text):
  class WordpieceTokenizer (line 242) | class WordpieceTokenizer(object):
    method __init__ (line 245) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 250) | def tokenize(self, text):
  class RoFormerTokenizer (line 299) | class RoFormerTokenizer(PreTrainedTokenizer):
    method __init__ (line 354) | def __init__(
    method do_lower_case (line 409) | def do_lower_case(self):
    method vocab_size (line 413) | def vocab_size(self):
    method __getstate__ (line 416) | def __getstate__(self):
    method __setstate__ (line 421) | def __setstate__(self, d):
    method get_vocab (line 427) | def get_vocab(self):
    method _tokenize (line 430) | def _tokenize(self, text, use_jieba=True):
    method _convert_token_to_id (line 452) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 456) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 460) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 465) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 490) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 518) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 547) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/roformer/tokenization_roformer_fast.py
  class RoFormerTokenizerFast (line 71) | class RoFormerTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 97) | def __init__(
    method __getstate__ (line 137) | def __getstate__(self):
    method __setstate__ (line 142) | def __setstate__(self, d):
    method build_inputs_with_special_tokens (line 147) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 171) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 200) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method save_pretrained (line 204) | def save_pretrained(

FILE: transformers/models/roformer/tokenization_utils.py
  class JiebaPreTokenizer (line 22) | class JiebaPreTokenizer:
    method __init__ (line 23) | def __init__(self, vocab) -> None:
    method jieba_split (line 40) | def jieba_split(self, i: int, normalized_string: NormalizedString) -> ...
    method pre_tokenize (line 67) | def pre_tokenize(self, pretok: PreTokenizedString):

FILE: transformers/models/rwkv/configuration_rwkv.py
  class RwkvConfig (line 38) | class RwkvConfig(PretrainedConfig):
    method __init__ (line 99) | def __init__(

FILE: transformers/models/rwkv/convert_rwkv_checkpoint_to_hf.py
  function convert_state_dict (line 50) | def convert_state_dict(state_dict):
  function convert_rmkv_checkpoint_to_hf_format (line 81) | def convert_rmkv_checkpoint_to_hf_format(

FILE: transformers/models/rwkv/modeling_rwkv.py
  function load_wkv_cuda_kernel (line 64) | def load_wkv_cuda_kernel(context_length):
  class RwkvLinearAttention (line 96) | class RwkvLinearAttention(torch.autograd.Function):
    method forward (line 98) | def forward(ctx, time_decay, time_first, key, value, state=None, retur...
    method backward (line 162) | def backward(ctx, g_output, g_state=None):
  function rwkv_linear_attention_cpu (line 202) | def rwkv_linear_attention_cpu(time_decay, time_first, key, value, state=...
  function rwkv_linear_attention (line 246) | def rwkv_linear_attention(time_decay, time_first, key, value, state=None...
  class RwkvSelfAttention (line 257) | class RwkvSelfAttention(nn.Module):
    method __init__ (line 258) | def __init__(self, config, layer_id=0):
    method extract_key_value (line 288) | def extract_key_value(self, hidden, state=None):
    method forward (line 307) | def forward(self, hidden, state=None, use_cache=False):
  class RwkvFeedForward (line 327) | class RwkvFeedForward(nn.Module):
    method __init__ (line 328) | def __init__(self, config, layer_id=0):
    method forward (line 345) | def forward(self, hidden, state=None):
  class RwkvBlock (line 365) | class RwkvBlock(nn.Module):
    method __init__ (line 366) | def __init__(self, config, layer_id):
    method forward (line 380) | def forward(self, hidden, state=None, use_cache=False, output_attentio...
  class RwkvPreTrainedModel (line 399) | class RwkvPreTrainedModel(PreTrainedModel):
    method _init_weights (line 410) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 467) | def _set_gradient_checkpointing(self, module, value=False):
  class RwkvOutput (line 473) | class RwkvOutput(ModelOutput):
  class RwkvCausalLMOutput (line 503) | class RwkvCausalLMOutput(ModelOutput):
  class RwkvModel (line 598) | class RwkvModel(RwkvPreTrainedModel):
    method __init__ (line 599) | def __init__(self, config):
    method get_input_embeddings (line 611) | def get_input_embeddings(self):
    method set_input_embeddings (line 614) | def set_input_embeddings(self, new_embeddings):
    method forward (line 623) | def forward(
    method _rescale_layers (line 698) | def _rescale_layers(self):
  class RwkvForCausalLM (line 734) | class RwkvForCausalLM(RwkvPreTrainedModel):
    method __init__ (line 735) | def __init__(self, config):
    method get_output_embeddings (line 743) | def get_output_embeddings(self):
    method set_output_embeddings (line 746) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 749) | def prepare_inputs_for_generation(self, input_ids, state=None, inputs_...
    method forward (line 769) | def forward(

FILE: transformers/models/sam/configuration_sam.py
  class SamPromptEncoderConfig (line 32) | class SamPromptEncoderConfig(PretrainedConfig):
    method __init__ (line 57) | def __init__(
  class SamMaskDecoderConfig (line 79) | class SamMaskDecoderConfig(PretrainedConfig):
    method __init__ (line 113) | def __init__(
  class SamVisionConfig (line 140) | class SamVisionConfig(PretrainedConfig):
    method __init__ (line 200) | def __init__(
  class SamConfig (line 254) | class SamConfig(PretrainedConfig):
    method __init__ (line 307) | def __init__(
    method to_dict (line 332) | def to_dict(self):

FILE: transformers/models/sam/convert_sam_original_to_hf_format.py
  function replace_keys (line 61) | def replace_keys(state_dict):
  function convert_sam_checkpoint (line 91) | def convert_sam_checkpoint(model_name, pytorch_dump_folder, push_to_hub,...

FILE: transformers/models/sam/image_processing_sam.py
  class SamImageProcessor (line 63) | class SamImageProcessor(BaseImageProcessor):
    method __init__ (line 107) | def __init__(
    method pad_image (line 141) | def pad_image(
    method _get_preprocess_shape (line 169) | def _get_preprocess_shape(self, old_shape: Tuple[int, int], longest_ed...
    method resize (line 180) | def resize(
    method rescale (line 215) | def rescale(
    method normalize (line 235) | def normalize(
    method preprocess (line 258) | def preprocess(
    method post_process_masks (line 387) | def post_process_masks(
    method _post_process_masks_pt (line 442) | def _post_process_masks_pt(
    method _post_process_masks_tf (line 489) | def _post_process_masks_tf(
    method post_process_for_mask_generation (line 531) | def post_process_for_mask_generation(
    method generate_crop_boxes (line 554) | def generate_crop_boxes(
    method filter_masks (line 615) | def filter_masks(
    method _filter_masks_pt (line 676) | def _filter_masks_pt(
    method _filter_masks_tf (line 756) | def _filter_masks_tf(
  function _compute_stability_score_pt (line 834) | def _compute_stability_score_pt(masks: "torch.Tensor", mask_threshold: f...
  function _compute_stability_score_tf (line 845) | def _compute_stability_score_tf(masks: "tf.Tensor", mask_threshold: floa...
  function _build_point_grid (line 856) | def _build_point_grid(n_per_side: int) -> np.ndarray:
  function _normalize_coordinates (line 866) | def _normalize_coordinates(
  function _generate_crop_boxes (line 894) | def _generate_crop_boxes(
  function _generate_per_layer_crops (line 947) | def _generate_per_layer_crops(crop_n_layers, overlap_ratio, original_size):
  function _generate_crop_images (line 981) | def _generate_crop_images(crop_boxes, image, points_grid, layer_idxs, ta...
  function _pad_masks (line 1009) | def _pad_masks(masks, crop_box: List[int], orig_height: int, orig_width:...
  function _pad_masks_tf (line 1019) | def _pad_masks_tf(masks, crop_box: List[int], orig_height: int, orig_wid...
  function _is_box_near_crop_edge (line 1029) | def _is_box_near_crop_edge(boxes, crop_box, orig_box, atol=20.0):
  function _is_box_near_crop_edge_tf (line 1047) | def _is_box_near_crop_edge_tf(boxes, crop_box, orig_box, atol=20.0):
  function _batched_mask_to_box (line 1065) | def _batched_mask_to_box(masks: "torch.Tensor"):
  function _batched_mask_to_box_tf (line 1114) | def _batched_mask_to_box_tf(masks: "tf.Tensor"):
  function _mask_to_rle_pytorch (line 1162) | def _mask_to_rle_pytorch(input_mask: "torch.Tensor"):
  function _mask_to_rle_tf (line 1185) | def _mask_to_rle_tf(input_mask: "tf.Tensor"):
  function _rle_to_mask (line 1208) | def _rle_to_mask(rle: Dict[str, Any]) -> np.ndarray:
  function _postprocess_for_mg (line 1222) | def _postprocess_for_mg(rle_masks, iou_scores, mask_boxes, amg_crops_nms...
  function _postprocess_for_mg_tf (line 1251) | def _postprocess_for_mg_tf(rle_masks, iou_scores, mask_boxes, amg_crops_...

FILE: transformers/models/sam/modeling_sam.py
  class SamVisionEncoderOutput (line 49) | class SamVisionEncoderOutput(ModelOutput):
  class SamImageSegmentationOutput (line 79) | class SamImageSegmentationOutput(ModelOutput):
  class SamPatchEmbeddings (line 114) | class SamPatchEmbeddings(nn.Module):
    method __init__ (line 121) | def __init__(self, config):
    method forward (line 135) | def forward(self, pixel_values):
  class SamMLPBlock (line 149) | class SamMLPBlock(nn.Module):
    method __init__ (line 150) | def __init__(self, config):
    method forward (line 156) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class SamLayerNorm (line 164) | class SamLayerNorm(nn.Module):
    method __init__ (line 170) | def __init__(self, normalized_shape, eps=1e-6, data_format="channels_l...
    method forward (line 180) | def forward(self, x: torch.Tensor) -> torch.Tensor:
  class SamAttention (line 194) | class SamAttention(nn.Module):
    method __init__ (line 200) | def __init__(self, config, downsample_rate=None):
    method _separate_heads (line 216) | def _separate_heads(self, hidden_states: Tensor, num_attention_heads: ...
    method _recombine_heads (line 222) | def _recombine_heads(self, hidden_states: Tensor, point_batch_size: in...
    method forward (line 227) | def forward(self, query: Tensor, key: Tensor, value: Tensor, attention...
  class SamTwoWayAttentionBlock (line 257) | class SamTwoWayAttentionBlock(nn.Module):
    method __init__ (line 258) | def __init__(self, config, attention_downsample_rate: int = 2, skip_fi...
    method forward (line 291) | def forward(
  class SamTwoWayTransformer (line 344) | class SamTwoWayTransformer(nn.Module):
    method __init__ (line 345) | def __init__(self, config: SamMaskDecoderConfig):
    method forward (line 358) | def forward(
  class SamFeedForward (line 415) | class SamFeedForward(nn.Module):
    method __init__ (line 416) | def __init__(
    method forward (line 427) | def forward(self, hidden_states):
  class SamMaskDecoder (line 439) | class SamMaskDecoder(nn.Module):
    method __init__ (line 440) | def __init__(self, config: SamMaskDecoderConfig):
    method forward (line 468) | def forward(
  class SamPositionalEmbedding (line 565) | class SamPositionalEmbedding(nn.Module):
    method __init__ (line 566) | def __init__(self, config):
    method forward (line 571) | def forward(self, input_coords, input_shape=None):
  class SamMaskEmbedding (line 588) | class SamMaskEmbedding(nn.Module):
    method __init__ (line 589) | def __init__(self, config: SamPromptEncoderConfig):
    method forward (line 603) | def forward(self, masks):
  class SamPromptEncoder (line 615) | class SamPromptEncoder(nn.Module):
    method __init__ (line 616) | def __init__(self, config: SamPromptEncoderConfig, shared_patch_embedd...
    method _embed_points (line 631) | def _embed_points(self, points: torch.Tensor, labels: torch.Tensor, pa...
    method _embed_boxes (line 669) | def _embed_boxes(self, boxes: torch.Tensor) -> torch.Tensor:
    method forward (line 680) | def forward(
  class SamVisionAttention (line 727) | class SamVisionAttention(nn.Module):
    method __init__ (line 730) | def __init__(self, config, window_size):
    method get_rel_pos (line 755) | def get_rel_pos(self, q_size: int, k_size: int, rel_pos: torch.Tensor)...
    method add_decomposed_rel_pos (line 787) | def add_decomposed_rel_pos(
    method forward (line 832) | def forward(self, hidden_states: torch.Tensor, output_attentions=False...
  class SamVisionLayer (line 867) | class SamVisionLayer(nn.Module):
    method __init__ (line 868) | def __init__(self, config, window_size):
    method window_partition (line 876) | def window_partition(self, hidden_states: torch.Tensor, window_size: i...
    method window_unpartition (line 900) | def window_unpartition(
    method forward (line 930) | def forward(
  class SamVisionNeck (line 962) | class SamVisionNeck(nn.Module):
    method __init__ (line 963) | def __init__(self, config: SamVisionConfig):
    method forward (line 972) | def forward(self, hidden_states):
  class SamVisionEncoder (line 982) | class SamVisionEncoder(nn.Module):
    method __init__ (line 983) | def __init__(self, config: SamVisionConfig):
    method get_input_embeddings (line 1014) | def get_input_embeddings(self):
    method forward (line 1017) | def forward(
  class SamPreTrainedModel (line 1084) | class SamPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1089) | def _init_weights(self, module):
  class SamModel (line 1192) | class SamModel(SamPreTrainedModel):
    method __init__ (line 1195) | def __init__(self, config):
    method get_input_embeddings (line 1205) | def get_input_embeddings(self):
    method get_image_wide_positional_embeddings (line 1208) | def get_image_wide_positional_embeddings(self):
    method get_image_embeddings (line 1222) | def get_image_embeddings(
    method get_prompt_embeddings (line 1253) | def get_prompt_embeddings(
    method forward (line 1286) | def forward(

FILE: transformers/models/sam/modeling_tf_sam.py
  class TFSamVisionEncoderOutput (line 52) | class TFSamVisionEncoderOutput(ModelOutput):
  class TFSamImageSegmentationOutput (line 82) | class TFSamImageSegmentationOutput(ModelOutput):
  class TFSamPatchEmbeddings (line 117) | class TFSamPatchEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 124) | def __init__(self, config, **kwargs):
    method call (line 140) | def call(self, pixel_values):
  class TFSamMLPBlock (line 154) | class TFSamMLPBlock(tf.keras.layers.Layer):
    method __init__ (line 155) | def __init__(self, config, **kwargs):
    method call (line 161) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFSamLayerNorm (line 168) | class TFSamLayerNorm(tf.keras.layers.Layer):
    method __init__ (line 174) | def __init__(self, normalized_shape, eps=1e-6, data_format="channels_l...
    method build (line 182) | def build(self, input_shape):
    method call (line 187) | def call(self, x: tf.Tensor) -> tf.Tensor:
  class TFSamAttention (line 195) | class TFSamAttention(tf.keras.layers.Layer):
    method __init__ (line 201) | def __init__(self, config, downsample_rate=None, **kwargs):
    method _separate_heads (line 217) | def _separate_heads(self, hidden_states: tf.Tensor, num_attention_head...
    method _recombine_heads (line 225) | def _recombine_heads(self, hidden_states: tf.Tensor, point_batch_size:...
    method call (line 233) | def call(self, query: tf.Tensor, key: tf.Tensor, value: tf.Tensor) -> ...
  class TFSamTwoWayAttentionBlock (line 261) | class TFSamTwoWayAttentionBlock(tf.keras.layers.Layer):
    method __init__ (line 262) | def __init__(self, config, attention_downsample_rate: int = 2, skip_fi...
    method call (line 299) | def call(
  class TFSamTwoWayTransformer (line 349) | class TFSamTwoWayTransformer(tf.keras.layers.Layer):
    method __init__ (line 350) | def __init__(self, config: SamMaskDecoderConfig, **kwargs):
    method call (line 365) | def call(
  class TFSamFeedForward (line 416) | class TFSamFeedForward(tf.keras.layers.Layer):
    method __init__ (line 417) | def __init__(
    method call (line 431) | def call(self, hidden_states):
  class TFSamMaskDecoder (line 443) | class TFSamMaskDecoder(tf.keras.layers.Layer):
    method __init__ (line 444) | def __init__(self, config: SamMaskDecoderConfig, **kwargs):
    method build (line 486) | def build(self, input_shape):
    method call (line 493) | def call(
  class TFSamPositionalEmbedding (line 570) | class TFSamPositionalEmbedding(tf.keras.layers.Layer):
    method __init__ (line 571) | def __init__(self, config, **kwargs):
    method build (line 576) | def build(self, input_shape):
    method call (line 586) | def call(self, input_coords, input_shape=None):
  class TFSamMaskEmbedding (line 608) | class TFSamMaskEmbedding(tf.keras.layers.Layer):
    method __init__ (line 609) | def __init__(self, config: SamPromptEncoderConfig, **kwargs):
    method call (line 619) | def call(self, masks):
    method build (line 632) | def build(self, input_shape):
  class TFSamPromptEncoder (line 652) | class TFSamPromptEncoder(tf.keras.layers.Layer):
    method __init__ (line 653) | def __init__(self, config: SamPromptEncoderConfig, shared_patch_embedd...
    method build (line 667) | def build(self, input_shape):
    method _embed_points (line 696) | def _embed_points(self, points: tf.Tensor, labels: tf.Tensor, pad: boo...
    method _embed_boxes (line 724) | def _embed_boxes(self, boxes: tf.Tensor) -> tf.Tensor:
    method call (line 738) | def call(
  class TFSamVisionAttention (line 788) | class TFSamVisionAttention(tf.keras.layers.Layer):
    method __init__ (line 791) | def __init__(self, config, window_size, **kwargs):
    method build (line 814) | def build(self, input_shape):
    method get_rel_pos (line 825) | def get_rel_pos(self, q_size: int, k_size: int, rel_pos: tf.Tensor) ->...
    method add_decomposed_rel_pos (line 861) | def add_decomposed_rel_pos(
    method call (line 906) | def call(self, hidden_states: tf.Tensor, output_attentions=False, trai...
  class TFSamVisionLayer (line 943) | class TFSamVisionLayer(tf.keras.layers.Layer):
    method __init__ (line 944) | def __init__(self, config, window_size, **kwargs):
    method window_partition (line 952) | def window_partition(self, hidden_states: tf.Tensor, window_size: int)...
    method window_unpartition (line 970) | def window_unpartition(
    method call (line 987) | def call(
  class TFSamVisionNeck (line 1019) | class TFSamVisionNeck(tf.keras.layers.Layer):
    method __init__ (line 1020) | def __init__(self, config: SamVisionConfig, **kwargs):
    method call (line 1040) | def call(self, hidden_states):
  class TFSamVisionEncoder (line 1050) | class TFSamVisionEncoder(tf.keras.layers.Layer):
    method __init__ (line 1051) | def __init__(self, config: SamVisionConfig, **kwargs):
    method build (line 1071) | def build(self, input_shape):
    method get_input_embeddings (line 1087) | def get_input_embeddings(self):
    method call (line 1090) | def call(
  class TFSamPreTrainedModel (line 1145) | class TFSamPreTrainedModel(TFPreTrainedModel):
  class TFSamModel (line 1236) | class TFSamModel(TFSamPreTrainedModel):
    method __init__ (line 1239) | def __init__(self, config, **kwargs):
    method get_input_embeddings (line 1250) | def get_input_embeddings(self):
    method get_image_wide_positional_embeddings (line 1253) | def get_image_wide_positional_embeddings(self):
    method get_image_embeddings (line 1264) | def get_image_embeddings(
    method get_prompt_embeddings (line 1294) | def get_prompt_embeddings(
    method call (line 1328) | def call(
    method serving_output (line 1454) | def serving_output(self, output: TFSamImageSegmentationOutput) -> TFSa...

FILE: transformers/models/sam/processing_sam.py
  class SamProcessor (line 35) | class SamProcessor(ProcessorMixin):
    method __init__ (line 50) | def __init__(self, image_processor):
    method __call__ (line 56) | def __call__(
    method _normalize_and_convert (line 98) | def _normalize_and_convert(
    method _pad_points_and_labels (line 173) | def _pad_points_and_labels(self, input_points, input_labels):
    method _normalize_coordinates (line 189) | def _normalize_coordinates(
    method _check_and_preprocess_points (line 210) | def _check_and_preprocess_points(
    method model_input_names (line 258) | def model_input_names(self):
    method post_process_masks (line 262) | def post_process_masks(self, *args, **kwargs):

FILE: transformers/models/segformer/configuration_segformer.py
  class SegformerConfig (line 38) | class SegformerConfig(PretrainedConfig):
    method __init__ (line 105) | def __init__(
  class SegformerOnnxConfig (line 157) | class SegformerOnnxConfig(OnnxConfig):
    method inputs (line 161) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 169) | def atol_for_validation(self) -> float:
    method default_onnx_opset (line 173) | def default_onnx_opset(self) -> int:

FILE: transformers/models/segformer/convert_segformer_original_to_pytorch.py
  function rename_keys (line 41) | def rename_keys(state_dict, encoder_only=False):
  function read_in_k_v (line 92) | def read_in_k_v(state_dict, config):
  function prepare_img (line 113) | def prepare_img():
  function convert_segformer_checkpoint (line 121) | def convert_segformer_checkpoint(model_name, checkpoint_path, pytorch_du...

FILE: transformers/models/segformer/feature_extraction_segformer.py
  class SegformerFeatureExtractor (line 26) | class SegformerFeatureExtractor(SegformerImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/segformer/image_processing_segformer.py
  class SegformerImageProcessor (line 47) | class SegformerImageProcessor(BaseImageProcessor):
    method __init__ (line 85) | def __init__(
    method reduce_labels (line 120) | def reduce_labels(self):
    method from_dict (line 129) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method resize (line 140) | def resize(
    method rescale (line 168) | def rescale(
    method normalize (line 188) | def normalize(
    method reduce_label (line 211) | def reduce_label(self, label: ImageInput) -> np.ndarray:
    method _preprocess (line 219) | def _preprocess(
    method _preprocess_image (line 246) | def _preprocess_image(
    method _preprocess_mask (line 278) | def _preprocess_mask(
    method __call__ (line 308) | def __call__(self, images, segmentation_maps=None, **kwargs):
    method preprocess (line 317) | def preprocess(
    method post_process_semantic_segmentation (line 442) | def post_process_semantic_segmentation(self, outputs, target_sizes: Li...

FILE: transformers/models/segformer/modeling_segformer.py
  class SegFormerImageClassifierOutput (line 60) | class SegFormerImageClassifierOutput(ImageClassifierOutput):
  function drop_path (line 88) | def drop_path(input, drop_prob: float = 0.0, training: bool = False, sca...
  class SegformerDropPath (line 109) | class SegformerDropPath(nn.Module):
    method __init__ (line 112) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 116) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 119) | def extra_repr(self) -> str:
  class SegformerOverlapPatchEmbeddings (line 123) | class SegformerOverlapPatchEmbeddings(nn.Module):
    method __init__ (line 126) | def __init__(self, patch_size, stride, num_channels, hidden_size):
    method forward (line 138) | def forward(self, pixel_values):
  class SegformerEfficientSelfAttention (line 148) | class SegformerEfficientSelfAttention(nn.Module):
    method __init__ (line 152) | def __init__(self, config, hidden_size, num_attention_heads, sequence_...
    method transpose_for_scores (line 179) | def transpose_for_scores(self, hidden_states):
    method forward (line 184) | def forward(
  class SegformerSelfOutput (line 229) | class SegformerSelfOutput(nn.Module):
    method __init__ (line 230) | def __init__(self, config, hidden_size):
    method forward (line 235) | def forward(self, hidden_states, input_tensor):
  class SegformerAttention (line 241) | class SegformerAttention(nn.Module):
    method __init__ (line 242) | def __init__(self, config, hidden_size, num_attention_heads, sequence_...
    method prune_heads (line 253) | def prune_heads(self, heads):
    method forward (line 271) | def forward(self, hidden_states, height, width, output_attentions=False):
  class SegformerDWConv (line 279) | class SegformerDWConv(nn.Module):
    method __init__ (line 280) | def __init__(self, dim=768):
    method forward (line 284) | def forward(self, hidden_states, height, width):
  class SegformerMixFFN (line 293) | class SegformerMixFFN(nn.Module):
    method __init__ (line 294) | def __init__(self, config, in_features, hidden_features=None, out_feat...
    method forward (line 306) | def forward(self, hidden_states, height, width):
  class SegformerLayer (line 316) | class SegformerLayer(nn.Module):
    method __init__ (line 319) | def __init__(self, config, hidden_size, num_attention_heads, drop_path...
    method forward (line 333) | def forward(self, hidden_states, height, width, output_attentions=False):
  class SegformerEncoder (line 359) | class SegformerEncoder(nn.Module):
    method __init__ (line 360) | def __init__(self, config):
    method forward (line 408) | def forward(
  class SegformerPreTrainedModel (line 450) | class SegformerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 460) | def _init_weights(self, module):
  class SegformerModel (line 510) | class SegformerModel(SegformerPreTrainedModel):
    method __init__ (line 511) | def __init__(self, config):
    method _prune_heads (line 521) | def _prune_heads(self, heads_to_prune):
    method forward (line 537) | def forward(
  class SegformerForImageClassification (line 575) | class SegformerForImageClassification(SegformerPreTrainedModel):
    method __init__ (line 576) | def __init__(self, config):
    method forward (line 595) | def forward(
  class SegformerMLP (line 666) | class SegformerMLP(nn.Module):
    method __init__ (line 671) | def __init__(self, config: SegformerConfig, input_dim):
    method forward (line 675) | def forward(self, hidden_states: torch.Tensor):
  class SegformerDecodeHead (line 681) | class SegformerDecodeHead(SegformerPreTrainedModel):
    method __init__ (line 682) | def __init__(self, config):
    method forward (line 706) | def forward(self, encoder_hidden_states: torch.FloatTensor) -> torch.T...
  class SegformerForSemanticSegmentation (line 743) | class SegformerForSemanticSegmentation(SegformerPreTrainedModel):
    method __init__ (line 744) | def __init__(self, config):
    method forward (line 754) | def forward(

FILE: transformers/models/segformer/modeling_tf_segformer.py
  class TFSegformerDropPath (line 59) | class TFSegformerDropPath(tf.keras.layers.Layer):
    method __init__ (line 65) | def __init__(self, drop_path, **kwargs):
    method call (line 69) | def call(self, x, training=None):
  class TFSegformerOverlapPatchEmbeddings (line 79) | class TFSegformerOverlapPatchEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 82) | def __init__(self, patch_size, stride, hidden_size, **kwargs):
    method call (line 91) | def call(self, pixel_values: tf.Tensor) -> Tuple[tf.Tensor, int, int]:
  class TFSegformerEfficientSelfAttention (line 103) | class TFSegformerEfficientSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 107) | def __init__(
    method transpose_for_scores (line 142) | def transpose_for_scores(self, tensor: tf.Tensor) -> tf.Tensor:
    method call (line 152) | def call(
  class TFSegformerSelfOutput (line 200) | class TFSegformerSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 201) | def __init__(self, config: SegformerConfig, hidden_size: int, **kwargs):
    method call (line 206) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFSegformerAttention (line 212) | class TFSegformerAttention(tf.keras.layers.Layer):
    method __init__ (line 213) | def __init__(
    method call (line 231) | def call(
  class TFSegformerDWConv (line 241) | class TFSegformerDWConv(tf.keras.layers.Layer):
    method __init__ (line 242) | def __init__(self, dim: int = 768, **kwargs):
    method call (line 248) | def call(self, hidden_states: tf.Tensor, height: int, width: int) -> t...
  class TFSegformerMixFFN (line 261) | class TFSegformerMixFFN(tf.keras.layers.Layer):
    method __init__ (line 262) | def __init__(
    method call (line 281) | def call(self, hidden_states: tf.Tensor, height: int, width: int, trai...
  class TFSegformerLayer (line 291) | class TFSegformerLayer(tf.keras.layers.Layer):
    method __init__ (line 294) | def __init__(
    method call (line 318) | def call(
  class TFSegformerEncoder (line 351) | class TFSegformerEncoder(tf.keras.layers.Layer):
    method __init__ (line 352) | def __init__(self, config: SegformerConfig, **kwargs):
    method call (line 402) | def call(
  class TFSegformerMainLayer (line 454) | class TFSegformerMainLayer(tf.keras.layers.Layer):
    method __init__ (line 457) | def __init__(self, config: SegformerConfig, **kwargs):
    method call (line 465) | def call(
  class TFSegformerPreTrainedModel (line 513) | class TFSegformerPreTrainedModel(TFPreTrainedModel):
    method input_signature (line 524) | def input_signature(self):
  class TFSegformerModel (line 574) | class TFSegformerModel(TFSegformerPreTrainedModel):
    method __init__ (line 575) | def __init__(self, config: SegformerConfig, *inputs, **kwargs):
    method call (line 591) | def call(
  class TFSegformerForImageClassification (line 616) | class TFSegformerForImageClassification(TFSegformerPreTrainedModel, TFSe...
    method __init__ (line 617) | def __init__(self, config: SegformerConfig, *inputs, **kwargs):
    method call (line 634) | def call(
  class TFSegformerMLP (line 672) | class TFSegformerMLP(tf.keras.layers.Layer):
    method __init__ (line 677) | def __init__(self, config: SegformerConfig, **kwargs):
    method call (line 681) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFSegformerDecodeHead (line 690) | class TFSegformerDecodeHead(TFSegformerPreTrainedModel):
    method __init__ (line 691) | def __init__(self, config: SegformerConfig, **kwargs):
    method call (line 712) | def call(self, encoder_hidden_states, training: bool = False):
  class TFSegformerForSemanticSegmentation (line 750) | class TFSegformerForSemanticSegmentation(TFSegformerPreTrainedModel):
    method __init__ (line 751) | def __init__(self, config: SegformerConfig, **kwargs):
    method hf_compute_loss (line 756) | def hf_compute_loss(self, logits, labels):
    method call (line 779) | def call(

FILE: transformers/models/sew/configuration_sew.py
  class SEWConfig (line 32) | class SEWConfig(PretrainedConfig):
    method __init__ (line 157) | def __init__(
    method inputs_to_logits_ratio (line 254) | def inputs_to_logits_ratio(self):

FILE: transformers/models/sew/convert_sew_original_pytorch_checkpoint_to_pytorch.py
  function set_recursively (line 62) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function recursively_load_weights (line 90) | def recursively_load_weights(fairseq_model, hf_model, is_finetuned):
  function load_conv_layer (line 134) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function convert_config (line 174) | def convert_config(model, is_finetuned):
  function convert_sew_checkpoint (line 224) | def convert_sew_checkpoint(

FILE: transformers/models/sew/modeling_sew.py
  function _compute_mask_indices (line 67) | def _compute_mask_indices(
  class SEWNoLayerNormConvLayer (line 187) | class SEWNoLayerNormConvLayer(nn.Module):
    method __init__ (line 188) | def __init__(self, config, layer_id=0):
    method forward (line 202) | def forward(self, hidden_states):
  class SEWLayerNormConvLayer (line 209) | class SEWLayerNormConvLayer(nn.Module):
    method __init__ (line 210) | def __init__(self, config, layer_id=0):
    method forward (line 225) | def forward(self, hidden_states):
  class SEWGroupNormConvLayer (line 237) | class SEWGroupNormConvLayer(nn.Module):
    method __init__ (line 238) | def __init__(self, config, layer_id=0):
    method forward (line 254) | def forward(self, hidden_states):
  class SEWPositionalConvEmbedding (line 261) | class SEWPositionalConvEmbedding(nn.Module):
    method __init__ (line 262) | def __init__(self, config):
    method forward (line 286) | def forward(self, hidden_states):
  class SEWSamePadLayer (line 295) | class SEWSamePadLayer(nn.Module):
    method __init__ (line 296) | def __init__(self, num_conv_pos_embeddings):
    method forward (line 300) | def forward(self, hidden_states):
  class SEWUpsampling (line 306) | class SEWUpsampling(nn.Module):
    method __init__ (line 307) | def __init__(self, config):
    method forward (line 313) | def forward(self, hidden_states):
  class SEWFeatureEncoder (line 329) | class SEWFeatureEncoder(nn.Module):
    method __init__ (line 332) | def __init__(self, config):
    method _freeze_parameters (line 349) | def _freeze_parameters(self):
    method forward (line 354) | def forward(self, input_values):
  class SEWFeatureExtractor (line 380) | class SEWFeatureExtractor(SEWFeatureEncoder):
    method __init__ (line 381) | def __init__(self, config):
  class SEWAttention (line 392) | class SEWAttention(nn.Module):
    method __init__ (line 395) | def __init__(
    method _shape (line 422) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 425) | def forward(
  class SEWFeedForward (line 547) | class SEWFeedForward(nn.Module):
    method __init__ (line 548) | def __init__(self, config):
    method forward (line 561) | def forward(self, hidden_states):
  class SEWEncoderLayer (line 572) | class SEWEncoderLayer(nn.Module):
    method __init__ (line 573) | def __init__(self, config):
    method forward (line 586) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class SEWEncoder (line 606) | class SEWEncoder(nn.Module):
    method __init__ (line 607) | def __init__(self, config):
    method forward (line 618) | def forward(
  class SEWPreTrainedModel (line 716) | class SEWPreTrainedModel(PreTrainedModel):
    method _init_weights (line 728) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 760) | def _set_gradient_checkpointing(self, module, value=False):
    method _get_feat_extract_output_lengths (line 764) | def _get_feat_extract_output_lengths(self, input_lengths: Union[torch....
    method _get_feature_vector_attention_mask (line 779) | def _get_feature_vector_attention_mask(self, feature_vector_length: in...
  class SEWModel (line 842) | class SEWModel(SEWPreTrainedModel):
    method __init__ (line 843) | def __init__(self, config: SEWConfig):
    method _mask_hidden_states (line 863) | def _mask_hidden_states(
    method forward (line 917) | def forward(
  class SEWForCTC (line 971) | class SEWForCTC(SEWPreTrainedModel):
    method __init__ (line 972) | def __init__(self, config, target_lang=None):
    method freeze_feature_extractor (line 1000) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1012) | def freeze_feature_encoder(self):
    method forward (line 1027) | def forward(
  class SEWForSequenceClassification (line 1107) | class SEWForSequenceClassification(SEWPreTrainedModel):
    method __init__ (line 1108) | def __init__(self, config):
    method freeze_feature_extractor (line 1125) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1137) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1144) | def freeze_base_model(self):
    method forward (line 1161) | def forward(

FILE: transformers/models/sew_d/configuration_sew_d.py
  class SEWDConfig (line 32) | class SEWDConfig(PretrainedConfig):
    method __init__ (line 172) | def __init__(
    method inputs_to_logits_ratio (line 281) | def inputs_to_logits_ratio(self):

FILE: transformers/models/sew_d/convert_sew_d_original_pytorch_checkpoint_to_pytorch.py
  function set_recursively (line 64) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function recursively_load_weights (line 92) | def recursively_load_weights(fairseq_model, hf_model, is_finetuned):
  function load_conv_layer (line 138) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function convert_config (line 178) | def convert_config(model, is_finetuned):
  function convert_sew_checkpoint (line 236) | def convert_sew_checkpoint(

FILE: transformers/models/sew_d/modeling_sew_d.py
  function _compute_mask_indices (line 73) | def _compute_mask_indices(
  function make_log_bucket_position (line 193) | def make_log_bucket_position(relative_pos, bucket_size, max_position):
  function build_relative_position (line 209) | def build_relative_position(query_size, key_size, bucket_size=-1, max_po...
  function c2p_dynamic_expand (line 241) | def c2p_dynamic_expand(c2p_pos, query_layer, relative_pos):
  function p2c_dynamic_expand (line 247) | def p2c_dynamic_expand(c2p_pos, query_layer, key_layer):
  function pos_dynamic_expand (line 253) | def pos_dynamic_expand(pos_index, p2c_att, key_layer):
  function get_mask (line 258) | def get_mask(input, local_context):
  class SEWDNoLayerNormConvLayer (line 278) | class SEWDNoLayerNormConvLayer(nn.Module):
    method __init__ (line 279) | def __init__(self, config, layer_id=0):
    method forward (line 293) | def forward(self, hidden_states):
  class SEWDLayerNormConvLayer (line 300) | class SEWDLayerNormConvLayer(nn.Module):
    method __init__ (line 301) | def __init__(self, config, layer_id=0):
    method forward (line 316) | def forward(self, hidden_states):
  class SEWDGroupNormConvLayer (line 328) | class SEWDGroupNormConvLayer(nn.Module):
    method __init__ (line 329) | def __init__(self, config, layer_id=0):
    method forward (line 345) | def forward(self, hidden_states):
  class SEWDPositionalConvEmbedding (line 353) | class SEWDPositionalConvEmbedding(nn.Module):
    method __init__ (line 354) | def __init__(self, config):
    method forward (line 378) | def forward(self, hidden_states):
  class SEWDSamePadLayer (line 387) | class SEWDSamePadLayer(nn.Module):
    method __init__ (line 388) | def __init__(self, num_conv_pos_embeddings):
    method forward (line 392) | def forward(self, hidden_states):
  class SEWDUpsampling (line 399) | class SEWDUpsampling(nn.Module):
    method __init__ (line 400) | def __init__(self, config):
    method forward (line 406) | def forward(self, hidden_states):
  class SEWDFeatureEncoder (line 422) | class SEWDFeatureEncoder(nn.Module):
    method __init__ (line 425) | def __init__(self, config):
    method _freeze_parameters (line 442) | def _freeze_parameters(self):
    method forward (line 447) | def forward(self, input_values):
  class SEWDFeatureExtractor (line 473) | class SEWDFeatureExtractor(SEWDFeatureEncoder):
    method __init__ (line 474) | def __init__(self, config):
  class ContextPooler (line 485) | class ContextPooler(nn.Module):
    method __init__ (line 486) | def __init__(self, config):
    method forward (line 492) | def forward(self, hidden_states):
    method output_dim (line 503) | def output_dim(self):
  class XSoftmax (line 508) | class XSoftmax(torch.autograd.Function):
    method forward (line 537) | def forward(self, input, mask, dim):
    method backward (line 548) | def backward(self, grad_output):
    method symbolic (line 554) | def symbolic(g, self, mask, dim):
  class DropoutContext (line 572) | class DropoutContext(object):
    method __init__ (line 573) | def __init__(self):
  class XDropout (line 581) | class XDropout(torch.autograd.Function):
    method forward (line 585) | def forward(ctx, input, local_ctx):
    method backward (line 595) | def backward(ctx, grad_output):
    method symbolic (line 603) | def symbolic(g: torch._C.Graph, input: torch._C.Value, local_ctx: Unio...
  class StableDropout (line 621) | class StableDropout(nn.Module):
    method __init__ (line 629) | def __init__(self, drop_prob):
    method forward (line 635) | def forward(self, x):
    method clear_context (line 646) | def clear_context(self):
    method init_context (line 650) | def init_context(self, reuse_mask=True, scale=1):
    method get_context (line 658) | def get_context(self):
  class SEWDSelfOutput (line 671) | class SEWDSelfOutput(nn.Module):
    method __init__ (line 672) | def __init__(self, config):
    method forward (line 678) | def forward(self, hidden_states, input_tensor):
  class DisentangledSelfAttention (line 686) | class DisentangledSelfAttention(nn.Module):
    method __init__ (line 697) | def __init__(self, config):
    method transpose_for_scores (line 735) | def transpose_for_scores(self, x, attention_heads):
    method forward (line 740) | def forward(
    method disentangled_attention_bias (line 824) | def disentangled_attention_bias(self, query_layer, key_layer, relative...
  class SEWDAttention (line 908) | class SEWDAttention(nn.Module):
    method __init__ (line 909) | def __init__(self, config):
    method forward (line 915) | def forward(
  class SEWDIntermediate (line 945) | class SEWDIntermediate(nn.Module):
    method __init__ (line 946) | def __init__(self, config):
    method forward (line 954) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class SEWDOutput (line 961) | class SEWDOutput(nn.Module):
    method __init__ (line 962) | def __init__(self, config):
    method forward (line 969) | def forward(self, hidden_states, input_tensor):
  class SEWDLayer (line 977) | class SEWDLayer(nn.Module):
    method __init__ (line 978) | def __init__(self, config):
    method forward (line 984) | def forward(
  class ConvLayer (line 1012) | class ConvLayer(nn.Module):
    method __init__ (line 1013) | def __init__(self, config):
    method forward (line 1025) | def forward(self, hidden_states, residual_states, input_mask):
  class SEWDTransformerEncoder (line 1049) | class SEWDTransformerEncoder(nn.Module):
    method __init__ (line 1052) | def __init__(self, config):
    method get_rel_embedding (line 1079) | def get_rel_embedding(self):
    method get_attention_mask (line 1085) | def get_attention_mask(self, attention_mask):
    method get_rel_pos (line 1094) | def get_rel_pos(self, hidden_states, query_states=None, relative_pos=N...
    method forward (line 1106) | def forward(
  class SEWDEncoder (line 1188) | class SEWDEncoder(nn.Module):
    method __init__ (line 1189) | def __init__(self, config):
    method forward (line 1198) | def forward(
  class SEWDPreTrainedModel (line 1251) | class SEWDPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1263) | def _init_weights(self, module):
    method _get_feat_extract_output_lengths (line 1299) | def _get_feat_extract_output_lengths(self, input_lengths: Union[torch....
    method _get_feature_vector_attention_mask (line 1314) | def _get_feature_vector_attention_mask(self, feature_vector_length: in...
    method _set_gradient_checkpointing (line 1326) | def _set_gradient_checkpointing(self, module, value=False):
  class SEWDModel (line 1382) | class SEWDModel(SEWDPreTrainedModel):
    method __init__ (line 1383) | def __init__(self, config: SEWDConfig):
    method _mask_hidden_states (line 1403) | def _mask_hidden_states(
    method forward (line 1457) | def forward(
  class SEWDForCTC (line 1511) | class SEWDForCTC(SEWDPreTrainedModel):
    method __init__ (line 1512) | def __init__(self, config, target_lang=None):
    method freeze_feature_extractor (line 1540) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1552) | def freeze_feature_encoder(self):
    method forward (line 1567) | def forward(
  class SEWDForSequenceClassification (line 1647) | class SEWDForSequenceClassification(SEWDPreTrainedModel):
    method __init__ (line 1648) | def __init__(self, config):
    method freeze_feature_extractor (line 1665) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1677) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1684) | def freeze_base_model(self):
    method forward (line 1701) | def forward(

FILE: transformers/models/speech_encoder_decoder/configuration_speech_encoder_decoder.py
  class SpeechEncoderDecoderConfig (line 27) | class SpeechEncoderDecoderConfig(PretrainedConfig):
    method __init__ (line 76) | def __init__(self, **kwargs):
    method from_encoder_decoder_configs (line 94) | def from_encoder_decoder_configs(
    method to_dict (line 110) | def to_dict(self):

FILE: transformers/models/speech_encoder_decoder/convert_mbart_wav2vec2_seq2seq_original_to_pytorch.py
  function set_recursively (line 69) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function recursively_load_weights_wav2vec2 (line 97) | def recursively_load_weights_wav2vec2(fairseq_model, hf_model):
  function load_conv_layer (line 143) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function load_adapter (line 183) | def load_adapter(full_name, value, adapter, unused_weights):
  function make_linear_from_emb (line 237) | def make_linear_from_emb(emb):
  function convert_wav2vec2_checkpoint (line 245) | def convert_wav2vec2_checkpoint(

FILE: transformers/models/speech_encoder_decoder/convert_speech_to_text_wav2vec2_seq2seq_original_to_pytorch.py
  function set_recursively (line 71) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function recursively_load_weights_wav2vec2 (line 99) | def recursively_load_weights_wav2vec2(fairseq_model, hf_model):
  function load_conv_layer (line 149) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function make_linear_from_emb (line 189) | def make_linear_from_emb(emb):
  function create_vocab_dict (line 196) | def create_vocab_dict(dict_path):
  function convert_wav2vec2_checkpoint (line 215) | def convert_wav2vec2_checkpoint(

FILE: transformers/models/speech_encoder_decoder/modeling_flax_speech_encoder_decoder.py
  class FlaxSpeechEncoderDecoderModule (line 200) | class FlaxSpeechEncoderDecoderModule(nn.Module):
    method setup (line 204) | def setup(self):
    method _get_feat_extract_output_lengths (line 230) | def _get_feat_extract_output_lengths(
    method _get_encoder_module (line 253) | def _get_encoder_module(self):
    method _get_projection_module (line 256) | def _get_projection_module(self):
    method _get_decoder_module (line 259) | def _get_decoder_module(self):
    method __call__ (line 262) | def __call__(
  class FlaxSpeechEncoderDecoderModel (line 329) | class FlaxSpeechEncoderDecoderModel(FlaxPreTrainedModel):
    method __init__ (line 341) | def __init__(
    method init_weights (line 377) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 420) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method _get_feat_extract_output_lengths (line 461) | def _get_feat_extract_output_lengths(
    method encode (line 468) | def encode(
    method decode (line 538) | def decode(
    method __call__ (line 662) | def __call__(
    method prepare_inputs_for_generation (line 744) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 777) | def update_inputs_for_generation(self, model_outputs, model_kwargs):
    method from_encoder_decoder_pretrained (line 783) | def from_encoder_decoder_pretrained(

FILE: transformers/models/speech_encoder_decoder/modeling_speech_encoder_decoder.py
  function shift_tokens_right (line 154) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  class SpeechEncoderDecoderModel (line 173) | class SpeechEncoderDecoderModel(PreTrainedModel):
    method __init__ (line 185) | def __init__(
    method _set_gradient_checkpointing (line 252) | def _set_gradient_checkpointing(self, module, value=False):
    method get_encoder (line 257) | def get_encoder(self):
    method get_decoder (line 260) | def get_decoder(self):
    method get_output_embeddings (line 263) | def get_output_embeddings(self):
    method set_output_embeddings (line 266) | def set_output_embeddings(self, new_embeddings):
    method freeze_feature_encoder (line 269) | def freeze_feature_encoder(self):
    method from_pretrained (line 277) | def from_pretrained(cls, *args, **kwargs):
    method from_encoder_decoder_pretrained (line 288) | def from_encoder_decoder_pretrained(
    method forward (line 444) | def forward(
    method prepare_decoder_input_ids_from_labels (line 582) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method prepare_inputs_for_generation (line 585) | def prepare_inputs_for_generation(
    method resize_token_embeddings (line 600) | def resize_token_embeddings(self, *args, **kwargs):
    method _reorder_cache (line 606) | def _reorder_cache(self, past_key_values, beam_idx):

FILE: transformers/models/speech_to_text/configuration_speech_to_text.py
  class Speech2TextConfig (line 31) | class Speech2TextConfig(PretrainedConfig):
    method __init__ (line 116) | def __init__(

FILE: transformers/models/speech_to_text/convert_s2t_fairseq_to_tfms.py
  function remove_ignore_keys_ (line 23) | def remove_ignore_keys_(state_dict):
  function rename_keys (line 38) | def rename_keys(s_dict):
  function make_linear_from_emb (line 47) | def make_linear_from_emb(emb):
  function convert_fairseq_s2t_checkpoint_to_tfms (line 54) | def convert_fairseq_s2t_checkpoint_to_tfms(checkpoint_path, pytorch_dump...

FILE: transformers/models/speech_to_text/feature_extraction_speech_to_text.py
  class Speech2TextFeatureExtractor (line 33) | class Speech2TextFeatureExtractor(SequenceFeatureExtractor):
    method __init__ (line 62) | def __init__(
    method _extract_fbank_features (line 80) | def _extract_fbank_features(
    method utterance_cmvn (line 94) | def utterance_cmvn(
    method normalize (line 117) | def normalize(
    method __call__ (line 126) | def __call__(

FILE: transformers/models/speech_to_text/modeling_speech_to_text.py
  function shift_tokens_right (line 50) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 67) | def _make_causal_mask(
  function _expand_mask (line 85) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class Conv1dSubsampler (line 99) | class Conv1dSubsampler(nn.Module):
    method __init__ (line 105) | def __init__(self, config):
    method forward (line 125) | def forward(self, input_features):
  class Speech2TextSinusoidalPositionalEmbedding (line 134) | class Speech2TextSinusoidalPositionalEmbedding(nn.Module):
    method __init__ (line 137) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method make_weights (line 144) | def make_weights(self, num_embeddings: int, embedding_dim: int, paddin...
    method get_embedding (line 155) | def get_embedding(num_embeddings: int, embedding_dim: int, padding_idx...
    method forward (line 173) | def forward(self, input_ids: torch.Tensor, past_key_values_length: int...
    method create_position_ids_from_input_ids (line 187) | def create_position_ids_from_input_ids(
  class Speech2TextAttention (line 205) | class Speech2TextAttention(nn.Module):
    method __init__ (line 208) | def __init__(
    method _shape (line 235) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 238) | def forward(
  class Speech2TextEncoderLayer (line 360) | class Speech2TextEncoderLayer(nn.Module):
    method __init__ (line 361) | def __init__(self, config: Speech2TextConfig):
    method forward (line 377) | def forward(
  class Speech2TextDecoderLayer (line 429) | class Speech2TextDecoderLayer(nn.Module):
    method __init__ (line 430) | def __init__(self, config: Speech2TextConfig):
    method forward (line 456) | def forward(
  class Speech2TextPreTrainedModel (line 546) | class Speech2TextPreTrainedModel(PreTrainedModel):
    method _init_weights (line 552) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 563) | def _set_gradient_checkpointing(self, module, value=False):
    method _get_feat_extract_output_lengths (line 567) | def _get_feat_extract_output_lengths(self, input_lengths: torch.LongTe...
    method _get_feature_vector_attention_mask (line 576) | def _get_feature_vector_attention_mask(self, feature_vector_length, at...
  class Speech2TextEncoder (line 697) | class Speech2TextEncoder(Speech2TextPreTrainedModel):
    method __init__ (line 707) | def __init__(self, config: Speech2TextConfig):
    method forward (line 732) | def forward(
  class Speech2TextDecoder (line 853) | class Speech2TextDecoder(Speech2TextPreTrainedModel):
    method __init__ (line 862) | def __init__(self, config: Speech2TextConfig):
    method get_input_embeddings (line 886) | def get_input_embeddings(self):
    method set_input_embeddings (line 889) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 892) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 913) | def forward(
  class Speech2TextModel (line 1131) | class Speech2TextModel(Speech2TextPreTrainedModel):
    method __init__ (line 1132) | def __init__(self, config: Speech2TextConfig):
    method get_input_embeddings (line 1141) | def get_input_embeddings(self):
    method set_input_embeddings (line 1144) | def set_input_embeddings(self, value):
    method get_encoder (line 1147) | def get_encoder(self):
    method get_decoder (line 1150) | def get_decoder(self):
    method forward (line 1155) | def forward(
  class Speech2TextForConditionalGeneration (line 1262) | class Speech2TextForConditionalGeneration(Speech2TextPreTrainedModel):
    method __init__ (line 1276) | def __init__(self, config: Speech2TextConfig):
    method get_encoder (line 1284) | def get_encoder(self):
    method get_decoder (line 1287) | def get_decoder(self):
    method resize_token_embeddings (line 1290) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method get_output_embeddings (line 1294) | def get_output_embeddings(self):
    method set_output_embeddings (line 1297) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1302) | def forward(
    method prepare_inputs_for_generation (line 1399) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1427) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/speech_to_text/modeling_tf_speech_to_text.py
  function shift_tokens_right (line 68) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int, decoder_...
  function _make_causal_mask (line 93) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 111) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFConv1dSubsampler (line 124) | class TFConv1dSubsampler(tf.keras.layers.Layer):
    method __init__ (line 130) | def __init__(self, config: Speech2TextConfig, **kwargs):
    method call (line 149) | def call(self, input_features: tf.Tensor) -> tf.Tensor:
  class TFSpeech2TextSinusoidalPositionalEmbedding (line 170) | class TFSpeech2TextSinusoidalPositionalEmbedding(tf.keras.layers.Layer):
    method __init__ (line 173) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method _get_embedding (line 181) | def _get_embedding(num_embeddings: int, embedding_dim: int, padding_id...
    method build (line 198) | def build(self, input_shape: tf.TensorShape):
    method call (line 211) | def call(self, input_ids: tf.Tensor, past_key_values_length: int = 0) ...
    method create_position_ids_from_input_ids (line 224) | def create_position_ids_from_input_ids(
  class TFSpeech2TextAttention (line 241) | class TFSpeech2TextAttention(tf.keras.layers.Layer):
    method __init__ (line 244) | def __init__(
    method _shape (line 272) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 275) | def call(
  class TFSpeech2TextEncoderLayer (line 394) | class TFSpeech2TextEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 395) | def __init__(self, config: Speech2TextConfig, **kwargs):
    method call (line 409) | def call(
  class TFSpeech2TextDecoderLayer (line 449) | class TFSpeech2TextDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 450) | def __init__(self, config: Speech2TextConfig, **kwargs):
    method call (line 478) | def call(
  class TFSpeech2TextPreTrainedModel (line 561) | class TFSpeech2TextPreTrainedModel(TFPreTrainedModel):
    method _get_feat_extract_output_lengths (line 566) | def _get_feat_extract_output_lengths(self, input_lengths: tf.Tensor):
    method input_signature (line 576) | def input_signature(self):
  class TFSpeech2TextEncoder (line 716) | class TFSpeech2TextEncoder(tf.keras.layers.Layer):
    method __init__ (line 726) | def __init__(self, config: Speech2TextConfig, **kwargs):
    method _get_feat_extract_output_lengths (line 749) | def _get_feat_extract_output_lengths(self, input_lengths: tf.Tensor):
    method _get_feature_vector_attention_mask (line 758) | def _get_feature_vector_attention_mask(self, feature_vector_length, at...
    method call (line 778) | def call(
  class TFSpeech2TextDecoder (line 885) | class TFSpeech2TextDecoder(tf.keras.layers.Layer):
    method __init__ (line 894) | def __init__(self, config: Speech2TextConfig, **kwargs):
    method get_embed_tokens (line 916) | def get_embed_tokens(self):
    method set_embed_tokens (line 919) | def set_embed_tokens(self, embed_tokens):
    method call (line 923) | def call(
  class TFSpeech2TextMainLayer (line 1107) | class TFSpeech2TextMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1110) | def __init__(self, config: Speech2TextConfig, **kwargs):
    method get_input_embeddings (line 1117) | def get_input_embeddings(self):
    method set_input_embeddings (line 1120) | def set_input_embeddings(self, new_embeddings):
    method call (line 1124) | def call(
  class TFSpeech2TextModel (line 1215) | class TFSpeech2TextModel(TFSpeech2TextPreTrainedModel):
    method __init__ (line 1216) | def __init__(self, config: Speech2TextConfig, *inputs, **kwargs):
    method get_encoder (line 1221) | def get_encoder(self):
    method get_decoder (line 1224) | def get_decoder(self):
    method call (line 1234) | def call(
    method serving_output (line 1273) | def serving_output(self, output):
  class TFSpeech2TextForConditionalGeneration (line 1297) | class TFSpeech2TextForConditionalGeneration(TFSpeech2TextPreTrainedModel...
    method __init__ (line 1298) | def __init__(self, config: Speech2TextConfig):
    method get_encoder (line 1305) | def get_encoder(self):
    method get_decoder (line 1308) | def get_decoder(self):
    method resize_token_embeddings (line 1311) | def resize_token_embeddings(self, new_num_tokens: int) -> tf.Variable:
    method get_output_embeddings (line 1315) | def get_output_embeddings(self):
    method set_output_embeddings (line 1318) | def set_output_embeddings(self, new_embeddings):
    method call (line 1324) | def call(
    method serving_output (line 1427) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 1446) | def prepare_inputs_for_generation(

FILE: transformers/models/speech_to_text/processing_speech_to_text.py
  class Speech2TextProcessor (line 24) | class Speech2TextProcessor(ProcessorMixin):
    method __init__ (line 42) | def __init__(self, feature_extractor, tokenizer):
    method __call__ (line 47) | def __call__(self, *args, **kwargs):
    method batch_decode (line 86) | def batch_decode(self, *args, **kwargs):
    method decode (line 93) | def decode(self, *args, **kwargs):
    method as_target_processor (line 101) | def as_target_processor(self):

FILE: transformers/models/speech_to_text/tokenization_speech_to_text.py
  class Speech2TextTokenizer (line 59) | class Speech2TextTokenizer(PreTrainedTokenizer):
    method __init__ (line 113) | def __init__(
    method vocab_size (line 164) | def vocab_size(self) -> int:
    method tgt_lang (line 168) | def tgt_lang(self) -> str:
    method tgt_lang (line 172) | def tgt_lang(self, new_tgt_lang) -> None:
    method set_tgt_lang_special_tokens (line 176) | def set_tgt_lang_special_tokens(self, tgt_lang: str) -> None:
    method _tokenize (line 181) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 184) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 187) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 191) | def convert_tokens_to_string(self, tokens: List[str]) -> str:
    method build_inputs_with_special_tokens (line 207) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method get_special_tokens_mask (line 214) | def get_special_tokens_mask(
    method get_vocab (line 244) | def get_vocab(self) -> Dict:
    method __getstate__ (line 249) | def __getstate__(self) -> Dict:
    method __setstate__ (line 254) | def __setstate__(self, d: Dict) -> None:
    method save_vocabulary (line 263) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  function load_spm (line 285) | def load_spm(path: str, sp_model_kwargs: Dict[str, Any]) -> sentencepiec...
  function load_json (line 291) | def load_json(path: str) -> Union[Dict, List]:
  function save_json (line 296) | def save_json(data, path: str) -> None:

FILE: transformers/models/speech_to_text_2/configuration_speech_to_text_2.py
  class Speech2Text2Config (line 31) | class Speech2Text2Config(PretrainedConfig):
    method __init__ (line 93) | def __init__(

FILE: transformers/models/speech_to_text_2/modeling_speech_to_text_2.py
  function _make_causal_mask (line 47) | def _make_causal_mask(
  function _expand_mask (line 65) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class Speech2Text2SinusoidalPositionalEmbedding (line 80) | class Speech2Text2SinusoidalPositionalEmbedding(nn.Module):
    method __init__ (line 83) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method make_weights (line 90) | def make_weights(self, num_embeddings: int, embedding_dim: int, paddin...
    method get_embedding (line 101) | def get_embedding(num_embeddings: int, embedding_dim: int, padding_idx...
    method forward (line 119) | def forward(self, input_ids: torch.Tensor, past_key_values_length: int...
    method create_position_ids_from_input_ids (line 133) | def create_position_ids_from_input_ids(
  class Speech2Text2Attention (line 151) | class Speech2Text2Attention(nn.Module):
    method __init__ (line 154) | def __init__(
    method _shape (line 181) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 184) | def forward(
  class Speech2Text2DecoderLayer (line 305) | class Speech2Text2DecoderLayer(nn.Module):
    method __init__ (line 306) | def __init__(self, config: Speech2Text2Config):
    method forward (line 335) | def forward(
  class Speech2Text2PreTrainedModel (line 425) | class Speech2Text2PreTrainedModel(PreTrainedModel):
    method _init_weights (line 430) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 441) | def _set_gradient_checkpointing(self, module, value=False):
  class Speech2Text2Decoder (line 463) | class Speech2Text2Decoder(Speech2Text2PreTrainedModel):
    method __init__ (line 472) | def __init__(self, config: Speech2Text2Config):
    method get_input_embeddings (line 494) | def get_input_embeddings(self):
    method set_input_embeddings (line 497) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 500) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 521) | def forward(
  class Speech2Text2DecoderWrapper (line 739) | class Speech2Text2DecoderWrapper(Speech2Text2PreTrainedModel):
    method __init__ (line 745) | def __init__(self, config):
    method forward (line 749) | def forward(self, *args, **kwargs):
  class Speech2Text2ForCausalLM (line 758) | class Speech2Text2ForCausalLM(Speech2Text2PreTrainedModel):
    method __init__ (line 761) | def __init__(self, config):
    method get_input_embeddings (line 773) | def get_input_embeddings(self):
    method set_input_embeddings (line 776) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 779) | def get_output_embeddings(self):
    method set_output_embeddings (line 782) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 785) | def set_decoder(self, decoder):
    method get_decoder (line 788) | def get_decoder(self):
    method forward (line 792) | def forward(
    method prepare_inputs_for_generation (line 958) | def prepare_inputs_for_generation(
    method _reorder_cache (line 976) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/speech_to_text_2/processing_speech_to_text_2.py
  class Speech2Text2Processor (line 24) | class Speech2Text2Processor(ProcessorMixin):
    method __init__ (line 41) | def __init__(self, feature_extractor, tokenizer):
    method __call__ (line 46) | def __call__(self, *args, **kwargs):
    method batch_decode (line 85) | def batch_decode(self, *args, **kwargs):
    method decode (line 92) | def decode(self, *args, **kwargs):
    method as_target_processor (line 100) | def as_target_processor(self):

FILE: transformers/models/speech_to_text_2/tokenization_speech_to_text_2.py
  function get_pairs (line 56) | def get_pairs(word):
  class Speech2Text2Tokenizer (line 73) | class Speech2Text2Tokenizer(PreTrainedTokenizer):
    method __init__ (line 102) | def __init__(
    method vocab_size (line 142) | def vocab_size(self) -> int:
    method get_vocab (line 145) | def get_vocab(self) -> Dict:
    method bpe (line 148) | def bpe(self, token):
    method _tokenize (line 197) | def _tokenize(self, text):
    method _convert_token_to_id (line 220) | def _convert_token_to_id(self, token: str) -> int:
    method _convert_id_to_token (line 224) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 229) | def convert_tokens_to_string(self, tokens: List[str]) -> str:
    method save_vocabulary (line 241) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/speecht5/configuration_speecht5.py
  class SpeechT5Config (line 37) | class SpeechT5Config(PretrainedConfig):
    method __init__ (line 200) | def __init__(
    method inputs_to_logits_ratio (line 345) | def inputs_to_logits_ratio(self):
  class SpeechT5HifiGanConfig (line 349) | class SpeechT5HifiGanConfig(PretrainedConfig):
    method __init__ (line 403) | def __init__(

FILE: transformers/models/speecht5/convert_hifigan.py
  function load_weights (line 29) | def load_weights(checkpoint, hf_model, config):
  function convert_hifigan_checkpoint (line 59) | def convert_hifigan_checkpoint(

FILE: transformers/models/speecht5/convert_speecht5_original_pytorch_checkpoint_to_pytorch.py
  function set_recursively (line 154) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function should_ignore (line 189) | def should_ignore(name, ignore_keys):
  function recursively_load_weights (line 203) | def recursively_load_weights(fairseq_dict, hf_model, task):
  function load_conv_layer (line 275) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function convert_speecht5_checkpoint (line 320) | def convert_speecht5_checkpoint(

FILE: transformers/models/speecht5/feature_extraction_speecht5.py
  class SpeechT5FeatureExtractor (line 32) | class SpeechT5FeatureExtractor(SequenceFeatureExtractor):
    method __init__ (line 79) | def __init__(
    method zero_mean_unit_var_norm (line 142) | def zero_mean_unit_var_norm(
    method _extract_mel_features (line 163) | def _extract_mel_features(
    method __call__ (line 182) | def __call__(
    method _process_audio (line 299) | def _process_audio(
    method to_dict (line 386) | def to_dict(self) -> Dict[str, Any]:

FILE: transformers/models/speecht5/modeling_speecht5.py
  function shift_tokens_right (line 60) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function shift_spectrograms_right (line 76) | def shift_spectrograms_right(input_values: torch.Tensor, reduction_facto...
  function _make_causal_mask (line 94) | def _make_causal_mask(
  function _expand_mask (line 112) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  function _compute_mask_indices (line 127) | def _compute_mask_indices(
  class SpeechT5NoLayerNormConvLayer (line 247) | class SpeechT5NoLayerNormConvLayer(nn.Module):
    method __init__ (line 248) | def __init__(self, config, layer_id=0):
    method forward (line 262) | def forward(self, hidden_states):
  class SpeechT5LayerNormConvLayer (line 269) | class SpeechT5LayerNormConvLayer(nn.Module):
    method __init__ (line 270) | def __init__(self, config, layer_id=0):
    method forward (line 285) | def forward(self, hidden_states):
  class SpeechT5GroupNormConvLayer (line 297) | class SpeechT5GroupNormConvLayer(nn.Module):
    method __init__ (line 298) | def __init__(self, config, layer_id=0):
    method forward (line 314) | def forward(self, hidden_states):
  class SpeechT5SinusoidalPositionalEmbedding (line 322) | class SpeechT5SinusoidalPositionalEmbedding(nn.Module):
    method __init__ (line 325) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method make_weights (line 332) | def make_weights(self, num_embeddings: int, embedding_dim: int, paddin...
    method get_embedding (line 343) | def get_embedding(num_embeddings: int, embedding_dim: int, padding_idx...
    method forward (line 361) | def forward(self, input_ids: torch.Tensor, past_key_values_length: int...
    method create_position_ids_from_input_ids (line 375) | def create_position_ids_from_input_ids(
  class SpeechT5PositionalConvEmbedding (line 393) | class SpeechT5PositionalConvEmbedding(nn.Module):
    method __init__ (line 394) | def __init__(self, config):
    method forward (line 421) | def forward(self, hidden_states):
  class SpeechT5ScaledPositionalEncoding (line 432) | class SpeechT5ScaledPositionalEncoding(nn.Module):
    method __init__ (line 437) | def __init__(self, dropout, dim, max_len=5000):
    method forward (line 450) | def forward(self, emb):
  class SpeechT5RelativePositionalEncoding (line 456) | class SpeechT5RelativePositionalEncoding(torch.nn.Module):
    method __init__ (line 457) | def __init__(self, dim, max_length=1000):
    method forward (line 463) | def forward(self, hidden_states):
  class SpeechT5SamePadLayer (line 476) | class SpeechT5SamePadLayer(nn.Module):
    method __init__ (line 477) | def __init__(self, num_conv_pos_embeddings):
    method forward (line 481) | def forward(self, hidden_states):
  class SpeechT5FeatureEncoder (line 488) | class SpeechT5FeatureEncoder(nn.Module):
    method __init__ (line 491) | def __init__(self, config):
    method _freeze_parameters (line 510) | def _freeze_parameters(self):
    method forward (line 515) | def forward(self, input_values):
  class SpeechT5FeatureProjection (line 542) | class SpeechT5FeatureProjection(nn.Module):
    method __init__ (line 543) | def __init__(self, config):
    method forward (line 549) | def forward(self, hidden_states):
  class SpeechT5SpeechEncoderPrenet (line 557) | class SpeechT5SpeechEncoderPrenet(nn.Module):
    method __init__ (line 558) | def __init__(self, config):
    method freeze_feature_encoder (line 575) | def freeze_feature_encoder(self):
    method forward (line 578) | def forward(
    method _get_feature_vector_attention_mask (line 613) | def _get_feature_vector_attention_mask(self, feature_vector_length: in...
    method _get_feat_extract_output_lengths (line 629) | def _get_feat_extract_output_lengths(self, input_lengths: Union[torch....
    method _mask_hidden_states (line 645) | def _mask_hidden_states(
  class SpeechT5SpeechDecoderPrenet (line 692) | class SpeechT5SpeechDecoderPrenet(nn.Module):
    method __init__ (line 693) | def __init__(self, config):
    method forward (line 717) | def forward(
  class SpeechT5BatchNormConvLayer (line 744) | class SpeechT5BatchNormConvLayer(nn.Module):
    method __init__ (line 745) | def __init__(self, config, layer_id=0):
    method forward (line 775) | def forward(self, hidden_states):
  class SpeechT5SpeechDecoderPostnet (line 784) | class SpeechT5SpeechDecoderPostnet(nn.Module):
    method __init__ (line 785) | def __init__(self, config):
    method forward (line 796) | def forward(self, hidden_states: torch.Tensor):
    method postnet (line 802) | def postnet(self, hidden_states: torch.Tensor):
  class SpeechT5TextEncoderPrenet (line 809) | class SpeechT5TextEncoderPrenet(nn.Module):
    method __init__ (line 810) | def __init__(self, config):
    method get_input_embeddings (line 820) | def get_input_embeddings(self):
    method set_input_embeddings (line 823) | def set_input_embeddings(self, value):
    method forward (line 826) | def forward(self, input_ids: torch.Tensor):
  class SpeechT5TextDecoderPrenet (line 832) | class SpeechT5TextDecoderPrenet(nn.Module):
    method __init__ (line 833) | def __init__(self, config):
    method get_input_embeddings (line 847) | def get_input_embeddings(self):
    method set_input_embeddings (line 850) | def set_input_embeddings(self, value):
    method forward (line 853) | def forward(
  class SpeechT5TextDecoderPostnet (line 875) | class SpeechT5TextDecoderPostnet(nn.Module):
    method __init__ (line 876) | def __init__(self, config):
    method forward (line 881) | def forward(self, hidden_states: torch.Tensor):
    method get_output_embeddings (line 884) | def get_output_embeddings(self):
    method set_output_embeddings (line 887) | def set_output_embeddings(self, new_embeddings):
  class SpeechT5Attention (line 891) | class SpeechT5Attention(nn.Module):
    method __init__ (line 897) | def __init__(
    method _shape (line 924) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 927) | def forward(
  class SpeechT5FeedForward (line 1051) | class SpeechT5FeedForward(nn.Module):
    method __init__ (line 1052) | def __init__(self, config, intermediate_size):
    method forward (line 1065) | def forward(self, hidden_states):
  class SpeechT5EncoderLayer (line 1075) | class SpeechT5EncoderLayer(nn.Module):
    method __init__ (line 1076) | def __init__(self, config: SpeechT5Config):
    method forward (line 1089) | def forward(
  class SpeechT5DecoderLayer (line 1136) | class SpeechT5DecoderLayer(nn.Module):
    method __init__ (line 1137) | def __init__(self, config: SpeechT5Config):
    method forward (line 1159) | def forward(
  class SpeechT5PreTrainedModel (line 1244) | class SpeechT5PreTrainedModel(PreTrainedModel):
    method _init_weights (line 1257) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 1287) | def _set_gradient_checkpointing(self, module, value=False):
  class SpeechT5Encoder (line 1292) | class SpeechT5Encoder(SpeechT5PreTrainedModel):
    method __init__ (line 1297) | def __init__(self, config: SpeechT5Config):
    method forward (line 1314) | def forward(
  class SpeechT5EncoderWithSpeechPrenet (line 1433) | class SpeechT5EncoderWithSpeechPrenet(SpeechT5PreTrainedModel):
    method __init__ (line 1439) | def __init__(self, config: SpeechT5Config):
    method forward (line 1448) | def forward(
  class SpeechT5EncoderWithTextPrenet (line 1471) | class SpeechT5EncoderWithTextPrenet(SpeechT5PreTrainedModel):
    method __init__ (line 1476) | def __init__(self, config: SpeechT5Config):
    method get_input_embeddings (line 1485) | def get_input_embeddings(self):
    method set_input_embeddings (line 1488) | def set_input_embeddings(self, value):
    method forward (line 1491) | def forward(
  class SpeechT5EncoderWithoutPrenet (line 1514) | class SpeechT5EncoderWithoutPrenet(SpeechT5PreTrainedModel):
    method __init__ (line 1520) | def __init__(self, config: SpeechT5Config):
    method forward (line 1528) | def forward(
  class SpeechT5Decoder (line 1547) | class SpeechT5Decoder(SpeechT5PreTrainedModel):
    method __init__ (line 1552) | def __init__(self, config: SpeechT5Config):
    method _prepare_decoder_attention_mask (line 1564) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 1587) | def forward(
  class SpeechT5DecoderWithSpeechPrenet (line 1781) | class SpeechT5DecoderWithSpeechPrenet(SpeechT5PreTrainedModel):
    method __init__ (line 1787) | def __init__(self, config: SpeechT5Config):
    method forward (line 1796) | def forward(
  class SpeechT5DecoderWithTextPrenet (line 1830) | class SpeechT5DecoderWithTextPrenet(SpeechT5PreTrainedModel):
    method __init__ (line 1835) | def __init__(self, config: SpeechT5Config):
    method get_input_embeddings (line 1844) | def get_input_embeddings(self):
    method set_input_embeddings (line 1847) | def set_input_embeddings(self, value):
    method forward (line 1850) | def forward(
  class SpeechT5DecoderWithoutPrenet (line 1883) | class SpeechT5DecoderWithoutPrenet(SpeechT5PreTrainedModel):
    method __init__ (line 1889) | def __init__(self, config: SpeechT5Config):
    method forward (line 1897) | def forward(
  class SpeechT5GuidedMultiheadAttentionLoss (line 1927) | class SpeechT5GuidedMultiheadAttentionLoss(nn.Module):
    method __init__ (line 1933) | def __init__(self, config: SpeechT5Config):
    method forward (line 1938) | def forward(
    method _make_guided_attention_masks (line 1963) | def _make_guided_attention_masks(self, input_masks, output_masks, devi...
    method _make_guided_attention_mask (line 1975) | def _make_guided_attention_mask(input_length, output_length, sigma, de...
  class SpeechT5SpectrogramLoss (line 1986) | class SpeechT5SpectrogramLoss(nn.Module):
    method __init__ (line 1991) | def __init__(self, config: SpeechT5Config):
    method forward (line 2003) | def forward(
  class SpeechT5Model (line 2178) | class SpeechT5Model(SpeechT5PreTrainedModel):
    method __init__ (line 2179) | def __init__(
    method get_input_embeddings (line 2193) | def get_input_embeddings(self):
    method set_input_embeddings (line 2200) | def set_input_embeddings(self, value):
    method get_encoder (line 2206) | def get_encoder(self):
    method get_decoder (line 2209) | def get_decoder(self):
    method freeze_feature_encoder (line 2212) | def freeze_feature_encoder(self):
    method forward (line 2222) | def forward(
  class SpeechT5ForSpeechToText (line 2326) | class SpeechT5ForSpeechToText(SpeechT5PreTrainedModel):
    method __init__ (line 2335) | def __init__(self, config: SpeechT5Config):
    method get_encoder (line 2355) | def get_encoder(self):
    method get_decoder (line 2358) | def get_decoder(self):
    method freeze_feature_encoder (line 2361) | def freeze_feature_encoder(self):
    method resize_token_embeddings (line 2368) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method get_output_embeddings (line 2372) | def get_output_embeddings(self):
    method set_output_embeddings (line 2375) | def set_output_embeddings(self, new_embeddings):
    method forward (line 2380) | def forward(
    method prepare_inputs_for_generation (line 2507) | def prepare_inputs_for_generation(
    method _reorder_cache (line 2535) | def _reorder_cache(past_key_values, beam_idx):
  function _generate_speech (line 2542) | def _generate_speech(
  class SpeechT5ForTextToSpeech (line 2637) | class SpeechT5ForTextToSpeech(SpeechT5PreTrainedModel):
    method __init__ (line 2643) | def __init__(self, config: SpeechT5Config):
    method get_encoder (line 2663) | def get_encoder(self):
    method get_decoder (line 2666) | def get_decoder(self):
    method forward (line 2671) | def forward(
    method generate_speech (line 2796) | def generate_speech(
  class SpeechT5ForSpeechToSpeech (line 2858) | class SpeechT5ForSpeechToSpeech(SpeechT5PreTrainedModel):
    method __init__ (line 2866) | def __init__(self, config: SpeechT5Config):
    method get_encoder (line 2878) | def get_encoder(self):
    method get_decoder (line 2881) | def get_decoder(self):
    method freeze_feature_encoder (line 2884) | def freeze_feature_encoder(self):
    method forward (line 2893) | def forward(
    method generate_speech (line 3012) | def generate_speech(
  class HifiGanResidualBlock (line 3090) | class HifiGanResidualBlock(nn.Module):
    method __init__ (line 3091) | def __init__(self, channels, kernel_size=3, dilation=(1, 3, 5), leaky_...
    method get_padding (line 3122) | def get_padding(self, kernel_size, dilation=1):
    method apply_weight_norm (line 3125) | def apply_weight_norm(self):
    method remove_weight_norm (line 3131) | def remove_weight_norm(self):
    method forward (line 3137) | def forward(self, hidden_states):
  class SpeechT5HifiGan (line 3152) | class SpeechT5HifiGan(PreTrainedModel):
    method __init__ (line 3156) | def __init__(self, config: SpeechT5HifiGanConfig):
    method _init_weights (line 3194) | def _init_weights(self, module):
    method apply_weight_norm (line 3201) | def apply_weight_norm(self):
    method remove_weight_norm (line 3209) | def remove_weight_norm(self):
    method forward (line 3217) | def forward(self, spectrogram: torch.FloatTensor) -> torch.FloatTensor:

FILE: transformers/models/speecht5/processing_speecht5.py
  class SpeechT5Processor (line 20) | class SpeechT5Processor(ProcessorMixin):
    method __init__ (line 36) | def __init__(self, feature_extractor, tokenizer):
    method __call__ (line 39) | def __call__(self, *args, **kwargs):
    method pad (line 110) | def pad(self, *args, **kwargs):
    method batch_decode (line 170) | def batch_decode(self, *args, **kwargs):
    method decode (line 177) | def decode(self, *args, **kwargs):

FILE: transformers/models/speecht5/tokenization_speecht5.py
  class SpeechT5Tokenizer (line 47) | class SpeechT5Tokenizer(PreTrainedTokenizer):
    method __init__ (line 93) | def __init__(
    method vocab_size (line 120) | def vocab_size(self):
    method get_vocab (line 123) | def get_vocab(self):
    method __getstate__ (line 128) | def __getstate__(self):
    method __setstate__ (line 133) | def __setstate__(self, d):
    method _tokenize (line 143) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 147) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 151) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 156) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 170) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method get_special_tokens_mask (line 177) | def get_special_tokens_mask(
    method save_vocabulary (line 190) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/splinter/configuration_splinter.py
  class SplinterConfig (line 32) | class SplinterConfig(PretrainedConfig):
    method __init__ (line 93) | def __init__(

FILE: transformers/models/splinter/modeling_splinter.py
  class SplinterEmbeddings (line 49) | class SplinterEmbeddings(nn.Module):
    method __init__ (line 52) | def __init__(self, config):
    method forward (line 67) | def forward(
  class SplinterSelfAttention (line 102) | class SplinterSelfAttention(nn.Module):
    method __init__ (line 103) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 129) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 134) | def forward(
  class SplinterSelfOutput (line 237) | class SplinterSelfOutput(nn.Module):
    method __init__ (line 238) | def __init__(self, config):
    method forward (line 244) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class SplinterAttention (line 252) | class SplinterAttention(nn.Module):
    method __init__ (line 253) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 259) | def prune_heads(self, heads):
    method forward (line 277) | def forward(
  class SplinterIntermediate (line 302) | class SplinterIntermediate(nn.Module):
    method __init__ (line 303) | def __init__(self, config):
    method forward (line 311) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class SplinterOutput (line 318) | class SplinterOutput(nn.Module):
    method __init__ (line 319) | def __init__(self, config):
    method forward (line 325) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class SplinterLayer (line 333) | class SplinterLayer(nn.Module):
    method __init__ (line 334) | def __init__(self, config):
    method forward (line 348) | def forward(
    method feed_forward_chunk (line 413) | def feed_forward_chunk(self, attention_output):
  class SplinterEncoder (line 420) | class SplinterEncoder(nn.Module):
    method __init__ (line 421) | def __init__(self, config):
    method forward (line 427) | def forward(
  class SplinterPreTrainedModel (line 518) | class SplinterPreTrainedModel(PreTrainedModel):
    method _init_weights (line 530) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 546) | def _set_gradient_checkpointing(self, module, value=False):
  class SplinterModel (line 616) | class SplinterModel(SplinterPreTrainedModel):
    method __init__ (line 623) | def __init__(self, config):
    method get_input_embeddings (line 633) | def get_input_embeddings(self):
    method set_input_embeddings (line 636) | def set_input_embeddings(self, value):
    method _prune_heads (line 639) | def _prune_heads(self, heads_to_prune):
    method forward (line 653) | def forward(
  class SplinterFullyConnectedLayer (line 774) | class SplinterFullyConnectedLayer(nn.Module):
    method __init__ (line 775) | def __init__(self, input_dim, output_dim, hidden_act="gelu"):
    method forward (line 785) | def forward(self, inputs: torch.Tensor) -> torch.Tensor:
  class QuestionAwareSpanSelectionHead (line 792) | class QuestionAwareSpanSelectionHead(nn.Module):
    method __init__ (line 798) | def __init__(self, config):
    method forward (line 809) | def forward(self, inputs, positions):
  class SplinterForQuestionAnswering (line 837) | class SplinterForQuestionAnswering(SplinterPreTrainedModel):
    method __init__ (line 838) | def __init__(self, config):
    method forward (line 854) | def forward(
  class SplinterForPreTrainingOutput (line 952) | class SplinterForPreTrainingOutput(ModelOutput):
  class SplinterForPreTraining (line 991) | class SplinterForPreTraining(SplinterPreTrainedModel):
    method __init__ (line 992) | def __init__(self, config):
    method forward (line 1005) | def forward(
    method _prepare_question_positions (line 1105) | def _prepare_question_positions(self, input_ids: torch.Tensor) -> torc...

FILE: transformers/models/splinter/tokenization_splinter.py
  function load_vocab (line 55) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 66) | def whitespace_tokenize(text):
  class SplinterTokenizer (line 75) | class SplinterTokenizer(PreTrainedTokenizer):
    method __init__ (line 124) | def __init__(
    method question_token_id (line 173) | def question_token_id(self):
    method do_lower_case (line 181) | def do_lower_case(self):
    method vocab_size (line 185) | def vocab_size(self):
    method get_vocab (line 188) | def get_vocab(self):
    method _tokenize (line 191) | def _tokenize(self, text):
    method _convert_token_to_id (line 204) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 208) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 212) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 217) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 249) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 277) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 306) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 327) | class BasicTokenizer(object):
    method __init__ (line 347) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 355) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 392) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 403) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 425) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 438) | def _is_chinese_char(self, cp):
    method _clean_text (line 462) | def _clean_text(self, text):
  class WordpieceTokenizer (line 476) | class WordpieceTokenizer(object):
    method __init__ (line 479) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 484) | def tokenize(self, text):

FILE: transformers/models/splinter/tokenization_splinter_fast.py
  class SplinterTokenizerFast (line 55) | class SplinterTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 103) | def __init__(
    method question_token_id (line 146) | def question_token_id(self):
    method build_inputs_with_special_tokens (line 153) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 185) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 214) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/squeezebert/configuration_squeezebert.py
  class SqueezeBertConfig (line 37) | class SqueezeBertConfig(PretrainedConfig):
    method __init__ (line 115) | def __init__(
  class SqueezeBertOnnxConfig (line 163) | class SqueezeBertOnnxConfig(OnnxConfig):
    method inputs (line 165) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/squeezebert/modeling_squeezebert.py
  class SqueezeBertEmbeddings (line 52) | class SqueezeBertEmbeddings(nn.Module):
    method __init__ (line 55) | def __init__(self, config):
    method forward (line 69) | def forward(self, input_ids=None, token_type_ids=None, position_ids=No...
  class MatMulWrapper (line 94) | class MatMulWrapper(nn.Module):
    method __init__ (line 100) | def __init__(self):
    method forward (line 103) | def forward(self, mat1, mat2):
  class SqueezeBertLayerNorm (line 114) | class SqueezeBertLayerNorm(nn.LayerNorm):
    method __init__ (line 121) | def __init__(self, hidden_size, eps=1e-12):
    method forward (line 124) | def forward(self, x):
  class ConvDropoutLayerNorm (line 130) | class ConvDropoutLayerNorm(nn.Module):
    method __init__ (line 135) | def __init__(self, cin, cout, groups, dropout_prob):
    method forward (line 142) | def forward(self, hidden_states, input_tensor):
  class ConvActivation (line 150) | class ConvActivation(nn.Module):
    method __init__ (line 155) | def __init__(self, cin, cout, groups, act):
    method forward (line 160) | def forward(self, x):
  class SqueezeBertSelfAttention (line 165) | class SqueezeBertSelfAttention(nn.Module):
    method __init__ (line 166) | def __init__(self, config, cin, q_groups=1, k_groups=1, v_groups=1):
    method transpose_for_scores (line 190) | def transpose_for_scores(self, x):
    method transpose_key_for_scores (line 199) | def transpose_key_for_scores(self, x):
    method transpose_output (line 209) | def transpose_output(self, x):
    method forward (line 219) | def forward(self, hidden_states, attention_mask, output_attentions):
  class SqueezeBertModule (line 255) | class SqueezeBertModule(nn.Module):
    method __init__ (line 256) | def __init__(self, config):
    method forward (line 282) | def forward(self, hidden_states, attention_mask, output_attentions):
  class SqueezeBertEncoder (line 297) | class SqueezeBertEncoder(nn.Module):
    method __init__ (line 298) | def __init__(self, config):
    method forward (line 309) | def forward(
  class SqueezeBertPooler (line 358) | class SqueezeBertPooler(nn.Module):
    method __init__ (line 359) | def __init__(self, config):
    method forward (line 364) | def forward(self, hidden_states):
  class SqueezeBertPredictionHeadTransform (line 373) | class SqueezeBertPredictionHeadTransform(nn.Module):
    method __init__ (line 374) | def __init__(self, config):
    method forward (line 383) | def forward(self, hidden_states):
  class SqueezeBertLMPredictionHead (line 390) | class SqueezeBertLMPredictionHead(nn.Module):
    method __init__ (line 391) | def __init__(self, config):
    method forward (line 404) | def forward(self, hidden_states):
  class SqueezeBertOnlyMLMHead (line 410) | class SqueezeBertOnlyMLMHead(nn.Module):
    method __init__ (line 411) | def __init__(self, config):
    method forward (line 415) | def forward(self, sequence_output):
  class SqueezeBertPreTrainedModel (line 420) | class SqueezeBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 430) | def _init_weights(self, module):
  class SqueezeBertModel (line 546) | class SqueezeBertModel(SqueezeBertPreTrainedModel):
    method __init__ (line 547) | def __init__(self, config):
    method get_input_embeddings (line 557) | def get_input_embeddings(self):
    method set_input_embeddings (line 560) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 563) | def _prune_heads(self, heads_to_prune):
    method forward (line 577) | def forward(
  class SqueezeBertForMaskedLM (line 645) | class SqueezeBertForMaskedLM(SqueezeBertPreTrainedModel):
    method __init__ (line 652) | def __init__(self, config):
    method get_output_embeddings (line 661) | def get_output_embeddings(self):
    method set_output_embeddings (line 664) | def set_output_embeddings(self, new_embeddings):
    method forward (line 673) | def forward(
  class SqueezeBertForSequenceClassification (line 733) | class SqueezeBertForSequenceClassification(SqueezeBertPreTrainedModel):
    method __init__ (line 734) | def __init__(self, config):
    method forward (line 752) | def forward(
  class SqueezeBertForMultipleChoice (line 832) | class SqueezeBertForMultipleChoice(SqueezeBertPreTrainedModel):
    method __init__ (line 833) | def __init__(self, config):
    method forward (line 851) | def forward(
  class SqueezeBertForTokenClassification (line 925) | class SqueezeBertForTokenClassification(SqueezeBertPreTrainedModel):
    method __init__ (line 926) | def __init__(self, config):
    method forward (line 943) | def forward(
  class SqueezeBertForQuestionAnswering (line 1003) | class SqueezeBertForQuestionAnswering(SqueezeBertPreTrainedModel):
    method __init__ (line 1004) | def __init__(self, config):
    method forward (line 1020) | def forward(

FILE: transformers/models/squeezebert/tokenization_squeezebert.py
  function load_vocab (line 57) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 69) | def whitespace_tokenize(text):
  class SqueezeBertTokenizer (line 79) | class SqueezeBertTokenizer(PreTrainedTokenizer):
    method __init__ (line 126) | def __init__(
    method do_lower_case (line 173) | def do_lower_case(self):
    method vocab_size (line 177) | def vocab_size(self):
    method get_vocab (line 180) | def get_vocab(self):
    method _tokenize (line 183) | def _tokenize(self, text):
    method _convert_token_to_id (line 196) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 200) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 204) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 209) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 234) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 262) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 291) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
  class BasicTokenizer (line 313) | class BasicTokenizer(object):
    method __init__ (line 333) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 341) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 378) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 389) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 411) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 424) | def _is_chinese_char(self, cp):
    method _clean_text (line 448) | def _clean_text(self, text):
  class WordpieceTokenizer (line 462) | class WordpieceTokenizer(object):
    method __init__ (line 465) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 470) | def tokenize(self, text):

FILE: transformers/models/squeezebert/tokenization_squeezebert_fast.py
  class SqueezeBertTokenizerFast (line 69) | class SqueezeBertTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 115) | def __init__(
    method build_inputs_with_special_tokens (line 157) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method create_token_type_ids_from_sequences (line 181) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 210) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/swiftformer/configuration_swiftformer.py
  class SwiftFormerConfig (line 34) | class SwiftFormerConfig(PretrainedConfig):
    method __init__ (line 90) | def __init__(
  class SwiftFormerOnnxConfig (line 123) | class SwiftFormerOnnxConfig(OnnxConfig):
    method inputs (line 127) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 135) | def atol_for_validation(self) -> float:

FILE: transformers/models/swiftformer/convert_swiftformer_original_to_hf.py
  function prepare_img (line 42) | def prepare_img():
  function get_expected_output (line 48) | def get_expected_output(swiftformer_name):
  function rename_key (line 62) | def rename_key(dct, old, new):
  function create_rename_keys (line 67) | def create_rename_keys(state_dict):
  function convert_swiftformer_checkpoint (line 90) | def convert_swiftformer_checkpoint(swiftformer_name, pytorch_dump_folder...

FILE: transformers/models/swiftformer/modeling_swiftformer.py
  class SwiftFormerPatchEmbedding (line 61) | class SwiftFormerPatchEmbedding(nn.Module):
    method __init__ (line 70) | def __init__(self, config: SwiftFormerConfig):
    method forward (line 84) | def forward(self, x):
  function drop_path (line 89) | def drop_path(x, drop_prob: float = 0.0, training: bool = False):
  class SwiftFormerDropPath (line 110) | class SwiftFormerDropPath(nn.Module):
    method __init__ (line 113) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 117) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 120) | def extra_repr(self) -> str:
  class SwiftFormerEmbeddings (line 124) | class SwiftFormerEmbeddings(nn.Module):
    method __init__ (line 133) | def __init__(self, config: SwiftFormerConfig, index: int):
    method forward (line 151) | def forward(self, x):
  class SwiftFormerConvEncoder (line 157) | class SwiftFormerConvEncoder(nn.Module):
    method __init__ (line 166) | def __init__(self, config: SwiftFormerConfig, dim: int):
    method forward (line 178) | def forward(self, x):
  class SwiftFormerMlp (line 189) | class SwiftFormerMlp(nn.Module):
    method __init__ (line 198) | def __init__(self, config: SwiftFormerConfig, in_features: int):
    method forward (line 208) | def forward(self, x):
  class SwiftFormerEfficientAdditiveAttention (line 218) | class SwiftFormerEfficientAdditiveAttention(nn.Module):
    method __init__ (line 227) | def __init__(self, config: SwiftFormerConfig, dim: int = 512):
    method forward (line 238) | def forward(self, x):
  class SwiftFormerLocalRepresentation (line 258) | class SwiftFormerLocalRepresentation(nn.Module):
    method __init__ (line 267) | def __init__(self, config: SwiftFormerConfig, dim: int):
    method forward (line 278) | def forward(self, x):
  class SwiftFormerEncoderBlock (line 289) | class SwiftFormerEncoderBlock(nn.Module):
    method __init__ (line 299) | def __init__(self, config: SwiftFormerConfig, dim: int, drop_path: flo...
    method forward (line 318) | def forward(self, x):
  class SwiftFormerStage (line 340) | class SwiftFormerStage(nn.Module):
    method __init__ (line 350) | def __init__(self, config: SwiftFormerConfig, index: int) -> None:
    method forward (line 368) | def forward(self, input):
  class SwiftFormerEncoder (line 374) | class SwiftFormerEncoder(nn.Module):
    method __init__ (line 375) | def __init__(self, config: SwiftFormerConfig) -> None:
    method forward (line 397) | def forward(
  class SwiftFormerPreTrainedModel (line 424) | class SwiftFormerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 435) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 445) | def _set_gradient_checkpointing(self, module: SwiftFormerEncoder, valu...
  class SwiftFormerModel (line 479) | class SwiftFormerModel(SwiftFormerPreTrainedModel):
    method __init__ (line 480) | def __init__(self, config: SwiftFormerConfig):
    method forward (line 498) | def forward(
  class SwiftFormerForImageClassification (line 536) | class SwiftFormerForImageClassification(SwiftFormerPreTrainedModel):
    method __init__ (line 537) | def __init__(self, config: SwiftFormerConfig) -> None:
    method forward (line 560) | def forward(

FILE: transformers/models/swin/configuration_swin.py
  class SwinConfig (line 38) | class SwinConfig(BackboneConfigMixin, PretrainedConfig):
    method __init__ (line 115) | def __init__(
  class SwinOnnxConfig (line 167) | class SwinOnnxConfig(OnnxConfig):
    method inputs (line 171) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 179) | def atol_for_validation(self) -> float:

FILE: transformers/models/swin/convert_swin_simmim_to_pytorch.py
  function get_swin_config (line 28) | def get_swin_config(model_name):
  function rename_key (line 52) | def rename_key(name):
  function convert_state_dict (line 85) | def convert_state_dict(orig_state_dict, model):
  function convert_swin_checkpoint (line 123) | def convert_swin_checkpoint(model_name, checkpoint_path, pytorch_dump_fo...

FILE: transformers/models/swin/convert_swin_timm_to_pytorch.py
  function get_swin_config (line 13) | def get_swin_config(swin_name):
  function rename_key (line 59) | def rename_key(name):
  function convert_state_dict (line 92) | def convert_state_dict(orig_state_dict, model):
  function convert_swin_checkpoint (line 130) | def convert_swin_checkpoint(swin_name, pytorch_dump_folder_path):

FILE: transformers/models/swin/modeling_swin.py
  class SwinEncoderOutput (line 68) | class SwinEncoderOutput(ModelOutput):
  class SwinModelOutput (line 101) | class SwinModelOutput(ModelOutput):
  class SwinMaskedImageModelingOutput (line 137) | class SwinMaskedImageModelingOutput(ModelOutput):
    method logits (line 172) | def logits(self):
  class SwinImageClassifierOutput (line 182) | class SwinImageClassifierOutput(ModelOutput):
  function window_partition (line 217) | def window_partition(input_feature, window_size):
  function window_reverse (line 229) | def window_reverse(windows, window_size, height, width):
  class SwinEmbeddings (line 239) | class SwinEmbeddings(nn.Module):
    method __init__ (line 244) | def __init__(self, config, use_mask_token=False):
    method forward (line 260) | def forward(
  class SwinPatchEmbeddings (line 281) | class SwinPatchEmbeddings(nn.Module):
    method __init__ (line 288) | def __init__(self, config):
    method maybe_pad (line 303) | def maybe_pad(self, pixel_values, height, width):
    method forward (line 312) | def forward(self, pixel_values: Optional[torch.FloatTensor]) -> Tuple[...
  class SwinPatchMerging (line 328) | class SwinPatchMerging(nn.Module):
    method __init__ (line 341) | def __init__(self, input_resolution: Tuple[int], dim: int, norm_layer:...
    method maybe_pad (line 348) | def maybe_pad(self, input_feature, height, width):
    method forward (line 356) | def forward(self, input_feature: torch.Tensor, input_dimensions: Tuple...
  function drop_path (line 383) | def drop_path(input, drop_prob=0.0, training=False, scale_by_keep=True):
  class SwinDropPath (line 404) | class SwinDropPath(nn.Module):
    method __init__ (line 407) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 411) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 414) | def extra_repr(self) -> str:
  class SwinSelfAttention (line 418) | class SwinSelfAttention(nn.Module):
    method __init__ (line 419) | def __init__(self, config, dim, num_heads, window_size):
    method transpose_for_scores (line 456) | def transpose_for_scores(self, x):
    method forward (line 461) | def forward(
  class SwinSelfOutput (line 518) | class SwinSelfOutput(nn.Module):
    method __init__ (line 519) | def __init__(self, config, dim):
    method forward (line 524) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class SwinAttention (line 531) | class SwinAttention(nn.Module):
    method __init__ (line 532) | def __init__(self, config, dim, num_heads, window_size):
    method prune_heads (line 538) | def prune_heads(self, heads):
    method forward (line 556) | def forward(
  class SwinIntermediate (line 569) | class SwinIntermediate(nn.Module):
    method __init__ (line 570) | def __init__(self, config, dim):
    method forward (line 578) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class SwinOutput (line 584) | class SwinOutput(nn.Module):
    method __init__ (line 585) | def __init__(self, config, dim):
    method forward (line 590) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class SwinLayer (line 596) | class SwinLayer(nn.Module):
    method __init__ (line 597) | def __init__(self, config, dim, input_resolution, num_heads, shift_siz...
    method set_shift_and_window_size (line 610) | def set_shift_and_window_size(self, input_resolution):
    method get_attn_mask (line 616) | def get_attn_mask(self, height, width, dtype):
    method maybe_pad (line 644) | def maybe_pad(self, hidden_states, height, width):
    method forward (line 651) | def forward(
  class SwinStage (line 719) | class SwinStage(nn.Module):
    method __init__ (line 720) | def __init__(self, config, dim, input_resolution, depth, num_heads, dr...
    method forward (line 745) | def forward(
  class SwinEncoder (line 778) | class SwinEncoder(nn.Module):
    method __init__ (line 779) | def __init__(self, config, grid_size):
    method forward (line 801) | def forward(
  class SwinPreTrainedModel (line 881) | class SwinPreTrainedModel(PreTrainedModel):
    method _init_weights (line 892) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 904) | def _set_gradient_checkpointing(self, module, value=False):
  class SwinModel (line 946) | class SwinModel(SwinPreTrainedModel):
    method __init__ (line 947) | def __init__(self, config, add_pooling_layer=True, use_mask_token=False):
    method get_input_embeddings (line 962) | def get_input_embeddings(self):
    method _prune_heads (line 965) | def _prune_heads(self, heads_to_prune):
    method forward (line 981) | def forward(
  class SwinForMaskedImageModeling (line 1055) | class SwinForMaskedImageModeling(SwinPreTrainedModel):
    method __init__ (line 1056) | def __init__(self, config):
    method forward (line 1074) | def forward(
  class SwinForImageClassification (line 1166) | class SwinForImageClassification(SwinPreTrainedModel):
    method __init__ (line 1167) | def __init__(self, config):
    method forward (line 1188) | def forward(
  class SwinBackbone (line 1259) | class SwinBackbone(SwinPreTrainedModel, BackboneMixin):
    method __init__ (line 1260) | def __init__(self, config: SwinConfig):
    method get_input_embeddings (line 1277) | def get_input_embeddings(self):
    method forward (line 1280) | def forward(

FILE: transformers/models/swin/modeling_tf_swin.py
  class TFSwinEncoderOutput (line 73) | class TFSwinEncoderOutput(ModelOutput):
  class TFSwinModelOutput (line 106) | class TFSwinModelOutput(ModelOutput):
  class TFSwinMaskedImageModelingOutput (line 142) | class TFSwinMaskedImageModelingOutput(ModelOutput):
    method logits (line 177) | def logits(self):
  class TFSwinImageClassifierOutput (line 187) | class TFSwinImageClassifierOutput(ModelOutput):
  function window_partition (line 222) | def window_partition(input_feature: tf.Tensor, window_size: int) -> tf.T...
  function window_reverse (line 236) | def window_reverse(windows: tf.Tensor, window_size: int, height: int, wi...
  function drop_path (line 251) | def drop_path(
  class TFSwinEmbeddings (line 270) | class TFSwinEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 275) | def __init__(self, config: SwinConfig, use_mask_token: bool = False, *...
    method build (line 287) | def build(self, input_shape: tf.TensorShape) -> None:
    method call (line 301) | def call(
  class TFSwinPatchEmbeddings (line 325) | class TFSwinPatchEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 330) | def __init__(self, config, **kwargs):
    method maybe_pad (line 351) | def maybe_pad(self, pixel_values: tf.Tensor, height: int, width: int) ...
    method call (line 360) | def call(self, pixel_values: tf.Tensor, training: bool = False) -> Tup...
  class TFSwinPatchMerging (line 385) | class TFSwinPatchMerging(tf.keras.layers.Layer):
    method __init__ (line 398) | def __init__(
    method maybe_pad (line 411) | def maybe_pad(self, input_feature: tf.Tensor, height: int, width: int)...
    method call (line 419) | def call(self, input_feature: tf.Tensor, input_dimensions: Tuple[int, ...
  class TFSwinDropPath (line 447) | class TFSwinDropPath(tf.keras.layers.Layer):
    method __init__ (line 450) | def __init__(self, drop_prob: float = None, scale_by_keep: bool = True...
    method call (line 455) | def call(self, input: tf.Tensor, training: bool = False) -> tf.Tensor:
  class TFSwinSelfAttention (line 459) | class TFSwinSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 460) | def __init__(self, config: SwinConfig, dim: int, num_heads: int, **kwa...
    method build (line 496) | def build(self, input_shape: tf.TensorShape) -> None:
    method transpose_for_scores (line 526) | def transpose_for_scores(self, x: tf.Tensor) -> tf.Tensor:
    method call (line 531) | def call(
  class TFSwinSelfOutput (line 595) | class TFSwinSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 596) | def __init__(self, config: SwinConfig, dim: int, **kwargs) -> None:
    method call (line 601) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFSwinAttention (line 607) | class TFSwinAttention(tf.keras.layers.Layer):
    method __init__ (line 608) | def __init__(self, config: SwinConfig, dim: int, num_heads: int, **kwa...
    method prune_heads (line 614) | def prune_heads(self, heads):
    method call (line 621) | def call(
  class TFSwinIntermediate (line 635) | class TFSwinIntermediate(tf.keras.layers.Layer):
    method __init__ (line 636) | def __init__(self, config: SwinConfig, dim: int, **kwargs) -> None:
    method call (line 644) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFSwinOutput (line 650) | class TFSwinOutput(tf.keras.layers.Layer):
    method __init__ (line 651) | def __init__(self, config: SwinConfig, dim: int, **kwargs) -> None:
    method call (line 656) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFSwinLayer (line 662) | class TFSwinLayer(tf.keras.layers.Layer):
    method __init__ (line 663) | def __init__(
    method get_attn_mask (line 688) | def get_attn_mask(self, height: int, width: int, window_size: int, shi...
    method maybe_pad (line 716) | def maybe_pad(
    method call (line 726) | def call(
  class TFSwinStage (line 793) | class TFSwinStage(tf.keras.layers.Layer):
    method __init__ (line 794) | def __init__(
    method call (line 833) | def call(
  class TFSwinEncoder (line 865) | class TFSwinEncoder(tf.keras.layers.Layer):
    method __init__ (line 866) | def __init__(self, config: SwinConfig, grid_size: Tuple[int, int], **k...
    method call (line 887) | def call(
  class TFSwinPreTrainedModel (line 945) | class TFSwinPreTrainedModel(TFPreTrainedModel):
    method _set_gradient_checkpointing (line 956) | def _set_gradient_checkpointing(self, module, value=False) -> None:
  function normalize_data_format (line 995) | def normalize_data_format(value: str) -> str:
  class AdaptiveAveragePooling1D (line 1010) | class AdaptiveAveragePooling1D(tf.keras.layers.Layer):
    method __init__ (line 1032) | def __init__(
    method call (line 1044) | def call(self, inputs: tf.Tensor, *args) -> None:
    method compute_output_shape (line 1056) | def compute_output_shape(self, input_shape: Iterable[int]) -> tf.Tenso...
    method get_config (line 1064) | def get_config(self) -> Dict[str, Any]:
  class TFSwinMainLayer (line 1074) | class TFSwinMainLayer(tf.keras.layers.Layer):
    method __init__ (line 1077) | def __init__(
    method get_input_embeddings (line 1091) | def get_input_embeddings(self) -> TFSwinPatchEmbeddings:
    method _prune_heads (line 1094) | def _prune_heads(self, heads_to_prune: Dict[int, List]):
    method get_head_mask (line 1102) | def get_head_mask(self, head_mask: Optional[Any]) -> List:
    method call (line 1108) | def call(
  class TFSwinModel (line 1173) | class TFSwinModel(TFSwinPreTrainedModel):
    method __init__ (line 1174) | def __init__(
    method call (line 1190) | def call(
  class TFSwinPixelShuffle (line 1226) | class TFSwinPixelShuffle(tf.keras.layers.Layer):
    method __init__ (line 1229) | def __init__(self, upscale_factor: int, **kwargs) -> None:
    method call (line 1235) | def call(self, x: tf.Tensor) -> tf.Tensor:
  class TFSwinDecoder (line 1252) | class TFSwinDecoder(tf.keras.layers.Layer):
    method __init__ (line 1253) | def __init__(self, config: SwinConfig, **kwargs):
    method call (line 1260) | def call(self, x: tf.Tensor) -> tf.Tensor:
  class TFSwinForMaskedImageModeling (line 1276) | class TFSwinForMaskedImageModeling(TFSwinPreTrainedModel):
    method __init__ (line 1277) | def __init__(self, config: SwinConfig):
    method call (line 1287) | def call(
  class TFSwinForImageClassification (line 1388) | class TFSwinForImageClassification(TFSwinPreTrainedModel, TFSequenceClas...
    method __init__ (line 1389) | def __init__(self, config: SwinConfig):
    method call (line 1410) | def call(

FILE: transformers/models/swin2sr/configuration_swin2sr.py
  class Swin2SRConfig (line 30) | class Swin2SRConfig(PretrainedConfig):
    method __init__ (line 106) | def __init__(

FILE: transformers/models/swin2sr/convert_swin2sr_original_to_pytorch.py
  function get_config (line 27) | def get_config(checkpoint_url):
  function rename_key (line 55) | def rename_key(name, config):
  function convert_state_dict (line 126) | def convert_state_dict(orig_state_dict, config):
  function convert_swin2sr_checkpoint (line 163) | def convert_swin2sr_checkpoint(checkpoint_url, pytorch_dump_folder_path,...

FILE: transformers/models/swin2sr/image_processing_swin2sr.py
  class Swin2SRImageProcessor (line 30) | class Swin2SRImageProcessor(BaseImageProcessor):
    method __init__ (line 45) | def __init__(
    method rescale (line 60) | def rescale(
    method pad (line 82) | def pad(self, image: np.ndarray, size: int, data_format: Optional[Unio...
    method preprocess (line 106) | def preprocess(

FILE: transformers/models/swin2sr/modeling_swin2sr.py
  class Swin2SREncoderOutput (line 59) | class Swin2SREncoderOutput(ModelOutput):
  function window_partition (line 85) | def window_partition(input_feature, window_size):
  function window_reverse (line 98) | def window_reverse(windows, window_size, height, width):
  function drop_path (line 109) | def drop_path(input, drop_prob=0.0, training=False, scale_by_keep=True):
  class Swin2SRDropPath (line 130) | class Swin2SRDropPath(nn.Module):
    method __init__ (line 133) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 137) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 140) | def extra_repr(self) -> str:
  class Swin2SREmbeddings (line 144) | class Swin2SREmbeddings(nn.Module):
    method __init__ (line 149) | def __init__(self, config):
    method forward (line 163) | def forward(self, pixel_values: Optional[torch.FloatTensor]) -> Tuple[...
  class Swin2SRPatchEmbeddings (line 174) | class Swin2SRPatchEmbeddings(nn.Module):
    method __init__ (line 175) | def __init__(self, config, normalize_patches=True):
    method forward (line 189) | def forward(self, embeddings: Optional[torch.FloatTensor]) -> Tuple[to...
  class Swin2SRPatchUnEmbeddings (line 201) | class Swin2SRPatchUnEmbeddings(nn.Module):
    method __init__ (line 204) | def __init__(self, config):
    method forward (line 209) | def forward(self, embeddings, x_size):
  class Swin2SRPatchMerging (line 216) | class Swin2SRPatchMerging(nn.Module):
    method __init__ (line 229) | def __init__(self, input_resolution: Tuple[int], dim: int, norm_layer:...
    method maybe_pad (line 236) | def maybe_pad(self, input_feature, height, width):
    method forward (line 244) | def forward(self, input_feature: torch.Tensor, input_dimensions: Tuple...
  class Swin2SRSelfAttention (line 271) | class Swin2SRSelfAttention(nn.Module):
    method __init__ (line 272) | def __init__(self, config, dim, num_heads, window_size, pretrained_win...
    method transpose_for_scores (line 331) | def transpose_for_scores(self, x):
    method forward (line 336) | def forward(
  class Swin2SRSelfOutput (line 399) | class Swin2SRSelfOutput(nn.Module):
    method __init__ (line 400) | def __init__(self, config, dim):
    method forward (line 405) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class Swin2SRAttention (line 413) | class Swin2SRAttention(nn.Module):
    method __init__ (line 414) | def __init__(self, config, dim, num_heads, window_size, pretrained_win...
    method prune_heads (line 428) | def prune_heads(self, heads):
    method forward (line 446) | def forward(
  class Swin2SRIntermediate (line 460) | class Swin2SRIntermediate(nn.Module):
    method __init__ (line 461) | def __init__(self, config, dim):
    method forward (line 469) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Swin2SROutput (line 476) | class Swin2SROutput(nn.Module):
    method __init__ (line 477) | def __init__(self, config, dim):
    method forward (line 482) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Swin2SRLayer (line 489) | class Swin2SRLayer(nn.Module):
    method __init__ (line 490) | def __init__(self, config, dim, input_resolution, num_heads, shift_siz...
    method set_shift_and_window_size (line 512) | def set_shift_and_window_size(self, input_resolution):
    method get_attn_mask (line 536) | def get_attn_mask(self, height, width, dtype):
    method maybe_pad (line 564) | def maybe_pad(self, hidden_states, height, width):
    method forward (line 571) | def forward(
  class Swin2SRStage (line 635) | class Swin2SRStage(nn.Module):
    method __init__ (line 640) | def __init__(self, config, dim, input_resolution, depth, num_heads, dr...
    method forward (line 674) | def forward(
  class Swin2SREncoder (line 706) | class Swin2SREncoder(nn.Module):
    method __init__ (line 707) | def __init__(self, config, grid_size):
    method forward (line 729) | def forward(
  class Swin2SRPreTrainedModel (line 784) | class Swin2SRPreTrainedModel(PreTrainedModel):
    method _init_weights (line 795) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 805) | def _set_gradient_checkpointing(self, module, value=False):
  class Swin2SRModel (line 847) | class Swin2SRModel(Swin2SRPreTrainedModel):
    method __init__ (line 848) | def __init__(self, config):
    method get_input_embeddings (line 870) | def get_input_embeddings(self):
    method _prune_heads (line 873) | def _prune_heads(self, heads_to_prune):
    method pad_and_normalize (line 881) | def pad_and_normalize(self, pixel_values):
    method forward (line 904) | def forward(
  class Upsample (line 960) | class Upsample(nn.Module):
    method __init__ (line 970) | def __init__(self, scale, num_features):
    method forward (line 985) | def forward(self, hidden_state):
  class UpsampleOneStep (line 998) | class UpsampleOneStep(nn.Module):
    method __init__ (line 1010) | def __init__(self, scale, in_channels, out_channels):
    method forward (line 1016) | def forward(self, x):
  class PixelShuffleUpsampler (line 1023) | class PixelShuffleUpsampler(nn.Module):
    method __init__ (line 1024) | def __init__(self, config, num_features):
    method forward (line 1031) | def forward(self, sequence_output):
  class NearestConvUpsampler (line 1040) | class NearestConvUpsampler(nn.Module):
    method __init__ (line 1041) | def __init__(self, config, num_features):
    method forward (line 1054) | def forward(self, sequence_output):
  class PixelShuffleAuxUpsampler (line 1067) | class PixelShuffleAuxUpsampler(nn.Module):
    method __init__ (line 1068) | def __init__(self, config, num_features):
    method forward (line 1080) | def forward(self, sequence_output, bicubic, height, width):
  class Swin2SRForImageSuperResolution (line 1101) | class Swin2SRForImageSuperResolution(Swin2SRPreTrainedModel):
    method __init__ (line 1102) | def __init__(self, config):
    method forward (line 1130) | def forward(

FILE: transformers/models/swinv2/configuration_swinv2.py
  class Swinv2Config (line 30) | class Swinv2Config(PretrainedConfig):
    method __init__ (line 99) | def __init__(

FILE: transformers/models/swinv2/convert_swinv2_timm_to_pytorch.py
  function get_swinv2_config (line 30) | def get_swinv2_config(swinv2_name):
  function rename_key (line 92) | def rename_key(name):
  function convert_state_dict (line 132) | def convert_state_dict(orig_state_dict, model):
  function convert_swinv2_checkpoint (line 170) | def convert_swinv2_checkpoint(swinv2_name, pytorch_dump_folder_path):

FILE: transformers/models/swinv2/modeling_swinv2.py
  class Swinv2EncoderOutput (line 68) | class Swinv2EncoderOutput(ModelOutput):
  class Swinv2ModelOutput (line 102) | class Swinv2ModelOutput(ModelOutput):
  class Swinv2MaskedImageModelingOutput (line 139) | class Swinv2MaskedImageModelingOutput(ModelOutput):
    method logits (line 174) | def logits(self):
  class Swinv2ImageClassifierOutput (line 185) | class Swinv2ImageClassifierOutput(ModelOutput):
  function window_partition (line 221) | def window_partition(input_feature, window_size):
  function window_reverse (line 234) | def window_reverse(windows, window_size, height, width):
  function drop_path (line 245) | def drop_path(input, drop_prob=0.0, training=False, scale_by_keep=True):
  class Swinv2DropPath (line 266) | class Swinv2DropPath(nn.Module):
    method __init__ (line 269) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 273) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 276) | def extra_repr(self) -> str:
  class Swinv2Embeddings (line 281) | class Swinv2Embeddings(nn.Module):
    method __init__ (line 286) | def __init__(self, config, use_mask_token=False):
    method forward (line 302) | def forward(
  class Swinv2PatchEmbeddings (line 324) | class Swinv2PatchEmbeddings(nn.Module):
    method __init__ (line 331) | def __init__(self, config):
    method maybe_pad (line 346) | def maybe_pad(self, pixel_values, height, width):
    method forward (line 355) | def forward(self, pixel_values: Optional[torch.FloatTensor]) -> Tuple[...
  class Swinv2PatchMerging (line 371) | class Swinv2PatchMerging(nn.Module):
    method __init__ (line 384) | def __init__(self, input_resolution: Tuple[int], dim: int, norm_layer:...
    method maybe_pad (line 391) | def maybe_pad(self, input_feature, height, width):
    method forward (line 399) | def forward(self, input_feature: torch.Tensor, input_dimensions: Tuple...
  class Swinv2SelfAttention (line 425) | class Swinv2SelfAttention(nn.Module):
    method __init__ (line 426) | def __init__(self, config, dim, num_heads, window_size, pretrained_win...
    method transpose_for_scores (line 485) | def transpose_for_scores(self, x):
    method forward (line 490) | def forward(
  class Swinv2SelfOutput (line 553) | class Swinv2SelfOutput(nn.Module):
    method __init__ (line 554) | def __init__(self, config, dim):
    method forward (line 559) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class Swinv2Attention (line 566) | class Swinv2Attention(nn.Module):
    method __init__ (line 567) | def __init__(self, config, dim, num_heads, window_size, pretrained_win...
    method prune_heads (line 581) | def prune_heads(self, heads):
    method forward (line 599) | def forward(
  class Swinv2Intermediate (line 613) | class Swinv2Intermediate(nn.Module):
    method __init__ (line 614) | def __init__(self, config, dim):
    method forward (line 622) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Swinv2Output (line 629) | class Swinv2Output(nn.Module):
    method __init__ (line 630) | def __init__(self, config, dim):
    method forward (line 635) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class Swinv2Layer (line 641) | class Swinv2Layer(nn.Module):
    method __init__ (line 642) | def __init__(self, config, dim, input_resolution, num_heads, shift_siz...
    method set_shift_and_window_size (line 664) | def set_shift_and_window_size(self, input_resolution):
    method get_attn_mask (line 688) | def get_attn_mask(self, height, width, dtype):
    method maybe_pad (line 716) | def maybe_pad(self, hidden_states, height, width):
    method forward (line 723) | def forward(
  class Swinv2Stage (line 787) | class Swinv2Stage(nn.Module):
    method __init__ (line 788) | def __init__(
    method forward (line 817) | def forward(
  class Swinv2Encoder (line 850) | class Swinv2Encoder(nn.Module):
    method __init__ (line 851) | def __init__(self, config, grid_size, pretrained_window_sizes=(0, 0, 0...
    method forward (line 877) | def forward(
  class Swinv2PreTrainedModel (line 958) | class Swinv2PreTrainedModel(PreTrainedModel):
    method _init_weights (line 969) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 981) | def _set_gradient_checkpointing(self, module, value=False):
  class Swinv2Model (line 1024) | class Swinv2Model(Swinv2PreTrainedModel):
    method __init__ (line 1025) | def __init__(self, config, add_pooling_layer=True, use_mask_token=False):
    method get_input_embeddings (line 1040) | def get_input_embeddings(self):
    method _prune_heads (line 1043) | def _prune_heads(self, heads_to_prune):
    method forward (line 1059) | def forward(
  class Swinv2ForMaskedImageModeling (line 1135) | class Swinv2ForMaskedImageModeling(Swinv2PreTrainedModel):
    method __init__ (line 1136) | def __init__(self, config):
    method forward (line 1154) | def forward(
  class Swinv2ForImageClassification (line 1247) | class Swinv2ForImageClassification(Swinv2PreTrainedModel):
    method __init__ (line 1248) | def __init__(self, config):
    method forward (line 1269) | def forward(

FILE: transformers/models/switch_transformers/configuration_switch_transformers.py
  class SwitchTransformersConfig (line 27) | class SwitchTransformersConfig(PretrainedConfig):
    method __init__ (line 101) | def __init__(

FILE: transformers/models/switch_transformers/convert_big_switch.py
  function rename_base_flax_keys (line 19) | def rename_base_flax_keys(flax_key_tuple, flax_tensor):
  function get_key_and_tensorstore_dict (line 37) | def get_key_and_tensorstore_dict(layer, checkpoint_info, switch_checkpoi...
  function rename_and_save_block (line 62) | def rename_and_save_block(current_block, save_path):
  function shard_on_the_fly (line 71) | def shard_on_the_fly(switch_checkpoint_path, dump_path, max_shard_size, ...
  function sanity_check (line 179) | def sanity_check():

FILE: transformers/models/switch_transformers/convert_switch_transformers_original_flax_checkpoint_to_pytorch.py
  function rename_keys (line 54) | def rename_keys(s_dict):
  function convert_gin_to_config (line 121) | def convert_gin_to_config(gin_file, num_experts):
  function convert_flax_checkpoint_to_pytorch (line 142) | def convert_flax_checkpoint_to_pytorch(

FILE: transformers/models/switch_transformers/modeling_switch_transformers.py
  function router_z_loss_func (line 72) | def router_z_loss_func(router_logits: torch.Tensor) -> float:
  function load_balancing_loss_func (line 92) | def load_balancing_loss_func(router_probs: torch.Tensor, expert_indices:...
  class SwitchTransformersTop1Router (line 131) | class SwitchTransformersTop1Router(nn.Module):
    method __init__ (line 142) | def __init__(self, config: SwitchTransformersConfig):
    method _compute_router_probabilities (line 151) | def _compute_router_probabilities(self, hidden_states: torch.Tensor) -...
    method _cast_classifier (line 193) | def _cast_classifier(self):
    method forward (line 201) | def forward(self, hidden_states: torch.Tensor) -> Tuple:
  class SwitchTransformersLayerNorm (line 234) | class SwitchTransformersLayerNorm(nn.Module):
    method __init__ (line 235) | def __init__(self, hidden_size, eps=1e-6):
    method forward (line 243) | def forward(self, hidden_states):
  class SwitchTransformersDenseActDense (line 263) | class SwitchTransformersDenseActDense(nn.Module):
    method __init__ (line 264) | def __init__(self, config: SwitchTransformersConfig):
    method forward (line 271) | def forward(self, hidden_states):
  class SwitchTransformersDenseGatedActDense (line 286) | class SwitchTransformersDenseGatedActDense(nn.Module):
    method __init__ (line 287) | def __init__(self, config: SwitchTransformersConfig):
    method forward (line 295) | def forward(self, hidden_states):
  class SwitchTransformersSparseMLP (line 304) | class SwitchTransformersSparseMLP(nn.Module):
    method __init__ (line 309) | def __init__(self, config: SwitchTransformersConfig, expert_class: nn....
    method forward (line 319) | def forward(self, hidden_states):
  class SwitchTransformersLayerFF (line 347) | class SwitchTransformersLayerFF(nn.Module):
    method __init__ (line 359) | def __init__(self, config: SwitchTransformersConfig, is_sparse=False):
    method forward (line 372) | def forward(self, hidden_states, output_router_logits):
  class SwitchTransformersAttention (line 390) | class SwitchTransformersAttention(nn.Module):
    method __init__ (line 391) | def __init__(self, config: SwitchTransformersConfig, has_relative_atte...
    method prune_heads (line 414) | def prune_heads(self, heads):
    method _relative_position_bucket (line 431) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 478) | def compute_bias(self, query_length, key_length, device=None):
    method forward (line 495) | def forward(
  class SwitchTransformersLayerSelfAttention (line 626) | class SwitchTransformersLayerSelfAttention(nn.Module):
    method __init__ (line 627) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 635) | def forward(
  class SwitchTransformersLayerCrossAttention (line 661) | class SwitchTransformersLayerCrossAttention(nn.Module):
    method __init__ (line 662) | def __init__(self, config):
    method forward (line 668) | def forward(
  class SwitchTransformersBlock (line 697) | class SwitchTransformersBlock(nn.Module):
    method __init__ (line 698) | def __init__(self, config, has_relative_attention_bias=False, is_spars...
    method forward (line 711) | def forward(
  class SwitchTransformersPreTrainedModel (line 818) | class SwitchTransformersPreTrainedModel(PreTrainedModel):
    method dummy_inputs (line 830) | def dummy_inputs(self):
    method _init_weights (line 840) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 897) | def _set_gradient_checkpointing(self, module, value=False):
    method _shift_right (line 901) | def _shift_right(self, input_ids):
  class SwitchTransformersStack (line 929) | class SwitchTransformersStack(SwitchTransformersPreTrainedModel):
    method __init__ (line 930) | def __init__(self, config, embed_tokens=None):
    method get_input_embeddings (line 959) | def get_input_embeddings(self):
    method set_input_embeddings (line 962) | def set_input_embeddings(self, new_embeddings):
    method forward (line 965) | def forward(
  class SwitchTransformersModel (line 1339) | class SwitchTransformersModel(SwitchTransformersPreTrainedModel):
    method __init__ (line 1342) | def __init__(self, config: SwitchTransformersConfig):
    method get_input_embeddings (line 1363) | def get_input_embeddings(self):
    method set_input_embeddings (line 1366) | def set_input_embeddings(self, new_embeddings):
    method get_encoder (line 1371) | def get_encoder(self):
    method get_decoder (line 1374) | def get_decoder(self):
    method _prune_heads (line 1377) | def _prune_heads(self, heads_to_prune):
    method forward (line 1387) | def forward(
  class SwitchTransformersForConditionalGeneration (line 1507) | class SwitchTransformersForConditionalGeneration(SwitchTransformersPreTr...
    method __init__ (line 1514) | def __init__(self, config: SwitchTransformersConfig):
    method get_input_embeddings (line 1543) | def get_input_embeddings(self):
    method set_input_embeddings (line 1546) | def set_input_embeddings(self, new_embeddings):
    method set_output_embeddings (line 1551) | def set_output_embeddings(self, new_embeddings):
    method get_output_embeddings (line 1554) | def get_output_embeddings(self):
    method get_encoder (line 1557) | def get_encoder(self):
    method get_decoder (line 1560) | def get_decoder(self):
    method forward (line 1565) | def forward(
    method _unpack_router_logits (line 1746) | def _unpack_router_logits(self, router_outputs):
    method prepare_inputs_for_generation (line 1756) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1783) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 1786) | def _reorder_cache(self, past_key_values, beam_idx):
  class SwitchTransformersEncoderModel (line 1824) | class SwitchTransformersEncoderModel(SwitchTransformersPreTrainedModel):
    method __init__ (line 1827) | def __init__(self, config: SwitchTransformersConfig):
    method get_input_embeddings (line 1842) | def get_input_embeddings(self):
    method set_input_embeddings (line 1845) | def set_input_embeddings(self, new_embeddings):
    method get_encoder (line 1849) | def get_encoder(self):
    method _prune_heads (line 1852) | def _prune_heads(self, heads_to_prune):
    method forward (line 1862) | def forward(

FILE: transformers/models/t5/configuration_t5.py
  class T5Config (line 34) | class T5Config(PretrainedConfig):
    method __init__ (line 82) | def __init__(
  class T5OnnxConfig (line 143) | class T5OnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 145) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method default_onnx_opset (line 164) | def default_onnx_opset(self) -> int:

FILE: transformers/models/t5/convert_t5_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 27) | def convert_tf_checkpoint_to_pytorch(tf_checkpoint_path, config_file, py...

FILE: transformers/models/t5/convert_t5x_checkpoint_to_flax.py
  function convert_t5x_checkpoint_to_flax (line 25) | def convert_t5x_checkpoint_to_flax(t5x_checkpoint_path, config_name, fla...

FILE: transformers/models/t5/convert_t5x_checkpoint_to_pytorch.py
  function t5x_attention_lookup (line 45) | def t5x_attention_lookup(params, i, prefix, layer_name="attention"):
  function t5x_mlp_lookup (line 54) | def t5x_mlp_lookup(params, i, prefix, split_mlp_wi=False):
  function t5x_layer_norm_lookup (line 67) | def t5x_layer_norm_lookup(params, i, prefix, layer_name):
  function convert_t5x_to_pytorch (line 72) | def convert_t5x_to_pytorch(variables: dict, *, num_layers: int, is_encod...
  function make_state_dict (line 157) | def make_state_dict(converted_params, is_encoder_only: bool):
  function load_t5x_weights_in_t5 (line 177) | def load_t5x_weights_in_t5(model, config, t5x_checkpoint_path, is_encode...
  function convert_t5x_checkpoint_to_pytorch (line 185) | def convert_t5x_checkpoint_to_pytorch(

FILE: transformers/models/t5/modeling_flax_t5.py
  function shift_tokens_right (line 59) | def shift_tokens_right(input_ids: np.array, pad_token_id: int, decoder_s...
  class FlaxT5LayerNorm (line 71) | class FlaxT5LayerNorm(nn.Module):
    method setup (line 77) | def setup(self):
    method __call__ (line 80) | def __call__(self, hidden_states):
  class FlaxT5DenseActDense (line 91) | class FlaxT5DenseActDense(nn.Module):
    method setup (line 95) | def setup(self):
    method __call__ (line 114) | def __call__(self, hidden_states, deterministic=True):
  class FlaxT5DenseGatedActDense (line 122) | class FlaxT5DenseGatedActDense(nn.Module):
    method setup (line 126) | def setup(self):
    method __call__ (line 151) | def __call__(self, hidden_states, deterministic):
  class FlaxT5LayerFF (line 160) | class FlaxT5LayerFF(nn.Module):
    method setup (line 164) | def setup(self):
    method __call__ (line 173) | def __call__(self, hidden_states, deterministic=True):
  class FlaxT5Attention (line 180) | class FlaxT5Attention(nn.Module):
    method setup (line 186) | def setup(self):
    method _relative_position_bucket (line 233) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 268) | def compute_bias(self, query_length, key_length):
    method _split_heads (line 285) | def _split_heads(self, hidden_states):
    method _merge_heads (line 288) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 292) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method _create_position_bias (line 324) | def _create_position_bias(
    method __call__ (line 348) | def __call__(
  class FlaxT5LayerSelfAttention (line 464) | class FlaxT5LayerSelfAttention(nn.Module):
    method setup (line 469) | def setup(self):
    method __call__ (line 479) | def __call__(
  class FlaxT5LayerCrossAttention (line 502) | class FlaxT5LayerCrossAttention(nn.Module):
    method setup (line 506) | def setup(self):
    method __call__ (line 513) | def __call__(
  class FlaxT5Block (line 535) | class FlaxT5Block(nn.Module):
    method setup (line 540) | def setup(self):
    method __call__ (line 557) | def __call__(
  class FlaxT5LayerCollection (line 608) | class FlaxT5LayerCollection(nn.Module):
    method setup (line 613) | def setup(self):
    method __call__ (line 618) | def __call__(
  class FlaxT5BlockCollection (line 643) | class FlaxT5BlockCollection(nn.Module):
    method setup (line 648) | def setup(self):
    method __call__ (line 672) | def __call__(
  class FlaxT5Stack (line 729) | class FlaxT5Stack(nn.Module):
    method setup (line 735) | def setup(self):
    method __call__ (line 746) | def __call__(
  class FlaxT5PreTrainedModel (line 927) | class FlaxT5PreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 937) | def __init__(
    method enable_gradient_checkpointing (line 950) | def enable_gradient_checkpointing(self):
    method init_weights (line 957) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 987) | def __call__(
    method init_cache (line 1036) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 1074) | def encode(
    method decode (line 1132) | def decode(
  class FlaxT5Module (line 1278) | class FlaxT5Module(nn.Module):
    method _get_encoder_module (line 1283) | def _get_encoder_module(self):
    method _get_decoder_module (line 1286) | def _get_decoder_module(self):
    method setup (line 1289) | def setup(self):
    method __call__ (line 1316) | def __call__(
  class FlaxT5Model (line 1367) | class FlaxT5Model(FlaxT5PreTrainedModel):
  class FlaxT5EncoderModule (line 1408) | class FlaxT5EncoderModule(nn.Module):
    method setup (line 1413) | def setup(self):
    method __call__ (line 1432) | def __call__(
  class FlaxT5EncoderModel (line 1454) | class FlaxT5EncoderModel(FlaxT5PreTrainedModel):
    method __call__ (line 1458) | def __call__(
  class FlaxT5ForConditionalGenerationModule (line 1495) | class FlaxT5ForConditionalGenerationModule(nn.Module):
    method _get_encoder_module (line 1500) | def _get_encoder_module(self):
    method _get_decoder_module (line 1503) | def _get_decoder_module(self):
    method setup (line 1506) | def setup(self):
    method __call__ (line 1539) | def __call__(
  class FlaxT5ForConditionalGeneration (line 1605) | class FlaxT5ForConditionalGeneration(FlaxT5PreTrainedModel):
    method decode (line 1610) | def decode(
    method prepare_inputs_for_generation (line 1739) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 1768) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/t5/modeling_t5.py
  function load_tf_weights_in_t5 (line 74) | def load_tf_weights_in_t5(model, config, tf_checkpoint_path):
  class T5LayerNorm (line 237) | class T5LayerNorm(nn.Module):
    method __init__ (line 238) | def __init__(self, hidden_size, eps=1e-6):
    method forward (line 246) | def forward(self, hidden_states):
  class T5DenseActDense (line 278) | class T5DenseActDense(nn.Module):
    method __init__ (line 279) | def __init__(self, config: T5Config):
    method forward (line 286) | def forward(self, hidden_states):
  class T5DenseGatedActDense (line 300) | class T5DenseGatedActDense(nn.Module):
    method __init__ (line 301) | def __init__(self, config: T5Config):
    method forward (line 309) | def forward(self, hidden_states):
  class T5LayerFF (line 329) | class T5LayerFF(nn.Module):
    method __init__ (line 330) | def __init__(self, config: T5Config):
    method forward (line 340) | def forward(self, hidden_states):
  class T5Attention (line 347) | class T5Attention(nn.Module):
    method __init__ (line 348) | def __init__(self, config: T5Config, has_relative_attention_bias=False):
    method prune_heads (line 371) | def prune_heads(self, heads):
    method _relative_position_bucket (line 388) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 435) | def compute_bias(self, query_length, key_length, device=None):
    method forward (line 452) | def forward(
  class T5LayerSelfAttention (line 582) | class T5LayerSelfAttention(nn.Module):
    method __init__ (line 583) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 589) | def forward(
  class T5LayerCrossAttention (line 614) | class T5LayerCrossAttention(nn.Module):
    method __init__ (line 615) | def __init__(self, config):
    method forward (line 621) | def forward(
  class T5Block (line 650) | class T5Block(nn.Module):
    method __init__ (line 651) | def __init__(self, config, has_relative_attention_bias=False):
    method forward (line 661) | def forward(
  class T5PreTrainedModel (line 774) | class T5PreTrainedModel(PreTrainedModel):
    method dummy_inputs (line 789) | def dummy_inputs(self):
    method _init_weights (line 799) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 843) | def _set_gradient_checkpointing(self, module, value=False):
    method _shift_right (line 847) | def _shift_right(self, input_ids):
  class T5Stack (line 875) | class T5Stack(T5PreTrainedModel):
    method __init__ (line 876) | def __init__(self, config, embed_tokens=None):
    method parallelize (line 896) | def parallelize(self, device_map=None):
    method deparallelize (line 924) | def deparallelize(self):
    method get_input_embeddings (line 939) | def get_input_embeddings(self):
    method set_input_embeddings (line 942) | def set_input_embeddings(self, new_embeddings):
    method forward (line 945) | def forward(
  class T5Model (line 1324) | class T5Model(T5PreTrainedModel):
    method __init__ (line 1333) | def __init__(self, config: T5Config):
    method parallelize (line 1357) | def parallelize(self, device_map=None):
    method deparallelize (line 1376) | def deparallelize(self):
    method get_input_embeddings (line 1389) | def get_input_embeddings(self):
    method set_input_embeddings (line 1392) | def set_input_embeddings(self, new_embeddings):
    method get_encoder (line 1397) | def get_encoder(self):
    method get_decoder (line 1400) | def get_decoder(self):
    method _prune_heads (line 1403) | def _prune_heads(self, heads_to_prune):
    method forward (line 1413) | def forward(
  class T5ForConditionalGeneration (line 1527) | class T5ForConditionalGeneration(T5PreTrainedModel):
    method __init__ (line 1537) | def __init__(self, config: T5Config):
    method parallelize (line 1565) | def parallelize(self, device_map=None):
    method deparallelize (line 1585) | def deparallelize(self):
    method get_input_embeddings (line 1599) | def get_input_embeddings(self):
    method set_input_embeddings (line 1602) | def set_input_embeddings(self, new_embeddings):
    method set_output_embeddings (line 1607) | def set_output_embeddings(self, new_embeddings):
    method get_output_embeddings (line 1610) | def get_output_embeddings(self):
    method get_encoder (line 1613) | def get_encoder(self):
    method get_decoder (line 1616) | def get_decoder(self):
    method forward (line 1621) | def forward(
    method prepare_inputs_for_generation (line 1774) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1803) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 1806) | def _reorder_cache(self, past_key_values, beam_idx):
  class T5EncoderModel (line 1841) | class T5EncoderModel(T5PreTrainedModel):
    method __init__ (line 1844) | def __init__(self, config: T5Config):
    method parallelize (line 1861) | def parallelize(self, device_map=None):
    method deparallelize (line 1879) | def deparallelize(self):
    method get_input_embeddings (line 1890) | def get_input_embeddings(self):
    method set_input_embeddings (line 1893) | def set_input_embeddings(self, new_embeddings):
    method get_encoder (line 1897) | def get_encoder(self):
    method _prune_heads (line 1900) | def _prune_heads(self, heads_to_prune):
    method forward (line 1910) | def forward(

FILE: transformers/models/t5/modeling_tf_t5.py
  class TFT5LayerNorm (line 77) | class TFT5LayerNorm(tf.keras.layers.Layer):
    method __init__ (line 78) | def __init__(self, epsilon=1e-6, **kwargs):
    method build (line 85) | def build(self, input_shape):
    method call (line 90) | def call(self, hidden_states):
  class TFT5DenseActDense (line 96) | class TFT5DenseActDense(tf.keras.layers.Layer):
    method __init__ (line 97) | def __init__(self, config, **kwargs):
    method call (line 114) | def call(self, hidden_states, training=False):
  class TFT5DenseGatedActDense (line 122) | class TFT5DenseGatedActDense(tf.keras.layers.Layer):
    method __init__ (line 123) | def __init__(self, config, **kwargs):
    method call (line 143) | def call(self, hidden_states, training=False):
  class TFT5LayerFF (line 152) | class TFT5LayerFF(tf.keras.layers.Layer):
    method __init__ (line 153) | def __init__(self, config, **kwargs):
    method call (line 163) | def call(self, hidden_states, training=False):
  class TFT5Attention (line 170) | class TFT5Attention(tf.keras.layers.Layer):
    method __init__ (line 173) | def __init__(self, config, has_relative_attention_bias=False, **kwargs):
    method build (line 221) | def build(self, input_shape):
    method prune_heads (line 232) | def prune_heads(self, heads):
    method _relative_position_bucket (line 236) | def _relative_position_bucket(relative_position, bidirectional=True, n...
    method compute_bias (line 280) | def compute_bias(self, query_length, key_length):
    method call (line 299) | def call(
  class TFT5LayerSelfAttention (line 434) | class TFT5LayerSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 435) | def __init__(self, config, has_relative_attention_bias=False, **kwargs):
    method call (line 445) | def call(
  class TFT5LayerCrossAttention (line 472) | class TFT5LayerCrossAttention(tf.keras.layers.Layer):
    method __init__ (line 473) | def __init__(self, config, **kwargs):
    method call (line 483) | def call(
  class TFT5Block (line 514) | class TFT5Block(tf.keras.layers.Layer):
    method __init__ (line 515) | def __init__(self, config, has_relative_attention_bias=False, **kwargs):
    method call (line 536) | def call(
  class TFT5MainLayer (line 622) | class TFT5MainLayer(tf.keras.layers.Layer):
    method __init__ (line 625) | def __init__(self, config, embed_tokens=None, **kwargs):
    method _prune_heads (line 646) | def _prune_heads(self, heads_to_prune):
    method call (line 650) | def call(
  class TFT5PreTrainedModel (line 857) | class TFT5PreTrainedModel(TFPreTrainedModel):
    method get_input_embeddings (line 868) | def get_input_embeddings(self):
    method set_input_embeddings (line 871) | def set_input_embeddings(self, value):
    method _shift_right (line 877) | def _shift_right(self, input_ids):
  class TFT5Model (line 1093) | class TFT5Model(TFT5PreTrainedModel):
    method __init__ (line 1094) | def __init__(self, config, *inputs, **kwargs):
    method get_encoder (line 1115) | def get_encoder(self):
    method get_decoder (line 1118) | def get_decoder(self):
    method call (line 1124) | def call(
  class TFT5ForConditionalGeneration (line 1226) | class TFT5ForConditionalGeneration(TFT5PreTrainedModel, TFCausalLanguage...
    method __init__ (line 1227) | def __init__(self, config, *inputs, **kwargs):
    method get_output_embeddings (line 1254) | def get_output_embeddings(self):
    method set_output_embeddings (line 1262) | def set_output_embeddings(self, value):
    method get_encoder (line 1275) | def get_encoder(self):
    method get_decoder (line 1278) | def get_decoder(self):
    method call (line 1284) | def call(
    method serving_output (line 1424) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 1443) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 1471) | def prepare_decoder_input_ids_from_labels(self, labels: tf.Tensor):
  class TFT5EncoderModel (line 1479) | class TFT5EncoderModel(TFT5PreTrainedModel):
    method __init__ (line 1480) | def __init__(self, config, *inputs, **kwargs):
    method get_encoder (line 1495) | def get_encoder(self):
    method call (line 1501) | def call(

FILE: transformers/models/t5/tokenization_t5.py
  class T5Tokenizer (line 55) | class T5Tokenizer(PreTrainedTokenizer):
    method __init__ (line 114) | def __init__(
    method _eventually_correct_t5_max_length (line 157) | def _eventually_correct_t5_max_length(pretrained_model_name_or_path, m...
    method vocab_size (line 179) | def vocab_size(self):
    method get_vocab (line 182) | def get_vocab(self):
    method get_special_tokens_mask (line 187) | def get_special_tokens_mask(
    method get_sentinel_tokens (line 215) | def get_sentinel_tokens(self):
    method get_sentinel_token_ids (line 220) | def get_sentinel_token_ids(self):
    method _add_eos_if_not_present (line 223) | def _add_eos_if_not_present(self, token_ids: List[int]) -> List[int]:
    method create_token_type_ids_from_sequences (line 234) | def create_token_type_ids_from_sequences(
    method build_inputs_with_special_tokens (line 256) | def build_inputs_with_special_tokens(
    method __getstate__ (line 282) | def __getstate__(self):
    method __setstate__ (line 287) | def __setstate__(self, d):
    method _tokenize (line 297) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 301) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 309) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 317) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 336) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/t5/tokenization_t5_fast.py
  class T5TokenizerFast (line 66) | class T5TokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 109) | def __init__(
    method _eventually_correct_t5_max_length (line 149) | def _eventually_correct_t5_max_length(pretrained_model_name_or_path, m...
    method save_vocabulary (line 170) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 190) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 216) | def create_token_type_ids_from_sequences(
    method get_sentinel_tokens (line 238) | def get_sentinel_tokens(self):
    method get_sentinel_token_ids (line 243) | def get_sentinel_token_ids(self):

FILE: transformers/models/table_transformer/configuration_table_transformer.py
  class TableTransformerConfig (line 37) | class TableTransformerConfig(PretrainedConfig):
    method __init__ (line 144) | def __init__(
    method num_attention_heads (line 233) | def num_attention_heads(self) -> int:
    method hidden_size (line 237) | def hidden_size(self) -> int:
  class TableTransformerOnnxConfig (line 242) | class TableTransformerOnnxConfig(OnnxConfig):
    method inputs (line 246) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 255) | def atol_for_validation(self) -> float:
    method default_onnx_opset (line 259) | def default_onnx_opset(self) -> int:

FILE: transformers/models/table_transformer/convert_table_transformer_original_pytorch_checkpoint_to_pytorch.py
  function rename_key (line 115) | def rename_key(state_dict, old, new):
  function rename_backbone_keys (line 120) | def rename_backbone_keys(state_dict):
  function read_in_q_k_v (line 132) | def read_in_q_k_v(state_dict):
  function resize (line 173) | def resize(image, checkpoint_url):
  function normalize (line 183) | def normalize(image):
  function convert_table_transformer_checkpoint (line 190) | def convert_table_transformer_checkpoint(checkpoint_url, pytorch_dump_fo...

FILE: transformers/models/table_transformer/modeling_table_transformer.py
  class TableTransformerDecoderOutput (line 66) | class TableTransformerDecoderOutput(BaseModelOutputWithCrossAttentions):
  class TableTransformerModelOutput (line 98) | class TableTransformerModelOutput(Seq2SeqModelOutput):
  class TableTransformerObjectDetectionOutput (line 140) | class TableTransformerObjectDetectionOutput(ModelOutput):
  class TableTransformerFrozenBatchNorm2d (line 203) | class TableTransformerFrozenBatchNorm2d(nn.Module):
    method __init__ (line 211) | def __init__(self, n):
    method _load_from_state_dict (line 218) | def _load_from_state_dict(
    method forward (line 229) | def forward(self, x):
  function replace_batch_norm (line 243) | def replace_batch_norm(m, name=""):
  class TableTransformerConvEncoder (line 259) | class TableTransformerConvEncoder(nn.Module):
    method __init__ (line 267) | def __init__(self, config):
    method forward (line 306) | def forward(self, pixel_values: torch.Tensor, pixel_mask: torch.Tensor):
  class TableTransformerConvModel (line 319) | class TableTransformerConvModel(nn.Module):
    method __init__ (line 324) | def __init__(self, conv_encoder, position_embedding):
    method forward (line 329) | def forward(self, pixel_values, pixel_mask):
  function _expand_mask (line 340) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, target_len: Opt...
  class TableTransformerSinePositionEmbedding (line 355) | class TableTransformerSinePositionEmbedding(nn.Module):
    method __init__ (line 361) | def __init__(self, embedding_dim=64, temperature=10000, normalize=Fals...
    method forward (line 372) | def forward(self, pixel_values, pixel_mask):
  class TableTransformerLearnedPositionEmbedding (line 393) | class TableTransformerLearnedPositionEmbedding(nn.Module):
    method __init__ (line 398) | def __init__(self, embedding_dim=256):
    method forward (line 403) | def forward(self, pixel_values, pixel_mask=None):
  function build_position_encoding (line 417) | def build_position_encoding(config):
  class TableTransformerAttention (line 431) | class TableTransformerAttention(nn.Module):
    method __init__ (line 438) | def __init__(
    method _shape (line 463) | def _shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method with_pos_embed (line 466) | def with_pos_embed(self, tensor: torch.Tensor, position_embeddings: Op...
    method forward (line 469) | def forward(
  class TableTransformerEncoderLayer (line 562) | class TableTransformerEncoderLayer(nn.Module):
    method __init__ (line 564) | def __init__(self, config: TableTransformerConfig):
    method forward (line 580) | def forward(
  class TableTransformerDecoderLayer (line 635) | class TableTransformerDecoderLayer(nn.Module):
    method __init__ (line 637) | def __init__(self, config: TableTransformerConfig):
    method forward (line 663) | def forward(
  class TableTransformerClassificationHead (line 745) | class TableTransformerClassificationHead(nn.Module):
    method __init__ (line 748) | def __init__(self, input_dim: int, inner_dim: int, num_classes: int, p...
    method forward (line 754) | def forward(self, hidden_states: torch.Tensor):
  class TableTransformerPreTrainedModel (line 763) | class TableTransformerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 768) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 785) | def _set_gradient_checkpointing(self, module, value=False):
  class TableTransformerEncoder (line 844) | class TableTransformerEncoder(TableTransformerPreTrainedModel):
    method __init__ (line 859) | def __init__(self, config: TableTransformerConfig):
    method forward (line 872) | def forward(
  class TableTransformerDecoder (line 956) | class TableTransformerDecoder(TableTransformerPreTrainedModel):
    method __init__ (line 971) | def __init__(self, config: TableTransformerConfig):
    method forward (line 984) | def forward(
  class TableTransformerModel (line 1144) | class TableTransformerModel(TableTransformerPreTrainedModel):
    method __init__ (line 1146) | def __init__(self, config: TableTransformerConfig):
    method get_encoder (line 1165) | def get_encoder(self):
    method get_decoder (line 1168) | def get_decoder(self):
    method freeze_backbone (line 1171) | def freeze_backbone(self):
    method unfreeze_backbone (line 1175) | def unfreeze_backbone(self):
    method forward (line 1181) | def forward(
  class TableTransformerForObjectDetection (line 1313) | class TableTransformerForObjectDetection(TableTransformerPreTrainedModel):
    method __init__ (line 1315) | def __init__(self, config: TableTransformerConfig):
    method _set_aux_loss (line 1334) | def _set_aux_loss(self, outputs_class, outputs_coord):
    method forward (line 1342) | def forward(
  function dice_loss (line 1477) | def dice_loss(inputs, targets, num_boxes):
  function sigmoid_focal_loss (line 1497) | def sigmoid_focal_loss(inputs, targets, num_boxes, alpha: float = 0.25, ...
  class TableTransformerLoss (line 1529) | class TableTransformerLoss(nn.Module):
    method __init__ (line 1554) | def __init__(self, matcher, num_classes, eos_coef, losses):
    method loss_labels (line 1565) | def loss_labels(self, outputs, targets, indices, num_boxes):
    method loss_cardinality (line 1587) | def loss_cardinality(self, outputs, targets, indices, num_boxes):
    method loss_boxes (line 1602) | def loss_boxes(self, outputs, targets, indices, num_boxes):
    method loss_masks (line 1626) | def loss_masks(self, outputs, targets, indices, num_boxes):
    method _get_source_permutation_idx (line 1659) | def _get_source_permutation_idx(self, indices):
    method _get_target_permutation_idx (line 1665) | def _get_target_permutation_idx(self, indices):
    method get_loss (line 1671) | def get_loss(self, loss, outputs, targets, indices, num_boxes):
    method forward (line 1682) | def forward(self, outputs, targets):
  class TableTransformerMLPPredictionHead (line 1728) | class TableTransformerMLPPredictionHead(nn.Module):
    method __init__ (line 1737) | def __init__(self, input_dim, hidden_dim, output_dim, num_layers):
    method forward (line 1743) | def forward(self, x):
  class TableTransformerHungarianMatcher (line 1750) | class TableTransformerHungarianMatcher(nn.Module):
    method __init__ (line 1767) | def __init__(self, class_cost: float = 1, bbox_cost: float = 1, giou_c...
    method forward (line 1778) | def forward(self, outputs, targets):
  function _upcast (line 1829) | def _upcast(t: Tensor) -> Tensor:
  function box_area (line 1838) | def box_area(boxes: Tensor) -> Tensor:
  function box_iou (line 1855) | def box_iou(boxes1, boxes2):
  function generalized_box_iou (line 1872) | def generalized_box_iou(boxes1, boxes2):
  function _max_by_axis (line 1897) | def _max_by_axis(the_list):
  class NestedTensor (line 1907) | class NestedTensor(object):
    method __init__ (line 1908) | def __init__(self, tensors, mask: Optional[Tensor]):
    method to (line 1912) | def to(self, device):
    method decompose (line 1921) | def decompose(self):
    method __repr__ (line 1924) | def __repr__(self):
  function nested_tensor_from_tensor_list (line 1929) | def nested_tensor_from_tensor_list(tensor_list: List[Tensor]):

FILE: transformers/models/tapas/configuration_tapas.py
  class TapasConfig (line 45) | class TapasConfig(PretrainedConfig):
    method __init__ (line 157) | def __init__(

FILE: transformers/models/tapas/convert_tapas_original_tf_checkpoint_to_pytorch.py
  function convert_tf_checkpoint_to_pytorch (line 35) | def convert_tf_checkpoint_to_pytorch(

FILE: transformers/models/tapas/modeling_tapas.py
  class TableQuestionAnsweringOutput (line 91) | class TableQuestionAnsweringOutput(ModelOutput):
  function load_tf_weights_in_tapas (line 120) | def load_tf_weights_in_tapas(model, config, tf_checkpoint_path):
  class TapasEmbeddings (line 254) | class TapasEmbeddings(nn.Module):
    method __init__ (line 260) | def __init__(self, config):
    method forward (line 281) | def forward(self, input_ids=None, token_type_ids=None, position_ids=No...
  class TapasSelfAttention (line 333) | class TapasSelfAttention(nn.Module):
    method __init__ (line 334) | def __init__(self, config):
    method transpose_for_scores (line 353) | def transpose_for_scores(self, x):
    method forward (line 358) | def forward(
  class TapasSelfOutput (line 429) | class TapasSelfOutput(nn.Module):
    method __init__ (line 430) | def __init__(self, config):
    method forward (line 436) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class TapasAttention (line 443) | class TapasAttention(nn.Module):
    method __init__ (line 444) | def __init__(self, config):
    method prune_heads (line 451) | def prune_heads(self, heads):
    method forward (line 470) | def forward(
  class TapasIntermediate (line 495) | class TapasIntermediate(nn.Module):
    method __init__ (line 496) | def __init__(self, config):
    method forward (line 504) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class TapasOutput (line 511) | class TapasOutput(nn.Module):
    method __init__ (line 512) | def __init__(self, config):
    method forward (line 518) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class TapasLayer (line 525) | class TapasLayer(nn.Module):
    method __init__ (line 526) | def __init__(self, config):
    method forward (line 541) | def forward(
    method feed_forward_chunk (line 607) | def feed_forward_chunk(self, attention_output):
  class TapasEncoder (line 613) | class TapasEncoder(nn.Module):
    method __init__ (line 614) | def __init__(self, config):
    method forward (line 620) | def forward(
  class TapasPooler (line 682) | class TapasPooler(nn.Module):
    method __init__ (line 683) | def __init__(self, config):
    method forward (line 688) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class TapasPredictionHeadTransform (line 698) | class TapasPredictionHeadTransform(nn.Module):
    method __init__ (line 699) | def __init__(self, config):
    method forward (line 708) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class TapasLMPredictionHead (line 716) | class TapasLMPredictionHead(nn.Module):
    method __init__ (line 717) | def __init__(self, config):
    method forward (line 730) | def forward(self, hidden_states):
  class TapasOnlyMLMHead (line 737) | class TapasOnlyMLMHead(nn.Module):
    method __init__ (line 738) | def __init__(self, config):
    method forward (line 742) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class TapasPreTrainedModel (line 747) | class TapasPreTrainedModel(PreTrainedModel):
    method _init_weights (line 758) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 774) | def _set_gradient_checkpointing(self, module, value=False):
  class TapasModel (line 841) | class TapasModel(TapasPreTrainedModel):
    method __init__ (line 852) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 864) | def get_input_embeddings(self):
    method set_input_embeddings (line 867) | def set_input_embeddings(self, value):
    method _prune_heads (line 870) | def _prune_heads(self, heads_to_prune):
    method forward (line 880) | def forward(
  class TapasForMaskedLM (line 993) | class TapasForMaskedLM(TapasPreTrainedModel):
    method __init__ (line 998) | def __init__(self, config):
    method get_output_embeddings (line 1007) | def get_output_embeddings(self):
    method set_output_embeddings (line 1010) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1015) | def forward(
  class TapasForQuestionAnswering (line 1109) | class TapasForQuestionAnswering(TapasPreTrainedModel):
    method __init__ (line 1110) | def __init__(self, config: TapasConfig):
    method forward (line 1146) | def forward(
  class TapasForSequenceClassification (line 1454) | class TapasForSequenceClassification(TapasPreTrainedModel):
    method __init__ (line 1455) | def __init__(self, config):
    method forward (line 1468) | def forward(
  class AverageApproximationFunction (line 1574) | class AverageApproximationFunction(str, enum.Enum):
  class IndexMap (line 1583) | class IndexMap(object):
    method __init__ (line 1586) | def __init__(self, indices, num_segments, batch_dims=0):
    method batch_shape (line 1605) | def batch_shape(self):
  class ProductIndexMap (line 1609) | class ProductIndexMap(IndexMap):
    method __init__ (line 1612) | def __init__(self, outer_index, inner_index):
    method project_outer (line 1637) | def project_outer(self, index):
    method project_inner (line 1642) | def project_inner(self, index):
  function gather (line 1654) | def gather(values, index, name="segmented_gather"):
  function flatten (line 1687) | def flatten(index, name="segmented_flatten"):
  function range_index_map (line 1716) | def range_index_map(batch_shape, num_segments, name="range_index_map"):
  function _segment_reduce (line 1757) | def _segment_reduce(values, index, segment_reduce_fn, name):
  function reduce_sum (line 1805) | def reduce_sum(values, index, name="segmented_reduce_sum"):
  function reduce_mean (line 1832) | def reduce_mean(values, index, name="segmented_reduce_mean"):
  function reduce_max (line 1861) | def reduce_max(values, index, name="segmented_reduce_max"):
  function reduce_min (line 1888) | def reduce_min(values, index, name="segmented_reduce_min"):
  function compute_column_logits (line 1918) | def compute_column_logits(
  function _single_column_cell_selection_loss (line 1970) | def _single_column_cell_selection_loss(token_logits, column_logits, labe...
  function compute_token_logits (line 2077) | def compute_token_logits(sequence_output, temperature, output_weights, o...
  function _calculate_aggregate_mask (line 2099) | def _calculate_aggregate_mask(answer, pooled_output, cell_selection_pref...
  function _calculate_aggregation_loss_known (line 2150) | def _calculate_aggregation_loss_known(
  function _calculate_aggregation_loss_unknown (line 2196) | def _calculate_aggregation_loss_unknown(logits_aggregation, aggregate_ma...
  function _calculate_aggregation_loss (line 2220) | def _calculate_aggregation_loss(
  function _calculate_expected_result (line 2258) | def _calculate_expected_result(
  function huber_loss (line 2346) | def huber_loss(input, target, delta: float = 1.0):
  function _calculate_regression_loss (line 2351) | def _calculate_regression_loss(

FILE: transformers/models/tapas/modeling_tf_tapas.py
  class TFTableQuestionAnsweringOutput (line 116) | class TFTableQuestionAnsweringOutput(ModelOutput):
  class TFTapasEmbeddings (line 145) | class TFTapasEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 151) | def __init__(self, config: TapasConfig, **kwargs):
    method build (line 163) | def build(self, input_shape: tf.TensorShape):
    method call (line 191) | def call(
  class TFTapasSelfAttention (line 255) | class TFTapasSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 256) | def __init__(self, config: TapasConfig, **kwargs):
    method transpose_for_scores (line 283) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 290) | def call(
  class TFTapasSelfOutput (line 373) | class TFTapasSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 374) | def __init__(self, config: TapasConfig, **kwargs):
    method call (line 383) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFTapasAttention (line 392) | class TFTapasAttention(tf.keras.layers.Layer):
    method __init__ (line 393) | def __init__(self, config: TapasConfig, **kwargs):
    method prune_heads (line 399) | def prune_heads(self, heads):
    method call (line 402) | def call(
  class TFTapasIntermediate (line 433) | class TFTapasIntermediate(tf.keras.layers.Layer):
    method __init__ (line 434) | def __init__(self, config: TapasConfig, **kwargs):
    method call (line 446) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFTapasOutput (line 454) | class TFTapasOutput(tf.keras.layers.Layer):
    method __init__ (line 455) | def __init__(self, config: TapasConfig, **kwargs):
    method call (line 464) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFTapasLayer (line 473) | class TFTapasLayer(tf.keras.layers.Layer):
    method __init__ (line 474) | def __init__(self, config: TapasConfig, **kwargs):
    method call (line 487) | def call(
  class TFTapasEncoder (line 560) | class TFTapasEncoder(tf.keras.layers.Layer):
    method __init__ (line 561) | def __init__(self, config: TapasConfig, **kwargs):
    method call (line 566) | def call(
  class TFTapasPooler (line 630) | class TFTapasPooler(tf.keras.layers.Layer):
    method __init__ (line 631) | def __init__(self, config: TapasConfig, **kwargs):
    method call (line 641) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFTapasPredictionHeadTransform (line 651) | class TFTapasPredictionHeadTransform(tf.keras.layers.Layer):
    method __init__ (line 652) | def __init__(self, config: TapasConfig, **kwargs):
    method call (line 668) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFTapasLMPredictionHead (line 677) | class TFTapasLMPredictionHead(tf.keras.layers.Layer):
    method __init__ (line 678) | def __init__(self, config: TapasConfig, input_embeddings: tf.keras.lay...
    method build (line 690) | def build(self, input_shape: tf.TensorShape):
    method get_output_embeddings (line 695) | def get_output_embeddings(self) -> tf.keras.layers.Layer:
    method set_output_embeddings (line 698) | def set_output_embeddings(self, value: tf.Variable):
    method get_bias (line 702) | def get_bias(self) -> Dict[str, tf.Variable]:
    method set_bias (line 705) | def set_bias(self, value: tf.Variable):
    method call (line 709) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFTapasMLMHead (line 721) | class TFTapasMLMHead(tf.keras.layers.Layer):
    method __init__ (line 722) | def __init__(self, config: TapasConfig, input_embeddings: tf.keras.lay...
    method call (line 727) | def call(self, sequence_output: tf.Tensor) -> tf.Tensor:
  class TFTapasMainLayer (line 734) | class TFTapasMainLayer(tf.keras.layers.Layer):
    method __init__ (line 737) | def __init__(self, config: TapasConfig, add_pooling_layer: bool = True...
    method get_input_embeddings (line 747) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 750) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 754) | def _prune_heads(self, heads_to_prune):
    method call (line 762) | def call(
  class TFTapasPreTrainedModel (line 856) | class TFTapasPreTrainedModel(TFPreTrainedModel):
    method input_signature (line 866) | def input_signature(self):
  class TFTapasModel (line 974) | class TFTapasModel(TFTapasPreTrainedModel):
    method __init__ (line 975) | def __init__(self, config: TapasConfig, *inputs, **kwargs):
    method call (line 983) | def call(
  class TFTapasForMaskedLM (line 1038) | class TFTapasForMaskedLM(TFTapasPreTrainedModel, TFMaskedLanguageModelin...
    method __init__ (line 1039) | def __init__(self, config: TapasConfig, *inputs, **kwargs):
    method get_lm_head (line 1051) | def get_lm_head(self) -> tf.keras.layers.Layer:
    method call (line 1057) | def call(
  class TFTapasComputeTokenLogits (line 1133) | class TFTapasComputeTokenLogits(tf.keras.layers.Layer):
    method __init__ (line 1134) | def __init__(self, config: TapasConfig, **kwargs):
    method call (line 1153) | def call(self, sequence_output: tf.Tensor) -> tf.Tensor:
  class TFTapasComputeColumnLogits (line 1169) | class TFTapasComputeColumnLogits(tf.keras.layers.Layer):
    method __init__ (line 1170) | def __init__(self, config: TapasConfig, **kwargs):
    method call (line 1187) | def call(self, sequence_output, cell_index, cell_mask, allow_empty_col...
  class TFTapasForQuestionAnswering (line 1238) | class TFTapasForQuestionAnswering(TFTapasPreTrainedModel):
    method __init__ (line 1239) | def __init__(self, config: TapasConfig, *inputs, **kwargs):
    method call (line 1263) | def call(
  class TFTapasForSequenceClassification (line 1563) | class TFTapasForSequenceClassification(TFTapasPreTrainedModel, TFSequenc...
    method __init__ (line 1564) | def __init__(self, config: TapasConfig, *inputs, **kwargs):
    method call (line 1577) | def call(
  class AverageApproximationFunction (line 1661) | class AverageApproximationFunction(str, enum.Enum):
  class IndexMap (line 1670) | class IndexMap(object):
    method __init__ (line 1673) | def __init__(self, indices, num_segments, batch_dims=0):
    method batch_shape (line 1689) | def batch_shape(self):
  class ProductIndexMap (line 1693) | class ProductIndexMap(IndexMap):
    method __init__ (line 1696) | def __init__(self, outer_index, inner_index):
    method project_outer (line 1722) | def project_outer(self, index):
    method project_inner (line 1730) | def project_inner(self, index):
  function gather (line 1739) | def gather(values, index, name="segmented_gather"):
  function flatten (line 1755) | def flatten(index, name="segmented_flatten"):
  function range_index_map (line 1778) | def range_index_map(batch_shape, num_segments, name="range_index_map"):
  function _segment_reduce (line 1806) | def _segment_reduce(values, index, segment_reduce_fn, name):
  function reduce_mean (line 1841) | def reduce_mean(values, index, name="segmented_reduce_mean"):
  function reduce_sum (line 1864) | def reduce_sum(values, index, name="segmented_reduce_sum"):
  function reduce_max (line 1887) | def reduce_max(values, index, name="segmented_reduce_max"):
  function reduce_min (line 1909) | def reduce_min(values, index, name="segmented_reduce_min"):
  function _single_column_cell_selection_loss (line 1914) | def _single_column_cell_selection_loss(token_logits, column_logits, labe...
  function _calculate_aggregate_mask (line 1987) | def _calculate_aggregate_mask(answer, pooled_output, cell_selection_pref...
  function _calculate_aggregation_loss_known (line 2031) | def _calculate_aggregation_loss_known(
  function _calculate_aggregation_loss_unknown (line 2077) | def _calculate_aggregation_loss_unknown(logits_aggregation, aggregate_ma...
  function _calculate_aggregation_loss (line 2101) | def _calculate_aggregation_loss(
  function _calculate_expected_result (line 2139) | def _calculate_expected_result(
  function _calculate_regression_loss (line 2218) | def _calculate_regression_loss(

FILE: transformers/models/tapas/tokenization_tapas.py
  class TapasTruncationStrategy (line 137) | class TapasTruncationStrategy(ExplicitEnum):
  class TokenCoordinates (line 150) | class TokenCoordinates:
  class TokenizedTable (line 157) | class TokenizedTable:
  class SerializedExample (line 163) | class SerializedExample:
  function _is_inner_wordpiece (line 170) | def _is_inner_wordpiece(token: Text):
  function load_vocab (line 174) | def load_vocab(vocab_file):
  function whitespace_tokenize (line 185) | def whitespace_tokenize(text):
  class TapasTokenizer (line 236) | class TapasTokenizer(PreTrainedTokenizer):
    method __init__ (line 320) | def __init__(
    method do_lower_case (line 405) | def do_lower_case(self):
    method vocab_size (line 409) | def vocab_size(self):
    method get_vocab (line 412) | def get_vocab(self):
    method _tokenize (line 415) | def _tokenize(self, text):
    method _convert_token_to_id (line 430) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 434) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 438) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 443) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method create_attention_mask_from_sequences (line 463) | def create_attention_mask_from_sequences(self, query_ids: List[int], t...
    method create_segment_token_type_ids_from_sequences (line 477) | def create_segment_token_type_ids_from_sequences(
    method create_column_token_type_ids_from_sequences (line 494) | def create_column_token_type_ids_from_sequences(
    method create_row_token_type_ids_from_sequences (line 511) | def create_row_token_type_ids_from_sequences(
    method build_inputs_with_special_tokens (line 528) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 547) | def get_special_tokens_mask(
    method __call__ (line 576) | def __call__(
    method batch_encode_plus (line 693) | def batch_encode_plus(
    method _get_question_tokens (line 789) | def _get_question_tokens(self, query):
    method _batch_encode_plus (line 802) | def _batch_encode_plus(
    method _batch_prepare_for_model (line 859) | def _batch_prepare_for_model(
    method encode (line 931) | def encode(
    method encode_plus (line 974) | def encode_plus(
    method _encode_plus (line 1057) | def _encode_plus(
    method prepare_for_model (line 1113) | def prepare_for_model(
    method _get_truncated_table_rows (line 1305) | def _get_truncated_table_rows(
    method _tokenize_table (line 1362) | def _tokenize_table(
    method _question_encoding_cost (line 1407) | def _question_encoding_cost(self, question_tokens):
    method _get_token_budget (line 1411) | def _get_token_budget(self, question_tokens, max_length=None):
    method _get_table_values (line 1425) | def _get_table_values(self, table, num_columns, num_rows, num_tokens) ...
    method _get_table_boundaries (line 1444) | def _get_table_boundaries(self, table):
    method _get_table_cost (line 1457) | def _get_table_cost(self, table, num_columns, num_rows, num_tokens):
    method _get_max_num_tokens (line 1460) | def _get_max_num_tokens(self, question_tokens, tokenized_table, num_co...
    method _get_num_columns (line 1479) | def _get_num_columns(self, table):
    method _get_num_rows (line 1485) | def _get_num_rows(self, table, drop_rows_to_fit):
    method _serialize_text (line 1494) | def _serialize_text(self, question_tokens):
    method _serialize (line 1515) | def _serialize(
    method _get_column_values (line 1545) | def _get_column_values(self, table, col_index):
    method _get_cell_token_indexes (line 1553) | def _get_cell_token_indexes(self, column_ids, row_ids, column_id, row_...
    method _get_numeric_column_ranks (line 1558) | def _get_numeric_column_ranks(self, column_ids, row_ids, table):
    method _get_numeric_sort_key_fn (line 1593) | def _get_numeric_sort_key_fn(self, table_numeric_values, value):
    method _get_numeric_relations (line 1614) | def _get_numeric_relations(self, question, column_ids, row_ids, table):
    method _get_numeric_values (line 1655) | def _get_numeric_values(self, table, column_ids, row_ids):
    method _get_numeric_values_scale (line 1678) | def _get_numeric_values_scale(self, table, column_ids, row_ids):
    method _pad_to_seq_length (line 1699) | def _pad_to_seq_length(self, inputs):
    method _get_all_answer_ids_from_coordinates (line 1705) | def _get_all_answer_ids_from_coordinates(
    method _get_all_answer_ids (line 1725) | def _get_all_answer_ids(self, column_ids, row_ids, answer_coordinates):
    method _find_tokens (line 1740) | def _find_tokens(self, text, segment):
    method _find_answer_coordinates_from_answer_text (line 1751) | def _find_answer_coordinates_from_answer_text(
    method _find_answer_ids_from_answer_texts (line 1771) | def _find_answer_ids_from_answer_texts(
    method _get_answer_ids (line 1809) | def _get_answer_ids(self, column_ids, row_ids, answer_coordinates):
    method get_answer_ids (line 1817) | def get_answer_ids(self, column_ids, row_ids, tokenized_table, answer_...
    method _pad (line 1827) | def _pad(
    method _get_cell_token_probs (line 1921) | def _get_cell_token_probs(self, probabilities, segment_ids, row_ids, c...
    method _get_mean_cell_probs (line 1929) | def _get_mean_cell_probs(self, probabilities, segment_ids, row_ids, co...
    method convert_logits_to_predictions (line 1938) | def convert_logits_to_predictions(self, data, logits, logits_agg=None,...
  class BasicTokenizer (line 2040) | class BasicTokenizer(object):
    method __init__ (line 2060) | def __init__(self, do_lower_case=True, never_split=None, tokenize_chin...
    method tokenize (line 2068) | def tokenize(self, text, never_split=None):
    method _run_strip_accents (line 2105) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 2116) | def _run_split_on_punc(self, text, never_split=None):
    method _tokenize_chinese_chars (line 2138) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 2151) | def _is_chinese_char(self, cp):
    method _clean_text (line 2175) | def _clean_text(self, text):
  class WordpieceTokenizer (line 2190) | class WordpieceTokenizer(object):
    method __init__ (line 2193) | def __init__(self, vocab, unk_token, max_input_chars_per_word=100):
    method tokenize (line 2198) | def tokenize(self, text):
  class Relation (line 2256) | class Relation(enum.Enum):
  class Date (line 2269) | class Date:
  class NumericValue (line 2276) | class NumericValue:
  class NumericValueSpan (line 2282) | class NumericValueSpan:
  class Cell (line 2289) | class Cell:
  class Question (line 2295) | class Question:
  function _process_date_pattern (line 2354) | def _process_date_pattern(dp):
  function _process_date_patterns (line 2368) | def _process_date_patterns():
  function _get_numeric_value_from_date (line 2422) | def _get_numeric_value_from_date(date, mask):
  function _get_span_length_key (line 2437) | def _get_span_length_key(span):
  function _get_numeric_value_from_float (line 2442) | def _get_numeric_value_from_float(value):
  function _parse_date (line 2448) | def _parse_date(text):
  function _parse_number (line 2465) | def _parse_number(text):
  function get_all_spans (line 2483) | def get_all_spans(text, max_ngram_length):
  function normalize_for_match (line 2504) | def normalize_for_match(text):
  function format_text (line 2508) | def format_text(text):
  function parse_text (line 2522) | def parse_text(text):
  function _get_value_type (line 2596) | def _get_value_type(numeric_value):
  function _get_value_as_primitive_value (line 2604) | def _get_value_as_primitive_value(numeric_value):
  function _get_all_types (line 2622) | def _get_all_types(numeric_values):
  function get_numeric_sort_key_fn (line 2626) | def get_numeric_sort_key_fn(numeric_values):
  function _consolidate_numeric_values (line 2673) | def _consolidate_numeric_values(row_index_to_values, min_consolidation_f...
  function _get_numeric_values (line 2720) | def _get_numeric_values(text):
  function _get_column_values (line 2726) | def _get_column_values(table, col_index):
  function get_numeric_relation (line 2742) | def get_numeric_relation(value, other_value, sort_key_fn):
  function add_numeric_values_to_question (line 2755) | def add_numeric_values_to_question(question):
  function filter_invalid_unicode (line 2763) | def filter_invalid_unicode(text):
  function filter_invalid_unicode_from_table (line 2768) | def filter_invalid_unicode_from_table(table):
  function add_numeric_table_values (line 2794) | def add_numeric_table_values(table, min_consolidation_fraction=0.7, debu...

FILE: transformers/models/tapex/tokenization_tapex.py
  class TapexTruncationStrategy (line 57) | class TapexTruncationStrategy(ExplicitEnum):
  function bytes_to_unicode (line 120) | def bytes_to_unicode():
  function get_pairs (line 142) | def get_pairs(word):
  class IndexedRowTableLinearize (line 155) | class IndexedRowTableLinearize:
    method process_table (line 160) | def process_table(self, table_content: Dict):
    method process_header (line 173) | def process_header(self, headers: List):
    method process_row (line 179) | def process_row(self, row: List, row_index: int):
  class TapexTokenizer (line 194) | class TapexTokenizer(PreTrainedTokenizer):
    method __init__ (line 272) | def __init__(
    method build_inputs_with_special_tokens (line 337) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 360) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 385) | def create_token_type_ids_from_sequences(
    method prepare_for_tokenization (line 406) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method vocab_size (line 413) | def vocab_size(self):
    method get_vocab (line 416) | def get_vocab(self):
    method bpe (line 419) | def bpe(self, token):
    method _tokenize (line 461) | def _tokenize(self, text):
    method _convert_token_to_id (line 471) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 475) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 479) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 485) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method __call__ (line 515) | def __call__(
    method source_call_func (line 592) | def source_call_func(
    method batch_encode_plus (line 680) | def batch_encode_plus(
    method _batch_encode_plus (line 737) | def _batch_encode_plus(
    method _batch_prepare_for_model (line 796) | def _batch_prepare_for_model(
    method encode (line 867) | def encode(
    method encode_plus (line 899) | def encode_plus(
    method _encode_plus (line 947) | def _encode_plus(
    method target_call_func (line 1005) | def target_call_func(
    method target_batch_encode_plus (line 1070) | def target_batch_encode_plus(
    method _target_batch_encode_plus (line 1123) | def _target_batch_encode_plus(
    method target_encode (line 1183) | def target_encode(
    method target_encode_plus (line 1214) | def target_encode_plus(
    method _target_encode_plus (line 1265) | def _target_encode_plus(
    method prepare_table_query (line 1319) | def prepare_table_query(
    method truncate_table_cells (line 1363) | def truncate_table_cells(self, table_content: Dict, question: str, ans...
    method truncate_cell (line 1379) | def truncate_cell(self, cell_value):
    method truncate_table_rows (line 1394) | def truncate_table_rows(
    method estimate_delete_ratio (line 1423) | def estimate_delete_ratio(self, table_content: Dict, question: str, ma...
    method delete_unrelated_rows (line 1449) | def delete_unrelated_rows(self, table_content: Dict, question: str, an...

FILE: transformers/models/time_series_transformer/configuration_time_series_transformer.py
  class TimeSeriesTransformerConfig (line 33) | class TimeSeriesTransformerConfig(PretrainedConfig):
    method __init__ (line 137) | def __init__(
    method _number_of_features (line 225) | def _number_of_features(self) -> int:

FILE: transformers/models/time_series_transformer/modeling_time_series_transformer.py
  class TimeSeriesFeatureEmbedder (line 50) | class TimeSeriesFeatureEmbedder(nn.Module):
    method __init__ (line 61) | def __init__(self, cardinalities: List[int], embedding_dims: List[int]...
    method forward (line 67) | def forward(self, features: torch.Tensor) -> torch.Tensor:
  class TimeSeriesStdScaler (line 84) | class TimeSeriesStdScaler(nn.Module):
    method __init__ (line 98) | def __init__(self, dim: int, keepdim: bool = False, minimum_scale: flo...
    method forward (line 107) | def forward(self, data: torch.Tensor, weights: torch.Tensor) -> Tuple[...
  class TimeSeriesMeanScaler (line 117) | class TimeSeriesMeanScaler(nn.Module):
    method __init__ (line 133) | def __init__(
    method forward (line 143) | def forward(
  class TimeSeriesNOPScaler (line 174) | class TimeSeriesNOPScaler(nn.Module):
    method __init__ (line 185) | def __init__(self, dim: int, keepdim: bool = False):
    method forward (line 190) | def forward(
  function nll (line 198) | def nll(input: torch.distributions.Distribution, target: torch.Tensor) -...
  function weighted_average (line 205) | def weighted_average(input_tensor: torch.Tensor, weights: Optional[torch...
  function _make_causal_mask (line 230) | def _make_causal_mask(
  function _expand_mask (line 248) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class TimeSeriesSinusoidalPositionalEmbedding (line 263) | class TimeSeriesSinusoidalPositionalEmbedding(nn.Embedding):
    method __init__ (line 266) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method _init_weight (line 271) | def _init_weight(out: nn.Parameter) -> nn.Parameter:
    method forward (line 288) | def forward(self, input_ids_shape: torch.Size, past_key_values_length:...
  class TimeSeriesValueEmbedding (line 297) | class TimeSeriesValueEmbedding(nn.Module):
    method __init__ (line 298) | def __init__(self, feature_size, d_model):
    method forward (line 302) | def forward(self, x):
  class TimeSeriesTransformerAttention (line 307) | class TimeSeriesTransformerAttention(nn.Module):
    method __init__ (line 310) | def __init__(
    method _shape (line 337) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 340) | def forward(
  class TimeSeriesTransformerEncoderLayer (line 462) | class TimeSeriesTransformerEncoderLayer(nn.Module):
    method __init__ (line 463) | def __init__(self, config: TimeSeriesTransformerConfig):
    method forward (line 479) | def forward(
  class TimeSeriesTransformerDecoderLayer (line 531) | class TimeSeriesTransformerDecoderLayer(nn.Module):
    method __init__ (line 532) | def __init__(self, config: TimeSeriesTransformerConfig):
    method forward (line 558) | def forward(
  class TimeSeriesTransformerPreTrainedModel (line 648) | class TimeSeriesTransformerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 654) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 667) | def _set_gradient_checkpointing(self, module, value=False):
  class TimeSeriesTransformerEncoder (line 843) | class TimeSeriesTransformerEncoder(TimeSeriesTransformerPreTrainedModel):
    method __init__ (line 852) | def __init__(self, config: TimeSeriesTransformerConfig):
    method forward (line 871) | def forward(
  class TimeSeriesTransformerDecoder (line 981) | class TimeSeriesTransformerDecoder(TimeSeriesTransformerPreTrainedModel):
    method __init__ (line 990) | def __init__(self, config: TimeSeriesTransformerConfig):
    method _prepare_decoder_attention_mask (line 1008) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 1031) | def forward(
  class TimeSeriesTransformerModel (line 1228) | class TimeSeriesTransformerModel(TimeSeriesTransformerPreTrainedModel):
    method __init__ (line 1229) | def __init__(self, config: TimeSeriesTransformerConfig):
    method _past_length (line 1253) | def _past_length(self) -> int:
    method get_lagged_subsequences (line 1256) | def get_lagged_subsequences(
    method create_network_inputs (line 1288) | def create_network_inputs(
    method get_encoder (line 1360) | def get_encoder(self):
    method get_decoder (line 1363) | def get_decoder(self):
    method forward (line 1368) | def forward(
  class TimeSeriesTransformerForPrediction (line 1489) | class TimeSeriesTransformerForPrediction(TimeSeriesTransformerPreTrained...
    method __init__ (line 1490) | def __init__(self, config: TimeSeriesTransformerConfig):
    method output_params (line 1513) | def output_params(self, dec_output):
    method get_encoder (line 1516) | def get_encoder(self):
    method get_decoder (line 1519) | def get_decoder(self):
    method output_distribution (line 1523) | def output_distribution(self, params, loc=None, scale=None, trailing_n...
    method forward (line 1531) | def forward(
    method generate (line 1664) | def generate(

FILE: transformers/models/timesformer/configuration_timesformer.py
  class TimesformerConfig (line 28) | class TimesformerConfig(PretrainedConfig):
    method __init__ (line 90) | def __init__(

FILE: transformers/models/timesformer/convert_timesformer_to_pytorch.py
  function get_timesformer_config (line 28) | def get_timesformer_config(model_name):
  function rename_key (line 58) | def rename_key(name):
  function convert_state_dict (line 103) | def convert_state_dict(orig_state_dict, config):
  function prepare_video (line 130) | def prepare_video():
  function convert_timesformer_checkpoint (line 138) | def convert_timesformer_checkpoint(checkpoint_url, pytorch_dump_folder_p...

FILE: transformers/models/timesformer/modeling_timesformer.py
  class TimesformerPatchEmbeddings (line 46) | class TimesformerPatchEmbeddings(nn.Module):
    method __init__ (line 49) | def __init__(self, config):
    method forward (line 65) | def forward(self, pixel_values):
  class TimesformerEmbeddings (line 75) | class TimesformerEmbeddings(nn.Module):
    method __init__ (line 80) | def __init__(self, config):
    method forward (line 100) | def forward(self, pixel_values):
  function drop_path (line 156) | def drop_path(input: torch.Tensor, drop_prob: float = 0.0, training: boo...
  class TimeSformerDropPath (line 177) | class TimeSformerDropPath(nn.Module):
    method __init__ (line 180) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 184) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 187) | def extra_repr(self) -> str:
  class TimesformerSelfAttention (line 192) | class TimesformerSelfAttention(nn.Module):
    method __init__ (line 193) | def __init__(self, config: TimesformerConfig):
    method forward (line 206) | def forward(self, hidden_states, output_attentions: bool = False):
  class TimesformerSelfOutput (line 226) | class TimesformerSelfOutput(nn.Module):
    method __init__ (line 232) | def __init__(self, config: TimesformerConfig) -> None:
    method forward (line 237) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class TimeSformerAttention (line 244) | class TimeSformerAttention(nn.Module):
    method __init__ (line 245) | def __init__(self, config: TimesformerConfig) -> None:
    method forward (line 250) | def forward(
  class TimesformerIntermediate (line 264) | class TimesformerIntermediate(nn.Module):
    method __init__ (line 265) | def __init__(self, config: TimesformerConfig) -> None:
    method forward (line 275) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class TimesformerOutput (line 283) | class TimesformerOutput(nn.Module):
    method __init__ (line 284) | def __init__(self, config: TimesformerConfig) -> None:
    method forward (line 289) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class TimesformerLayer (line 297) | class TimesformerLayer(nn.Module):
    method __init__ (line 298) | def __init__(self, config: TimesformerConfig, layer_index: int) -> None:
    method forward (line 326) | def forward(self, hidden_states: torch.Tensor, output_attentions: bool...
  class TimesformerEncoder (line 420) | class TimesformerEncoder(nn.Module):
    method __init__ (line 421) | def __init__(self, config: TimesformerConfig) -> None:
    method forward (line 427) | def forward(
  class TimesformerPreTrainedModel (line 473) | class TimesformerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 484) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 497) | def _set_gradient_checkpointing(self, module, value=False):
  class TimesformerModel (line 534) | class TimesformerModel(TimesformerPreTrainedModel):
    method __init__ (line 535) | def __init__(self, config):
    method get_input_embeddings (line 547) | def get_input_embeddings(self):
    method _prune_heads (line 550) | def _prune_heads(self, heads_to_prune):
    method forward (line 560) | def forward(
  class TimesformerForVideoClassification (line 667) | class TimesformerForVideoClassification(TimesformerPreTrainedModel):
    method __init__ (line 668) | def __init__(self, config):
    method forward (line 682) | def forward(

FILE: transformers/models/timm_backbone/configuration_timm_backbone.py
  class TimmBackboneConfig (line 25) | class TimmBackboneConfig(PretrainedConfig):
    method __init__ (line 63) | def __init__(

FILE: transformers/models/timm_backbone/modeling_timm_backbone.py
  class TimmBackbone (line 33) | class TimmBackbone(PreTrainedModel, BackboneMixin):
    method __init__ (line 43) | def __init__(self, config, **kwargs):
    method from_pretrained (line 80) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
    method _init_weights (line 103) | def _init_weights(self, module):
    method forward (line 109) | def forward(

FILE: transformers/models/trajectory_transformer/configuration_trajectory_transformer.py
  class TrajectoryTransformerConfig (line 31) | class TrajectoryTransformerConfig(PretrainedConfig):
    method __init__ (line 111) | def __init__(

FILE: transformers/models/trajectory_transformer/convert_trajectory_transformer_original_pytorch_checkpoint_to_pytorch.py
  class Parser (line 23) | class Parser(utils.Parser):
  function convert_trajectory_transformer_original_pytorch_checkpoint_to_pytorch (line 28) | def convert_trajectory_transformer_original_pytorch_checkpoint_to_pytorc...

FILE: transformers/models/trajectory_transformer/modeling_trajectory_transformer.py
  function load_tf_weights_in_trajectory_transformer (line 50) | def load_tf_weights_in_trajectory_transformer(model, config, tf_checkpoi...
  class TrajectoryTransformerOutput (line 124) | class TrajectoryTransformerOutput(ModelOutput):
  class TrajectoryTransformerPreTrainedModel (line 154) | class TrajectoryTransformerPreTrainedModel(PreTrainedModel):
    method _set_gradient_checkpointing (line 166) | def _set_gradient_checkpointing(self, module, value=False):
    method _init_weights (line 170) | def _init_weights(self, module):
  class EinLinear (line 229) | class EinLinear(nn.Module):
    method __init__ (line 230) | def __init__(self, n_models, in_features, out_features, bias):
    method reset_parameters (line 241) | def reset_parameters(self):
    method forward (line 249) | def forward(self, input):
  class CausalSelfAttention (line 262) | class CausalSelfAttention(nn.Module):
    method __init__ (line 263) | def __init__(self, config):
    method forward (line 295) | def forward(
  class Block (line 357) | class Block(nn.Module):
    method __init__ (line 358) | def __init__(self, config):
    method forward (line 370) | def forward(
  class TrajectoryTransformerModel (line 406) | class TrajectoryTransformerModel(TrajectoryTransformerPreTrainedModel):
    method __init__ (line 409) | def __init__(self, config):
    method get_block_size (line 440) | def get_block_size(self):
    method offset_tokens (line 443) | def offset_tokens(self, trajectories):
    method pad_to_full_observation (line 455) | def pad_to_full_observation(self, hidden_states):
    method forward (line 471) | def forward(

FILE: transformers/models/transfo_xl/configuration_transfo_xl.py
  class TransfoXLConfig (line 29) | class TransfoXLConfig(PretrainedConfig):
    method __init__ (line 116) | def __init__(
    method max_position_embeddings (line 178) | def max_position_embeddings(self):
    method max_position_embeddings (line 184) | def max_position_embeddings(self, value):

FILE: transformers/models/transfo_xl/convert_transfo_xl_original_tf_checkpoint_to_pytorch.py
  function convert_transfo_xl_checkpoint_to_pytorch (line 41) | def convert_transfo_xl_checkpoint_to_pytorch(

FILE: transformers/models/transfo_xl/modeling_tf_transfo_xl.py
  class TFPositionalEmbedding (line 59) | class TFPositionalEmbedding(tf.keras.layers.Layer):
    method __init__ (line 60) | def __init__(self, demb, **kwargs):
    method call (line 65) | def call(self, pos_seq, bsz=None):
  class TFPositionwiseFF (line 76) | class TFPositionwiseFF(tf.keras.layers.Layer):
    method __init__ (line 77) | def __init__(self, d_model, d_inner, dropout, pre_lnorm=False, layer_n...
    method call (line 95) | def call(self, inp, training=False):
  class TFRelPartialLearnableMultiHeadAttn (line 119) | class TFRelPartialLearnableMultiHeadAttn(tf.keras.layers.Layer):
    method __init__ (line 120) | def __init__(
    method build (line 170) | def build(self, input_shape):
    method _rel_shift (line 180) | def _rel_shift(self, x):
    method call (line 190) | def call(self, w, r, attn_mask, mems, head_mask, output_attentions, tr...
  class TFRelPartialLearnableDecoderLayer (line 271) | class TFRelPartialLearnableDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 272) | def __init__(
    method call (line 314) | def call(self, dec_inp, r, dec_attn_mask, mems, head_mask, output_atte...
  class TFTransfoEmbeddings (line 323) | class TFTransfoEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 324) | def __init__(self, vocab_size, emb_size, init_std, **kwargs):
    method build (line 331) | def build(self, input_shape):
    method call (line 340) | def call(self, inputs):
  class TFAdaptiveEmbedding (line 344) | class TFAdaptiveEmbedding(tf.keras.layers.Layer):
    method __init__ (line 345) | def __init__(self, n_token, d_embed, d_proj, cutoffs, div_val=1, init_...
    method build (line 378) | def build(self, input_shape):
    method call (line 392) | def call(self, inp):
  class TFTransfoXLMainLayer (line 421) | class TFTransfoXLMainLayer(tf.keras.layers.Layer):
    method __init__ (line 424) | def __init__(self, config, **kwargs):
    method build (line 487) | def build(self, input_shape):
    method get_input_embeddings (line 497) | def get_input_embeddings(self):
    method set_input_embeddings (line 500) | def set_input_embeddings(self, value):
    method backward_compatible (line 503) | def backward_compatible(self):
    method reset_memory_length (line 506) | def reset_memory_length(self, mem_len):
    method _prune_heads (line 509) | def _prune_heads(self, heads):
    method init_mems (line 512) | def init_mems(self, bsz):
    method _update_mems (line 523) | def _update_mems(self, hids, mems, mlen, qlen):
    method call (line 544) | def call(
  class TFTransfoXLPreTrainedModel (line 678) | class TFTransfoXLPreTrainedModel(TFPreTrainedModel):
  class TFTransfoXLModelOutput (line 689) | class TFTransfoXLModelOutput(ModelOutput):
  class TFTransfoXLLMHeadModelOutput (line 720) | class TFTransfoXLLMHeadModelOutput(ModelOutput):
  class TFTransfoXLSequenceClassifierOutputWithPast (line 753) | class TFTransfoXLSequenceClassifierOutputWithPast(ModelOutput):
  class TFTransfoXLModel (line 871) | class TFTransfoXLModel(TFTransfoXLPreTrainedModel):
    method __init__ (line 872) | def __init__(self, config, *inputs, **kwargs):
    method call (line 883) | def call(
  class TFTransfoXLLMHeadModel (line 915) | class TFTransfoXLLMHeadModel(TFTransfoXLPreTrainedModel):
    method __init__ (line 916) | def __init__(self, config):
    method _resize_token_embeddings (line 929) | def _resize_token_embeddings(self, new_num_tokens):
    method get_output_embeddings (line 932) | def get_output_embeddings(self):
    method reset_memory_length (line 938) | def reset_memory_length(self, mem_len):
    method init_mems (line 941) | def init_mems(self, bsz):
    method call (line 951) | def call(
    method prepare_inputs_for_generation (line 995) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
  class TFTransfoXLForSequenceClassification (line 1022) | class TFTransfoXLForSequenceClassification(TFTransfoXLPreTrainedModel, T...
    method __init__ (line 1023) | def __init__(self, config, *inputs, **kwargs):
    method get_output_embeddings (line 1034) | def get_output_embeddings(self):
    method call (line 1044) | def call(

FILE: transformers/models/transfo_xl/modeling_tf_transfo_xl_utilities.py
  class TFAdaptiveSoftmaxMask (line 26) | class TFAdaptiveSoftmaxMask(tf.keras.layers.Layer):
    method __init__ (line 27) | def __init__(self, vocab_size, d_embed, d_proj, cutoffs, div_val=1, ke...
    method build (line 46) | def build(self, input_shape):
    method _logit (line 105) | def _logit(x, W, b, proj=None):
    method _gather_logprob (line 112) | def _gather_logprob(logprob, target):
    method call (line 118) | def call(self, hidden, target, return_mean=True, training=False):

FILE: transformers/models/transfo_xl/modeling_transfo_xl.py
  function build_tf_to_pytorch_map (line 51) | def build_tf_to_pytorch_map(model, config):
  function load_tf_weights_in_transfo_xl (line 119) | def load_tf_weights_in_transfo_xl(model, config, tf_path):
  class PositionalEmbedding (line 178) | class PositionalEmbedding(nn.Module):
    method __init__ (line 179) | def __init__(self, demb):
    method forward (line 187) | def forward(self, pos_seq, bsz=None):
  class PositionwiseFF (line 197) | class PositionwiseFF(nn.Module):
    method __init__ (line 198) | def __init__(self, d_model, d_inner, dropout, pre_lnorm=False, layer_n...
    method forward (line 217) | def forward(self, inp):
  class RelPartialLearnableMultiHeadAttn (line 234) | class RelPartialLearnableMultiHeadAttn(nn.Module):
    method __init__ (line 235) | def __init__(
    method _rel_shift (line 275) | def _rel_shift(self, x):
    method forward (line 287) | def forward(self, w, r, attn_mask=None, mems=None, head_mask=None, out...
  class RelPartialLearnableDecoderLayer (line 372) | class RelPartialLearnableDecoderLayer(nn.Module):
    method __init__ (line 373) | def __init__(self, n_head, d_model, d_head, d_inner, dropout, layer_no...
    method forward (line 383) | def forward(self, dec_inp, r, dec_attn_mask=None, mems=None, head_mask...
  class AdaptiveEmbedding (line 399) | class AdaptiveEmbedding(nn.Module):
    method __init__ (line 400) | def __init__(self, n_token, d_embed, d_proj, cutoffs, div_val=1, sampl...
    method forward (line 427) | def forward(self, inp):
  class TransfoXLPreTrainedModel (line 459) | class TransfoXLPreTrainedModel(PreTrainedModel):
    method _init_weight (line 469) | def _init_weight(self, weight):
    method _init_bias (line 475) | def _init_bias(self, bias):
    method _init_weights (line 478) | def _init_weights(self, m):
    method resize_token_embeddings (line 518) | def resize_token_embeddings(self, new_num_tokens: Optional[int] = None...
    method _get_new_num_tokens_layer (line 557) | def _get_new_num_tokens_layer(self, new_num_tokens, layer):
    method _get_embedding_shapes (line 570) | def _get_embedding_shapes(self):
    method _resize_token_embeddings (line 574) | def _resize_token_embeddings(self, new_num_tokens, layer=-1):
    method _resize_cutoffs (line 585) | def _resize_cutoffs(self, new_num_tokens, new_emb_size, new_embedding_...
  class TransfoXLModelOutput (line 600) | class TransfoXLModelOutput(ModelOutput):
  class TransfoXLSequenceClassifierOutputWithPast (line 631) | class TransfoXLSequenceClassifierOutputWithPast(ModelOutput):
  class TransfoXLLMHeadModelOutput (line 665) | class TransfoXLLMHeadModelOutput(ModelOutput):
    method logits (line 701) | def logits(self):
  class TransfoXLModel (line 764) | class TransfoXLModel(TransfoXLPreTrainedModel):
    method __init__ (line 765) | def __init__(self, config):
    method get_input_embeddings (line 820) | def get_input_embeddings(self):
    method set_input_embeddings (line 823) | def set_input_embeddings(self, new_embeddings):
    method backward_compatible (line 826) | def backward_compatible(self):
    method reset_memory_length (line 829) | def reset_memory_length(self, mem_len):
    method _prune_heads (line 832) | def _prune_heads(self, heads):
    method init_mems (line 836) | def init_mems(self, bsz):
    method _update_mems (line 848) | def _update_mems(self, hids, mems, mlen, qlen):
    method forward (line 873) | def forward(
  class TransfoXLLMHeadModel (line 1004) | class TransfoXLLMHeadModel(TransfoXLPreTrainedModel):
    method __init__ (line 1007) | def __init__(self, config):
    method tie_weights (line 1033) | def tie_weights(self):
    method reset_memory_length (line 1054) | def reset_memory_length(self, mem_len):
    method init_mems (line 1057) | def init_mems(self, bsz):
    method forward (line 1066) | def forward(
    method get_output_embeddings (line 1141) | def get_output_embeddings(self):
    method prepare_inputs_for_generation (line 1148) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _resize_cutoffs (line 1160) | def _resize_cutoffs(self, new_num_tokens, new_emb_size, new_embedding_...
    method _reorder_cache (line 1168) | def _reorder_cache(mems: List[torch.Tensor], beam_idx: torch.Tensor) -...
  class TransfoXLForSequenceClassification (line 1192) | class TransfoXLForSequenceClassification(TransfoXLPreTrainedModel):
    method __init__ (line 1195) | def __init__(self, config):
    method forward (line 1209) | def forward(

FILE: transformers/models/transfo_xl/modeling_transfo_xl_utilities.py
  class ProjectedAdaptiveLogSoftmax (line 29) | class ProjectedAdaptiveLogSoftmax(nn.Module):
    method __init__ (line 30) | def __init__(self, n_token, d_embed, d_proj, cutoffs, div_val=1, keep_...
    method _compute_logit (line 71) | def _compute_logit(self, hidden, weight, bias, proj):
    method forward (line 85) | def forward(self, hidden, labels=None, keep_order=False):
    method log_prob (line 194) | def log_prob(self, hidden):

FILE: transformers/models/transfo_xl/tokenization_transfo_xl.py
  function tokenize_numbers (line 76) | def tokenize_numbers(text_array: List[str]) -> List[str]:
  function detokenize_numbers (line 102) | def detokenize_numbers(text: str) -> str:
  class TransfoXLTokenizer (line 123) | class TransfoXLTokenizer(PreTrainedTokenizer):
    method __init__ (line 168) | def __init__(
    method do_lower_case (line 260) | def do_lower_case(self):
    method _compile_space_around_punctuation_pattern (line 263) | def _compile_space_around_punctuation_pattern(self):
    method count_file (line 268) | def count_file(self, path, verbose=False, add_eos=False):
    method count_sents (line 284) | def count_sents(self, sents, verbose=False):
    method _build_from_file (line 295) | def _build_from_file(self, vocab_file):
    method save_vocabulary (line 310) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_vocab (line 322) | def build_vocab(self):
    method encode_file (line 343) | def encode_file(self, path, ordered=False, verbose=False, add_eos=True...
    method encode_sents (line 361) | def encode_sents(self, sents, ordered=False, verbose=False):
    method add_special (line 375) | def add_special(self, sym):
    method add_symbol (line 381) | def add_symbol(self, sym):
    method move_added_token (line 386) | def move_added_token(self, token: str, target_idx: int):
    method moses_punct_norm (line 413) | def moses_punct_norm(self, text):
    method moses_tokenize (line 416) | def moses_tokenize(self, text):
    method moses_pipeline (line 421) | def moses_pipeline(self, text: str) -> List[str]:
    method _convert_id_to_token (line 446) | def _convert_id_to_token(self, idx):
    method _convert_token_to_id (line 451) | def _convert_token_to_id(self, sym):
    method convert_tokens_to_string (line 468) | def convert_tokens_to_string(self, tokens):
    method convert_to_tensor (line 477) | def convert_to_tensor(self, symbols):
    method vocab_size (line 481) | def vocab_size(self):
    method get_vocab (line 484) | def get_vocab(self):
    method _tokenize (line 487) | def _tokenize(self, line, add_eos=False, add_double_eos=False):
  class LMOrderedIterator (line 507) | class LMOrderedIterator(object):
    method __init__ (line 508) | def __init__(self, data, bsz, bptt, device="cpu", ext_len=None):
    method get_batch (line 530) | def get_batch(self, i, bptt=None):
    method get_fixlen_iter (line 546) | def get_fixlen_iter(self, start=0):
    method get_varlen_iter (line 550) | def get_varlen_iter(self, start=0, std=5, min_len=5, max_deviation=3):
    method __iter__ (line 562) | def __iter__(self):
  class LMShuffledIterator (line 566) | class LMShuffledIterator(object):
    method __init__ (line 567) | def __init__(self, data, bsz, bptt, device="cpu", ext_len=None, shuffl...
    method get_sent_stream (line 580) | def get_sent_stream(self):
    method stream_iterator (line 589) | def stream_iterator(self, sent_stream):
    method __iter__ (line 636) | def __iter__(self):
  class LMMultiFileIterator (line 644) | class LMMultiFileIterator(LMShuffledIterator):
    method __init__ (line 645) | def __init__(self, paths, vocab, bsz, bptt, device="cpu", ext_len=None...
    method get_sent_stream (line 656) | def get_sent_stream(self, path):
    method __iter__ (line 664) | def __iter__(self):
  class TransfoXLCorpus (line 675) | class TransfoXLCorpus(object):
    method from_pretrained (line 678) | def from_pretrained(cls, pretrained_model_name_or_path, cache_dir=None...
    method __init__ (line 713) | def __init__(self, *args, **kwargs):
    method build_corpus (line 720) | def build_corpus(self, path, dataset):
    method get_iterator (line 754) | def get_iterator(self, split, *args, **kwargs):
  function get_lm_corpus (line 775) | def get_lm_corpus(datadir, dataset):

FILE: transformers/models/trocr/configuration_trocr.py
  class TrOCRConfig (line 31) | class TrOCRConfig(PretrainedConfig):
    method __init__ (line 102) | def __init__(

FILE: transformers/models/trocr/convert_trocr_unilm_to_pytorch.py
  function create_rename_keys (line 43) | def create_rename_keys(encoder_config, decoder_config):
  function read_in_q_k_v (line 88) | def read_in_q_k_v(state_dict, encoder_config):
  function rename_key (line 104) | def rename_key(dct, old, new):
  function prepare_img (line 110) | def prepare_img(checkpoint_url):
  function convert_tr_ocr_checkpoint (line 124) | def convert_tr_ocr_checkpoint(checkpoint_url, pytorch_dump_folder_path):

FILE: transformers/models/trocr/modeling_trocr.py
  function _make_causal_mask (line 47) | def _make_causal_mask(
  function _expand_mask (line 65) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class TrOCRLearnedPositionalEmbedding (line 80) | class TrOCRLearnedPositionalEmbedding(nn.Embedding):
    method __init__ (line 85) | def __init__(self, num_embeddings: int, embedding_dim: int):
    method forward (line 91) | def forward(self, input_ids: torch.Tensor, past_key_values_length: int...
  class TrOCRSinusoidalPositionalEmbedding (line 102) | class TrOCRSinusoidalPositionalEmbedding(nn.Module):
    method __init__ (line 105) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method get_embedding (line 114) | def get_embedding(num_embeddings: int, embedding_dim: int, padding_idx...
    method forward (line 133) | def forward(self, input_ids: torch.Tensor, past_key_values_length: int...
    method create_position_ids_from_input_ids (line 151) | def create_position_ids_from_input_ids(
  class TrOCRAttention (line 164) | class TrOCRAttention(nn.Module):
    method __init__ (line 167) | def __init__(
    method _shape (line 200) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 203) | def forward(
  class TrOCRDecoderLayer (line 313) | class TrOCRDecoderLayer(nn.Module):
    method __init__ (line 314) | def __init__(self, config: TrOCRConfig):
    method forward (line 348) | def forward(
  class TrOCRPreTrainedModel (line 442) | class TrOCRPreTrainedModel(PreTrainedModel):
    method _init_weights (line 447) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 458) | def _set_gradient_checkpointing(self, module, value=False):
  class TrOCRDecoder (line 480) | class TrOCRDecoder(TrOCRPreTrainedModel):
    method __init__ (line 488) | def __init__(self, config: TrOCRConfig):
    method get_input_embeddings (line 517) | def get_input_embeddings(self):
    method set_input_embeddings (line 520) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 523) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 544) | def forward(
  class TrOCRDecoderWrapper (line 771) | class TrOCRDecoderWrapper(TrOCRPreTrainedModel):
    method __init__ (line 777) | def __init__(self, config):
    method forward (line 781) | def forward(self, *args, **kwargs):
  class TrOCRForCausalLM (line 790) | class TrOCRForCausalLM(TrOCRPreTrainedModel):
    method __init__ (line 793) | def __init__(self, config):
    method get_input_embeddings (line 805) | def get_input_embeddings(self):
    method set_input_embeddings (line 808) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 811) | def get_output_embeddings(self):
    method set_output_embeddings (line 814) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 817) | def set_decoder(self, decoder):
    method get_decoder (line 820) | def get_decoder(self):
    method forward (line 824) | def forward(
    method prepare_inputs_for_generation (line 998) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1016) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/trocr/processing_trocr.py
  class TrOCRProcessor (line 24) | class TrOCRProcessor(ProcessorMixin):
    method __init__ (line 42) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 61) | def __call__(self, *args, **kwargs):
    method batch_decode (line 94) | def batch_decode(self, *args, **kwargs):
    method decode (line 101) | def decode(self, *args, **kwargs):
    method as_target_processor (line 109) | def as_target_processor(self):
    method feature_extractor_class (line 125) | def feature_extractor_class(self):
    method feature_extractor (line 133) | def feature_extractor(self):

FILE: transformers/models/tvlt/configuration_tvlt.py
  class TvltConfig (line 28) | class TvltConfig(PretrainedConfig):
    method __init__ (line 115) | def __init__(

FILE: transformers/models/tvlt/feature_extraction_tvlt.py
  class TvltFeatureExtractor (line 30) | class TvltFeatureExtractor(SequenceFeatureExtractor):
    method __init__ (line 59) | def __init__(
    method _np_extract_fbank_features (line 96) | def _np_extract_fbank_features(self, waveform: np.array) -> np.ndarray:
    method __call__ (line 116) | def __call__(

FILE: transformers/models/tvlt/image_processing_tvlt.py
  function make_batched (line 45) | def make_batched(videos) -> List[List[ImageInput]]:
  class TvltImageProcessor (line 68) | class TvltImageProcessor(BaseImageProcessor):
    method __init__ (line 119) | def __init__(
    method resize (line 155) | def resize(
    method center_crop (line 187) | def center_crop(
    method rescale (line 211) | def rescale(
    method normalize (line 232) | def normalize(
    method _preprocess_image (line 261) | def _preprocess_image(
    method preprocess (line 306) | def preprocess(

FILE: transformers/models/tvlt/modeling_tvlt.py
  class TvltModelOutput (line 55) | class TvltModelOutput(ModelOutput):
  class TvltDecoderOutput (line 96) | class TvltDecoderOutput(ModelOutput):
  class TvltForPreTrainingOutput (line 119) | class TvltForPreTrainingOutput(ModelOutput):
  function generate_pixel_mask_noise (line 152) | def generate_pixel_mask_noise(pixel_values, pixel_mask=None, mask_ratio=...
  function generate_audio_mask_noise (line 161) | def generate_audio_mask_noise(audio_values, audio_mask=None, mask_ratio=...
  function random_masking (line 179) | def random_masking(sequence, noise, len_keep, attention_masks=None):
  class TvltPixelEmbeddings (line 208) | class TvltPixelEmbeddings(nn.Module):
    method __init__ (line 211) | def __init__(self, config):
    method forward (line 223) | def forward(self, pixel_values, attention_masks=None):
  class TvltAudioEmbeddings (line 235) | class TvltAudioEmbeddings(nn.Module):
    method __init__ (line 238) | def __init__(self, config):
    method forward (line 252) | def forward(self, audio_values, attention_masks=None):
  class TvltPixelPatchEmbeddings (line 264) | class TvltPixelPatchEmbeddings(nn.Module):
    method __init__ (line 271) | def __init__(self, config):
    method forward (line 287) | def forward(self, pixel_values: torch.Tensor) -> torch.Tensor:
  class TvltAudioPatchEmbeddings (line 305) | class TvltAudioPatchEmbeddings(nn.Module):
    method __init__ (line 312) | def __init__(self, config):
    method forward (line 333) | def forward(self, audio_values: torch.Tensor) -> torch.Tensor:
  class TvltSelfAttention (line 350) | class TvltSelfAttention(nn.Module):
    method __init__ (line 351) | def __init__(self, config):
    method transpose_for_scores (line 369) | def transpose_for_scores(self, x):
    method forward (line 374) | def forward(self, hidden_states, attention_mask=None, head_mask=None, ...
  class TvltSelfOutput (line 411) | class TvltSelfOutput(nn.Module):
    method __init__ (line 417) | def __init__(self, config: TvltConfig) -> None:
    method forward (line 422) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class TvltAttention (line 430) | class TvltAttention(nn.Module):
    method __init__ (line 431) | def __init__(self, config):
    method prune_heads (line 437) | def prune_heads(self, heads):
    method forward (line 455) | def forward(self, hidden_states, attention_mask=None, head_mask=None, ...
  class TvltIntermediate (line 465) | class TvltIntermediate(nn.Module):
    method __init__ (line 466) | def __init__(self, config: TvltConfig) -> None:
    method forward (line 474) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class TvltOutput (line 482) | class TvltOutput(nn.Module):
    method __init__ (line 483) | def __init__(self, config: TvltConfig) -> None:
    method forward (line 488) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class TvltLayer (line 498) | class TvltLayer(nn.Module):
    method __init__ (line 501) | def __init__(self, config):
    method forward (line 511) | def forward(self, hidden_states, attention_mask=None, head_mask=None, ...
  class TvltEncoder (line 537) | class TvltEncoder(nn.Module):
    method __init__ (line 538) | def __init__(self, config):
    method forward (line 544) | def forward(
  class TvltPreTrainedModel (line 596) | class TvltPreTrainedModel(PreTrainedModel):
    method _init_weights (line 607) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 619) | def _set_gradient_checkpointing(self, module, value=False):
  class TvltModel (line 684) | class TvltModel(TvltPreTrainedModel):
    method __init__ (line 685) | def __init__(self, config):
    method get_input_embeddings (line 703) | def get_input_embeddings(self):
    method _prune_heads (line 706) | def _prune_heads(self, heads_to_prune):
    method forward (line 716) | def forward(
  class TvltDecoder (line 847) | class TvltDecoder(nn.Module):
    method __init__ (line 848) | def __init__(self, config):
    method forward (line 865) | def forward(
  class TvltForPreTraining (line 915) | class TvltForPreTraining(TvltPreTrainedModel):
    method __init__ (line 916) | def __init__(self, config):
    method patchify_pixel (line 970) | def patchify_pixel(self, pixel_values):
    method patchify_audio (line 998) | def patchify_audio(self, audio_values):
    method pixel_mae_loss (line 1025) | def pixel_mae_loss(self, pixel_values, pixel_predictions, mask):
    method audio_mae_loss (line 1032) | def audio_mae_loss(self, audio_values, audio_predictions, mask):
    method concatenate_mask (line 1039) | def concatenate_mask(self, mask_token, sequence, ids_restore):
    method forward (line 1050) | def forward(
  class TvltPooler (line 1189) | class TvltPooler(nn.Module):
    method __init__ (line 1190) | def __init__(self, config):
    method forward (line 1195) | def forward(self, hidden_states):
  class TvltMatchingHead (line 1202) | class TvltMatchingHead(nn.Module):
    method __init__ (line 1203) | def __init__(self, config):
    method forward (line 1208) | def forward(self, hidden_states):
  class TvltMAEHead (line 1213) | class TvltMAEHead(nn.Module):
    method __init__ (line 1214) | def __init__(self, config, output_dim=None):
    method forward (line 1219) | def forward(self, hidden_states):
  class TvltForAudioVisualClassification (line 1231) | class TvltForAudioVisualClassification(TvltPreTrainedModel):
    method __init__ (line 1232) | def __init__(self, config):
    method forward (line 1251) | def forward(

FILE: transformers/models/tvlt/processing_tvlt.py
  class TvltProcessor (line 22) | class TvltProcessor(ProcessorMixin):
    method __init__ (line 39) | def __init__(self, image_processor, feature_extractor):
    method __call__ (line 45) | def __call__(
    method model_input_names (line 85) | def model_input_names(self):

FILE: transformers/models/unispeech/configuration_unispeech.py
  class UniSpeechConfig (line 34) | class UniSpeechConfig(PretrainedConfig):
    method __init__ (line 183) | def __init__(
    method inputs_to_logits_ratio (line 302) | def inputs_to_logits_ratio(self):

FILE: transformers/models/unispeech/convert_unispeech_original_pytorch_checkpoint_to_pytorch.py
  function set_recursively (line 69) | def set_recursively(hf_pointer, key, value, full_name, weight_type, is_f...
  function recursively_load_weights (line 106) | def recursively_load_weights(fairseq_model, hf_model, is_finetuned):
  function load_conv_layer (line 150) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function convert_unispeech_checkpoint (line 191) | def convert_unispeech_checkpoint(

FILE: transformers/models/unispeech/modeling_unispeech.py
  class UniSpeechForPreTrainingOutput (line 67) | class UniSpeechForPreTrainingOutput(ModelOutput):
  function _compute_mask_indices (line 103) | def _compute_mask_indices(
  class UniSpeechNoLayerNormConvLayer (line 223) | class UniSpeechNoLayerNormConvLayer(nn.Module):
    method __init__ (line 224) | def __init__(self, config, layer_id=0):
    method forward (line 238) | def forward(self, hidden_states):
  class UniSpeechLayerNormConvLayer (line 245) | class UniSpeechLayerNormConvLayer(nn.Module):
    method __init__ (line 246) | def __init__(self, config, layer_id=0):
    method forward (line 261) | def forward(self, hidden_states):
  class UniSpeechGroupNormConvLayer (line 273) | class UniSpeechGroupNormConvLayer(nn.Module):
    method __init__ (line 274) | def __init__(self, config, layer_id=0):
    method forward (line 290) | def forward(self, hidden_states):
  class UniSpeechPositionalConvEmbedding (line 298) | class UniSpeechPositionalConvEmbedding(nn.Module):
    method __init__ (line 299) | def __init__(self, config):
    method forward (line 326) | def forward(self, hidden_states):
  class UniSpeechSamePadLayer (line 338) | class UniSpeechSamePadLayer(nn.Module):
    method __init__ (line 339) | def __init__(self, num_conv_pos_embeddings):
    method forward (line 343) | def forward(self, hidden_states):
  class UniSpeechFeatureEncoder (line 350) | class UniSpeechFeatureEncoder(nn.Module):
    method __init__ (line 353) | def __init__(self, config):
    method _freeze_parameters (line 373) | def _freeze_parameters(self):
    method forward (line 378) | def forward(self, input_values):
  class UniSpeechFeatureExtractor (line 404) | class UniSpeechFeatureExtractor(UniSpeechFeatureEncoder):
    method __init__ (line 405) | def __init__(self, config):
  class UniSpeechFeatureProjection (line 416) | class UniSpeechFeatureProjection(nn.Module):
    method __init__ (line 417) | def __init__(self, config):
    method forward (line 423) | def forward(self, hidden_states):
  class UniSpeechAttention (line 432) | class UniSpeechAttention(nn.Module):
    method __init__ (line 435) | def __init__(
    method _shape (line 462) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 465) | def forward(
  class UniSpeechFeedForward (line 587) | class UniSpeechFeedForward(nn.Module):
    method __init__ (line 588) | def __init__(self, config):
    method forward (line 601) | def forward(self, hidden_states):
  class UniSpeechEncoderLayer (line 612) | class UniSpeechEncoderLayer(nn.Module):
    method __init__ (line 613) | def __init__(self, config):
    method forward (line 626) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class UniSpeechAttnAdapterLayer (line 647) | class UniSpeechAttnAdapterLayer(nn.Module):
    method __init__ (line 648) | def __init__(self, config):
    method forward (line 662) | def forward(self, hidden_states: torch.FloatTensor):
  class UniSpeechEncoderLayerStableLayerNorm (line 673) | class UniSpeechEncoderLayerStableLayerNorm(nn.Module):
    method __init__ (line 674) | def __init__(self, config):
    method forward (line 692) | def forward(
  class UniSpeechEncoder (line 719) | class UniSpeechEncoder(nn.Module):
    method __init__ (line 720) | def __init__(self, config):
    method forward (line 729) | def forward(
  class UniSpeechEncoderStableLayerNorm (line 807) | class UniSpeechEncoderStableLayerNorm(nn.Module):
    method __init__ (line 808) | def __init__(self, config):
    method forward (line 819) | def forward(
  class UniSpeechGumbelVectorQuantizer (line 898) | class UniSpeechGumbelVectorQuantizer(nn.Module):
    method __init__ (line 904) | def __init__(self, config):
    method _compute_perplexity (line 925) | def _compute_perplexity(probs):
    method forward (line 930) | def forward(self, hidden_states):
  class UniSpeechPreTrainedModel (line 968) | class UniSpeechPreTrainedModel(PreTrainedModel):
    method _init_weights (line 980) | def _init_weights(self, module):
    method _get_feat_extract_output_lengths (line 1013) | def _get_feat_extract_output_lengths(self, input_lengths: Union[torch....
    method _get_feature_vector_attention_mask (line 1028) | def _get_feature_vector_attention_mask(self, feature_vector_length: in...
    method _set_gradient_checkpointing (line 1043) | def _set_gradient_checkpointing(self, module, value=False):
  class UniSpeechModel (line 1108) | class UniSpeechModel(UniSpeechPreTrainedModel):
    method __init__ (line 1109) | def __init__(self, config: UniSpeechConfig):
    method _mask_hidden_states (line 1127) | def _mask_hidden_states(
    method forward (line 1181) | def forward(
  class UniSpeechForPreTraining (line 1232) | class UniSpeechForPreTraining(UniSpeechPreTrainedModel):
    method __init__ (line 1233) | def __init__(self, config: UniSpeechConfig):
    method set_gumbel_temperature (line 1248) | def set_gumbel_temperature(self, temperature: int):
    method freeze_feature_extractor (line 1254) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1266) | def freeze_feature_encoder(self):
    method compute_contrastive_logits (line 1274) | def compute_contrastive_logits(
    method forward (line 1295) | def forward(
  class UniSpeechForCTC (line 1380) | class UniSpeechForCTC(UniSpeechPreTrainedModel):
    method __init__ (line 1381) | def __init__(self, config, target_lang=None):
    method freeze_feature_extractor (line 1409) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1421) | def freeze_feature_encoder(self):
    method forward (line 1436) | def forward(
  class UniSpeechForSequenceClassification (line 1515) | class UniSpeechForSequenceClassification(UniSpeechPreTrainedModel):
    method __init__ (line 1516) | def __init__(self, config):
    method freeze_feature_extractor (line 1534) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1547) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1555) | def freeze_base_model(self):
    method forward (line 1571) | def forward(

FILE: transformers/models/unispeech_sat/configuration_unispeech_sat.py
  class UniSpeechSatConfig (line 34) | class UniSpeechSatConfig(PretrainedConfig):
    method __init__ (line 193) | def __init__(
    method inputs_to_logits_ratio (line 320) | def inputs_to_logits_ratio(self):

FILE: transformers/models/unispeech_sat/convert_unispeech_original_s3prl_checkpoint_to_pytorch.py
  function convert_classification (line 36) | def convert_classification(base_model_name, hf_config, downstream_dict):
  function convert_diarization (line 45) | def convert_diarization(base_model_name, hf_config, downstream_dict):
  function convert_xvector (line 52) | def convert_xvector(base_model_name, hf_config, downstream_dict):
  function convert_s3prl_checkpoint (line 71) | def convert_s3prl_checkpoint(base_model_name, config_path, checkpoint_pa...

FILE: transformers/models/unispeech_sat/convert_unispeech_sat_original_pytorch_checkpoint_to_pytorch.py
  function set_recursively (line 64) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function recursively_load_weights (line 93) | def recursively_load_weights(fairseq_model, hf_model):
  function load_conv_layer (line 140) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function convert_unispeech_sat_checkpoint (line 185) | def convert_unispeech_sat_checkpoint(

FILE: transformers/models/unispeech_sat/modeling_unispeech_sat.py
  class UniSpeechSatForPreTrainingOutput (line 80) | class UniSpeechSatForPreTrainingOutput(ModelOutput):
  function _compute_mask_indices (line 117) | def _compute_mask_indices(
  class UniSpeechSatNoLayerNormConvLayer (line 237) | class UniSpeechSatNoLayerNormConvLayer(nn.Module):
    method __init__ (line 238) | def __init__(self, config, layer_id=0):
    method forward (line 252) | def forward(self, hidden_states):
  class UniSpeechSatLayerNormConvLayer (line 259) | class UniSpeechSatLayerNormConvLayer(nn.Module):
    method __init__ (line 260) | def __init__(self, config, layer_id=0):
    method forward (line 275) | def forward(self, hidden_states):
  class UniSpeechSatGroupNormConvLayer (line 287) | class UniSpeechSatGroupNormConvLayer(nn.Module):
    method __init__ (line 288) | def __init__(self, config, layer_id=0):
    method forward (line 304) | def forward(self, hidden_states):
  class UniSpeechSatPositionalConvEmbedding (line 312) | class UniSpeechSatPositionalConvEmbedding(nn.Module):
    method __init__ (line 313) | def __init__(self, config):
    method forward (line 340) | def forward(self, hidden_states):
  class UniSpeechSatSamePadLayer (line 352) | class UniSpeechSatSamePadLayer(nn.Module):
    method __init__ (line 353) | def __init__(self, num_conv_pos_embeddings):
    method forward (line 357) | def forward(self, hidden_states):
  class UniSpeechSatFeatureEncoder (line 364) | class UniSpeechSatFeatureEncoder(nn.Module):
    method __init__ (line 367) | def __init__(self, config):
    method _freeze_parameters (line 387) | def _freeze_parameters(self):
    method forward (line 392) | def forward(self, input_values):
  class UniSpeechSatFeatureExtractor (line 418) | class UniSpeechSatFeatureExtractor(UniSpeechSatFeatureEncoder):
    method __init__ (line 419) | def __init__(self, config):
  class UniSpeechSatFeatureProjection (line 430) | class UniSpeechSatFeatureProjection(nn.Module):
    method __init__ (line 431) | def __init__(self, config):
    method forward (line 437) | def forward(self, hidden_states):
  class UniSpeechSatAttention (line 446) | class UniSpeechSatAttention(nn.Module):
    method __init__ (line 449) | def __init__(
    method _shape (line 476) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 479) | def forward(
  class UniSpeechSatFeedForward (line 601) | class UniSpeechSatFeedForward(nn.Module):
    method __init__ (line 602) | def __init__(self, config):
    method forward (line 615) | def forward(self, hidden_states):
  class UniSpeechSatEncoderLayer (line 626) | class UniSpeechSatEncoderLayer(nn.Module):
    method __init__ (line 627) | def __init__(self, config):
    method forward (line 640) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class UniSpeechSatAttnAdapterLayer (line 661) | class UniSpeechSatAttnAdapterLayer(nn.Module):
    method __init__ (line 662) | def __init__(self, config):
    method forward (line 676) | def forward(self, hidden_states: torch.FloatTensor):
  class UniSpeechSatEncoderLayerStableLayerNorm (line 687) | class UniSpeechSatEncoderLayerStableLayerNorm(nn.Module):
    method __init__ (line 688) | def __init__(self, config):
    method forward (line 706) | def forward(
  class UniSpeechSatEncoder (line 733) | class UniSpeechSatEncoder(nn.Module):
    method __init__ (line 734) | def __init__(self, config):
    method forward (line 743) | def forward(
  class UniSpeechSatEncoderStableLayerNorm (line 821) | class UniSpeechSatEncoderStableLayerNorm(nn.Module):
    method __init__ (line 822) | def __init__(self, config):
    method forward (line 833) | def forward(
  class UniSpeechSatGumbelVectorQuantizer (line 912) | class UniSpeechSatGumbelVectorQuantizer(nn.Module):
    method __init__ (line 918) | def __init__(self, config):
    method _compute_perplexity (line 939) | def _compute_perplexity(probs, mask=None):
    method forward (line 944) | def forward(self, hidden_states):
  class UniSpeechSatPreTrainedModel (line 982) | class UniSpeechSatPreTrainedModel(PreTrainedModel):
    method _init_weights (line 994) | def _init_weights(self, module):
    method _get_feat_extract_output_lengths (line 1027) | def _get_feat_extract_output_lengths(self, input_lengths: Union[torch....
    method _get_feature_vector_attention_mask (line 1042) | def _get_feature_vector_attention_mask(self, feature_vector_length: in...
    method _set_gradient_checkpointing (line 1057) | def _set_gradient_checkpointing(self, module, value=False):
  class UniSpeechSatModel (line 1124) | class UniSpeechSatModel(UniSpeechSatPreTrainedModel):
    method __init__ (line 1125) | def __init__(self, config: UniSpeechSatConfig):
    method _mask_hidden_states (line 1142) | def _mask_hidden_states(
    method forward (line 1196) | def forward(
  class UniSpeechSatForPreTraining (line 1245) | class UniSpeechSatForPreTraining(UniSpeechSatPreTrainedModel):
    method __init__ (line 1246) | def __init__(self, config: UniSpeechSatConfig):
    method set_gumbel_temperature (line 1268) | def set_gumbel_temperature(self, temperature: int):
    method freeze_feature_extractor (line 1274) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1286) | def freeze_feature_encoder(self):
    method compute_contrastive_logits (line 1294) | def compute_contrastive_logits(
    method forward (line 1315) | def forward(
  class UniSpeechSatForCTC (line 1387) | class UniSpeechSatForCTC(UniSpeechSatPreTrainedModel):
    method __init__ (line 1388) | def __init__(self, config, target_lang=None):
    method freeze_feature_extractor (line 1416) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1428) | def freeze_feature_encoder(self):
    method forward (line 1443) | def forward(
  class UniSpeechSatForSequenceClassification (line 1522) | class UniSpeechSatForSequenceClassification(UniSpeechSatPreTrainedModel):
    method __init__ (line 1523) | def __init__(self, config):
    method freeze_feature_extractor (line 1541) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1554) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1562) | def freeze_base_model(self):
    method forward (line 1578) | def forward(
  class UniSpeechSatForAudioFrameClassification (line 1647) | class UniSpeechSatForAudioFrameClassification(UniSpeechSatPreTrainedModel):
    method __init__ (line 1648) | def __init__(self, config):
    method freeze_feature_extractor (line 1664) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1676) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1683) | def freeze_base_model(self):
    method forward (line 1699) | def forward(
  class AMSoftmaxLoss (line 1754) | class AMSoftmaxLoss(nn.Module):
    method __init__ (line 1755) | def __init__(self, input_dim, num_labels, scale=30.0, margin=0.4):
    method forward (line 1763) | def forward(self, hidden_states, labels):
  class TDNNLayer (line 1778) | class TDNNLayer(nn.Module):
    method __init__ (line 1779) | def __init__(self, config, layer_id=0):
    method forward (line 1789) | def forward(self, hidden_states):
  class UniSpeechSatForXVector (line 1811) | class UniSpeechSatForXVector(UniSpeechSatPreTrainedModel):
    method __init__ (line 1812) | def __init__(self, config):
    method freeze_feature_extractor (line 1831) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1843) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1850) | def freeze_base_model(self):
    method _get_tdnn_output_lengths (line 1858) | def _get_tdnn_output_lengths(self, input_lengths: Union[torch.LongTens...
    method forward (line 1881) | def forward(

FILE: transformers/models/upernet/configuration_upernet.py
  class UperNetConfig (line 27) | class UperNetConfig(PretrainedConfig):
    method __init__ (line 75) | def __init__(
    method to_dict (line 112) | def to_dict(self):

FILE: transformers/models/upernet/convert_convnext_upernet_to_pytorch.py
  function get_upernet_config (line 28) | def get_upernet_config(model_name):
  function create_rename_keys (line 72) | def create_rename_keys(config):
  function rename_key (line 116) | def rename_key(dct, old, new):
  function convert_upernet_checkpoint (line 121) | def convert_upernet_checkpoint(model_name, pytorch_dump_folder_path, pus...

FILE: transformers/models/upernet/convert_swin_upernet_to_pytorch.py
  function get_upernet_config (line 31) | def get_upernet_config(model_name):
  function create_rename_keys (line 82) | def create_rename_keys(config):
  function rename_key (line 128) | def rename_key(dct, old, new):
  function read_in_q_k_v (line 134) | def read_in_q_k_v(state_dict, backbone_config):
  function correct_unfold_reduction_order (line 159) | def correct_unfold_reduction_order(x):
  function reverse_correct_unfold_reduction_order (line 166) | def reverse_correct_unfold_reduction_order(x):
  function correct_unfold_norm_order (line 174) | def correct_unfold_norm_order(x):
  function reverse_correct_unfold_norm_order (line 184) | def reverse_correct_unfold_norm_order(x):
  function convert_upernet_checkpoint (line 191) | def convert_upernet_checkpoint(model_name, pytorch_dump_folder_path, pus...

FILE: transformers/models/upernet/modeling_upernet.py
  class UperNetConvModule (line 40) | class UperNetConvModule(nn.Module):
    method __init__ (line 46) | def __init__(
    method forward (line 67) | def forward(self, input: torch.Tensor) -> torch.Tensor:
  class UperNetPyramidPoolingBlock (line 75) | class UperNetPyramidPoolingBlock(nn.Module):
    method __init__ (line 76) | def __init__(self, pool_scale: int, in_channels: int, channels: int) -...
    method forward (line 85) | def forward(self, input: torch.Tensor) -> torch.Tensor:
  class UperNetPyramidPoolingModule (line 92) | class UperNetPyramidPoolingModule(nn.Module):
    method __init__ (line 107) | def __init__(self, pool_scales: Tuple[int, ...], in_channels: int, cha...
    method forward (line 119) | def forward(self, x: torch.Tensor) -> List[torch.Tensor]:
  class UperNetHead (line 130) | class UperNetHead(nn.Module):
    method __init__ (line 136) | def __init__(self, config, in_channels):
    method init_weights (line 175) | def init_weights(self):
    method _init_weights (line 178) | def _init_weights(self, module):
    method psp_forward (line 184) | def psp_forward(self, inputs):
    method forward (line 193) | def forward(self, encoder_hidden_states: torch.Tensor) -> torch.Tensor:
  class UperNetFCNHead (line 223) | class UperNetFCNHead(nn.Module):
    method __init__ (line 239) | def __init__(
    method init_weights (line 275) | def init_weights(self):
    method _init_weights (line 278) | def _init_weights(self, module):
    method forward (line 284) | def forward(self, encoder_hidden_states: torch.Tensor) -> torch.Tensor:
  class UperNetPreTrainedModel (line 294) | class UperNetPreTrainedModel(PreTrainedModel):
    method _init_weights (line 304) | def _init_weights(self, module):
    method init_weights (line 310) | def init_weights(self):
    method _set_gradient_checkpointing (line 316) | def _set_gradient_checkpointing(self, module, value=False):
  class UperNetForSemanticSegmentation (line 351) | class UperNetForSemanticSegmentation(UperNetPreTrainedModel):
    method __init__ (line 352) | def __init__(self, config):
    method forward (line 366) | def forward(

FILE: transformers/models/van/configuration_van.py
  class VanConfig (line 30) | class VanConfig(PretrainedConfig):
    method __init__ (line 82) | def __init__(

FILE: transformers/models/van/convert_van_to_pytorch.py
  class Tracker (line 43) | class Tracker:
    method _forward_hook (line 48) | def _forward_hook(self, m, inputs: Tensor, outputs: Tensor):
    method __call__ (line 54) | def __call__(self, x: Tensor):
    method parametrized (line 62) | def parametrized(self):
  class ModuleTransfer (line 68) | class ModuleTransfer:
    method __call__ (line 75) | def __call__(self, x: Tensor):
  function copy_parameters (line 98) | def copy_parameters(from_model: nn.Module, our_model: nn.Module) -> nn.M...
  function convert_weight_and_push (line 122) | def convert_weight_and_push(
  function convert_weights_and_push (line 167) | def convert_weights_and_push(save_directory: Path, model_name: str = Non...

FILE: transformers/models/van/modeling_van.py
  function drop_path (line 57) | def drop_path(input, drop_prob: float = 0.0, training: bool = False):
  class VanDropPath (line 78) | class VanDropPath(nn.Module):
    method __init__ (line 81) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 85) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 88) | def extra_repr(self) -> str:
  class VanOverlappingPatchEmbedder (line 92) | class VanOverlappingPatchEmbedder(nn.Module):
    method __init__ (line 99) | def __init__(self, in_channels: int, hidden_size: int, patch_size: int...
    method forward (line 106) | def forward(self, input: torch.Tensor) -> torch.Tensor:
  class VanMlpLayer (line 112) | class VanMlpLayer(nn.Module):
    method __init__ (line 118) | def __init__(
    method forward (line 134) | def forward(self, hidden_state: torch.Tensor) -> torch.Tensor:
  class VanLargeKernelAttention (line 144) | class VanLargeKernelAttention(nn.Module):
    method __init__ (line 149) | def __init__(self, hidden_size: int):
    method forward (line 157) | def forward(self, hidden_state: torch.Tensor) -> torch.Tensor:
  class VanLargeKernelAttentionLayer (line 164) | class VanLargeKernelAttentionLayer(nn.Module):
    method __init__ (line 169) | def __init__(self, hidden_size: int):
    method forward (line 173) | def forward(self, hidden_state: torch.Tensor) -> torch.Tensor:
  class VanSpatialAttentionLayer (line 179) | class VanSpatialAttentionLayer(nn.Module):
    method __init__ (line 185) | def __init__(self, hidden_size: int, hidden_act: str = "gelu"):
    method forward (line 198) | def forward(self, hidden_state: torch.Tensor) -> torch.Tensor:
  class VanLayerScaling (line 207) | class VanLayerScaling(nn.Module):
    method __init__ (line 212) | def __init__(self, hidden_size: int, initial_value: float = 1e-2):
    method forward (line 216) | def forward(self, hidden_state: torch.Tensor) -> torch.Tensor:
  class VanLayer (line 222) | class VanLayer(nn.Module):
    method __init__ (line 227) | def __init__(
    method forward (line 245) | def forward(self, hidden_state: torch.Tensor) -> torch.Tensor:
  class VanStage (line 265) | class VanStage(nn.Module):
    method __init__ (line 270) | def __init__(
    method forward (line 296) | def forward(self, hidden_state: torch.Tensor) -> torch.Tensor:
  class VanEncoder (line 308) | class VanEncoder(nn.Module):
    method __init__ (line 313) | def __init__(self, config: VanConfig):
    method forward (line 343) | def forward(
  class VanPreTrainedModel (line 363) | class VanPreTrainedModel(PreTrainedModel):
    method _init_weights (line 374) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 390) | def _set_gradient_checkpointing(self, module, value=False):
  class VanModel (line 425) | class VanModel(VanPreTrainedModel):
    method __init__ (line 426) | def __init__(self, config):
    method forward (line 443) | def forward(
  class VanForImageClassification (line 480) | class VanForImageClassification(VanPreTrainedModel):
    method __init__ (line 481) | def __init__(self, config):
    method forward (line 499) | def forward(

FILE: transformers/models/videomae/configuration_videomae.py
  class VideoMAEConfig (line 28) | class VideoMAEConfig(PretrainedConfig):
    method __init__ (line 99) | def __init__(

FILE: transformers/models/videomae/convert_videomae_to_pytorch.py
  function get_videomae_config (line 33) | def get_videomae_config(model_name):
  function set_architecture_configs (line 59) | def set_architecture_configs(model_name, config):
  function rename_key (line 91) | def rename_key(name):
  function convert_state_dict (line 138) | def convert_state_dict(orig_state_dict, config):
  function prepare_video (line 171) | def prepare_video():
  function convert_videomae_checkpoint (line 179) | def convert_videomae_checkpoint(checkpoint_url, pytorch_dump_folder_path...

FILE: transformers/models/videomae/feature_extraction_videomae.py
  class VideoMAEFeatureExtractor (line 26) | class VideoMAEFeatureExtractor(VideoMAEImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/videomae/image_processing_videomae.py
  function make_batched (line 50) | def make_batched(videos) -> List[List[ImageInput]]:
  class VideoMAEImageProcessor (line 63) | class VideoMAEImageProcessor(BaseImageProcessor):
    method __init__ (line 103) | def __init__(
    method resize (line 134) | def resize(
    method center_crop (line 166) | def center_crop(
    method rescale (line 190) | def rescale(
    method normalize (line 210) | def normalize(
    method _preprocess_image (line 233) | def _preprocess_image(
    method preprocess (line 279) | def preprocess(

FILE: transformers/models/videomae/modeling_videomae.py
  class VideoMAEDecoderOutput (line 57) | class VideoMAEDecoderOutput(ModelOutput):
  class VideoMAEForPreTrainingOutput (line 80) | class VideoMAEForPreTrainingOutput(ModelOutput):
  function get_sinusoid_encoding_table (line 107) | def get_sinusoid_encoding_table(n_position, d_hid):
  class VideoMAEEmbeddings (line 121) | class VideoMAEEmbeddings(nn.Module):
    method __init__ (line 127) | def __init__(self, config):
    method forward (line 136) | def forward(self, pixel_values, bool_masked_pos):
  class VideoMAEPatchEmbeddings (line 153) | class VideoMAEPatchEmbeddings(nn.Module):
    method __init__ (line 163) | def __init__(self, config):
    method forward (line 190) | def forward(self, pixel_values):
  class VideoMAESelfAttention (line 206) | class VideoMAESelfAttention(nn.Module):
    method __init__ (line 207) | def __init__(self, config: VideoMAEConfig) -> None:
    method transpose_for_scores (line 232) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 237) | def forward(
  class VideoMAESelfOutput (line 277) | class VideoMAESelfOutput(nn.Module):
    method __init__ (line 283) | def __init__(self, config: VideoMAEConfig) -> None:
    method forward (line 288) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class VideoMAEAttention (line 296) | class VideoMAEAttention(nn.Module):
    method __init__ (line 297) | def __init__(self, config: VideoMAEConfig) -> None:
    method prune_heads (line 303) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 321) | def forward(
  class VideoMAEIntermediate (line 336) | class VideoMAEIntermediate(nn.Module):
    method __init__ (line 337) | def __init__(self, config: VideoMAEConfig) -> None:
    method forward (line 345) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class VideoMAEOutput (line 353) | class VideoMAEOutput(nn.Module):
    method __init__ (line 354) | def __init__(self, config: VideoMAEConfig) -> None:
    method forward (line 359) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class VideoMAELayer (line 369) | class VideoMAELayer(nn.Module):
    method __init__ (line 372) | def __init__(self, config: VideoMAEConfig) -> None:
    method forward (line 382) | def forward(
  class VideoMAEEncoder (line 412) | class VideoMAEEncoder(nn.Module):
    method __init__ (line 413) | def __init__(self, config: VideoMAEConfig) -> None:
    method forward (line 419) | def forward(
  class VideoMAEPreTrainedModel (line 469) | class VideoMAEPreTrainedModel(PreTrainedModel):
    method _init_weights (line 480) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 492) | def _set_gradient_checkpointing(self, module, value=False):
  class VideoMAEModel (line 535) | class VideoMAEModel(VideoMAEPreTrainedModel):
    method __init__ (line 536) | def __init__(self, config):
    method get_input_embeddings (line 551) | def get_input_embeddings(self):
    method _prune_heads (line 554) | def _prune_heads(self, heads_to_prune):
    method forward (line 564) | def forward(
  class VideoMAEDecoder (line 681) | class VideoMAEDecoder(nn.Module):
    method __init__ (line 682) | def __init__(self, config, num_patches):
    method forward (line 704) | def forward(
  class VideoMAEForPreTraining (line 759) | class VideoMAEForPreTraining(VideoMAEPreTrainedModel):
    method __init__ (line 760) | def __init__(self, config):
    method forward (line 779) | def forward(
  class VideoMAEForVideoClassification (line 943) | class VideoMAEForVideoClassification(VideoMAEPreTrainedModel):
    method __init__ (line 944) | def __init__(self, config):
    method forward (line 959) | def forward(

FILE: transformers/models/vilt/configuration_vilt.py
  class ViltConfig (line 28) | class ViltConfig(PretrainedConfig):
    method __init__ (line 101) | def __init__(

FILE: transformers/models/vilt/convert_vilt_original_to_pytorch.py
  function create_rename_keys (line 45) | def create_rename_keys(config, vqa_model=False, nlvr_model=False, irtr_m...
  function read_in_q_k_v (line 134) | def read_in_q_k_v(state_dict, config):
  function remove_classification_head_ (line 157) | def remove_classification_head_(state_dict):
  function rename_key (line 163) | def rename_key(dct, old, new):
  function convert_vilt_checkpoint (line 169) | def convert_vilt_checkpoint(checkpoint_url, pytorch_dump_folder_path):

FILE: transformers/models/vilt/feature_extraction_vilt.py
  class ViltFeatureExtractor (line 26) | class ViltFeatureExtractor(ViltImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/vilt/image_processing_vilt.py
  function max_across_indices (line 46) | def max_across_indices(values: Iterable[Any]) -> List[Any]:
  function make_pixel_mask (line 53) | def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> ...
  function get_max_height_width (line 69) | def get_max_height_width(images: List[np.ndarray]) -> List[int]:
  function get_resize_output_image_size (line 84) | def get_resize_output_image_size(
  class ViltImageProcessor (line 111) | class ViltImageProcessor(BaseImageProcessor):
    method __init__ (line 153) | def __init__(
    method from_dict (line 186) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method resize (line 197) | def resize(
    method rescale (line 233) | def rescale(
    method normalize (line 253) | def normalize(
    method _pad_image (line 276) | def _pad_image(
    method pad (line 297) | def pad(
    method pad_and_create_pixel_mask (line 334) | def pad_and_create_pixel_mask(
    method preprocess (line 369) | def preprocess(

FILE: transformers/models/vilt/modeling_vilt.py
  class ViltForImagesAndTextClassificationOutput (line 65) | class ViltForImagesAndTextClassificationOutput(ModelOutput):
  class ViltEmbeddings (line 90) | class ViltEmbeddings(nn.Module):
    method __init__ (line 99) | def __init__(self, config):
    method visual_embed (line 114) | def visual_embed(self, pixel_values, pixel_mask, max_image_length=200):
    method forward (line 202) | def forward(
  class TextEmbeddings (line 244) | class TextEmbeddings(nn.Module):
    method __init__ (line 247) | def __init__(self, config):
    method forward (line 264) | def forward(self, input_ids=None, token_type_ids=None, position_ids=No...
  class ViltPatchEmbeddings (line 299) | class ViltPatchEmbeddings(nn.Module):
    method __init__ (line 304) | def __init__(self, config):
    method forward (line 319) | def forward(self, pixel_values):
  class ViltSelfAttention (line 329) | class ViltSelfAttention(nn.Module):
    method __init__ (line 330) | def __init__(self, config):
    method transpose_for_scores (line 348) | def transpose_for_scores(self, x):
    method forward (line 353) | def forward(self, hidden_states, attention_mask=None, head_mask=None, ...
  class ViltSelfOutput (line 390) | class ViltSelfOutput(nn.Module):
    method __init__ (line 396) | def __init__(self, config: ViltConfig) -> None:
    method forward (line 401) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ViltAttention (line 408) | class ViltAttention(nn.Module):
    method __init__ (line 409) | def __init__(self, config):
    method prune_heads (line 415) | def prune_heads(self, heads):
    method forward (line 433) | def forward(self, hidden_states, attention_mask=None, head_mask=None, ...
  class ViltIntermediate (line 443) | class ViltIntermediate(nn.Module):
    method __init__ (line 444) | def __init__(self, config: ViltConfig) -> None:
    method forward (line 452) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ViltOutput (line 460) | class ViltOutput(nn.Module):
    method __init__ (line 461) | def __init__(self, config: ViltConfig) -> None:
    method forward (line 466) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ViltLayer (line 475) | class ViltLayer(nn.Module):
    method __init__ (line 478) | def __init__(self, config):
    method forward (line 488) | def forward(self, hidden_states, attention_mask=None, head_mask=None, ...
  class ViltEncoder (line 513) | class ViltEncoder(nn.Module):
    method __init__ (line 514) | def __init__(self, config):
    method forward (line 520) | def forward(
  class ViltPreTrainedModel (line 572) | class ViltPreTrainedModel(PreTrainedModel):
    method _init_weights (line 583) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 599) | def _set_gradient_checkpointing(self, module, value=False):
  class ViltModel (line 730) | class ViltModel(ViltPreTrainedModel):
    method __init__ (line 731) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 744) | def get_input_embeddings(self):
    method set_input_embeddings (line 747) | def set_input_embeddings(self, value):
    method _prune_heads (line 750) | def _prune_heads(self, heads_to_prune):
    method forward (line 760) | def forward(
  class ViltPooler (line 874) | class ViltPooler(nn.Module):
    method __init__ (line 875) | def __init__(self, config):
    method forward (line 880) | def forward(self, hidden_states):
  class ViltForMaskedLM (line 895) | class ViltForMaskedLM(ViltPreTrainedModel):
    method __init__ (line 898) | def __init__(self, config):
    method get_output_embeddings (line 907) | def get_output_embeddings(self):
    method set_output_embeddings (line 910) | def set_output_embeddings(self, new_embeddings):
    method forward (line 915) | def forward(
  class ViltPredictionHeadTransform (line 1028) | class ViltPredictionHeadTransform(nn.Module):
    method __init__ (line 1029) | def __init__(self, config):
    method forward (line 1038) | def forward(self, hidden_states):
  class ViltMLMHead (line 1045) | class ViltMLMHead(nn.Module):
    method __init__ (line 1046) | def __init__(self, config, weight=None):
    method forward (line 1058) | def forward(self, x):
  class ViltForQuestionAnswering (line 1071) | class ViltForQuestionAnswering(ViltPreTrainedModel):
    method __init__ (line 1072) | def __init__(self, config):
    method forward (line 1091) | def forward(
  class ViltForImageAndTextRetrieval (line 1184) | class ViltForImageAndTextRetrieval(ViltPreTrainedModel):
    method __init__ (line 1185) | def __init__(self, config):
    method forward (line 1198) | def forward(
  class ViltForImagesAndTextClassification (line 1285) | class ViltForImagesAndTextClassification(ViltPreTrainedModel):
    method __init__ (line 1286) | def __init__(self, config):
    method forward (line 1306) | def forward(
  class ViltForTokenClassification (line 1427) | class ViltForTokenClassification(ViltPreTrainedModel):
    method __init__ (line 1430) | def __init__(self, config):
    method forward (line 1444) | def forward(

FILE: transformers/models/vilt/processing_vilt.py
  class ViltProcessor (line 27) | class ViltProcessor(ProcessorMixin):
    method __init__ (line 44) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 62) | def __call__(
    method batch_decode (line 112) | def batch_decode(self, *args, **kwargs):
    method decode (line 119) | def decode(self, *args, **kwargs):
    method model_input_names (line 127) | def model_input_names(self):
    method feature_extractor_class (line 133) | def feature_extractor_class(self):
    method feature_extractor (line 141) | def feature_extractor(self):

FILE: transformers/models/vision_encoder_decoder/configuration_vision_encoder_decoder.py
  class VisionEncoderDecoderConfig (line 34) | class VisionEncoderDecoderConfig(PretrainedConfig):
    method __init__ (line 83) | def __init__(self, **kwargs):
    method from_encoder_decoder_configs (line 101) | def from_encoder_decoder_configs(
    method to_dict (line 117) | def to_dict(self):
  class VisionEncoderDecoderEncoderOnnxConfig (line 131) | class VisionEncoderDecoderEncoderOnnxConfig(OnnxConfig):
    method inputs (line 135) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 143) | def atol_for_validation(self) -> float:
    method outputs (line 147) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
  class VisionEncoderDecoderDecoderOnnxConfig (line 151) | class VisionEncoderDecoderDecoderOnnxConfig(OnnxConfig):
    method inputs (line 153) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method generate_dummy_inputs (line 161) | def generate_dummy_inputs(
  class VisionEncoderDecoderOnnxConfig (line 186) | class VisionEncoderDecoderOnnxConfig(OnnxConfig):
    method inputs (line 188) | def inputs(self) -> None:
    method get_encoder_config (line 191) | def get_encoder_config(self, encoder_config: PretrainedConfig) -> Onnx...
    method get_decoder_config (line 204) | def get_decoder_config(

FILE: transformers/models/vision_encoder_decoder/modeling_flax_vision_encoder_decoder.py
  class FlaxVisionEncoderDecoderModule (line 170) | class FlaxVisionEncoderDecoderModule(nn.Module):
    method setup (line 174) | def setup(self):
    method _get_encoder_module (line 200) | def _get_encoder_module(self):
    method _get_projection_module (line 203) | def _get_projection_module(self):
    method _get_decoder_module (line 206) | def _get_decoder_module(self):
    method __call__ (line 209) | def __call__(
  class FlaxVisionEncoderDecoderModel (line 268) | class FlaxVisionEncoderDecoderModel(FlaxPreTrainedModel):
    method __init__ (line 280) | def __init__(
    method init_weights (line 313) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 353) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 396) | def encode(
    method decode (line 470) | def decode(
    method __call__ (line 600) | def __call__(
    method prepare_inputs_for_generation (line 687) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 718) | def update_inputs_for_generation(self, model_outputs, model_kwargs):
    method from_encoder_decoder_pretrained (line 724) | def from_encoder_decoder_pretrained(

FILE: transformers/models/vision_encoder_decoder/modeling_tf_vision_encoder_decoder.py
  function shift_tokens_right (line 149) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int, decoder_...
  class TFVisionEncoderDecoderModel (line 176) | class TFVisionEncoderDecoderModel(TFPreTrainedModel, TFCausalLanguageMod...
    method __init__ (line 188) | def __init__(
    method input_signature (line 256) | def input_signature(self):
    method get_encoder (line 277) | def get_encoder(self):
    method get_decoder (line 280) | def get_decoder(self):
    method get_input_embeddings (line 283) | def get_input_embeddings(self):
    method get_output_embeddings (line 286) | def get_output_embeddings(self):
    method set_output_embeddings (line 289) | def set_output_embeddings(self, new_embeddings):
    method from_pretrained (line 293) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
    method from_encoder_decoder_pretrained (line 339) | def from_encoder_decoder_pretrained(
    method call (line 491) | def call(
    method serving_output (line 662) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 689) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 707) | def prepare_decoder_input_ids_from_labels(self, labels: tf.Tensor):
    method resize_token_embeddings (line 710) | def resize_token_embeddings(self, *args, **kwargs):

FILE: transformers/models/vision_encoder_decoder/modeling_vision_encoder_decoder.py
  function shift_tokens_right (line 37) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  class VisionEncoderDecoderModel (line 151) | class VisionEncoderDecoderModel(PreTrainedModel):
    method __init__ (line 163) | def __init__(
    method _set_gradient_checkpointing (line 228) | def _set_gradient_checkpointing(self, module, value=False):
    method get_encoder (line 233) | def get_encoder(self):
    method get_decoder (line 236) | def get_decoder(self):
    method get_output_embeddings (line 239) | def get_output_embeddings(self):
    method set_output_embeddings (line 242) | def set_output_embeddings(self, new_embeddings):
    method from_pretrained (line 246) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
    method from_encoder_decoder_pretrained (line 366) | def from_encoder_decoder_pretrained(
    method forward (line 521) | def forward(
    method prepare_decoder_input_ids_from_labels (line 648) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method prepare_inputs_for_generation (line 651) | def prepare_inputs_for_generation(
    method resize_token_embeddings (line 666) | def resize_token_embeddings(self, *args, **kwargs):
    method _reorder_cache (line 672) | def _reorder_cache(self, past_key_values, beam_idx):

FILE: transformers/models/vision_text_dual_encoder/configuration_vision_text_dual_encoder.py
  class VisionTextDualEncoderConfig (line 28) | class VisionTextDualEncoderConfig(PretrainedConfig):
    method __init__ (line 78) | def __init__(self, projection_dim=512, logit_scale_init_value=2.6592, ...
    method from_vision_text_configs (line 106) | def from_vision_text_configs(cls, vision_config: PretrainedConfig, tex...
    method to_dict (line 117) | def to_dict(self):

FILE: transformers/models/vision_text_dual_encoder/modeling_flax_vision_text_dual_encoder.py
  class FlaxVisionTextDualEncoderModule (line 122) | class FlaxVisionTextDualEncoderModule(nn.Module):
    method setup (line 126) | def setup(self):
    method __call__ (line 157) | def __call__(
  class FlaxVisionTextDualEncoderModel (line 219) | class FlaxVisionTextDualEncoderModel(FlaxPreTrainedModel):
    method __init__ (line 223) | def __init__(
    method init_weights (line 243) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 269) | def __call__(
    method get_text_features (line 319) | def get_text_features(
    method get_image_features (line 381) | def get_image_features(
    method from_vision_text_pretrained (line 415) | def from_vision_text_pretrained(

FILE: transformers/models/vision_text_dual_encoder/modeling_tf_vision_text_dual_encoder.py
  function contrastive_loss (line 160) | def contrastive_loss(logits: tf.Tensor) -> tf.Tensor:
  function clip_loss (line 169) | def clip_loss(similarity: tf.Tensor) -> tf.Tensor:
  class TFVisionTextDualEncoderModel (line 176) | class TFVisionTextDualEncoderModel(TFPreTrainedModel):
    method __init__ (line 181) | def __init__(
    method build (line 224) | def build(self, input_shape=None):
    method from_pretrained (line 231) | def from_pretrained(cls, pretrained_model_name_or_path, *model_args, *...
    method get_text_features (line 258) | def get_text_features(
    method get_image_features (line 300) | def get_image_features(
    method call (line 344) | def call(
    method from_vision_text_pretrained (line 462) | def from_vision_text_pretrained(
    method dummy_inputs (line 600) | def dummy_inputs(self):

FILE: transformers/models/vision_text_dual_encoder/modeling_vision_text_dual_encoder.py
  function contrastive_loss (line 150) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  function clip_loss (line 155) | def clip_loss(similarity: torch.Tensor) -> torch.Tensor:
  class VisionTextDualEncoderModel (line 162) | class VisionTextDualEncoderModel(PreTrainedModel):
    method __init__ (line 166) | def __init__(
    method get_text_features (line 210) | def get_text_features(
    method get_image_features (line 252) | def get_image_features(
    method forward (line 295) | def forward(
    method from_pretrained (line 408) | def from_pretrained(cls, *args, **kwargs):
    method from_vision_text_pretrained (line 415) | def from_vision_text_pretrained(

FILE: transformers/models/vision_text_dual_encoder/processing_vision_text_dual_encoder.py
  class VisionTextDualEncoderProcessor (line 25) | class VisionTextDualEncoderProcessor(ProcessorMixin):
    method __init__ (line 44) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 62) | def __call__(self, text=None, images=None, return_tensors=None, **kwar...
    method batch_decode (line 115) | def batch_decode(self, *args, **kwargs):
    method decode (line 122) | def decode(self, *args, **kwargs):
    method model_input_names (line 130) | def model_input_names(self):
    method feature_extractor_class (line 136) | def feature_extractor_class(self):
    method feature_extractor (line 144) | def feature_extractor(self):

FILE: transformers/models/visual_bert/configuration_visual_bert.py
  class VisualBertConfig (line 43) | class VisualBertConfig(PretrainedConfig):
    method __init__ (line 113) | def __init__(

FILE: transformers/models/visual_bert/convert_visual_bert_original_pytorch_checkpoint_to_pytorch.py
  function load_state_dict (line 59) | def load_state_dict(checkpoint_path):
  function get_new_dict (line 64) | def get_new_dict(d, config, rename_keys_prefix=rename_keys_prefix):
  function convert_visual_bert_checkpoint (line 83) | def convert_visual_bert_checkpoint(checkpoint_path, pytorch_dump_folder_...

FILE: transformers/models/visual_bert/modeling_visual_bert.py
  class VisualBertEmbeddings (line 65) | class VisualBertEmbeddings(nn.Module):
    method __init__ (line 68) | def __init__(self, config):
    method forward (line 98) | def forward(
  class VisualBertSelfAttention (line 197) | class VisualBertSelfAttention(nn.Module):
    method __init__ (line 198) | def __init__(self, config):
    method transpose_for_scores (line 216) | def transpose_for_scores(self, x):
    method forward (line 221) | def forward(
  class VisualBertSelfOutput (line 266) | class VisualBertSelfOutput(nn.Module):
    method __init__ (line 267) | def __init__(self, config):
    method forward (line 273) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class VisualBertAttention (line 280) | class VisualBertAttention(nn.Module):
    method __init__ (line 281) | def __init__(self, config):
    method prune_heads (line 287) | def prune_heads(self, heads):
    method forward (line 305) | def forward(
  class VisualBertIntermediate (line 324) | class VisualBertIntermediate(nn.Module):
    method __init__ (line 325) | def __init__(self, config):
    method forward (line 333) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class VisualBertOutput (line 340) | class VisualBertOutput(nn.Module):
    method __init__ (line 341) | def __init__(self, config):
    method forward (line 347) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class VisualBertLayer (line 354) | class VisualBertLayer(nn.Module):
    method __init__ (line 355) | def __init__(self, config):
    method forward (line 363) | def forward(
    method feed_forward_chunk (line 387) | def feed_forward_chunk(self, attention_output):
  class VisualBertEncoder (line 393) | class VisualBertEncoder(nn.Module):
    method __init__ (line 394) | def __init__(self, config):
    method forward (line 400) | def forward(
  class VisualBertPooler (line 458) | class VisualBertPooler(nn.Module):
    method __init__ (line 459) | def __init__(self, config):
    method forward (line 464) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class VisualBertPredictionHeadTransform (line 474) | class VisualBertPredictionHeadTransform(nn.Module):
    method __init__ (line 475) | def __init__(self, config):
    method forward (line 484) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class VisualBertLMPredictionHead (line 492) | class VisualBertLMPredictionHead(nn.Module):
    method __init__ (line 493) | def __init__(self, config):
    method forward (line 506) | def forward(self, hidden_states):
  class VisualBertPreTrainingHeads (line 513) | class VisualBertPreTrainingHeads(nn.Module):
    method __init__ (line 514) | def __init__(self, config):
    method forward (line 519) | def forward(self, sequence_output, pooled_output):
  class VisualBertPreTrainedModel (line 525) | class VisualBertPreTrainedModel(PreTrainedModel):
    method _init_weights (line 536) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 549) | def _set_gradient_checkpointing(self, module, value=False):
  class VisualBertForPreTrainingOutput (line 555) | class VisualBertForPreTrainingOutput(ModelOutput):
  class VisualBertModel (line 677) | class VisualBertModel(VisualBertPreTrainedModel):
    method __init__ (line 685) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 702) | def get_input_embeddings(self):
    method set_input_embeddings (line 705) | def set_input_embeddings(self, value):
    method _prune_heads (line 708) | def _prune_heads(self, heads_to_prune):
    method forward (line 718) | def forward(
  class VisualBertForPreTraining (line 873) | class VisualBertForPreTraining(VisualBertPreTrainedModel):
    method __init__ (line 876) | def __init__(self, config):
    method get_output_embeddings (line 885) | def get_output_embeddings(self):
    method set_output_embeddings (line 888) | def set_output_embeddings(self, new_embeddings):
    method forward (line 893) | def forward(
  class VisualBertForMultipleChoice (line 1024) | class VisualBertForMultipleChoice(VisualBertPreTrainedModel):
    method __init__ (line 1025) | def __init__(self, config):
    method forward (line 1039) | def forward(
  class VisualBertForQuestionAnswering (line 1177) | class VisualBertForQuestionAnswering(VisualBertPreTrainedModel):
    method __init__ (line 1178) | def __init__(self, config):
    method forward (line 1191) | def forward(
  class VisualBertForVisualReasoning (line 1303) | class VisualBertForVisualReasoning(VisualBertPreTrainedModel):
    method __init__ (line 1304) | def __init__(self, config):
    method forward (line 1317) | def forward(
  class VisualBertRegionToPhraseAttention (line 1412) | class VisualBertRegionToPhraseAttention(nn.Module):
    method __init__ (line 1413) | def __init__(self, config):
    method transpose_for_scores (line 1430) | def transpose_for_scores(self, x):
    method forward (line 1435) | def forward(self, query, key, attention_mask):
  class VisualBertForRegionToPhraseAlignment (line 1463) | class VisualBertForRegionToPhraseAlignment(VisualBertPreTrainedModel):
    method __init__ (line 1466) | def __init__(self, config):
    method forward (line 1479) | def forward(

FILE: transformers/models/vit/configuration_vit.py
  class ViTConfig (line 35) | class ViTConfig(PretrainedConfig):
    method __init__ (line 93) | def __init__(
  class ViTOnnxConfig (line 129) | class ViTOnnxConfig(OnnxConfig):
    method inputs (line 133) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 141) | def atol_for_validation(self) -> float:

FILE: transformers/models/vit/convert_dino_to_pytorch.py
  function create_rename_keys (line 36) | def create_rename_keys(config, base_model=False):
  function read_in_q_k_v (line 87) | def read_in_q_k_v(state_dict, config, base_model=False):
  function remove_classification_head_ (line 113) | def remove_classification_head_(state_dict):
  function rename_key (line 119) | def rename_key(dct, old, new):
  function prepare_img (line 125) | def prepare_img():
  function convert_vit_checkpoint (line 132) | def convert_vit_checkpoint(model_name, pytorch_dump_folder_path, base_mo...

FILE: transformers/models/vit/convert_vit_timm_to_pytorch.py
  function create_rename_keys (line 37) | def create_rename_keys(config, base_model=False):
  function read_in_q_k_v (line 90) | def read_in_q_k_v(state_dict, config, base_model=False):
  function remove_classification_head_ (line 116) | def remove_classification_head_(state_dict):
  function rename_key (line 122) | def rename_key(dct, old, new):
  function prepare_img (line 128) | def prepare_img():
  function convert_vit_checkpoint (line 135) | def convert_vit_checkpoint(vit_name, pytorch_dump_folder_path):

FILE: transformers/models/vit/feature_extraction_vit.py
  class ViTFeatureExtractor (line 26) | class ViTFeatureExtractor(ViTImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/vit/image_processing_vit.py
  class ViTImageProcessor (line 39) | class ViTImageProcessor(BaseImageProcessor):
    method __init__ (line 72) | def __init__(
    method resize (line 96) | def resize(
    method rescale (line 130) | def rescale(
    method normalize (line 152) | def normalize(
    method preprocess (line 181) | def preprocess(

FILE: transformers/models/vit/modeling_flax_vit.py
  class FlaxViTPatchEmbeddings (line 87) | class FlaxViTPatchEmbeddings(nn.Module):
    method setup (line 91) | def setup(self):
    method __call__ (line 108) | def __call__(self, pixel_values):
  class FlaxViTEmbeddings (line 119) | class FlaxViTEmbeddings(nn.Module):
    method setup (line 125) | def setup(self):
    method __call__ (line 140) | def __call__(self, pixel_values, deterministic=True):
  class FlaxViTSelfAttention (line 152) | class FlaxViTSelfAttention(nn.Module):
    method setup (line 156) | def setup(self):
    method __call__ (line 188) | def __call__(self, hidden_states, deterministic: bool = True, output_a...
  class FlaxViTSelfOutput (line 223) | class FlaxViTSelfOutput(nn.Module):
    method setup (line 227) | def setup(self):
    method __call__ (line 237) | def __call__(self, hidden_states, input_tensor, deterministic: bool = ...
  class FlaxViTAttention (line 243) | class FlaxViTAttention(nn.Module):
    method setup (line 247) | def setup(self):
    method __call__ (line 251) | def __call__(self, hidden_states, deterministic=True, output_attention...
  class FlaxViTIntermediate (line 264) | class FlaxViTIntermediate(nn.Module):
    method setup (line 268) | def setup(self):
    method __call__ (line 278) | def __call__(self, hidden_states):
  class FlaxViTOutput (line 284) | class FlaxViTOutput(nn.Module):
    method setup (line 288) | def setup(self):
    method __call__ (line 298) | def __call__(self, hidden_states, attention_output, deterministic: boo...
  class FlaxViTLayer (line 305) | class FlaxViTLayer(nn.Module):
    method setup (line 309) | def setup(self):
    method __call__ (line 316) | def __call__(self, hidden_states, deterministic: bool = True, output_a...
  class FlaxViTLayerCollection (line 341) | class FlaxViTLayerCollection(nn.Module):
    method setup (line 345) | def setup(self):
    method __call__ (line 350) | def __call__(
  class FlaxViTEncoder (line 384) | class FlaxViTEncoder(nn.Module):
    method setup (line 388) | def setup(self):
    method __call__ (line 391) | def __call__(
  class FlaxViTPooler (line 408) | class FlaxViTPooler(nn.Module):
    method setup (line 412) | def setup(self):
    method __call__ (line 421) | def __call__(self, hidden_states):
  class FlaxViTPreTrainedModel (line 427) | class FlaxViTPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 438) | def __init__(
    method init_weights (line 452) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 472) | def __call__(
  class FlaxViTModule (line 505) | class FlaxViTModule(nn.Module):
    method setup (line 510) | def setup(self):
    method __call__ (line 516) | def __call__(
  class FlaxViTModel (line 555) | class FlaxViTModel(FlaxViTPreTrainedModel):
  class FlaxViTForImageClassificationModule (line 585) | class FlaxViTForImageClassificationModule(nn.Module):
    method setup (line 589) | def setup(self):
    method __call__ (line 599) | def __call__(
  class FlaxViTForImageClassification (line 638) | class FlaxViTForImageClassification(FlaxViTPreTrainedModel):

FILE: transformers/models/vit/modeling_tf_vit.py
  class TFViTEmbeddings (line 56) | class TFViTEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 62) | def __init__(self, config: ViTConfig, **kwargs):
    method build (line 69) | def build(self, input_shape: tf.TensorShape):
    method interpolate_pos_encoding (line 86) | def interpolate_pos_encoding(self, embeddings, height, width) -> tf.Te...
    method call (line 120) | def call(
  class TFViTPatchEmbeddings (line 145) | class TFViTPatchEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 152) | def __init__(self, config: ViTConfig, **kwargs):
    method call (line 178) | def call(
  class TFViTSelfAttention (line 209) | class TFViTSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 210) | def __init__(self, config: ViTConfig, **kwargs):
    method transpose_for_scores (line 235) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 242) | def call(
  class TFViTSelfOutput (line 284) | class TFViTSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 290) | def __init__(self, config: ViTConfig, **kwargs):
    method call (line 298) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFViTAttention (line 305) | class TFViTAttention(tf.keras.layers.Layer):
    method __init__ (line 306) | def __init__(self, config: ViTConfig, **kwargs):
    method prune_heads (line 312) | def prune_heads(self, heads):
    method call (line 315) | def call(
  class TFViTIntermediate (line 333) | class TFViTIntermediate(tf.keras.layers.Layer):
    method __init__ (line 334) | def __init__(self, config: ViTConfig, **kwargs):
    method call (line 346) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFViTOutput (line 353) | class TFViTOutput(tf.keras.layers.Layer):
    method __init__ (line 354) | def __init__(self, config: ViTConfig, **kwargs):
    method call (line 362) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFViTLayer (line 370) | class TFViTLayer(tf.keras.layers.Layer):
    method __init__ (line 373) | def __init__(self, config: ViTConfig, **kwargs):
    method call (line 387) | def call(
  class TFViTEncoder (line 420) | class TFViTEncoder(tf.keras.layers.Layer):
    method __init__ (line 421) | def __init__(self, config: ViTConfig, **kwargs):
    method call (line 426) | def call(
  class TFViTMainLayer (line 466) | class TFViTMainLayer(tf.keras.layers.Layer):
    method __init__ (line 469) | def __init__(self, config: ViTConfig, add_pooling_layer: bool = True, ...
    method get_input_embeddings (line 479) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method _prune_heads (line 482) | def _prune_heads(self, heads_to_prune):
    method call (line 490) | def call(
  class TFViTPreTrainedModel (line 543) | class TFViTPreTrainedModel(TFPreTrainedModel):
  class TFViTModel (line 631) | class TFViTModel(TFViTPreTrainedModel):
    method __init__ (line 632) | def __init__(self, config: ViTConfig, *inputs, add_pooling_layer=True,...
    method call (line 646) | def call(
  class TFViTPooler (line 669) | class TFViTPooler(tf.keras.layers.Layer):
    method __init__ (line 670) | def __init__(self, config: ViTConfig, **kwargs):
    method call (line 680) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFViTForImageClassification (line 704) | class TFViTForImageClassification(TFViTPreTrainedModel, TFSequenceClassi...
    method __init__ (line 705) | def __init__(self, config: ViTConfig, *inputs, **kwargs):
    method call (line 726) | def call(

FILE: transformers/models/vit/modeling_vit.py
  class ViTEmbeddings (line 66) | class ViTEmbeddings(nn.Module):
    method __init__ (line 71) | def __init__(self, config: ViTConfig, use_mask_token: bool = False) ->...
    method interpolate_pos_encoding (line 82) | def interpolate_pos_encoding(self, embeddings: torch.Tensor, height: i...
    method forward (line 115) | def forward(
  class ViTPatchEmbeddings (line 146) | class ViTPatchEmbeddings(nn.Module):
    method __init__ (line 153) | def __init__(self, config):
    method forward (line 168) | def forward(self, pixel_values: torch.Tensor, interpolate_pos_encoding...
  class ViTSelfAttention (line 185) | class ViTSelfAttention(nn.Module):
    method __init__ (line 186) | def __init__(self, config: ViTConfig) -> None:
    method transpose_for_scores (line 204) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 209) | def forward(
  class ViTSelfOutput (line 245) | class ViTSelfOutput(nn.Module):
    method __init__ (line 251) | def __init__(self, config: ViTConfig) -> None:
    method forward (line 256) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ViTAttention (line 263) | class ViTAttention(nn.Module):
    method __init__ (line 264) | def __init__(self, config: ViTConfig) -> None:
    method prune_heads (line 270) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 288) | def forward(
  class ViTIntermediate (line 302) | class ViTIntermediate(nn.Module):
    method __init__ (line 303) | def __init__(self, config: ViTConfig) -> None:
    method forward (line 311) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ViTOutput (line 318) | class ViTOutput(nn.Module):
    method __init__ (line 319) | def __init__(self, config: ViTConfig) -> None:
    method forward (line 324) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ViTLayer (line 333) | class ViTLayer(nn.Module):
    method __init__ (line 336) | def __init__(self, config: ViTConfig) -> None:
    method forward (line 346) | def forward(
  class ViTEncoder (line 375) | class ViTEncoder(nn.Module):
    method __init__ (line 376) | def __init__(self, config: ViTConfig) -> None:
    method forward (line 382) | def forward(
  class ViTPreTrainedModel (line 432) | class ViTPreTrainedModel(PreTrainedModel):
    method _init_weights (line 444) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 470) | def _set_gradient_checkpointing(self, module: ViTEncoder, value: bool ...
  class ViTModel (line 515) | class ViTModel(ViTPreTrainedModel):
    method __init__ (line 516) | def __init__(self, config: ViTConfig, add_pooling_layer: bool = True, ...
    method get_input_embeddings (line 529) | def get_input_embeddings(self) -> ViTPatchEmbeddings:
    method _prune_heads (line 532) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
    method forward (line 548) | def forward(
  class ViTPooler (line 610) | class ViTPooler(nn.Module):
    method __init__ (line 611) | def __init__(self, config: ViTConfig):
    method forward (line 616) | def forward(self, hidden_states):
  class ViTForMaskedImageModeling (line 637) | class ViTForMaskedImageModeling(ViTPreTrainedModel):
    method __init__ (line 638) | def __init__(self, config: ViTConfig) -> None:
    method forward (line 657) | def forward(
  class ViTForImageClassification (line 766) | class ViTForImageClassification(ViTPreTrainedModel):
    method __init__ (line 767) | def __init__(self, config: ViTConfig) -> None:
    method forward (line 786) | def forward(

FILE: transformers/models/vit_hybrid/configuration_vit_hybrid.py
  class ViTHybridConfig (line 34) | class ViTHybridConfig(PretrainedConfig):
    method __init__ (line 93) | def __init__(
    method to_dict (line 150) | def to_dict(self) -> Dict[str, any]:

FILE: transformers/models/vit_hybrid/convert_vit_hybrid_timm_to_pytorch.py
  function create_rename_keys (line 46) | def create_rename_keys(config, base_model=False):
  function read_in_q_k_v (line 121) | def read_in_q_k_v(state_dict, config, base_model=False):
  function remove_classification_head_ (line 147) | def remove_classification_head_(state_dict):
  function rename_key (line 153) | def rename_key(dct, old, new):
  function prepare_img (line 159) | def prepare_img():
  function convert_vit_checkpoint (line 166) | def convert_vit_checkpoint(vit_name, pytorch_dump_folder_path, push_to_h...

FILE: transformers/models/vit_hybrid/image_processing_vit_hybrid.py
  class ViTHybridImageProcessor (line 51) | class ViTHybridImageProcessor(BaseImageProcessor):
    method __init__ (line 91) | def __init__(
    method resize (line 124) | def resize(
    method center_crop (line 152) | def center_crop(
    method rescale (line 176) | def rescale(
    method normalize (line 196) | def normalize(
    method preprocess (line 219) | def preprocess(

FILE: transformers/models/vit_hybrid/modeling_vit_hybrid.py
  class ViTHybridEmbeddings (line 56) | class ViTHybridEmbeddings(nn.Module):
    method __init__ (line 62) | def __init__(self, config: ViTHybridConfig, use_mask_token: bool = Fal...
    method interpolate_pos_encoding (line 73) | def interpolate_pos_encoding(self, embeddings: torch.Tensor, height: i...
    method forward (line 107) | def forward(
  class ViTHybridPatchEmbeddings (line 138) | class ViTHybridPatchEmbeddings(nn.Module):
    method __init__ (line 145) | def __init__(self, config, feature_size=None):
    method forward (line 177) | def forward(self, pixel_values: torch.Tensor, interpolate_pos_encoding...
  class ViTHybridSelfAttention (line 197) | class ViTHybridSelfAttention(nn.Module):
    method __init__ (line 198) | def __init__(self, config: ViTHybridConfig) -> None:
    method transpose_for_scores (line 216) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 221) | def forward(
  class ViTHybridSelfOutput (line 258) | class ViTHybridSelfOutput(nn.Module):
    method __init__ (line 264) | def __init__(self, config: ViTHybridConfig) -> None:
    method forward (line 269) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ViTHybridAttention (line 277) | class ViTHybridAttention(nn.Module):
    method __init__ (line 278) | def __init__(self, config: ViTHybridConfig) -> None:
    method prune_heads (line 284) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 302) | def forward(
  class ViTHybridIntermediate (line 317) | class ViTHybridIntermediate(nn.Module):
    method __init__ (line 318) | def __init__(self, config: ViTHybridConfig) -> None:
    method forward (line 326) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ViTHybridOutput (line 334) | class ViTHybridOutput(nn.Module):
    method __init__ (line 335) | def __init__(self, config: ViTHybridConfig) -> None:
    method forward (line 340) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ViTHybridLayer (line 349) | class ViTHybridLayer(nn.Module):
    method __init__ (line 352) | def __init__(self, config: ViTHybridConfig) -> None:
    method forward (line 362) | def forward(
  class ViTHybridEncoder (line 393) | class ViTHybridEncoder(nn.Module):
    method __init__ (line 394) | def __init__(self, config: ViTHybridConfig) -> None:
    method forward (line 400) | def forward(
  class ViTHybridPreTrainedModel (line 451) | class ViTHybridPreTrainedModel(PreTrainedModel):
    method _init_weights (line 463) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 489) | def _set_gradient_checkpointing(self, module: ViTHybridEncoder, value:...
  class ViTHybridModel (line 533) | class ViTHybridModel(ViTHybridPreTrainedModel):
    method __init__ (line 534) | def __init__(self, config: ViTHybridConfig, add_pooling_layer: bool = ...
    method get_input_embeddings (line 547) | def get_input_embeddings(self) -> ViTHybridPatchEmbeddings:
    method _prune_heads (line 550) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
    method forward (line 566) | def forward(
  class ViTHybridPooler (line 629) | class ViTHybridPooler(nn.Module):
    method __init__ (line 630) | def __init__(self, config: ViTHybridConfig):
    method forward (line 635) | def forward(self, hidden_states):
  class ViTHybridForImageClassification (line 652) | class ViTHybridForImageClassification(ViTHybridPreTrainedModel):
    method __init__ (line 653) | def __init__(self, config: ViTHybridConfig) -> None:
    method forward (line 672) | def forward(

FILE: transformers/models/vit_mae/configuration_vit_mae.py
  class ViTMAEConfig (line 29) | class ViTMAEConfig(PretrainedConfig):
    method __init__ (line 98) | def __init__(

FILE: transformers/models/vit_mae/convert_vit_mae_to_pytorch.py
  function rename_key (line 26) | def rename_key(name):
  function convert_state_dict (line 69) | def convert_state_dict(orig_state_dict, config):
  function convert_vit_mae_checkpoint (line 105) | def convert_vit_mae_checkpoint(checkpoint_url, pytorch_dump_folder_path):

FILE: transformers/models/vit_mae/modeling_tf_vit_mae.py
  class TFViTMAEModelOutput (line 56) | class TFViTMAEModelOutput(ModelOutput):
  class TFViTMAEDecoderOutput (line 85) | class TFViTMAEDecoderOutput(ModelOutput):
  class TFViTMAEForPreTrainingOutput (line 108) | class TFViTMAEForPreTrainingOutput(ModelOutput):
  function get_2d_sincos_pos_embed (line 139) | def get_2d_sincos_pos_embed(embed_dim, grid_size, add_cls_token=False):
  function get_2d_sincos_pos_embed_from_grid (line 167) | def get_2d_sincos_pos_embed_from_grid(embed_dim, grid):
  function get_1d_sincos_pos_embed_from_grid (line 179) | def get_1d_sincos_pos_embed_from_grid(embed_dim, pos):
  class TFViTMAEEmbeddings (line 202) | class TFViTMAEEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 208) | def __init__(self, config: ViTMAEConfig, **kwargs):
    method build (line 216) | def build(self, input_shape: tf.TensorShape):
    method random_masking (line 238) | def random_masking(self, sequence: tf.Tensor, noise: tf.Tensor | None ...
    method call (line 279) | def call(self, pixel_values: tf.Tensor, noise: tf.Tensor = None) -> tf...
  class TFViTMAEPatchEmbeddings (line 296) | class TFViTMAEPatchEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 303) | def __init__(self, config: ViTMAEConfig, **kwargs):
    method call (line 327) | def call(self, pixel_values: tf.Tensor, training: bool = False) -> tf....
  class TFViTMAESelfAttention (line 357) | class TFViTMAESelfAttention(tf.keras.layers.Layer):
    method __init__ (line 358) | def __init__(self, config: ViTMAEConfig, **kwargs):
    method transpose_for_scores (line 383) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 390) | def call(
  class TFViTMAESelfOutput (line 433) | class TFViTMAESelfOutput(tf.keras.layers.Layer):
    method __init__ (line 439) | def __init__(self, config: ViTMAEConfig, **kwargs):
    method call (line 447) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFViTMAEAttention (line 455) | class TFViTMAEAttention(tf.keras.layers.Layer):
    method __init__ (line 456) | def __init__(self, config: ViTMAEConfig, **kwargs):
    method prune_heads (line 462) | def prune_heads(self, heads):
    method call (line 465) | def call(
  class TFViTMAEIntermediate (line 484) | class TFViTMAEIntermediate(tf.keras.layers.Layer):
    method __init__ (line 485) | def __init__(self, config: ViTMAEConfig, **kwargs):
    method call (line 497) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFViTMAEOutput (line 505) | class TFViTMAEOutput(tf.keras.layers.Layer):
    method __init__ (line 506) | def __init__(self, config: ViTMAEConfig, **kwargs):
    method call (line 514) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFViTMAELayer (line 523) | class TFViTMAELayer(tf.keras.layers.Layer):
    method __init__ (line 526) | def __init__(self, config: ViTMAEConfig, **kwargs):
    method call (line 540) | def call(
  class TFViTMAEEncoder (line 574) | class TFViTMAEEncoder(tf.keras.layers.Layer):
    method __init__ (line 575) | def __init__(self, config: ViTMAEConfig, **kwargs):
    method call (line 580) | def call(
  class TFViTMAEMainLayer (line 620) | class TFViTMAEMainLayer(tf.keras.layers.Layer):
    method __init__ (line 623) | def __init__(self, config: ViTMAEConfig, **kwargs):
    method get_input_embeddings (line 632) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method _prune_heads (line 635) | def _prune_heads(self, heads_to_prune):
    method call (line 643) | def call(
  class TFViTMAEPreTrainedModel (line 691) | class TFViTMAEPreTrainedModel(TFPreTrainedModel):
  class TFViTMAEModel (line 778) | class TFViTMAEModel(TFViTMAEPreTrainedModel):
    method __init__ (line 779) | def __init__(self, config: ViTMAEConfig, *inputs, **kwargs):
    method get_input_embeddings (line 784) | def get_input_embeddings(self):
    method call (line 790) | def call(
  class TFViTMAEDecoder (line 833) | class TFViTMAEDecoder(tf.keras.layers.Layer):
    method __init__ (line 834) | def __init__(self, config, num_patches, **kwargs):
    method build (line 856) | def build(self, input_shape: tf.TensorShape):
    method call (line 878) | def call(
  class TFViTMAEForPreTraining (line 939) | class TFViTMAEForPreTraining(TFViTMAEPreTrainedModel):
    method __init__ (line 940) | def __init__(self, config):
    method get_input_embeddings (line 951) | def get_input_embeddings(self):
    method _prune_heads (line 954) | def _prune_heads(self, heads_to_prune):
    method patchify (line 957) | def patchify(self, pixel_values):
    method unpatchify (line 1005) | def unpatchify(self, patchified_pixel_values):
    method forward_loss (line 1037) | def forward_loss(self, pixel_values, pred, mask):
    method call (line 1066) | def call(

FILE: transformers/models/vit_mae/modeling_vit_mae.py
  class ViTMAEModelOutput (line 55) | class ViTMAEModelOutput(ModelOutput):
  class ViTMAEDecoderOutput (line 84) | class ViTMAEDecoderOutput(ModelOutput):
  class ViTMAEForPreTrainingOutput (line 107) | class ViTMAEForPreTrainingOutput(ModelOutput):
  function get_2d_sincos_pos_embed (line 138) | def get_2d_sincos_pos_embed(embed_dim, grid_size, add_cls_token=False):
  function get_2d_sincos_pos_embed_from_grid (line 166) | def get_2d_sincos_pos_embed_from_grid(embed_dim, grid):
  function get_1d_sincos_pos_embed_from_grid (line 178) | def get_1d_sincos_pos_embed_from_grid(embed_dim, pos):
  class ViTMAEEmbeddings (line 199) | class ViTMAEEmbeddings(nn.Module):
    method __init__ (line 205) | def __init__(self, config):
    method initialize_weights (line 218) | def initialize_weights(self):
    method random_masking (line 232) | def random_masking(self, sequence, noise=None):
    method forward (line 264) | def forward(self, pixel_values, noise=None):
  class ViTMAEPatchEmbeddings (line 282) | class ViTMAEPatchEmbeddings(nn.Module):
    method __init__ (line 289) | def __init__(self, config):
    method forward (line 303) | def forward(self, pixel_values):
  class ViTMAESelfAttention (line 318) | class ViTMAESelfAttention(nn.Module):
    method __init__ (line 319) | def __init__(self, config: ViTMAEConfig) -> None:
    method transpose_for_scores (line 337) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 342) | def forward(
  class ViTMAESelfOutput (line 379) | class ViTMAESelfOutput(nn.Module):
    method __init__ (line 385) | def __init__(self, config: ViTMAEConfig) -> None:
    method forward (line 390) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ViTMAEAttention (line 398) | class ViTMAEAttention(nn.Module):
    method __init__ (line 399) | def __init__(self, config: ViTMAEConfig) -> None:
    method prune_heads (line 405) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 423) | def forward(
  class ViTMAEIntermediate (line 438) | class ViTMAEIntermediate(nn.Module):
    method __init__ (line 439) | def __init__(self, config: ViTMAEConfig) -> None:
    method forward (line 447) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ViTMAEOutput (line 455) | class ViTMAEOutput(nn.Module):
    method __init__ (line 456) | def __init__(self, config: ViTMAEConfig) -> None:
    method forward (line 461) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ViTMAELayer (line 471) | class ViTMAELayer(nn.Module):
    method __init__ (line 474) | def __init__(self, config: ViTMAEConfig) -> None:
    method forward (line 484) | def forward(
  class ViTMAEEncoder (line 514) | class ViTMAEEncoder(nn.Module):
    method __init__ (line 515) | def __init__(self, config: ViTMAEConfig) -> None:
    method forward (line 521) | def forward(
  class ViTMAEPreTrainedModel (line 571) | class ViTMAEPreTrainedModel(PreTrainedModel):
    method _init_weights (line 582) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 594) | def _set_gradient_checkpointing(self, module, value=False):
  class ViTMAEModel (line 637) | class ViTMAEModel(ViTMAEPreTrainedModel):
    method __init__ (line 638) | def __init__(self, config):
    method get_input_embeddings (line 650) | def get_input_embeddings(self):
    method _prune_heads (line 653) | def _prune_heads(self, heads_to_prune):
    method forward (line 663) | def forward(
  class ViTMAEDecoder (line 732) | class ViTMAEDecoder(nn.Module):
    method __init__ (line 733) | def __init__(self, config, num_patches):
    method initialize_weights (line 758) | def initialize_weights(self, num_patches):
    method forward (line 768) | def forward(
  class ViTMAEForPreTraining (line 849) | class ViTMAEForPreTraining(ViTMAEPreTrainedModel):
    method __init__ (line 850) | def __init__(self, config):
    method get_input_embeddings (line 860) | def get_input_embeddings(self):
    method _prune_heads (line 863) | def _prune_heads(self, heads_to_prune):
    method patchify (line 871) | def patchify(self, pixel_values):
    method unpatchify (line 902) | def unpatchify(self, patchified_pixel_values):
    method forward_loss (line 937) | def forward_loss(self, pixel_values, pred, mask):
    method forward (line 964) | def forward(

FILE: transformers/models/vit_msn/configuration_vit_msn.py
  class ViTMSNConfig (line 30) | class ViTMSNConfig(PretrainedConfig):
    method __init__ (line 86) | def __init__(

FILE: transformers/models/vit_msn/convert_msn_to_pytorch.py
  function create_rename_keys (line 33) | def create_rename_keys(config, base_model=False):
  function read_in_q_k_v (line 86) | def read_in_q_k_v(state_dict, config, base_model=False):
  function remove_classification_head_ (line 112) | def remove_classification_head_(state_dict):
  function remove_projection_head (line 118) | def remove_projection_head(state_dict):
  function rename_key (line 143) | def rename_key(dct, old, new):
  function convert_vit_msn_checkpoint (line 148) | def convert_vit_msn_checkpoint(checkpoint_url, pytorch_dump_folder_path):

FILE: transformers/models/vit_msn/modeling_vit_msn.py
  class ViTMSNEmbeddings (line 46) | class ViTMSNEmbeddings(nn.Module):
    method __init__ (line 51) | def __init__(self, config: ViTMSNConfig, use_mask_token: bool = False)...
    method interpolate_pos_encoding (line 62) | def interpolate_pos_encoding(self, embeddings: torch.Tensor, height: i...
    method forward (line 97) | def forward(
  class ViTMSNPatchEmbeddings (line 129) | class ViTMSNPatchEmbeddings(nn.Module):
    method __init__ (line 136) | def __init__(self, config):
    method forward (line 151) | def forward(self, pixel_values: torch.Tensor, interpolate_pos_encoding...
  class ViTMSNSelfAttention (line 169) | class ViTMSNSelfAttention(nn.Module):
    method __init__ (line 170) | def __init__(self, config: ViTMSNConfig) -> None:
    method transpose_for_scores (line 188) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 193) | def forward(
  class ViTMSNSelfOutput (line 230) | class ViTMSNSelfOutput(nn.Module):
    method __init__ (line 236) | def __init__(self, config: ViTMSNConfig) -> None:
    method forward (line 241) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ViTMSNAttention (line 249) | class ViTMSNAttention(nn.Module):
    method __init__ (line 250) | def __init__(self, config: ViTMSNConfig) -> None:
    method prune_heads (line 256) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 274) | def forward(
  class ViTMSNIntermediate (line 289) | class ViTMSNIntermediate(nn.Module):
    method __init__ (line 290) | def __init__(self, config: ViTMSNConfig) -> None:
    method forward (line 298) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class ViTMSNOutput (line 306) | class ViTMSNOutput(nn.Module):
    method __init__ (line 307) | def __init__(self, config: ViTMSNConfig) -> None:
    method forward (line 312) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class ViTMSNLayer (line 322) | class ViTMSNLayer(nn.Module):
    method __init__ (line 325) | def __init__(self, config: ViTMSNConfig) -> None:
    method forward (line 335) | def forward(
  class ViTMSNEncoder (line 365) | class ViTMSNEncoder(nn.Module):
    method __init__ (line 366) | def __init__(self, config: ViTMSNConfig) -> None:
    method forward (line 372) | def forward(
  class ViTMSNPreTrainedModel (line 422) | class ViTMSNPreTrainedModel(PreTrainedModel):
    method _init_weights (line 435) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 447) | def _set_gradient_checkpointing(self, module: ViTMSNEncoder, value: bo...
  class ViTMSNModel (line 492) | class ViTMSNModel(ViTMSNPreTrainedModel):
    method __init__ (line 493) | def __init__(self, config: ViTMSNConfig, use_mask_token: bool = False):
    method get_input_embeddings (line 505) | def get_input_embeddings(self) -> ViTMSNPatchEmbeddings:
    method _prune_heads (line 508) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
    method forward (line 518) | def forward(
  class ViTMSNForImageClassification (line 601) | class ViTMSNForImageClassification(ViTMSNPreTrainedModel):
    method __init__ (line 602) | def __init__(self, config: ViTMSNConfig) -> None:
    method forward (line 616) | def forward(

FILE: transformers/models/wav2vec2/configuration_wav2vec2.py
  class Wav2Vec2Config (line 32) | class Wav2Vec2Config(PretrainedConfig):
    method __init__ (line 206) | def __init__(
    method inputs_to_logits_ratio (line 345) | def inputs_to_logits_ratio(self):

FILE: transformers/models/wav2vec2/convert_wav2vec2_original_pytorch_checkpoint_to_pytorch.py
  function set_recursively (line 70) | def set_recursively(key, value, full_name, weight_type, hf_pointer):
  function rename_dict (line 117) | def rename_dict(key, value, full_name, weight_type, hf_dict):
  function load_wav2vec2_layer (line 144) | def load_wav2vec2_layer(name, value, hf_model=None, hf_dict=None):
  function recursively_load_weights (line 172) | def recursively_load_weights(fairseq_model, hf_model, is_headless):
  function load_conv_layer (line 197) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function convert_wav2vec2_checkpoint (line 242) | def convert_wav2vec2_checkpoint(

FILE: transformers/models/wav2vec2/convert_wav2vec2_original_s3prl_checkpoint_to_pytorch.py
  function convert_classification (line 36) | def convert_classification(base_model_name, hf_config, downstream_dict):
  function convert_diarization (line 45) | def convert_diarization(base_model_name, hf_config, downstream_dict):
  function convert_xvector (line 52) | def convert_xvector(base_model_name, hf_config, downstream_dict):
  function convert_s3prl_checkpoint (line 71) | def convert_s3prl_checkpoint(base_model_name, config_path, checkpoint_pa...

FILE: transformers/models/wav2vec2/feature_extraction_wav2vec2.py
  class Wav2Vec2FeatureExtractor (line 31) | class Wav2Vec2FeatureExtractor(SequenceFeatureExtractor):
    method __init__ (line 67) | def __init__(
    method zero_mean_unit_var_norm (line 81) | def zero_mean_unit_var_norm(
    method __call__ (line 102) | def __call__(

FILE: transformers/models/wav2vec2/modeling_flax_wav2vec2.py
  class FlaxWav2Vec2BaseModelOutput (line 45) | class FlaxWav2Vec2BaseModelOutput(ModelOutput):
  class FlaxWav2Vec2ForPreTrainingOutput (line 75) | class FlaxWav2Vec2ForPreTrainingOutput(ModelOutput):
  function _compute_mask_indices (line 109) | def _compute_mask_indices(
  function _sample_negative_indices (line 182) | def _sample_negative_indices(features_shape: Tuple, num_negatives: int, ...
  class FlaxWav2Vec2LayerNormConvLayer (line 289) | class FlaxWav2Vec2LayerNormConvLayer(nn.Module):
    method setup (line 294) | def setup(self):
    method __call__ (line 310) | def __call__(self, hidden_states):
  class FlaxConvWithWeightNorm (line 317) | class FlaxConvWithWeightNorm(nn.Module):
    method setup (line 321) | def setup(self):
    method _get_normed_weights (line 340) | def _get_normed_weights(self):
    method __call__ (line 346) | def __call__(self, hidden_states):
  class FlaxWav2Vec2PositionalConvEmbedding (line 353) | class FlaxWav2Vec2PositionalConvEmbedding(nn.Module):
    method setup (line 357) | def setup(self):
    method __call__ (line 362) | def __call__(self, hidden_states):
  class FlaxConvLayersCollection (line 375) | class FlaxConvLayersCollection(nn.Module):
    method setup (line 379) | def setup(self):
    method __call__ (line 393) | def __call__(self, hidden_states):
  class FlaxWav2Vec2FeatureEncoder (line 399) | class FlaxWav2Vec2FeatureEncoder(nn.Module):
    method setup (line 405) | def setup(self):
    method __call__ (line 408) | def __call__(self, input_values, freeze_feature_encoder=False):
  class FlaxWav2Vec2FeatureProjection (line 416) | class FlaxWav2Vec2FeatureProjection(nn.Module):
    method setup (line 420) | def setup(self):
    method __call__ (line 429) | def __call__(self, hidden_states, deterministic=True):
  class FlaxWav2Vec2Attention (line 436) | class FlaxWav2Vec2Attention(nn.Module):
    method setup (line 444) | def setup(self) -> None:
    method _split_heads (line 465) | def _split_heads(self, hidden_states):
    method _merge_heads (line 468) | def _merge_heads(self, hidden_states):
    method __call__ (line 471) | def __call__(
  class FlaxWav2Vec2FeedForward (line 527) | class FlaxWav2Vec2FeedForward(nn.Module):
    method setup (line 531) | def setup(self):
    method __call__ (line 551) | def __call__(self, hidden_states, deterministic=True):
  class FlaxWav2Vec2EncoderLayerStableLayerNorm (line 561) | class FlaxWav2Vec2EncoderLayerStableLayerNorm(nn.Module):
    method setup (line 565) | def setup(self):
    method __call__ (line 578) | def __call__(self, hidden_states, attention_mask=None, deterministic=T...
  class FlaxWav2Vec2EncoderLayerStableLayerNormCollection (line 598) | class FlaxWav2Vec2EncoderLayerStableLayerNormCollection(nn.Module):
    method setup (line 602) | def setup(self):
    method __call__ (line 608) | def __call__(
  class FlaxWav2Vec2StableLayerNormEncoder (line 646) | class FlaxWav2Vec2StableLayerNormEncoder(nn.Module):
    method setup (line 650) | def setup(self):
    method __call__ (line 656) | def __call__(
  class FlaxWav2Vec2GumbelVectorQuantizer (line 701) | class FlaxWav2Vec2GumbelVectorQuantizer(nn.Module):
    method setup (line 710) | def setup(self):
    method _compute_perplexity (line 733) | def _compute_perplexity(probs, mask=None):
    method __call__ (line 744) | def __call__(self, hidden_states, mask_time_indices=None, deterministi...
  class FlaxWav2Vec2Adapter (line 779) | class FlaxWav2Vec2Adapter(nn.Module):
    method setup (line 783) | def setup(self):
    method __call__ (line 797) | def __call__(self, hidden_states, deterministic=True):
  class FlaxWav2Vec2AdapterLayer (line 808) | class FlaxWav2Vec2AdapterLayer(nn.Module):
    method setup (line 812) | def setup(self):
    method __call__ (line 822) | def __call__(self, hidden_states):
  class FlaxWav2Vec2AdapterLayersCollection (line 829) | class FlaxWav2Vec2AdapterLayersCollection(nn.Module):
    method setup (line 833) | def setup(self):
    method __call__ (line 839) | def __call__(self, hidden_states):
  class FlaxWav2Vec2PreTrainedModel (line 846) | class FlaxWav2Vec2PreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 857) | def __init__(
    method init_weights (line 869) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 889) | def __call__(
    method _get_feat_extract_output_lengths (line 933) | def _get_feat_extract_output_lengths(
  class FlaxWav2Vec2Module (line 939) | class FlaxWav2Vec2Module(nn.Module):
    method setup (line 943) | def setup(self):
    method __call__ (line 957) | def __call__(
    method _get_feat_extract_output_lengths (line 1009) | def _get_feat_extract_output_lengths(
    method _get_feature_vector_attention_mask (line 1032) | def _get_feature_vector_attention_mask(
  class FlaxWav2Vec2Model (line 1055) | class FlaxWav2Vec2Model(FlaxWav2Vec2PreTrainedModel):
  class FlaxWav2Vec2ForCTCModule (line 1098) | class FlaxWav2Vec2ForCTCModule(nn.Module):
    method setup (line 1102) | def setup(self):
    method __call__ (line 1111) | def __call__(
    method _get_feat_extract_output_lengths (line 1143) | def _get_feat_extract_output_lengths(
  class FlaxWav2Vec2ForCTC (line 1173) | class FlaxWav2Vec2ForCTC(FlaxWav2Vec2PreTrainedModel):
  class FlaxWav2Vec2ForPreTrainingModule (line 1219) | class FlaxWav2Vec2ForPreTrainingModule(nn.Module):
    method setup (line 1223) | def setup(self):
    method __call__ (line 1239) | def __call__(
    method _get_feat_extract_output_lengths (line 1294) | def _get_feat_extract_output_lengths(
  class FlaxWav2Vec2ForPreTraining (line 1319) | class FlaxWav2Vec2ForPreTraining(FlaxWav2Vec2PreTrainedModel):
    method __call__ (line 1324) | def __call__(

FILE: transformers/models/wav2vec2/modeling_tf_wav2vec2.py
  class TFWav2Vec2BaseModelOutput (line 66) | class TFWav2Vec2BaseModelOutput(ModelOutput):
  function _sample_without_replacement (line 94) | def _sample_without_replacement(distribution, num_samples):
  function _scatter_values_on_batch_indices (line 104) | def _scatter_values_on_batch_indices(values, batch_indices, output_shape):
  function _compute_mask_indices (line 119) | def _compute_mask_indices(
  function _expand_mask (line 194) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFWav2Vec2GroupNorm (line 207) | class TFWav2Vec2GroupNorm(tf.keras.layers.Layer):
    method __init__ (line 212) | def __init__(
    method build (line 242) | def build(self, input_shape):
    method call (line 253) | def call(self, inputs):
    method get_config (line 269) | def get_config(self):
    method compute_output_shape (line 286) | def compute_output_shape(self, input_shape):
    method _reshape_into_groups (line 289) | def _reshape_into_groups(self, inputs, input_shape, tensor_input_shape):
    method _apply_normalization (line 301) | def _apply_normalization(self, reshaped_inputs, input_shape):
    method _get_reshaped_weights (line 324) | def _get_reshaped_weights(self, input_shape):
    method _check_if_input_shape_is_none (line 335) | def _check_if_input_shape_is_none(self, input_shape):
    method _set_number_of_groups_for_instance_norm (line 346) | def _set_number_of_groups_for_instance_norm(self, input_shape):
    method _check_size_of_dimensions (line 352) | def _check_size_of_dimensions(self, input_shape):
    method _check_axis (line 372) | def _check_axis(self):
    method _create_input_spec (line 378) | def _create_input_spec(self, input_shape):
    method _add_gamma_weight (line 382) | def _add_gamma_weight(self, input_shape):
    method _add_beta_weight (line 397) | def _add_beta_weight(self, input_shape):
    method _create_broadcast_shape (line 412) | def _create_broadcast_shape(self, input_shape):
  class TFWav2Vec2WeightNormConv1D (line 423) | class TFWav2Vec2WeightNormConv1D(tf.keras.layers.Conv1D):
    method __init__ (line 426) | def __init__(self, filters, kernel_size, groups, explicit_padding, **k...
    method _init_norm (line 441) | def _init_norm(self):
    method _normalize_kernel (line 446) | def _normalize_kernel(self):
    method build (line 451) | def build(self, input_shape):
    method call (line 470) | def call(self, inputs):
  class TFWav2Vec2NoLayerNormConvLayer (line 483) | class TFWav2Vec2NoLayerNormConvLayer(tf.keras.layers.Layer):
    method __init__ (line 484) | def __init__(self, config: Wav2Vec2Config, layer_id: int = 0, **kwargs...
    method call (line 498) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFWav2Vec2LayerNormConvLayer (line 504) | class TFWav2Vec2LayerNormConvLayer(tf.keras.layers.Layer):
    method __init__ (line 505) | def __init__(self, config: Wav2Vec2Config, layer_id: int = 0, **kwargs...
    method call (line 520) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFWav2Vec2GroupNormConvLayer (line 527) | class TFWav2Vec2GroupNormConvLayer(tf.keras.layers.Layer):
    method __init__ (line 528) | def __init__(self, config: Wav2Vec2Config, layer_id: int = 0, **kwargs...
    method call (line 545) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFWav2Vec2PositionalConvEmbedding (line 552) | class TFWav2Vec2PositionalConvEmbedding(tf.keras.layers.Layer):
    method __init__ (line 553) | def __init__(self, config: Wav2Vec2Config, **kwargs: Any) -> None:
    method call (line 565) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFWav2Vec2SamePadLayer (line 572) | class TFWav2Vec2SamePadLayer(tf.keras.layers.Layer):
    method __init__ (line 573) | def __init__(self, num_conv_pos_embeddings, **kwargs):
    method call (line 577) | def call(self, hidden_states):
  class TFWav2Vec2FeatureEncoder (line 583) | class TFWav2Vec2FeatureEncoder(tf.keras.layers.Layer):
    method __init__ (line 584) | def __init__(self, config: Wav2Vec2Config, **kwargs: Any) -> None:
    method call (line 603) | def call(self, input_values):
  class TFWav2Vec2FeatureExtractor (line 610) | class TFWav2Vec2FeatureExtractor(TFWav2Vec2FeatureEncoder):
    method __init__ (line 611) | def __init__(self, config, **kwargs):
  class TFWav2Vec2FeatureProjection (line 621) | class TFWav2Vec2FeatureProjection(tf.keras.layers.Layer):
    method __init__ (line 622) | def __init__(self, config: Wav2Vec2Config, **kwargs):
    method call (line 634) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFWav2Vec2Attention (line 642) | class TFWav2Vec2Attention(tf.keras.layers.Layer):
    method __init__ (line 645) | def __init__(
    method _shape (line 673) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 676) | def call(
  class TFWav2Vec2FeedForward (line 795) | class TFWav2Vec2FeedForward(tf.keras.layers.Layer):
    method __init__ (line 796) | def __init__(self, config: Wav2Vec2Config, **kwargs):
    method call (line 817) | def call(self, hidden_states: tf.Tensor, training: bool = False) -> tf...
  class TFWav2Vec2EncoderLayer (line 827) | class TFWav2Vec2EncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 828) | def __init__(self, config: Wav2Vec2Config, **kwargs):
    method call (line 844) | def call(
  class TFWav2Vec2EncoderLayerStableLayerNorm (line 870) | class TFWav2Vec2EncoderLayerStableLayerNorm(tf.keras.layers.Layer):
    method __init__ (line 871) | def __init__(self, config: Wav2Vec2Config, **kwargs):
    method call (line 887) | def call(
  class TFWav2Vec2Encoder (line 911) | class TFWav2Vec2Encoder(tf.keras.layers.Layer):
    method __init__ (line 912) | def __init__(self, config: Wav2Vec2Config, **kwargs):
    method call (line 920) | def call(
  class TFWav2Vec2EncoderStableLayerNorm (line 976) | class TFWav2Vec2EncoderStableLayerNorm(tf.keras.layers.Layer):
    method __init__ (line 977) | def __init__(self, config: Wav2Vec2Config, **kwargs):
    method call (line 987) | def call(
  class TFWav2Vec2MainLayer (line 1044) | class TFWav2Vec2MainLayer(tf.keras.layers.Layer):
    method __init__ (line 1047) | def __init__(self, config: Wav2Vec2Config, **kwargs):
    method build (line 1058) | def build(self, input_shape: tf.TensorShape):
    method _get_feat_extract_output_lengths (line 1065) | def _get_feat_extract_output_lengths(self, input_lengths: tf.Tensor):
    method _mask_hidden_states (line 1080) | def _mask_hidden_states(self, hidden_states: tf.Tensor, mask_time_indi...
    method call (line 1125) | def call(
  class TFWav2Vec2PreTrainedModel (line 1177) | class TFWav2Vec2PreTrainedModel(TFPreTrainedModel):
    method input_signature (line 1188) | def input_signature(self):
    method dummy_inputs (line 1195) | def dummy_inputs(self):
    method __init__ (line 1201) | def __init__(self, config, *inputs, **kwargs):
    method _get_feat_extract_output_lengths (line 1208) | def _get_feat_extract_output_lengths(self, input_lengths, add_adapter=...
    method _get_feature_vector_attention_mask (line 1225) | def _get_feature_vector_attention_mask(
  class TFWav2Vec2Model (line 1351) | class TFWav2Vec2Model(TFWav2Vec2PreTrainedModel):
    method __init__ (line 1352) | def __init__(self, config: Wav2Vec2Config, *inputs, **kwargs):
    method call (line 1360) | def call(
  class TFWav2Vec2ForCTC (line 1425) | class TFWav2Vec2ForCTC(TFWav2Vec2PreTrainedModel):
    method __init__ (line 1426) | def __init__(self, config: Wav2Vec2Config, *inputs, **kwargs):
    method freeze_feature_extractor (line 1433) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1445) | def freeze_feature_encoder(self):
    method call (line 1455) | def call(
  class TFWav2Vec2ForSequenceClassification (line 1574) | class TFWav2Vec2ForSequenceClassification(TFWav2Vec2PreTrainedModel):
    method __init__ (line 1575) | def __init__(self, config):
    method freeze_feature_extractor (line 1588) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1600) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1607) | def freeze_base_model(self):
    method call (line 1616) | def call(

FILE: transformers/models/wav2vec2/modeling_wav2vec2.py
  class Wav2Vec2ForPreTrainingOutput (line 100) | class Wav2Vec2ForPreTrainingOutput(ModelOutput):
  function _compute_mask_indices (line 141) | def _compute_mask_indices(
  function _sample_negative_indices (line 260) | def _sample_negative_indices(
  class Wav2Vec2NoLayerNormConvLayer (line 296) | class Wav2Vec2NoLayerNormConvLayer(nn.Module):
    method __init__ (line 297) | def __init__(self, config, layer_id=0):
    method forward (line 311) | def forward(self, hidden_states):
  class Wav2Vec2LayerNormConvLayer (line 317) | class Wav2Vec2LayerNormConvLayer(nn.Module):
    method __init__ (line 318) | def __init__(self, config, layer_id=0):
    method forward (line 333) | def forward(self, hidden_states):
  class Wav2Vec2GroupNormConvLayer (line 344) | class Wav2Vec2GroupNormConvLayer(nn.Module):
    method __init__ (line 345) | def __init__(self, config, layer_id=0):
    method forward (line 361) | def forward(self, hidden_states):
  class Wav2Vec2PositionalConvEmbedding (line 368) | class Wav2Vec2PositionalConvEmbedding(nn.Module):
    method __init__ (line 369) | def __init__(self, config):
    method forward (line 396) | def forward(self, hidden_states):
  class Wav2Vec2SamePadLayer (line 407) | class Wav2Vec2SamePadLayer(nn.Module):
    method __init__ (line 408) | def __init__(self, num_conv_pos_embeddings):
    method forward (line 412) | def forward(self, hidden_states):
  class Wav2Vec2FeatureEncoder (line 418) | class Wav2Vec2FeatureEncoder(nn.Module):
    method __init__ (line 421) | def __init__(self, config):
    method _freeze_parameters (line 440) | def _freeze_parameters(self):
    method forward (line 445) | def forward(self, input_values):
  class Wav2Vec2FeatureExtractor (line 471) | class Wav2Vec2FeatureExtractor(Wav2Vec2FeatureEncoder):
    method __init__ (line 472) | def __init__(self, config):
  class Wav2Vec2FeatureProjection (line 482) | class Wav2Vec2FeatureProjection(nn.Module):
    method __init__ (line 483) | def __init__(self, config):
    method forward (line 489) | def forward(self, hidden_states):
  class Wav2Vec2Attention (line 498) | class Wav2Vec2Attention(nn.Module):
    method __init__ (line 501) | def __init__(
    method _shape (line 528) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 531) | def forward(
  class Wav2Vec2FeedForward (line 652) | class Wav2Vec2FeedForward(nn.Module):
    method __init__ (line 653) | def __init__(self, config):
    method forward (line 666) | def forward(self, hidden_states):
  class Wav2Vec2EncoderLayer (line 676) | class Wav2Vec2EncoderLayer(nn.Module):
    method __init__ (line 677) | def __init__(self, config):
    method forward (line 690) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class Wav2Vec2EncoderLayerStableLayerNorm (line 710) | class Wav2Vec2EncoderLayerStableLayerNorm(nn.Module):
    method __init__ (line 711) | def __init__(self, config):
    method forward (line 729) | def forward(
  class Wav2Vec2Encoder (line 755) | class Wav2Vec2Encoder(nn.Module):
    method __init__ (line 756) | def __init__(self, config):
    method forward (line 765) | def forward(
  class Wav2Vec2EncoderStableLayerNorm (line 842) | class Wav2Vec2EncoderStableLayerNorm(nn.Module):
    method __init__ (line 843) | def __init__(self, config):
    method forward (line 854) | def forward(
  class Wav2Vec2GumbelVectorQuantizer (line 933) | class Wav2Vec2GumbelVectorQuantizer(nn.Module):
    method __init__ (line 939) | def __init__(self, config):
    method _compute_perplexity (line 960) | def _compute_perplexity(probs, mask=None):
    method forward (line 971) | def forward(self, hidden_states, mask_time_indices=None):
  class Wav2Vec2Adapter (line 1009) | class Wav2Vec2Adapter(nn.Module):
    method __init__ (line 1010) | def __init__(self, config):
    method forward (line 1023) | def forward(self, hidden_states):
  class Wav2Vec2AdapterLayer (line 1040) | class Wav2Vec2AdapterLayer(nn.Module):
    method __init__ (line 1041) | def __init__(self, config):
    method forward (line 1051) | def forward(self, hidden_states):
  class Wav2Vec2AttnAdapterLayer (line 1058) | class Wav2Vec2AttnAdapterLayer(nn.Module):
    method __init__ (line 1059) | def __init__(self, config):
    method forward (line 1073) | def forward(self, hidden_states: torch.FloatTensor):
  class Wav2Vec2PreTrainedModel (line 1083) | class Wav2Vec2PreTrainedModel(PreTrainedModel):
    method _init_weights (line 1095) | def _init_weights(self, module):
    method _get_feat_extract_output_lengths (line 1134) | def _get_feat_extract_output_lengths(
    method _get_feature_vector_attention_mask (line 1157) | def _get_feature_vector_attention_mask(
    method _set_gradient_checkpointing (line 1177) | def _set_gradient_checkpointing(self, module, value=False):
    method _get_adapters (line 1181) | def _get_adapters(self):
    method load_adapter (line 1197) | def load_adapter(self, target_lang: str, **kwargs):
  class Wav2Vec2Model (line 1424) | class Wav2Vec2Model(Wav2Vec2PreTrainedModel):
    method __init__ (line 1425) | def __init__(self, config: Wav2Vec2Config):
    method freeze_feature_extractor (line 1445) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1457) | def freeze_feature_encoder(self):
    method _mask_hidden_states (line 1464) | def _mask_hidden_states(
    method forward (line 1518) | def forward(
  class Wav2Vec2ForPreTraining (line 1572) | class Wav2Vec2ForPreTraining(Wav2Vec2PreTrainedModel):
    method __init__ (line 1573) | def __init__(self, config: Wav2Vec2Config):
    method set_gumbel_temperature (line 1586) | def set_gumbel_temperature(self, temperature: int):
    method freeze_feature_extractor (line 1592) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1604) | def freeze_feature_encoder(self):
    method compute_contrastive_logits (line 1612) | def compute_contrastive_logits(
    method forward (line 1634) | def forward(
  class Wav2Vec2ForMaskedLM (line 1794) | class Wav2Vec2ForMaskedLM(Wav2Vec2PreTrainedModel):
    method __init__ (line 1795) | def __init__(self, config):
    method forward (line 1810) | def forward(
  class Wav2Vec2ForCTC (line 1843) | class Wav2Vec2ForCTC(Wav2Vec2PreTrainedModel):
    method __init__ (line 1844) | def __init__(self, config, target_lang=None):
    method freeze_feature_extractor (line 1872) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1884) | def freeze_feature_encoder(self):
    method forward (line 1899) | def forward(
  class Wav2Vec2ForSequenceClassification (line 1978) | class Wav2Vec2ForSequenceClassification(Wav2Vec2PreTrainedModel):
    method __init__ (line 1979) | def __init__(self, config):
    method freeze_feature_extractor (line 1996) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 2008) | def freeze_feature_encoder(self):
    method freeze_base_model (line 2015) | def freeze_base_model(self):
    method forward (line 2032) | def forward(
  class Wav2Vec2ForAudioFrameClassification (line 2100) | class Wav2Vec2ForAudioFrameClassification(Wav2Vec2PreTrainedModel):
    method __init__ (line 2101) | def __init__(self, config):
    method freeze_feature_extractor (line 2117) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 2129) | def freeze_feature_encoder(self):
    method freeze_base_model (line 2136) | def freeze_base_model(self):
    method forward (line 2152) | def forward(
  class AMSoftmaxLoss (line 2206) | class AMSoftmaxLoss(nn.Module):
    method __init__ (line 2207) | def __init__(self, input_dim, num_labels, scale=30.0, margin=0.4):
    method forward (line 2215) | def forward(self, hidden_states, labels):
  class TDNNLayer (line 2229) | class TDNNLayer(nn.Module):
    method __init__ (line 2230) | def __init__(self, config, layer_id=0):
    method forward (line 2240) | def forward(self, hidden_states):
  class Wav2Vec2ForXVector (line 2261) | class Wav2Vec2ForXVector(Wav2Vec2PreTrainedModel):
    method __init__ (line 2262) | def __init__(self, config):
    method freeze_feature_extractor (line 2281) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 2293) | def freeze_feature_encoder(self):
    method freeze_base_model (line 2300) | def freeze_base_model(self):
    method _get_tdnn_output_lengths (line 2308) | def _get_tdnn_output_lengths(self, input_lengths: Union[torch.LongTens...
    method forward (line 2331) | def forward(

FILE: transformers/models/wav2vec2/processing_wav2vec2.py
  class Wav2Vec2Processor (line 26) | class Wav2Vec2Processor(ProcessorMixin):
    method __init__ (line 43) | def __init__(self, feature_extractor, tokenizer):
    method from_pretrained (line 49) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method __call__ (line 67) | def __call__(self, *args, **kwargs):
    method pad (line 105) | def pad(self, *args, **kwargs):
    method batch_decode (line 135) | def batch_decode(self, *args, **kwargs):
    method decode (line 142) | def decode(self, *args, **kwargs):
    method as_target_processor (line 150) | def as_target_processor(self):

FILE: transformers/models/wav2vec2/tokenization_wav2vec2.py
  class Wav2Vec2CTCTokenizerOutput (line 106) | class Wav2Vec2CTCTokenizerOutput(ModelOutput):
  class Wav2Vec2CTCTokenizer (line 127) | class Wav2Vec2CTCTokenizer(PreTrainedTokenizer):
    method __init__ (line 164) | def __init__(
    method set_target_lang (line 215) | def set_target_lang(self, target_lang: str):
    method word_delimiter_token (line 237) | def word_delimiter_token(self) -> str:
    method word_delimiter_token_id (line 247) | def word_delimiter_token_id(self) -> Optional[int]:
    method word_delimiter_token (line 257) | def word_delimiter_token(self, value):
    method word_delimiter_token_id (line 261) | def word_delimiter_token_id(self, value):
    method vocab_size (line 265) | def vocab_size(self) -> int:
    method get_vocab (line 268) | def get_vocab(self) -> Dict:
    method _tokenize (line 271) | def _tokenize(self, text, **kwargs):
    method _convert_token_to_id (line 280) | def _convert_token_to_id(self, token: str) -> int:
    method _convert_id_to_token (line 284) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 289) | def convert_tokens_to_string(
    method _compute_offsets (line 353) | def _compute_offsets(
    method _get_word_offsets (line 368) | def _get_word_offsets(
    method prepare_for_tokenization (line 402) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method _decode (line 407) | def _decode(
    method batch_decode (line 460) | def batch_decode(
    method decode (line 530) | def decode(
    method save_vocabulary (line 635) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method _add_tokens (line 648) | def _add_tokens(self, new_tokens: Union[List[str], List[AddedToken]], ...
  class Wav2Vec2Tokenizer (line 707) | class Wav2Vec2Tokenizer(PreTrainedTokenizer):
    method __init__ (line 767) | def __init__(
    method word_delimiter_token (line 810) | def word_delimiter_token(self) -> str:
    method word_delimiter_token_id (line 820) | def word_delimiter_token_id(self) -> Optional[int]:
    method word_delimiter_token (line 830) | def word_delimiter_token(self, value):
    method word_delimiter_token_id (line 834) | def word_delimiter_token_id(self, value):
    method __call__ (line 838) | def __call__(
    method vocab_size (line 896) | def vocab_size(self) -> int:
    method get_vocab (line 899) | def get_vocab(self) -> Dict:
    method _convert_token_to_id (line 902) | def _convert_token_to_id(self, token: str) -> int:
    method _convert_id_to_token (line 906) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 911) | def convert_tokens_to_string(self, tokens: List[str]) -> str:
    method _decode (line 929) | def _decode(
    method save_vocabulary (line 962) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/wav2vec2_conformer/configuration_wav2vec2_conformer.py
  class Wav2Vec2ConformerConfig (line 33) | class Wav2Vec2ConformerConfig(PretrainedConfig):
    method __init__ (line 212) | def __init__(
    method inputs_to_logits_ratio (line 359) | def inputs_to_logits_ratio(self):

FILE: transformers/models/wav2vec2_conformer/convert_wav2vec2_conformer_original_pytorch_checkpoint_to_pytorch.py
  function set_recursively (line 82) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function recursively_load_weights (line 119) | def recursively_load_weights(fairseq_model, hf_model, is_headless):
  function load_conv_layer (line 176) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function convert_wav2vec2_conformer_checkpoint (line 221) | def convert_wav2vec2_conformer_checkpoint(

FILE: transformers/models/wav2vec2_conformer/modeling_wav2vec2_conformer.py
  class Wav2Vec2ConformerForPreTrainingOutput (line 74) | class Wav2Vec2ConformerForPreTrainingOutput(ModelOutput):
  function _compute_mask_indices (line 116) | def _compute_mask_indices(
  function _sample_negative_indices (line 236) | def _sample_negative_indices(
  class Wav2Vec2ConformerNoLayerNormConvLayer (line 273) | class Wav2Vec2ConformerNoLayerNormConvLayer(nn.Module):
    method __init__ (line 274) | def __init__(self, config, layer_id=0):
    method forward (line 288) | def forward(self, hidden_states):
  class Wav2Vec2ConformerLayerNormConvLayer (line 295) | class Wav2Vec2ConformerLayerNormConvLayer(nn.Module):
    method __init__ (line 296) | def __init__(self, config, layer_id=0):
    method forward (line 311) | def forward(self, hidden_states):
  class Wav2Vec2ConformerGroupNormConvLayer (line 323) | class Wav2Vec2ConformerGroupNormConvLayer(nn.Module):
    method __init__ (line 324) | def __init__(self, config, layer_id=0):
    method forward (line 340) | def forward(self, hidden_states):
  class Wav2Vec2ConformerPositionalConvEmbedding (line 348) | class Wav2Vec2ConformerPositionalConvEmbedding(nn.Module):
    method __init__ (line 349) | def __init__(self, config):
    method forward (line 376) | def forward(self, hidden_states):
  class Wav2Vec2ConformerRotaryPositionalEmbedding (line 387) | class Wav2Vec2ConformerRotaryPositionalEmbedding(nn.Module):
    method __init__ (line 392) | def __init__(self, config):
    method forward (line 402) | def forward(self, hidden_states):
  class Wav2Vec2ConformerRelPositionalEmbedding (line 419) | class Wav2Vec2ConformerRelPositionalEmbedding(nn.Module):
    method __init__ (line 422) | def __init__(self, config):
    method extend_pe (line 429) | def extend_pe(self, x):
    method forward (line 460) | def forward(self, hidden_states: torch.Tensor):
  class Wav2Vec2ConformerSamePadLayer (line 470) | class Wav2Vec2ConformerSamePadLayer(nn.Module):
    method __init__ (line 471) | def __init__(self, num_conv_pos_embeddings):
    method forward (line 475) | def forward(self, hidden_states):
  class Wav2Vec2ConformerFeatureEncoder (line 482) | class Wav2Vec2ConformerFeatureEncoder(nn.Module):
    method __init__ (line 485) | def __init__(self, config):
    method _freeze_parameters (line 505) | def _freeze_parameters(self):
    method forward (line 510) | def forward(self, input_values):
  class Wav2Vec2ConformerFeatureProjection (line 537) | class Wav2Vec2ConformerFeatureProjection(nn.Module):
    method __init__ (line 538) | def __init__(self, config):
    method forward (line 544) | def forward(self, hidden_states):
  class Wav2Vec2ConformerFeedForward (line 553) | class Wav2Vec2ConformerFeedForward(nn.Module):
    method __init__ (line 554) | def __init__(self, config):
    method forward (line 567) | def forward(self, hidden_states):
  class Wav2Vec2ConformerConvolutionModule (line 577) | class Wav2Vec2ConformerConvolutionModule(nn.Module):
    method __init__ (line 580) | def __init__(self, config):
    method forward (line 615) | def forward(self, hidden_states):
  class Wav2Vec2ConformerSelfAttention (line 637) | class Wav2Vec2ConformerSelfAttention(nn.Module):
    method __init__ (line 642) | def __init__(self, config):
    method forward (line 664) | def forward(
    method _apply_rotary_embedding (line 726) | def _apply_rotary_embedding(self, hidden_states, relative_position_emb...
    method _apply_relative_embeddings (line 745) | def _apply_relative_embeddings(self, query, key, relative_position_emb...
  class Wav2Vec2ConformerEncoderLayer (line 785) | class Wav2Vec2ConformerEncoderLayer(nn.Module):
    method __init__ (line 788) | def __init__(self, config):
    method forward (line 810) | def forward(
  class Wav2Vec2ConformerEncoder (line 852) | class Wav2Vec2ConformerEncoder(nn.Module):
    method __init__ (line 853) | def __init__(self, config):
    method forward (line 870) | def forward(
  class Wav2Vec2ConformerGumbelVectorQuantizer (line 954) | class Wav2Vec2ConformerGumbelVectorQuantizer(nn.Module):
    method __init__ (line 960) | def __init__(self, config):
    method _compute_perplexity (line 981) | def _compute_perplexity(probs, mask=None):
    method forward (line 992) | def forward(self, hidden_states, mask_time_indices=None):
  class Wav2Vec2ConformerAdapter (line 1031) | class Wav2Vec2ConformerAdapter(nn.Module):
    method __init__ (line 1032) | def __init__(self, config):
    method forward (line 1045) | def forward(self, hidden_states):
  class Wav2Vec2ConformerAdapterLayer (line 1063) | class Wav2Vec2ConformerAdapterLayer(nn.Module):
    method __init__ (line 1064) | def __init__(self, config):
    method forward (line 1074) | def forward(self, hidden_states):
  class Wav2Vec2ConformerPreTrainedModel (line 1081) | class Wav2Vec2ConformerPreTrainedModel(PreTrainedModel):
    method _init_weights (line 1093) | def _init_weights(self, module):
    method _get_feat_extract_output_lengths (line 1137) | def _get_feat_extract_output_lengths(
    method _get_feature_vector_attention_mask (line 1160) | def _get_feature_vector_attention_mask(
    method _set_gradient_checkpointing (line 1180) | def _set_gradient_checkpointing(self, module, value=False):
  class Wav2Vec2ConformerModel (line 1246) | class Wav2Vec2ConformerModel(Wav2Vec2ConformerPreTrainedModel):
    method __init__ (line 1247) | def __init__(self, config: Wav2Vec2ConformerConfig):
    method freeze_feature_encoder (line 1265) | def freeze_feature_encoder(self):
    method _mask_hidden_states (line 1273) | def _mask_hidden_states(
    method forward (line 1328) | def forward(
  class Wav2Vec2ConformerForPreTraining (line 1384) | class Wav2Vec2ConformerForPreTraining(Wav2Vec2ConformerPreTrainedModel):
    method __init__ (line 1386) | def __init__(self, config: Wav2Vec2ConformerConfig):
    method set_gumbel_temperature (line 1400) | def set_gumbel_temperature(self, temperature: int):
    method freeze_feature_encoder (line 1407) | def freeze_feature_encoder(self):
    method compute_contrastive_logits (line 1416) | def compute_contrastive_logits(
    method forward (line 1439) | def forward(
  class Wav2Vec2ConformerForCTC (line 1605) | class Wav2Vec2ConformerForCTC(Wav2Vec2ConformerPreTrainedModel):
    method __init__ (line 1607) | def __init__(self, config, target_lang=None):
    method freeze_feature_encoder (line 1636) | def freeze_feature_encoder(self):
    method forward (line 1652) | def forward(
  class Wav2Vec2ConformerForSequenceClassification (line 1731) | class Wav2Vec2ConformerForSequenceClassification(Wav2Vec2ConformerPreTra...
    method __init__ (line 1733) | def __init__(self, config):
    method freeze_feature_encoder (line 1751) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1758) | def freeze_base_model(self):
    method forward (line 1774) | def forward(
  class Wav2Vec2ConformerForAudioFrameClassification (line 1842) | class Wav2Vec2ConformerForAudioFrameClassification(Wav2Vec2ConformerPreT...
    method __init__ (line 1844) | def __init__(self, config):
    method freeze_feature_encoder (line 1861) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1869) | def freeze_base_model(self):
    method forward (line 1885) | def forward(
  class AMSoftmaxLoss (line 1940) | class AMSoftmaxLoss(nn.Module):
    method __init__ (line 1941) | def __init__(self, input_dim, num_labels, scale=30.0, margin=0.4):
    method forward (line 1949) | def forward(self, hidden_states, labels):
  class TDNNLayer (line 1964) | class TDNNLayer(nn.Module):
    method __init__ (line 1965) | def __init__(self, config, layer_id=0):
    method forward (line 1975) | def forward(self, hidden_states):
  class Wav2Vec2ConformerForXVector (line 1996) | class Wav2Vec2ConformerForXVector(Wav2Vec2ConformerPreTrainedModel):
    method __init__ (line 1997) | def __init__(self, config):
    method freeze_feature_encoder (line 2017) | def freeze_feature_encoder(self):
    method freeze_base_model (line 2025) | def freeze_base_model(self):
    method _get_tdnn_output_lengths (line 2034) | def _get_tdnn_output_lengths(self, input_lengths: Union[torch.LongTens...
    method forward (line 2057) | def forward(

FILE: transformers/models/wav2vec2_phoneme/tokenization_wav2vec2_phoneme.py
  class Wav2Vec2PhonemeCTCTokenizerOutput (line 77) | class Wav2Vec2PhonemeCTCTokenizerOutput(ModelOutput):
  class Wav2Vec2PhonemeCTCTokenizer (line 94) | class Wav2Vec2PhonemeCTCTokenizer(PreTrainedTokenizer):
    method __init__ (line 132) | def __init__(
    method vocab_size (line 173) | def vocab_size(self) -> int:
    method get_vocab (line 176) | def get_vocab(self) -> Dict:
    method init_backend (line 179) | def init_backend(self, phonemizer_lang: str):
    method prepare_for_tokenization (line 191) | def prepare_for_tokenization(
    method _tokenize (line 235) | def _tokenize(self, text, **kwargs):
    method phonemize (line 256) | def phonemize(self, text: str, phonemizer_lang: Optional[str] = None) ...
    method word_delimiter_token (line 275) | def word_delimiter_token(self) -> str:
    method word_delimiter_token_id (line 284) | def word_delimiter_token_id(self) -> Optional[int]:
    method word_delimiter_token (line 294) | def word_delimiter_token(self, value):
    method word_delimiter_token_id (line 298) | def word_delimiter_token_id(self, value):
    method phone_delimiter_token (line 302) | def phone_delimiter_token(self) -> str:
    method phone_delimiter_token_id (line 312) | def phone_delimiter_token_id(self) -> Optional[int]:
    method phone_delimiter_token (line 322) | def phone_delimiter_token(self, value):
    method phone_delimiter_token_id (line 326) | def phone_delimiter_token_id(self, value):
    method _convert_token_to_id (line 329) | def _convert_token_to_id(self, token: str) -> int:
    method _convert_id_to_token (line 333) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 338) | def convert_tokens_to_string(
    method _compute_offsets (line 389) | def _compute_offsets(
    method _decode (line 408) | def _decode(
    method decode (line 455) | def decode(
    method batch_decode (line 511) | def batch_decode(
    method save_vocabulary (line 567) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method _add_tokens (line 580) | def _add_tokens(self, new_tokens: Union[List[str], List[AddedToken]], ...

FILE: transformers/models/wav2vec2_with_lm/processing_wav2vec2_with_lm.py
  class Wav2Vec2DecoderWithLMOutput (line 45) | class Wav2Vec2DecoderWithLMOutput(ModelOutput):
  class Wav2Vec2ProcessorWithLM (line 67) | class Wav2Vec2ProcessorWithLM(ProcessorMixin):
    method __init__ (line 83) | def __init__(
    method save_pretrained (line 108) | def save_pretrained(self, save_directory):
    method from_pretrained (line 113) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method _set_language_model_attribute (line 184) | def _set_language_model_attribute(decoder: "BeamSearchDecoderCTC", att...
    method language_model (line 188) | def language_model(self):
    method get_missing_alphabet_tokens (line 192) | def get_missing_alphabet_tokens(decoder, tokenizer):
    method __call__ (line 214) | def __call__(self, *args, **kwargs):
    method pad (line 253) | def pad(self, *args, **kwargs):
    method batch_decode (line 284) | def batch_decode(
    method decode (line 469) | def decode(
    method as_target_processor (line 634) | def as_target_processor(self):

FILE: transformers/models/wavlm/configuration_wavlm.py
  class WavLMConfig (line 32) | class WavLMConfig(PretrainedConfig):
    method __init__ (line 195) | def __init__(
    method inputs_to_logits_ratio (line 335) | def inputs_to_logits_ratio(self):

FILE: transformers/models/wavlm/convert_wavlm_original_pytorch_checkpoint_to_pytorch.py
  function set_recursively (line 68) | def set_recursively(hf_pointer, key, value, full_name, weight_type):
  function recursively_load_weights (line 96) | def recursively_load_weights(fairseq_model, hf_model):
  function load_conv_layer (line 140) | def load_conv_layer(full_name, value, feature_extractor, unused_weights,...
  function convert_wavlm_checkpoint (line 181) | def convert_wavlm_checkpoint(checkpoint_path, pytorch_dump_folder_path, ...

FILE: transformers/models/wavlm/convert_wavlm_original_s3prl_checkpoint_to_pytorch.py
  function convert_classification (line 36) | def convert_classification(base_model_name, hf_config, downstream_dict):
  function convert_diarization (line 45) | def convert_diarization(base_model_name, hf_config, downstream_dict):
  function convert_xvector (line 52) | def convert_xvector(base_model_name, hf_config, downstream_dict):
  function convert_s3prl_checkpoint (line 71) | def convert_s3prl_checkpoint(base_model_name, config_path, checkpoint_pa...

FILE: transformers/models/wavlm/modeling_wavlm.py
  function _compute_mask_indices (line 76) | def _compute_mask_indices(
  class WavLMNoLayerNormConvLayer (line 196) | class WavLMNoLayerNormConvLayer(nn.Module):
    method __init__ (line 197) | def __init__(self, config, layer_id=0):
    method forward (line 211) | def forward(self, hidden_states):
  class WavLMLayerNormConvLayer (line 218) | class WavLMLayerNormConvLayer(nn.Module):
    method __init__ (line 219) | def __init__(self, config, layer_id=0):
    method forward (line 234) | def forward(self, hidden_states):
  class WavLMGroupNormConvLayer (line 246) | class WavLMGroupNormConvLayer(nn.Module):
    method __init__ (line 247) | def __init__(self, config, layer_id=0):
    method forward (line 263) | def forward(self, hidden_states):
  class WavLMPositionalConvEmbedding (line 271) | class WavLMPositionalConvEmbedding(nn.Module):
    method __init__ (line 272) | def __init__(self, config):
    method forward (line 299) | def forward(self, hidden_states):
  class WavLMSamePadLayer (line 311) | class WavLMSamePadLayer(nn.Module):
    method __init__ (line 312) | def __init__(self, num_conv_pos_embeddings):
    method forward (line 316) | def forward(self, hidden_states):
  class WavLMFeatureEncoder (line 323) | class WavLMFeatureEncoder(nn.Module):
    method __init__ (line 326) | def __init__(self, config):
    method _freeze_parameters (line 343) | def _freeze_parameters(self):
    method forward (line 348) | def forward(self, input_values):
  class WavLMFeatureExtractor (line 374) | class WavLMFeatureExtractor(WavLMFeatureEncoder):
    method __init__ (line 375) | def __init__(self, config):
  class WavLMFeatureProjection (line 386) | class WavLMFeatureProjection(nn.Module):
    method __init__ (line 387) | def __init__(self, config):
    method forward (line 393) | def forward(self, hidden_states):
  class WavLMAttention (line 401) | class WavLMAttention(nn.Module):
    method __init__ (line 404) | def __init__(
    method forward (line 440) | def forward(
    method torch_multi_head_self_attention (line 481) | def torch_multi_head_self_attention(
    method compute_bias (line 536) | def compute_bias(self, query_length: int, key_length: int) -> torch.Fl...
    method _relative_positions_bucket (line 546) | def _relative_positions_bucket(self, relative_positions: torch.FloatTe...
  class WavLMFeedForward (line 568) | class WavLMFeedForward(nn.Module):
    method __init__ (line 569) | def __init__(self, config):
    method forward (line 582) | def forward(self, hidden_states):
  class WavLMEncoderLayer (line 592) | class WavLMEncoderLayer(nn.Module):
    method __init__ (line 593) | def __init__(self, config: WavLMConfig, has_relative_position_bias: bo...
    method forward (line 608) | def forward(self, hidden_states, attention_mask=None, position_bias=No...
  class WavLMEncoderLayerStableLayerNorm (line 633) | class WavLMEncoderLayerStableLayerNorm(nn.Module):
    method __init__ (line 634) | def __init__(self, config: WavLMConfig, has_relative_position_bias: bo...
    method forward (line 649) | def forward(self, hidden_states, attention_mask=None, position_bias=No...
  class WavLMEncoder (line 670) | class WavLMEncoder(nn.Module):
    method __init__ (line 671) | def __init__(self, config):
    method forward (line 682) | def forward(
  class WavLMEncoderStableLayerNorm (line 758) | class WavLMEncoderStableLayerNorm(nn.Module):
    method __init__ (line 759) | def __init__(self, config):
    method forward (line 773) | def forward(
  class WavLMGumbelVectorQuantizer (line 847) | class WavLMGumbelVectorQuantizer(nn.Module):
    method __init__ (line 853) | def __init__(self, config):
    method _compute_perplexity (line 875) | def _compute_perplexity(probs):
    method forward (line 880) | def forward(self, hidden_states):
  class WavLMAdapter (line 918) | class WavLMAdapter(nn.Module):
    method __init__ (line 919) | def __init__(self, config):
    method forward (line 932) | def forward(self, hidden_states):
  class WavLMAdapterLayer (line 950) | class WavLMAdapterLayer(nn.Module):
    method __init__ (line 951) | def __init__(self, config):
    method forward (line 961) | def forward(self, hidden_states):
  class WavLMPreTrainedModel (line 968) | class WavLMPreTrainedModel(PreTrainedModel):
    method _init_weights (line 980) | def _init_weights(self, module):
    method _get_feat_extract_output_lengths (line 1013) | def _get_feat_extract_output_lengths(
    method _get_feature_vector_attention_mask (line 1036) | def _get_feature_vector_attention_mask(
    method _set_gradient_checkpointing (line 1056) | def _set_gradient_checkpointing(self, module, value=False):
  class WavLMModel (line 1122) | class WavLMModel(WavLMPreTrainedModel):
    method __init__ (line 1123) | def __init__(self, config: WavLMConfig):
    method freeze_feature_extractor (line 1143) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1155) | def freeze_feature_encoder(self):
    method _mask_hidden_states (line 1162) | def _mask_hidden_states(
    method forward (line 1216) | def forward(
  class WavLMForCTC (line 1274) | class WavLMForCTC(WavLMPreTrainedModel):
    method __init__ (line 1275) | def __init__(self, config, target_lang=None):
    method freeze_feature_extractor (line 1303) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1315) | def freeze_feature_encoder(self):
    method forward (line 1330) | def forward(
  class WavLMForSequenceClassification (line 1409) | class WavLMForSequenceClassification(WavLMPreTrainedModel):
    method __init__ (line 1410) | def __init__(self, config):
    method freeze_feature_extractor (line 1428) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1441) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1449) | def freeze_base_model(self):
    method forward (line 1465) | def forward(
  class WavLMForAudioFrameClassification (line 1534) | class WavLMForAudioFrameClassification(WavLMPreTrainedModel):
    method __init__ (line 1535) | def __init__(self, config):
    method freeze_feature_extractor (line 1551) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1563) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1570) | def freeze_base_model(self):
    method forward (line 1586) | def forward(
  class AMSoftmaxLoss (line 1641) | class AMSoftmaxLoss(nn.Module):
    method __init__ (line 1642) | def __init__(self, input_dim, num_labels, scale=30.0, margin=0.4):
    method forward (line 1650) | def forward(self, hidden_states, labels):
  class TDNNLayer (line 1665) | class TDNNLayer(nn.Module):
    method __init__ (line 1666) | def __init__(self, config, layer_id=0):
    method forward (line 1676) | def forward(self, hidden_states):
  class WavLMForXVector (line 1698) | class WavLMForXVector(WavLMPreTrainedModel):
    method __init__ (line 1699) | def __init__(self, config):
    method freeze_feature_extractor (line 1718) | def freeze_feature_extractor(self):
    method freeze_feature_encoder (line 1730) | def freeze_feature_encoder(self):
    method freeze_base_model (line 1737) | def freeze_base_model(self):
    method _get_tdnn_output_lengths (line 1745) | def _get_tdnn_output_lengths(self, input_lengths: Union[torch.LongTens...
    method forward (line 1768) | def forward(

FILE: transformers/models/whisper/configuration_whisper.py
  class WhisperConfig (line 62) | class WhisperConfig(PretrainedConfig):
    method __init__ (line 194) | def __init__(
  class WhisperOnnxConfig (line 280) | class WhisperOnnxConfig(OnnxSeq2SeqConfigWithPast):
    method inputs (line 282) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method generate_dummy_inputs (line 298) | def generate_dummy_inputs(
    method atol_for_validation (line 335) | def atol_for_validation(self) -> float:

FILE: transformers/models/whisper/convert_openai_to_hf.py
  function remove_ignore_keys_ (line 42) | def remove_ignore_keys_(state_dict):
  function rename_keys (line 72) | def rename_keys(s_dict):
  function make_linear_from_emb (line 86) | def make_linear_from_emb(emb):
  function _download (line 93) | def _download(url: str, root: str) -> bytes:
  function convert_openai_whisper_to_tfms (line 131) | def convert_openai_whisper_to_tfms(checkpoint_path, pytorch_dump_folder_...

FILE: transformers/models/whisper/english_normalizer.py
  function remove_symbols_and_diacritics (line 45) | def remove_symbols_and_diacritics(s: str, keep=""):
  function remove_symbols (line 68) | def remove_symbols(s: str):
  class BasicTextNormalizer (line 75) | class BasicTextNormalizer:
    method __init__ (line 76) | def __init__(self, remove_diacritics: bool = False, split_letters: boo...
    method __call__ (line 80) | def __call__(self, s: str):
  class EnglishNumberNormalizer (line 94) | class EnglishNumberNormalizer:
    method __init__ (line 105) | def __init__(self):
    method process_words (line 211) | def process_words(self, words: List[str]) -> Iterator[str]:
    method preprocess (line 434) | def preprocess(self, s: str):
    method postprocess (line 463) | def postprocess(self, s: str):
    method __call__ (line 488) | def __call__(self, s: str):
  class EnglishSpellingNormalizer (line 496) | class EnglishSpellingNormalizer:
    method __init__ (line 503) | def __init__(self, english_spelling_mapping):
    method __call__ (line 506) | def __call__(self, s: str):
  class EnglishTextNormalizer (line 510) | class EnglishTextNormalizer:
    method __init__ (line 511) | def __init__(self, english_spelling_mapping):
    method __call__ (line 571) | def __call__(self, s: str):

FILE: transformers/models/whisper/feature_extraction_whisper.py
  class WhisperFeatureExtractor (line 32) | class WhisperFeatureExtractor(SequenceFeatureExtractor):
    method __init__ (line 60) | def __init__(
    method _np_extract_fbank_features (line 94) | def _np_extract_fbank_features(self, waveform: np.array) -> np.ndarray:
    method zero_mean_unit_var_norm (line 115) | def zero_mean_unit_var_norm(
    method __call__ (line 136) | def __call__(
    method to_dict (line 266) | def to_dict(self) -> Dict[str, Any]:

FILE: transformers/models/whisper/modeling_flax_whisper.py
  class FlaxWhisperAttention (line 178) | class FlaxWhisperAttention(nn.Module):
    method setup (line 187) | def setup(self) -> None:
    method __call__ (line 212) | def __call__(
    method _split_heads (line 300) | def _split_heads(self, hidden_state) -> jnp.ndarray:
    method _merge_heads (line 303) | def _merge_heads(self, hidden_state) -> jnp.ndarray:
    method _concatenate_to_cache (line 307) | def _concatenate_to_cache(self, key, value, query, attention_mask) -> ...
  class FlaxWhisperEncoderLayer (line 338) | class FlaxWhisperEncoderLayer(nn.Module):
    method setup (line 342) | def setup(self) -> None:
    method __call__ (line 365) | def __call__(
  class FlaxWhisperEncoderLayerCollection (line 394) | class FlaxWhisperEncoderLayerCollection(nn.Module):
    method setup (line 399) | def setup(self):
    method __call__ (line 413) | def __call__(
  class FlaxWhisperDecoderLayer (line 457) | class FlaxWhisperDecoderLayer(nn.Module):
    method setup (line 461) | def setup(self) -> None:
    method __call__ (line 494) | def __call__(
  class FlaxWhisperDecoderLayerCollection (line 545) | class FlaxWhisperDecoderLayerCollection(nn.Module):
    method setup (line 550) | def setup(self):
    method __call__ (line 564) | def __call__(
  class FlaxWhisperEncoder (line 623) | class FlaxWhisperEncoder(nn.Module):
    method setup (line 628) | def setup(self) -> None:
    method __call__ (line 656) | def __call__(
  class FlaxWhisperDecoder (line 709) | class FlaxWhisperDecoder(nn.Module):
    method setup (line 714) | def setup(self) -> None:
    method __call__ (line 726) | def __call__(
  class FlaxWhisperModule (line 776) | class FlaxWhisperModule(nn.Module):
    method setup (line 781) | def setup(self) -> None:
    method __call__ (line 789) | def __call__(
    method _get_encoder_module (line 832) | def _get_encoder_module(self):
    method _get_decoder_module (line 835) | def _get_decoder_module(self):
  class FlaxWhisperPreTrainedModel (line 839) | class FlaxWhisperPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 845) | def __init__(
    method enable_gradient_checkpointing (line 858) | def enable_gradient_checkpointing(self):
    method init_weights (line 865) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 898) | def init_cache(self, batch_size, max_length, encoder_outputs):
    method encode (line 941) | def encode(
    method decode (line 997) | def decode(
    method __call__ (line 1111) | def __call__(
  class FlaxWhisperModel (line 1165) | class FlaxWhisperModel(FlaxWhisperPreTrainedModel):
  class FlaxWhisperForConditionalGenerationModule (line 1174) | class FlaxWhisperForConditionalGenerationModule(nn.Module):
    method setup (line 1179) | def setup(self) -> None:
    method _get_encoder_module (line 1190) | def _get_encoder_module(self):
    method _get_decoder_module (line 1193) | def _get_decoder_module(self):
    method __call__ (line 1196) | def __call__(
  class FlaxWhisperForConditionalGeneration (line 1244) | class FlaxWhisperForConditionalGeneration(FlaxWhisperPreTrainedModel):
    method decode (line 1250) | def decode(
    method generate (line 1382) | def generate(
    method prepare_inputs_for_generation (line 1447) | def prepare_inputs_for_generation(
    method update_inputs_for_generation (line 1478) | def update_inputs_for_generation(self, model_outputs, model_kwargs):
  class FlaxWhisperForAudioClassificationModule (line 1513) | class FlaxWhisperForAudioClassificationModule(nn.Module):
    method setup (line 1518) | def setup(self) -> None:
    method __call__ (line 1529) | def __call__(
  class FlaxWhisperForAudioClassification (line 1574) | class FlaxWhisperForAudioClassification(FlaxWhisperPreTrainedModel):
    method init_weights (line 1578) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method __call__ (line 1602) | def __call__(

FILE: transformers/models/whisper/modeling_tf_whisper.py
  function shift_tokens_right (line 60) | def shift_tokens_right(input_ids: tf.Tensor, pad_token_id: int, decoder_...
  function _make_causal_mask (line 85) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 103) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None):
  class TFWhisperPositionalEmbedding (line 116) | class TFWhisperPositionalEmbedding(tf.keras.layers.Layer):
    method __init__ (line 117) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method build (line 123) | def build(self, input_shape):
    method call (line 131) | def call(self, input_ids, past_key_values_length=0):
  class TFWhisperAttention (line 137) | class TFWhisperAttention(tf.keras.layers.Layer):
    method __init__ (line 140) | def __init__(
    method _shape (line 169) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 173) | def call(
  class TFWhisperEncoderLayer (line 293) | class TFWhisperEncoderLayer(tf.keras.layers.Layer):
    method __init__ (line 294) | def __init__(self, config: WhisperConfig, **kwargs):
    method call (line 308) | def call(
  class TFWhisperDecoderLayer (line 349) | class TFWhisperDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 350) | def __init__(self, config: WhisperConfig, **kwargs):
    method call (line 378) | def call(
  class TFWhisperPreTrainedModel (line 461) | class TFWhisperPreTrainedModel(TFPreTrainedModel):
    method _get_feat_extract_output_lengths (line 466) | def _get_feat_extract_output_lengths(self, input_lengths: tf.Tensor) -...
    method dummy_inputs (line 475) | def dummy_inputs(self) -> Dict[str, tf.Tensor]:
    method input_signature (line 490) | def input_signature(self):
  class TFWhisperEncoder (line 593) | class TFWhisperEncoder(tf.keras.layers.Layer):
    method __init__ (line 604) | def __init__(self, config: WhisperConfig, **kwargs):
    method call (line 629) | def call(
  class TFWhisperDecoder (line 725) | class TFWhisperDecoder(tf.keras.layers.Layer):
    method __init__ (line 734) | def __init__(self, config: WhisperConfig, **kwargs):
    method get_input_embeddings (line 758) | def get_input_embeddings(self):
    method set_input_embeddings (line 761) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 764) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method call (line 785) | def call(
  class TFWhisperMainLayer (line 969) | class TFWhisperMainLayer(tf.keras.layers.Layer):
    method __init__ (line 972) | def __init__(self, config: WhisperConfig, **kwargs):
    method get_input_embeddings (line 978) | def get_input_embeddings(self):
    method set_input_embeddings (line 981) | def set_input_embeddings(self, value):
    method get_encoder (line 984) | def get_encoder(self):
    method get_decoder (line 987) | def get_decoder(self):
    method call (line 993) | def call(
  class TFWhisperModel (line 1091) | class TFWhisperModel(TFWhisperPreTrainedModel):
    method __init__ (line 1092) | def __init__(self, config: WhisperConfig, **kwargs):
    method get_input_embeddings (line 1097) | def get_input_embeddings(self):
    method set_input_embeddings (line 1100) | def set_input_embeddings(self, value):
    method get_encoder (line 1103) | def get_encoder(self):
    method get_decoder (line 1106) | def get_decoder(self):
    method decoder (line 1109) | def decoder(self):
    method encoder (line 1112) | def encoder(self):
    method call (line 1118) | def call(
    method serving_output (line 1175) | def serving_output(self, output):
  class TFWhisperForConditionalGeneration (line 1199) | class TFWhisperForConditionalGeneration(TFWhisperPreTrainedModel, TFCaus...
    method __init__ (line 1210) | def __init__(self, config: WhisperConfig, **kwargs):
    method get_encoder (line 1214) | def get_encoder(self):
    method get_decoder (line 1217) | def get_decoder(self):
    method get_output_embeddings (line 1220) | def get_output_embeddings(self):
    method set_output_embeddings (line 1223) | def set_output_embeddings(self, value):
    method resize_token_embeddings (line 1226) | def resize_token_embeddings(self, new_num_tokens: int) -> tf.keras.lay...
    method call (line 1233) | def call(
    method serving_output (line 1328) | def serving_output(self, output):
    method prepare_inputs_for_generation (line 1347) | def prepare_inputs_for_generation(

FILE: transformers/models/whisper/modeling_whisper.py
  function shift_tokens_right (line 60) | def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decod...
  function _make_causal_mask (line 77) | def _make_causal_mask(
  function _expand_mask (line 95) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  function _compute_mask_indices (line 110) | def _compute_mask_indices(
  class WhisperPositionalEmbedding (line 229) | class WhisperPositionalEmbedding(nn.Embedding):
    method __init__ (line 230) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method forward (line 233) | def forward(self, input_ids, past_key_values_length=0):
  class WhisperAttention (line 237) | class WhisperAttention(nn.Module):
    method __init__ (line 240) | def __init__(
    method _shape (line 268) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 272) | def forward(
  class WhisperEncoderLayer (line 394) | class WhisperEncoderLayer(nn.Module):
    method __init__ (line 395) | def __init__(self, config: WhisperConfig):
    method forward (line 411) | def forward(
  class WhisperDecoderLayer (line 463) | class WhisperDecoderLayer(nn.Module):
    method __init__ (line 464) | def __init__(self, config: WhisperConfig):
    method forward (line 490) | def forward(
  class WhisperPreTrainedModel (line 580) | class WhisperPreTrainedModel(PreTrainedModel):
    method _init_weights (line 587) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 598) | def _set_gradient_checkpointing(self, module, value=False):
    method _get_feat_extract_output_lengths (line 602) | def _get_feat_extract_output_lengths(self, input_lengths: torch.LongTe...
  class WhisperEncoder (line 740) | class WhisperEncoder(WhisperPreTrainedModel):
    method __init__ (line 749) | def __init__(self, config: WhisperConfig):
    method _freeze_parameters (line 772) | def _freeze_parameters(self):
    method get_input_embeddings (line 777) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 780) | def set_input_embeddings(self, value: nn.Module):
    method forward (line 783) | def forward(
  class WhisperDecoder (line 886) | class WhisperDecoder(WhisperPreTrainedModel):
    method __init__ (line 894) | def __init__(self, config: WhisperConfig):
    method get_input_embeddings (line 914) | def get_input_embeddings(self):
    method set_input_embeddings (line 917) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 920) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 942) | def forward(
  class WhisperModel (line 1147) | class WhisperModel(WhisperPreTrainedModel):
    method __init__ (line 1150) | def __init__(self, config: WhisperConfig):
    method get_input_embeddings (line 1158) | def get_input_embeddings(self):
    method set_input_embeddings (line 1161) | def set_input_embeddings(self, value):
    method get_encoder (line 1164) | def get_encoder(self):
    method get_decoder (line 1167) | def get_decoder(self):
    method freeze_encoder (line 1170) | def freeze_encoder(self):
    method _mask_input_features (line 1177) | def _mask_input_features(
    method forward (line 1222) | def forward(
  class WhisperForConditionalGeneration (line 1317) | class WhisperForConditionalGeneration(WhisperPreTrainedModel):
    method __init__ (line 1328) | def __init__(self, config: WhisperConfig):
    method get_encoder (line 1336) | def get_encoder(self):
    method get_decoder (line 1339) | def get_decoder(self):
    method resize_token_embeddings (line 1342) | def resize_token_embeddings(self, new_num_tokens: int) -> nn.Embedding:
    method get_output_embeddings (line 1346) | def get_output_embeddings(self):
    method set_output_embeddings (line 1349) | def set_output_embeddings(self, new_embeddings):
    method get_input_embeddings (line 1352) | def get_input_embeddings(self) -> nn.Module:
    method freeze_encoder (line 1355) | def freeze_encoder(self):
    method forward (line 1364) | def forward(
    method generate (line 1460) | def generate(
    method prepare_inputs_for_generation (line 1673) | def prepare_inputs_for_generation(
    method _reorder_cache (line 1696) | def _reorder_cache(past_key_values, beam_idx):
  class WhisperForAudioClassification (line 1710) | class WhisperForAudioClassification(WhisperPreTrainedModel):
    method __init__ (line 1711) | def __init__(self, config):
    method freeze_encoder (line 1724) | def freeze_encoder(self):
    method get_input_embeddings (line 1731) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 1734) | def set_input_embeddings(self, value: nn.Module):
    method forward (line 1739) | def forward(

FILE: transformers/models/whisper/processing_whisper.py
  class WhisperProcessor (line 23) | class WhisperProcessor(ProcessorMixin):
    method __init__ (line 40) | def __init__(self, feature_extractor, tokenizer):
    method get_decoder_prompt_ids (line 45) | def get_decoder_prompt_ids(self, task=None, language=None, no_timestam...
    method __call__ (line 48) | def __call__(self, *args, **kwargs):
    method batch_decode (line 82) | def batch_decode(self, *args, **kwargs):
    method decode (line 89) | def decode(self, *args, **kwargs):
    method get_prompt_ids (line 96) | def get_prompt_ids(self, text: str, return_tensors="np"):

FILE: transformers/models/whisper/tokenization_whisper.py
  function bytes_to_unicode (line 51) | def bytes_to_unicode():
  function get_pairs (line 79) | def get_pairs(word):
  class WhisperTokenizer (line 214) | class WhisperTokenizer(PreTrainedTokenizer):
    method __init__ (line 257) | def __init__(
    method get_vocab (line 313) | def get_vocab(self):
    method vocab_size (line 319) | def vocab_size(self) -> int:
    method bpe (line 323) | def bpe(self, token):
    method set_prefix_tokens (line 365) | def set_prefix_tokens(self, language: str = None, task: str = None, pr...
    method prefix_tokens (line 390) | def prefix_tokens(self) -> List[int]:
    method build_inputs_with_special_tokens (line 425) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method get_special_tokens_mask (line 433) | def get_special_tokens_mask(
    method _tokenize (line 464) | def _tokenize(self, text):
    method _convert_token_to_id (line 475) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 479) | def _convert_id_to_token(self, index):
    method _normalize (line 486) | def _normalize(self, text):
    method _decode_with_timestamps (line 494) | def _decode_with_timestamps(self, token_ids, skip_special_tokens=False...
    method _compute_offsets (line 513) | def _compute_offsets(self, token_ids, time_precision=0.02):
    method decode (line 557) | def decode(
    method _decode (line 608) | def _decode(
    method convert_tokens_to_string (line 647) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 653) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method prepare_for_tokenization (line 692) | def prepare_for_tokenization(self, text, is_split_into_words=False, **...
    method _build_conversation_input_ids (line 699) | def _build_conversation_input_ids(self, conversation) -> List[int]:
    method get_decoder_prompt_ids (line 707) | def get_decoder_prompt_ids(self, task=None, language=None, no_timestam...
    method _decode_asr (line 717) | def _decode_asr(self, model_outputs, *, return_timestamps, return_lang...
    method get_prompt_ids (line 726) | def get_prompt_ids(self, text: str, return_tensors="np"):
    method _strip_prompt (line 741) | def _strip_prompt(token_ids: List[int], prompt_token_id: int, decoder_...
  function _decode_asr (line 752) | def _decode_asr(tokenizer, model_outputs, *, return_timestamps, return_l...
  function _find_longest_common_sequence (line 941) | def _find_longest_common_sequence(sequences):

FILE: transformers/models/whisper/tokenization_whisper_fast.py
  class WhisperTokenizerFast (line 88) | class WhisperTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 134) | def __init__(
    method _batch_encode_plus (line 181) | def _batch_encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _encode_plus (line 191) | def _encode_plus(self, *args, **kwargs) -> BatchEncoding:
    method _decode_with_timestamps (line 202) | def _decode_with_timestamps(self, token_ids, skip_special_tokens=False...
    method _compute_offsets (line 222) | def _compute_offsets(self, token_ids, time_precision=0.02):
    method decode (line 267) | def decode(
    method _decode (line 318) | def _decode(self, *args, normalize: bool = False, **kwargs) -> str:
    method _normalize (line 333) | def _normalize(self, text):
    method save_vocabulary (line 341) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method set_prefix_tokens (line 356) | def set_prefix_tokens(self, language: str = None, task: str = None, pr...
    method prefix_tokens (line 396) | def prefix_tokens(self) -> List[int]:
    method build_inputs_with_special_tokens (line 431) | def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=No...
    method get_special_tokens_mask (line 439) | def get_special_tokens_mask(
    method _build_conversation_input_ids (line 470) | def _build_conversation_input_ids(self, conversation) -> List[int]:
    method get_decoder_prompt_ids (line 479) | def get_decoder_prompt_ids(self, task=None, language=None, no_timestam...
    method _decode_asr (line 489) | def _decode_asr(self, model_outputs, *, return_timestamps, return_lang...
    method get_prompt_ids (line 499) | def get_prompt_ids(self, text: str, return_tensors="np"):
    method _strip_prompt (line 515) | def _strip_prompt(token_ids: List[int], prompt_token_id: int, decoder_...

FILE: transformers/models/x_clip/configuration_x_clip.py
  class XCLIPTextConfig (line 32) | class XCLIPTextConfig(PretrainedConfig):
    method __init__ (line 87) | def __init__(
    method from_pretrained (line 120) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class XCLIPVisionConfig (line 136) | class XCLIPVisionConfig(PretrainedConfig):
    method __init__ (line 201) | def __init__(
    method from_pretrained (line 245) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
  class XCLIPConfig (line 261) | class XCLIPConfig(PretrainedConfig):
    method __init__ (line 300) | def __init__(
    method from_text_vision_configs (line 406) | def from_text_vision_configs(cls, text_config: XCLIPTextConfig, vision...
    method to_dict (line 417) | def to_dict(self):

FILE: transformers/models/x_clip/convert_x_clip_original_pytorch_to_hf.py
  function get_xclip_config (line 35) | def get_xclip_config(model_name, num_frames):
  function rename_key (line 66) | def rename_key(name):
  function convert_state_dict (line 120) | def convert_state_dict(orig_state_dict, config):
  function prepare_video (line 202) | def prepare_video(num_frames):
  function convert_xclip_checkpoint (line 218) | def convert_xclip_checkpoint(model_name, pytorch_dump_folder_path=None, ...

FILE: transformers/models/x_clip/modeling_x_clip.py
  function _expand_mask (line 50) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  function contrastive_loss (line 66) | def contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
  function x_clip_loss (line 71) | def x_clip_loss(similarity: torch.Tensor) -> torch.Tensor:
  class XCLIPOutput (line 78) | class XCLIPOutput(ModelOutput):
    method to_tuple (line 111) | def to_tuple(self) -> Tuple[Any]:
  class XCLIPVisionEmbeddings (line 121) | class XCLIPVisionEmbeddings(nn.Module):
    method __init__ (line 122) | def __init__(self, config: XCLIPVisionConfig):
    method forward (line 144) | def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
  class XCLIPTextEmbeddings (line 156) | class XCLIPTextEmbeddings(nn.Module):
    method __init__ (line 157) | def __init__(self, config: XCLIPTextConfig):
    method forward (line 167) | def forward(
  class XCLIPAttention (line 188) | class XCLIPAttention(nn.Module):
    method __init__ (line 191) | def __init__(self, config):
    method _shape (line 210) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 213) | def forward(
  class XCLIPMLP (line 293) | class XCLIPMLP(nn.Module):
    method __init__ (line 294) | def __init__(self, config):
    method forward (line 301) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class XCLIPEncoderLayer (line 309) | class XCLIPEncoderLayer(nn.Module):
    method __init__ (line 310) | def __init__(self, config: XCLIPConfig):
    method forward (line 318) | def forward(
  function drop_path (line 360) | def drop_path(input, drop_prob: float = 0.0, training: bool = False):
  class XCLIPDropPath (line 381) | class XCLIPDropPath(nn.Module):
    method __init__ (line 384) | def __init__(self, drop_prob: Optional[float] = None) -> None:
    method forward (line 388) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    method extra_repr (line 391) | def extra_repr(self) -> str:
  class XCLIPVisionEncoderLayer (line 395) | class XCLIPVisionEncoderLayer(nn.Module):
    method __init__ (line 400) | def __init__(self, config: XCLIPConfig):
    method forward (line 416) | def forward(
  class XCLIPPreTrainedModel (line 475) | class XCLIPPreTrainedModel(PreTrainedModel):
    method _init_weights (line 486) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 535) | def _set_gradient_checkpointing(self, module, value=False):
  class XCLIPEncoder (line 637) | class XCLIPEncoder(nn.Module):
    method __init__ (line 646) | def __init__(self, config: XCLIPConfig):
    method forward (line 652) | def forward(
  function _make_causal_mask (line 741) | def _make_causal_mask(
  class XCLIPTextTransformer (line 758) | class XCLIPTextTransformer(nn.Module):
    method __init__ (line 759) | def __init__(self, config: XCLIPTextConfig):
    method forward (line 769) | def forward(
  class XCLIPTextModel (line 831) | class XCLIPTextModel(XCLIPPreTrainedModel):
    method __init__ (line 834) | def __init__(self, config: XCLIPTextConfig):
    method get_input_embeddings (line 840) | def get_input_embeddings(self) -> nn.Module:
    method set_input_embeddings (line 843) | def set_input_embeddings(self, value):
    method forward (line 848) | def forward(
  class XCLIPVisionEncoder (line 884) | class XCLIPVisionEncoder(nn.Module):
    method __init__ (line 893) | def __init__(self, config: XCLIPConfig):
    method forward (line 899) | def forward(
  class XCLIPVisionTransformer (line 987) | class XCLIPVisionTransformer(nn.Module):
    method __init__ (line 992) | def __init__(self, config: XCLIPVisionConfig):
    method forward (line 1004) | def forward(
  class XCLIPVisionModel (line 1046) | class XCLIPVisionModel(XCLIPPreTrainedModel):
    method __init__ (line 1050) | def __init__(self, config: XCLIPVisionConfig):
    method get_input_embeddings (line 1056) | def get_input_embeddings(self) -> nn.Module:
    method forward (line 1061) | def forward(
  class XCLIPMultiframeIntegrationTransformer (line 1143) | class XCLIPMultiframeIntegrationTransformer(nn.Module):
    method __init__ (line 1148) | def __init__(self, config: XCLIPVisionConfig):
    method forward (line 1154) | def forward(
  class XCLIPCrossAttention (line 1189) | class XCLIPCrossAttention(nn.Module):
    method __init__ (line 1192) | def __init__(self, config):
    method _shape (line 1208) | def _shape(self, tensor: torch.Tensor, seq_len: int, batch_size: int):
    method forward (line 1211) | def forward(self, queries, keys, values):
  class PromptGeneratorLayer (line 1241) | class PromptGeneratorLayer(nn.Module):
    method __init__ (line 1242) | def __init__(self, config):
    method forward (line 1256) | def forward(self, x, visual):
  class XCLIPPromptGenerator (line 1262) | class XCLIPPromptGenerator(nn.Module):
    method __init__ (line 1265) | def __init__(self, config):
    method forward (line 1272) | def forward(self, text, visual):
  class XCLIPModel (line 1281) | class XCLIPModel(XCLIPPreTrainedModel):
    method __init__ (line 1284) | def __init__(self, config: XCLIPConfig):
    method get_text_features (line 1329) | def get_text_features(
    method get_video_features (line 1376) | def get_video_features(
    method forward (line 1483) | def forward(

FILE: transformers/models/x_clip/processing_x_clip.py
  class XCLIPProcessor (line 25) | class XCLIPProcessor(ProcessorMixin):
    method __init__ (line 42) | def __init__(self, image_processor=None, tokenizer=None, **kwargs):
    method __call__ (line 60) | def __call__(self, text=None, videos=None, return_tensors=None, **kwar...
    method batch_decode (line 114) | def batch_decode(self, *args, **kwargs):
    method decode (line 121) | def decode(self, *args, **kwargs):
    method model_input_names (line 129) | def model_input_names(self):
    method feature_extractor_class (line 133) | def feature_extractor_class(self):
    method feature_extractor (line 141) | def feature_extractor(self):

FILE: transformers/models/xglm/configuration_xglm.py
  class XGLMConfig (line 29) | class XGLMConfig(PretrainedConfig):
    method __init__ (line 97) | def __init__(

FILE: transformers/models/xglm/convert_xglm_original_ckpt_to_trfms.py
  function remove_ignore_keys_ (line 10) | def remove_ignore_keys_(state_dict):
  function make_linear_from_emb (line 21) | def make_linear_from_emb(emb):
  function convert_fairseq_xglm_checkpoint_from_disk (line 28) | def convert_fairseq_xglm_checkpoint_from_disk(checkpoint_path):

FILE: transformers/models/xglm/modeling_flax_xglm.py
  function create_sinusoidal_positions (line 113) | def create_sinusoidal_positions(n_pos, dim, padding_idx=1):
  class FlaxXGLMAttention (line 127) | class FlaxXGLMAttention(nn.Module):
    method setup (line 136) | def setup(self) -> None:
    method _split_heads (line 163) | def _split_heads(self, hidden_states):
    method _merge_heads (line 166) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 170) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 202) | def __call__(
  class FlaxXGLMDecoderLayer (line 296) | class FlaxXGLMDecoderLayer(nn.Module):
    method setup (line 300) | def setup(self) -> None:
    method __call__ (line 336) | def __call__(
  class FlaxXGLMDecoderLayerCollection (line 387) | class FlaxXGLMDecoderLayerCollection(nn.Module):
    method setup (line 391) | def setup(self):
    method __call__ (line 397) | def __call__(
  class FlaxXGLMModule (line 456) | class FlaxXGLMModule(nn.Module):
    method setup (line 460) | def setup(self):
    method __call__ (line 483) | def __call__(
  class FlaxXGLMPreTrainedModel (line 540) | class FlaxXGLMPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 545) | def __init__(
    method init_weights (line 557) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 592) | def init_cache(self, batch_size, max_length):
    method __call__ (line 612) | def __call__(
  class FlaxXGLMModel (line 689) | class FlaxXGLMModel(FlaxXGLMPreTrainedModel):
  class FlaxXGLMForCausalLMModule (line 701) | class FlaxXGLMForCausalLMModule(nn.Module):
    method setup (line 705) | def setup(self):
    method __call__ (line 714) | def __call__(
  class FlaxXGLMForCausalLM (line 766) | class FlaxXGLMForCausalLM(FlaxXGLMPreTrainedModel):
    method prepare_inputs_for_generation (line 769) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 790) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/xglm/modeling_tf_xglm.py
  function create_sinusiodal_positions (line 66) | def create_sinusiodal_positions(num_positions: int, embedding_dim: int, ...
  function _create_position_ids_from_input_ids (line 89) | def _create_position_ids_from_input_ids(
  function _create_position_ids_from_inputs_embeds (line 102) | def _create_position_ids_from_inputs_embeds(
  function _make_causal_mask (line 120) | def _make_causal_mask(input_ids_shape: tf.TensorShape, past_key_values_l...
  function _expand_mask (line 138) | def _expand_mask(mask: tf.Tensor, tgt_len: Optional[int] = None, past_ke...
  class TFXGLMAttention (line 152) | class TFXGLMAttention(tf.keras.layers.Layer):
    method __init__ (line 155) | def __init__(
    method _shape (line 183) | def _shape(self, tensor: tf.Tensor, seq_len: int, bsz: int):
    method call (line 186) | def call(
  class TFXGLMDecoderLayer (line 305) | class TFXGLMDecoderLayer(tf.keras.layers.Layer):
    method __init__ (line 306) | def __init__(self, config: XGLMConfig, **kwargs: Any) -> None:
    method call (line 338) | def call(
  class TFXGLMMainLayer (line 420) | class TFXGLMMainLayer(tf.keras.layers.Layer):
    method __init__ (line 423) | def __init__(
    method get_input_embeddings (line 452) | def get_input_embeddings(self) -> TFSharedEmbeddings:
    method set_input_embeddings (line 455) | def set_input_embeddings(self, value: TFSharedEmbeddings) -> None:
    method _prepare_decoder_attention_mask (line 458) | def _prepare_decoder_attention_mask(
    method embed_positions (line 480) | def embed_positions(self, position_ids: np.ndarray | tf.Tensor | None ...
    method call (line 486) | def call(
  class TFXGLMPreTrainedModel (line 618) | class TFXGLMPreTrainedModel(TFPreTrainedModel):
  class TFXGLMModel (line 741) | class TFXGLMModel(TFXGLMPreTrainedModel):
    method __init__ (line 750) | def __init__(
    method call (line 764) | def call(
  class TFXGLMForCausalLM (line 808) | class TFXGLMForCausalLM(TFXGLMPreTrainedModel, TFCausalLanguageModelingL...
    method __init__ (line 818) | def __init__(
    method get_output_embeddings (line 831) | def get_output_embeddings(self):
    method set_output_embeddings (line 834) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 837) | def prepare_inputs_for_generation(self, inputs, past_key_values=None, ...
    method call (line 866) | def call(

FILE: transformers/models/xglm/modeling_xglm.py
  function _make_causal_mask (line 132) | def _make_causal_mask(
  function _expand_mask (line 150) | def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Option...
  class XGLMSinusoidalPositionalEmbedding (line 164) | class XGLMSinusoidalPositionalEmbedding(nn.Module):
    method __init__ (line 167) | def __init__(self, num_positions: int, embedding_dim: int, padding_idx...
    method make_weights (line 174) | def make_weights(self, num_embeddings: int, embedding_dim: int, paddin...
    method get_embedding (line 183) | def get_embedding(num_embeddings: int, embedding_dim: int, padding_idx...
    method forward (line 204) | def forward(self, position_ids: torch.Tensor = None, past_key_values_l...
  class XGLMAttention (line 216) | class XGLMAttention(nn.Module):
    method __init__ (line 219) | def __init__(
    method _shape (line 246) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 249) | def forward(
  class XGLMDecoderLayer (line 370) | class XGLMDecoderLayer(nn.Module):
    method __init__ (line 371) | def __init__(self, config: XGLMConfig):
    method forward (line 400) | def forward(
  class XGLMPreTrainedModel (line 490) | class XGLMPreTrainedModel(PreTrainedModel):
    method _init_weights (line 496) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 507) | def _set_gradient_checkpointing(self, module, value=False):
  class XGLMModel (line 516) | class XGLMModel(XGLMPreTrainedModel):
    method __init__ (line 525) | def __init__(self, config: XGLMConfig, embed_tokens: Optional[nn.Embed...
    method get_input_embeddings (line 550) | def get_input_embeddings(self):
    method set_input_embeddings (line 553) | def set_input_embeddings(self, value):
    method _prepare_decoder_attention_mask (line 556) | def _prepare_decoder_attention_mask(self, attention_mask, input_shape,...
    method forward (line 583) | def forward(
  class XGLMForCausalLM (line 750) | class XGLMForCausalLM(XGLMPreTrainedModel):
    method __init__ (line 761) | def __init__(self, config):
    method get_input_embeddings (line 769) | def get_input_embeddings(self):
    method set_input_embeddings (line 772) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 775) | def get_output_embeddings(self):
    method set_output_embeddings (line 778) | def set_output_embeddings(self, new_embeddings):
    method forward (line 787) | def forward(
    method prepare_inputs_for_generation (line 859) | def prepare_inputs_for_generation(
    method _reorder_cache (line 887) | def _reorder_cache(past_key_values, beam_idx):

FILE: transformers/models/xglm/tokenization_xglm.py
  class XGLMTokenizer (line 43) | class XGLMTokenizer(PreTrainedTokenizer):
    method __init__ (line 117) | def __init__(
    method __getstate__ (line 173) | def __getstate__(self):
    method __setstate__ (line 179) | def __setstate__(self, d):
    method build_inputs_with_special_tokens (line 189) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 214) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 242) | def create_token_type_ids_from_sequences(
    method vocab_size (line 267) | def vocab_size(self):
    method get_vocab (line 270) | def get_vocab(self):
    method _tokenize (line 275) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 278) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 287) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 293) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 298) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/xglm/tokenization_xglm_fast.py
  class XGLMTokenizerFast (line 49) | class XGLMTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 103) | def __init__(
    method build_inputs_with_special_tokens (line 139) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 164) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 188) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/xlm/configuration_xlm.py
  class XLMConfig (line 40) | class XLMConfig(PretrainedConfig):
    method __init__ (line 160) | def __init__(
  class XLMOnnxConfig (line 237) | class XLMOnnxConfig(OnnxConfig):
    method inputs (line 239) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/xlm/convert_xlm_original_pytorch_checkpoint_to_pytorch.py
  function convert_xlm_checkpoint_to_pytorch (line 31) | def convert_xlm_checkpoint_to_pytorch(xlm_checkpoint_path, pytorch_dump_...

FILE: transformers/models/xlm/modeling_tf_xlm.py
  function create_sinusoidal_embeddings (line 83) | def create_sinusoidal_embeddings(n_pos, dim, out):
  function get_masks (line 89) | def get_masks(slen, lengths, causal, padding_mask=None):
  class TFXLMMultiHeadAttention (line 118) | class TFXLMMultiHeadAttention(tf.keras.layers.Layer):
    method __init__ (line 121) | def __init__(self, n_heads, dim, config, **kwargs):
    method prune_heads (line 136) | def prune_heads(self, heads):
    method call (line 139) | def call(self, input, mask, kv, cache, head_mask, output_attentions, t...
  class TFXLMTransformerFFN (line 210) | class TFXLMTransformerFFN(tf.keras.layers.Layer):
    method __init__ (line 211) | def __init__(self, in_dim, dim_hidden, out_dim, config, **kwargs):
    method call (line 219) | def call(self, input, training=False):
  class TFXLMMainLayer (line 229) | class TFXLMMainLayer(tf.keras.layers.Layer):
    method __init__ (line 232) | def __init__(self, config, **kwargs):
    method build (line 319) | def build(self, input_shape):
    method get_input_embeddings (line 337) | def get_input_embeddings(self):
    method set_input_embeddings (line 340) | def set_input_embeddings(self, value):
    method _prune_heads (line 344) | def _prune_heads(self, heads_to_prune):
    method call (line 352) | def call(
  class TFXLMPreTrainedModel (line 517) | class TFXLMPreTrainedModel(TFPreTrainedModel):
    method dummy_inputs (line 527) | def dummy_inputs(self):
  class TFXLMWithLMHeadModelOutput (line 543) | class TFXLMWithLMHeadModelOutput(ModelOutput):
  class TFXLMModel (line 689) | class TFXLMModel(TFXLMPreTrainedModel):
    method __init__ (line 690) | def __init__(self, config, *inputs, **kwargs):
    method call (line 701) | def call(
  class TFXLMPredLayer (line 736) | class TFXLMPredLayer(tf.keras.layers.Layer):
    method __init__ (line 741) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 760) | def build(self, input_shape):
    method get_output_embeddings (line 766) | def get_output_embeddings(self):
    method set_output_embeddings (line 769) | def set_output_embeddings(self, value):
    method get_bias (line 773) | def get_bias(self):
    method set_bias (line 776) | def set_bias(self, value):
    method call (line 780) | def call(self, hidden_states):
  class TFXLMWithLMHeadModel (line 794) | class TFXLMWithLMHeadModel(TFXLMPreTrainedModel):
    method __init__ (line 795) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 802) | def get_lm_head(self):
    method get_prefix_bias_name (line 805) | def get_prefix_bias_name(self):
    method prepare_inputs_for_generation (line 809) | def prepare_inputs_for_generation(self, inputs, **kwargs):
    method call (line 830) | def call(
  class TFXLMForSequenceClassification (line 880) | class TFXLMForSequenceClassification(TFXLMPreTrainedModel, TFSequenceCla...
    method __init__ (line 881) | def __init__(self, config, *inputs, **kwargs):
    method call (line 895) | def call(
  class TFXLMForMultipleChoice (line 958) | class TFXLMForMultipleChoice(TFXLMPreTrainedModel, TFMultipleChoiceLoss):
    method __init__ (line 959) | def __init__(self, config, *inputs, **kwargs):
    method dummy_inputs (line 969) | def dummy_inputs(self):
    method call (line 994) | def call(
  class TFXLMForTokenClassification (line 1077) | class TFXLMForTokenClassification(TFXLMPreTrainedModel, TFTokenClassific...
    method __init__ (line 1078) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1095) | def call(
  class TFXLMForQuestionAnsweringSimple (line 1157) | class TFXLMForQuestionAnsweringSimple(TFXLMPreTrainedModel, TFQuestionAn...
    method __init__ (line 1158) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1172) | def call(

FILE: transformers/models/xlm/modeling_xlm.py
  function create_sinusoidal_embeddings (line 71) | def create_sinusoidal_embeddings(n_pos, dim, out):
  function get_masks (line 79) | def get_masks(slen, lengths, causal, padding_mask=None):
  class MultiHeadAttention (line 104) | class MultiHeadAttention(nn.Module):
    method __init__ (line 107) | def __init__(self, n_heads, dim, config):
    method prune_heads (line 121) | def prune_heads(self, heads):
    method forward (line 136) | def forward(self, input, mask, kv=None, cache=None, head_mask=None, ou...
  class TransformerFFN (line 200) | class TransformerFFN(nn.Module):
    method __init__ (line 201) | def __init__(self, in_dim, dim_hidden, out_dim, config):
    method forward (line 210) | def forward(self, input):
    method ff_chunk (line 213) | def ff_chunk(self, input):
  class XLMPreTrainedModel (line 221) | class XLMPreTrainedModel(PreTrainedModel):
    method __init__ (line 231) | def __init__(self, *inputs, **kwargs):
    method dummy_inputs (line 235) | def dummy_inputs(self):
    method _init_weights (line 244) | def _init_weights(self, module):
  class XLMForQuestionAnsweringOutput (line 262) | class XLMForQuestionAnsweringOutput(ModelOutput):
  class XLMModel (line 393) | class XLMModel(XLMPreTrainedModel):
    method __init__ (line 396) | def __init__(self, config):
    method get_input_embeddings (line 466) | def get_input_embeddings(self):
    method set_input_embeddings (line 469) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 472) | def _prune_heads(self, heads_to_prune):
    method forward (line 486) | def forward(
  class XLMPredLayer (line 623) | class XLMPredLayer(nn.Module):
    method __init__ (line 628) | def __init__(self, config):
    method forward (line 646) | def forward(self, x, y=None):
  class XLMWithLMHeadModel (line 672) | class XLMWithLMHeadModel(XLMPreTrainedModel):
    method __init__ (line 675) | def __init__(self, config):
    method get_output_embeddings (line 683) | def get_output_embeddings(self):
    method set_output_embeddings (line 686) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 689) | def prepare_inputs_for_generation(self, input_ids, **kwargs):
    method forward (line 709) | def forward(
  class XLMForSequenceClassification (line 769) | class XLMForSequenceClassification(XLMPreTrainedModel):
    method __init__ (line 770) | def __init__(self, config):
    method forward (line 787) | def forward(
  class XLMForQuestionAnsweringSimple (line 871) | class XLMForQuestionAnsweringSimple(XLMPreTrainedModel):
    method __init__ (line 872) | def __init__(self, config):
    method forward (line 887) | def forward(
  class XLMForQuestionAnswering (line 975) | class XLMForQuestionAnswering(XLMPreTrainedModel):
    method __init__ (line 976) | def __init__(self, config):
    method forward (line 987) | def forward(
  class XLMForTokenClassification (line 1096) | class XLMForTokenClassification(XLMPreTrainedModel):
    method __init__ (line 1097) | def __init__(self, config):
    method forward (line 1114) | def forward(
  class XLMForMultipleChoice (line 1180) | class XLMForMultipleChoice(XLMPreTrainedModel):
    method __init__ (line 1181) | def __init__(self, config, *inputs, **kwargs):
    method forward (line 1197) | def forward(

FILE: transformers/models/xlm/tokenization_xlm.py
  function get_pairs (line 428) | def get_pairs(word):
  function lowercase_and_remove_accent (line 441) | def lowercase_and_remove_accent(text):
  function replace_unicode_punct (line 458) | def replace_unicode_punct(text):
  function remove_non_printing_char (line 501) | def remove_non_printing_char(text):
  function romanian_preprocessing (line 514) | def romanian_preprocessing(text):
  class XLMTokenizer (line 528) | class XLMTokenizer(PreTrainedTokenizer):
    method __init__ (line 589) | def __init__(
    method do_lower_case (line 665) | def do_lower_case(self):
    method moses_punct_norm (line 668) | def moses_punct_norm(self, text, lang):
    method moses_tokenize (line 676) | def moses_tokenize(self, text, lang):
    method moses_pipeline (line 684) | def moses_pipeline(self, text, lang):
    method ja_tokenize (line 690) | def ja_tokenize(self, text):
    method vocab_size (line 712) | def vocab_size(self):
    method get_vocab (line 715) | def get_vocab(self):
    method bpe (line 718) | def bpe(self, token):
    method _tokenize (line 762) | def _tokenize(self, text, lang="en", bypass_tokenizer=False):
    method _convert_token_to_id (line 858) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 862) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 866) | def convert_tokens_to_string(self, tokens):
    method build_inputs_with_special_tokens (line 871) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 898) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 926) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 955) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method __getstate__ (line 983) | def __getstate__(self):
    method __setstate__ (line 988) | def __setstate__(self, d):

FILE: transformers/models/xlm_prophetnet/configuration_xlm_prophetnet.py
  class XLMProphetNetConfig (line 33) | class XLMProphetNetConfig(PretrainedConfig):
    method __init__ (line 109) | def __init__(
    method num_hidden_layers (line 176) | def num_hidden_layers(self) -> int:
    method num_hidden_layers (line 180) | def num_hidden_layers(self, value):

FILE: transformers/models/xlm_prophetnet/modeling_xlm_prophetnet.py
  function softmax (line 183) | def softmax(hidden_state, dim, onnx_trace=False):
  function ngram_attention_bias (line 191) | def ngram_attention_bias(sequence_length, ngram, device, dtype):
  function compute_relative_buckets (line 209) | def compute_relative_buckets(num_buckets, max_distance, relative_positio...
  function compute_all_stream_relative_buckets (line 237) | def compute_all_stream_relative_buckets(num_buckets, max_distance, posit...
  class XLMProphetNetSeq2SeqLMOutput (line 262) | class XLMProphetNetSeq2SeqLMOutput(ModelOutput):
    method decoder_cross_attentions (line 337) | def decoder_cross_attentions(self):
  class XLMProphetNetSeq2SeqModelOutput (line 348) | class XLMProphetNetSeq2SeqModelOutput(ModelOutput):
    method decoder_cross_attentions (line 424) | def decoder_cross_attentions(self):
  class XLMProphetNetDecoderModelOutput (line 435) | class XLMProphetNetDecoderModelOutput(ModelOutput):
  class XLMProphetNetDecoderLMOutput (line 496) | class XLMProphetNetDecoderLMOutput(ModelOutput):
  class XLMProphetNetPreTrainedModel (line 558) | class XLMProphetNetPreTrainedModel(PreTrainedModel):
    method _init_weights (line 563) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 573) | def _set_gradient_checkpointing(self, module, value=False):
    method _shift_right (line 577) | def _shift_right(self, input_ids):
  class XLMProphetNetPositionalEmbeddings (line 601) | class XLMProphetNetPositionalEmbeddings(nn.Embedding):
    method __init__ (line 608) | def __init__(self, config: XLMProphetNetConfig) -> None:
    method forward (line 612) | def forward(self, inputs_shape, device, attention_mask=None, past_key_...
    method _forward (line 640) | def _forward(self, position_ids):
  class XLMProphetNetAttention (line 645) | class XLMProphetNetAttention(nn.Module):
    method __init__ (line 648) | def __init__(
    method _shape (line 672) | def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
    method forward (line 675) | def forward(
  class XLMProphetNetFeedForward (line 775) | class XLMProphetNetFeedForward(nn.Module):
    method __init__ (line 780) | def __init__(self, config: XLMProphetNetConfig, ffn_dim: int):
    method forward (line 788) | def forward(self, hidden_states):
  class XLMProphetNetNgramSelfAttention (line 799) | class XLMProphetNetNgramSelfAttention(nn.Module):
    method __init__ (line 800) | def __init__(self, config: XLMProphetNetConfig):
    method _shape (line 829) | def _shape(self, tensor, seq_len, batch_size):
    method prepare_for_onnx_export_ (line 832) | def prepare_for_onnx_export_(self):
    method forward (line 835) | def forward(
    method get_main_relative_pos_embeddings (line 1013) | def get_main_relative_pos_embeddings(
    method get_predict_relative_pos_embeddings (line 1058) | def get_predict_relative_pos_embeddings(
  class XLMProphetNetEncoderLayer (line 1119) | class XLMProphetNetEncoderLayer(nn.Module):
    method __init__ (line 1124) | def __init__(self, config: XLMProphetNetConfig):
    method forward (line 1134) | def forward(
  class XLMProphetNetDecoderLayer (line 1163) | class XLMProphetNetDecoderLayer(nn.Module):
    method __init__ (line 1168) | def __init__(self, config: XLMProphetNetConfig):
    method forward (line 1183) | def forward(
  class XLMProphetNetEncoder (line 1252) | class XLMProphetNetEncoder(XLMProphetNetPreTrainedModel):
    method __init__ (line 1259) | def __init__(self, config: XLMProphetNetConfig, word_embeddings: nn.Em...
    method get_input_embeddings (line 1276) | def get_input_embeddings(self):
    method set_input_embeddings (line 1279) | def set_input_embeddings(self, value):
    method forward (line 1284) | def forward(
  class XLMProphetNetDecoder (line 1393) | class XLMProphetNetDecoder(XLMProphetNetPreTrainedModel):
    method __init__ (line 1400) | def __init__(self, config: XLMProphetNetConfig, word_embeddings: Optio...
    method get_input_embeddings (line 1424) | def get_input_embeddings(self):
    method set_input_embeddings (line 1427) | def set_input_embeddings(self, value):
    method forward (line 1432) | def forward(
    method compute_buffered_relative_buckets (line 1683) | def compute_buffered_relative_buckets(self, position_ids):
    method prepare_attention_mask (line 1705) | def prepare_attention_mask(self, hidden_states, attention_mask):
    method prepare_predict_attention_mask (line 1729) | def prepare_predict_attention_mask(self, hidden_states, attention_mask):
  class XLMProphetNetModel (line 1770) | class XLMProphetNetModel(XLMProphetNetPreTrainedModel):
    method __init__ (line 1773) | def __init__(self, config: XLMProphetNetConfig):
    method get_input_embeddings (line 1790) | def get_input_embeddings(self):
    method set_input_embeddings (line 1793) | def set_input_embeddings(self, value):
    method get_encoder (line 1798) | def get_encoder(self):
    method get_decoder (line 1801) | def get_decoder(self):
    method forward (line 1806) | def forward(
  class XLMProphetNetForConditionalGeneration (line 1900) | class XLMProphetNetForConditionalGeneration(XLMProphetNetPreTrainedModel):
    method __init__ (line 1907) | def __init__(self, config: XLMProphetNetConfig):
    method get_output_embeddings (line 1918) | def get_output_embeddings(self):
    method set_output_embeddings (line 1921) | def set_output_embeddings(self, new_embeddings):
    method get_input_embeddings (line 1924) | def get_input_embeddings(self):
    method forward (line 1929) | def forward(
    method _compute_loss (line 2035) | def _compute_loss(self, logits, labels, ignore_index=-100):
    method prepare_inputs_for_generation (line 2063) | def prepare_inputs_for_generation(
    method prepare_decoder_input_ids_from_labels (line 2092) | def prepare_decoder_input_ids_from_labels(self, labels: torch.Tensor):
    method _reorder_cache (line 2097) | def _reorder_cache(past_key_values, beam_idx):
    method get_encoder (line 2106) | def get_encoder(self):
    method get_decoder (line 2109) | def get_decoder(self):
  class XLMProphetNetForCausalLM (line 2119) | class XLMProphetNetForCausalLM(XLMProphetNetPreTrainedModel):
    method __init__ (line 2122) | def __init__(self, config: XLMProphetNetConfig):
    method get_input_embeddings (line 2138) | def get_input_embeddings(self):
    method set_input_embeddings (line 2141) | def set_input_embeddings(self, value):
    method get_output_embeddings (line 2144) | def get_output_embeddings(self):
    method set_output_embeddings (line 2147) | def set_output_embeddings(self, new_embeddings):
    method set_decoder (line 2150) | def set_decoder(self, decoder):
    method get_decoder (line 2153) | def get_decoder(self):
    method forward (line 2158) | def forward(
    method _compute_loss (line 2290) | def _compute_loss(self, logits, labels, ignore_index=-100):
    method prepare_inputs_for_generation (line 2318) | def prepare_inputs_for_generation(
    method _reorder_cache (line 2344) | def _reorder_cache(past_key_values, beam_idx):
  class XLMProphetNetDecoderWrapper (line 2352) | class XLMProphetNetDecoderWrapper(XLMProphetNetPreTrainedModel):
    method __init__ (line 2358) | def __init__(self, config: XLMProphetNetConfig):
    method forward (line 2362) | def forward(self, *args, **kwargs):

FILE: transformers/models/xlm_prophetnet/tokenization_xlm_prophetnet.py
  function load_vocab (line 48) | def load_vocab(vocab_file):
  class XLMProphetNetTokenizer (line 59) | class XLMProphetNetTokenizer(PreTrainedTokenizer):
    method __init__ (line 133) | def __init__(
    method __getstate__ (line 192) | def __getstate__(self):
    method __setstate__ (line 197) | def __setstate__(self, d):
    method get_special_tokens_mask (line 215) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 243) | def create_token_type_ids_from_sequences(
    method vocab_size (line 268) | def vocab_size(self):
    method get_vocab (line 271) | def get_vocab(self):
    method _tokenize (line 276) | def _tokenize(self, text: str) -> str:
    method _convert_token_to_id (line 279) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 288) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 294) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 299) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method build_inputs_with_special_tokens (line 316) | def build_inputs_with_special_tokens(

FILE: transformers/models/xlm_roberta/configuration_xlm_roberta.py
  class XLMRobertaConfig (line 45) | class XLMRobertaConfig(PretrainedConfig):
    method __init__ (line 115) | def __init__(
  class XLMRobertaOnnxConfig (line 157) | class XLMRobertaOnnxConfig(OnnxConfig):
    method inputs (line 159) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/xlm_roberta/modeling_flax_xlm_roberta.py
  function create_position_ids_from_input_ids (line 62) | def create_position_ids_from_input_ids(input_ids, padding_idx):
  class FlaxXLMRobertaEmbeddings (line 147) | class FlaxXLMRobertaEmbeddings(nn.Module):
    method setup (line 153) | def setup(self):
    method __call__ (line 175) | def __call__(self, input_ids, token_type_ids, position_ids, attention_...
  class FlaxXLMRobertaSelfAttention (line 191) | class FlaxXLMRobertaSelfAttention(nn.Module):
    method setup (line 196) | def setup(self):
    method _split_heads (line 225) | def _split_heads(self, hidden_states):
    method _merge_heads (line 228) | def _merge_heads(self, hidden_states):
    method _concatenate_to_cache (line 233) | def _concatenate_to_cache(self, key, value, query, attention_mask):
    method __call__ (line 264) | def __call__(
  class FlaxXLMRobertaSelfOutput (line 363) | class FlaxXLMRobertaSelfOutput(nn.Module):
    method setup (line 367) | def setup(self):
    method __call__ (line 376) | def __call__(self, hidden_states, input_tensor, deterministic: bool = ...
  class FlaxXLMRobertaAttention (line 384) | class FlaxXLMRobertaAttention(nn.Module):
    method setup (line 389) | def setup(self):
    method __call__ (line 393) | def __call__(
  class FlaxXLMRobertaIntermediate (line 427) | class FlaxXLMRobertaIntermediate(nn.Module):
    method setup (line 431) | def setup(self):
    method __call__ (line 439) | def __call__(self, hidden_states):
  class FlaxXLMRobertaOutput (line 446) | class FlaxXLMRobertaOutput(nn.Module):
    method setup (line 450) | def setup(self):
    method __call__ (line 459) | def __call__(self, hidden_states, attention_output, deterministic: boo...
  class FlaxXLMRobertaLayer (line 467) | class FlaxXLMRobertaLayer(nn.Module):
    method setup (line 471) | def setup(self):
    method __call__ (line 478) | def __call__(
  class FlaxXLMRobertaLayerCollection (line 525) | class FlaxXLMRobertaLayerCollection(nn.Module):
    method setup (line 530) | def setup(self):
    method __call__ (line 543) | def __call__(
  class FlaxXLMRobertaEncoder (line 608) | class FlaxXLMRobertaEncoder(nn.Module):
    method setup (line 613) | def setup(self):
    method __call__ (line 620) | def __call__(
  class FlaxXLMRobertaPooler (line 648) | class FlaxXLMRobertaPooler(nn.Module):
    method setup (line 652) | def setup(self):
    method __call__ (line 659) | def __call__(self, hidden_states):
  class FlaxXLMRobertaLMHead (line 666) | class FlaxXLMRobertaLMHead(nn.Module):
    method setup (line 671) | def setup(self):
    method __call__ (line 686) | def __call__(self, hidden_states, shared_embedding=None):
  class FlaxXLMRobertaClassificationHead (line 702) | class FlaxXLMRobertaClassificationHead(nn.Module):
    method setup (line 706) | def setup(self):
    method __call__ (line 724) | def __call__(self, hidden_states, deterministic=True):
  class FlaxXLMRobertaPreTrainedModel (line 735) | class FlaxXLMRobertaPreTrainedModel(FlaxPreTrainedModel):
    method __init__ (line 746) | def __init__(
    method enable_gradient_checkpointing (line 760) | def enable_gradient_checkpointing(self):
    method init_weights (line 767) | def init_weights(self, rng: jax.random.PRNGKey, input_shape: Tuple, pa...
    method init_cache (line 810) | def init_cache(self, batch_size, max_length):
    method __call__ (line 830) | def __call__(
  class FlaxXLMRobertaModule (line 928) | class FlaxXLMRobertaModule(nn.Module):
    method setup (line 934) | def setup(self):
    method __call__ (line 943) | def __call__(
  class FlaxXLMRobertaModel (line 1003) | class FlaxXLMRobertaModel(FlaxXLMRobertaPreTrainedModel):
  class FlaxXLMRobertaForMaskedLMModule (line 1011) | class FlaxXLMRobertaForMaskedLMModule(nn.Module):
    method setup (line 1016) | def setup(self):
    method __call__ (line 1025) | def __call__(
  class FlaxXLMRobertaForMaskedLM (line 1070) | class FlaxXLMRobertaForMaskedLM(FlaxXLMRobertaPreTrainedModel):
  class FlaxXLMRobertaForSequenceClassificationModule (line 1084) | class FlaxXLMRobertaForSequenceClassificationModule(nn.Module):
    method setup (line 1089) | def setup(self):
    method __call__ (line 1098) | def __call__(
  class FlaxXLMRobertaForSequenceClassification (line 1143) | class FlaxXLMRobertaForSequenceClassification(FlaxXLMRobertaPreTrainedMo...
  class FlaxXLMRobertaForMultipleChoiceModule (line 1156) | class FlaxXLMRobertaForMultipleChoiceModule(nn.Module):
    method setup (line 1161) | def setup(self):
    method __call__ (line 1170) | def __call__(
  class FlaxXLMRobertaForMultipleChoice (line 1224) | class FlaxXLMRobertaForMultipleChoice(FlaxXLMRobertaPreTrainedModel):
  class FlaxXLMRobertaForTokenClassificationModule (line 1240) | class FlaxXLMRobertaForTokenClassificationModule(nn.Module):
    method setup (line 1245) | def setup(self):
    method __call__ (line 1260) | def __call__(
  class FlaxXLMRobertaForTokenClassification (line 1306) | class FlaxXLMRobertaForTokenClassification(FlaxXLMRobertaPreTrainedModel):
  class FlaxXLMRobertaForQuestionAnsweringModule (line 1319) | class FlaxXLMRobertaForQuestionAnsweringModule(nn.Module):
    method setup (line 1324) | def setup(self):
    method __call__ (line 1333) | def __call__(
  class FlaxXLMRobertaForQuestionAnswering (line 1383) | class FlaxXLMRobertaForQuestionAnswering(FlaxXLMRobertaPreTrainedModel):
  class FlaxXLMRobertaForCausalLMModule (line 1396) | class FlaxXLMRobertaForCausalLMModule(nn.Module):
    method setup (line 1401) | def setup(self):
    method __call__ (line 1410) | def __call__(
  class FlaxXLMRobertaForCausalLM (line 1469) | class FlaxXLMRobertaForCausalLM(FlaxXLMRobertaPreTrainedModel):
    method prepare_inputs_for_generation (line 1472) | def prepare_inputs_for_generation(self, input_ids, max_length, attenti...
    method update_inputs_for_generation (line 1493) | def update_inputs_for_generation(self, model_outputs, model_kwargs):

FILE: transformers/models/xlm_roberta/modeling_tf_xlm_roberta.py
  class TFXLMRobertaEmbeddings (line 165) | class TFXLMRobertaEmbeddings(tf.keras.layers.Layer):
    method __init__ (line 170) | def __init__(self, config, **kwargs):
    method build (line 181) | def build(self, input_shape: tf.TensorShape):
    method create_position_ids_from_input_ids (line 205) | def create_position_ids_from_input_ids(self, input_ids, past_key_value...
    method call (line 219) | def call(
  class TFXLMRobertaPooler (line 266) | class TFXLMRobertaPooler(tf.keras.layers.Layer):
    method __init__ (line 267) | def __init__(self, config: XLMRobertaConfig, **kwargs):
    method call (line 277) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFXLMRobertaSelfAttention (line 287) | class TFXLMRobertaSelfAttention(tf.keras.layers.Layer):
    method __init__ (line 288) | def __init__(self, config: XLMRobertaConfig, **kwargs):
    method transpose_for_scores (line 315) | def transpose_for_scores(self, tensor: tf.Tensor, batch_size: int) -> ...
    method call (line 322) | def call(
  class TFXLMRobertaSelfOutput (line 405) | class TFXLMRobertaSelfOutput(tf.keras.layers.Layer):
    method __init__ (line 406) | def __init__(self, config: XLMRobertaConfig, **kwargs):
    method call (line 415) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFXLMRobertaAttention (line 424) | class TFXLMRobertaAttention(tf.keras.layers.Layer):
    method __init__ (line 425) | def __init__(self, config: XLMRobertaConfig, **kwargs):
    method prune_heads (line 431) | def prune_heads(self, heads):
    method call (line 434) | def call(
  class TFXLMRobertaIntermediate (line 465) | class TFXLMRobertaIntermediate(tf.keras.layers.Layer):
    method __init__ (line 466) | def __init__(self, config: XLMRobertaConfig, **kwargs):
    method call (line 478) | def call(self, hidden_states: tf.Tensor) -> tf.Tensor:
  class TFXLMRobertaOutput (line 486) | class TFXLMRobertaOutput(tf.keras.layers.Layer):
    method __init__ (line 487) | def __init__(self, config: XLMRobertaConfig, **kwargs):
    method call (line 496) | def call(self, hidden_states: tf.Tensor, input_tensor: tf.Tensor, trai...
  class TFXLMRobertaLayer (line 505) | class TFXLMRobertaLayer(tf.keras.layers.Layer):
    method __init__ (line 506) | def __init__(self, config: XLMRobertaConfig, **kwargs):
    method call (line 519) | def call(
  class TFXLMRobertaEncoder (line 592) | class TFXLMRobertaEncoder(tf.keras.layers.Layer):
    method __init__ (line 593) | def __init__(self, config: XLMRobertaConfig, **kwargs):
    method call (line 598) | def call(
  class TFXLMRobertaMainLayer (line 663) | class TFXLMRobertaMainLayer(tf.keras.layers.Layer):
    method __init__ (line 666) | def __init__(self, config, add_pooling_layer=True, **kwargs):
    method get_input_embeddings (line 683) | def get_input_embeddings(self) -> tf.keras.layers.Layer:
    method set_input_embeddings (line 687) | def set_input_embeddings(self, value: tf.Variable):
    method _prune_heads (line 692) | def _prune_heads(self, heads_to_prune):
    method call (line 701) | def call(
  class TFXLMRobertaPreTrainedModel (line 860) | class TFXLMRobertaPreTrainedModel(TFPreTrainedModel):
  class TFXLMRobertaModel (line 875) | class TFXLMRobertaModel(TFXLMRobertaPreTrainedModel):
    method __init__ (line 876) | def __init__(self, config, *inputs, **kwargs):
    method call (line 887) | def call(
  class TFXLMRobertaLMHead (line 945) | class TFXLMRobertaLMHead(tf.keras.layers.Layer):
    method __init__ (line 948) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 963) | def build(self, input_shape):
    method get_output_embeddings (line 968) | def get_output_embeddings(self):
    method set_output_embeddings (line 971) | def set_output_embeddings(self, value):
    method get_bias (line 975) | def get_bias(self):
    method set_bias (line 978) | def set_bias(self, value):
    method call (line 982) | def call(self, hidden_states):
  class TFXLMRobertaForMaskedLM (line 999) | class TFXLMRobertaForMaskedLM(TFXLMRobertaPreTrainedModel, TFMaskedLangu...
    method __init__ (line 1003) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 1009) | def get_lm_head(self):
    method get_prefix_bias_name (line 1012) | def get_prefix_bias_name(self):
    method call (line 1026) | def call(
  class TFXLMRobertaForCausalLM (line 1081) | class TFXLMRobertaForCausalLM(TFXLMRobertaPreTrainedModel, TFCausalLangu...
    method __init__ (line 1085) | def __init__(self, config: XLMRobertaConfig, *inputs, **kwargs):
    method get_lm_head (line 1094) | def get_lm_head(self):
    method get_prefix_bias_name (line 1097) | def get_prefix_bias_name(self):
    method prepare_inputs_for_generation (line 1102) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method call (line 1121) | def call(
  class TFXLMRobertaClassificationHead (line 1204) | class TFXLMRobertaClassificationHead(tf.keras.layers.Layer):
    method __init__ (line 1207) | def __init__(self, config, **kwargs):
    method call (line 1223) | def call(self, features, training=False):
  class TFXLMRobertaForSequenceClassification (line 1240) | class TFXLMRobertaForSequenceClassification(TFXLMRobertaPreTrainedModel,...
    method __init__ (line 1244) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1260) | def call(
  class TFXLMRobertaForMultipleChoice (line 1317) | class TFXLMRobertaForMultipleChoice(TFXLMRobertaPreTrainedModel, TFMulti...
    method __init__ (line 1322) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1340) | def call(
  class TFXLMRobertaForTokenClassification (line 1410) | class TFXLMRobertaForTokenClassification(TFXLMRobertaPreTrainedModel, TF...
    method __init__ (line 1415) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1437) | def call(
  class TFXLMRobertaForQuestionAnswering (line 1494) | class TFXLMRobertaForQuestionAnswering(TFXLMRobertaPreTrainedModel, TFQu...
    method __init__ (line 1498) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1516) | def call(

FILE: transformers/models/xlm_roberta/modeling_xlm_roberta.py
  class XLMRobertaEmbeddings (line 66) | class XLMRobertaEmbeddings(nn.Module):
    method __init__ (line 72) | def __init__(self, config):
    method forward (line 95) | def forward(
    method create_position_ids_from_inputs_embeds (line 135) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class XLMRobertaSelfAttention (line 154) | class XLMRobertaSelfAttention(nn.Module):
    method __init__ (line 155) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 181) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 186) | def forward(
  class XLMRobertaSelfOutput (line 289) | class XLMRobertaSelfOutput(nn.Module):
    method __init__ (line 290) | def __init__(self, config):
    method forward (line 296) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class XLMRobertaAttention (line 304) | class XLMRobertaAttention(nn.Module):
    method __init__ (line 305) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 311) | def prune_heads(self, heads):
    method forward (line 329) | def forward(
  class XLMRobertaIntermediate (line 354) | class XLMRobertaIntermediate(nn.Module):
    method __init__ (line 355) | def __init__(self, config):
    method forward (line 363) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class XLMRobertaOutput (line 370) | class XLMRobertaOutput(nn.Module):
    method __init__ (line 371) | def __init__(self, config):
    method forward (line 377) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class XLMRobertaLayer (line 385) | class XLMRobertaLayer(nn.Module):
    method __init__ (line 386) | def __init__(self, config):
    method forward (line 400) | def forward(
    method feed_forward_chunk (line 465) | def feed_forward_chunk(self, attention_output):
  class XLMRobertaEncoder (line 472) | class XLMRobertaEncoder(nn.Module):
    method __init__ (line 473) | def __init__(self, config):
    method forward (line 479) | def forward(
  class XLMRobertaPooler (line 571) | class XLMRobertaPooler(nn.Module):
    method __init__ (line 572) | def __init__(self, config):
    method forward (line 577) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class XLMRobertaPreTrainedModel (line 587) | class XLMRobertaPreTrainedModel(PreTrainedModel):
    method _init_weights (line 599) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 615) | def _set_gradient_checkpointing(self, module, value=False):
    method update_keys_to_ignore (line 619) | def update_keys_to_ignore(self, config, del_keys_to_ignore):
  class XLMRobertaModel (line 700) | class XLMRobertaModel(XLMRobertaPreTrainedModel):
    method __init__ (line 719) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 731) | def get_input_embeddings(self):
    method set_input_embeddings (line 734) | def set_input_embeddings(self, value):
    method _prune_heads (line 737) | def _prune_heads(self, heads_to_prune):
    method forward (line 752) | def forward(
  class XLMRobertaForCausalLM (line 887) | class XLMRobertaForCausalLM(XLMRobertaPreTrainedModel):
    method __init__ (line 892) | def __init__(self, config):
    method get_output_embeddings (line 907) | def get_output_embeddings(self):
    method set_output_embeddings (line 910) | def set_output_embeddings(self, new_embeddings):
    method forward (line 915) | def forward(
    method prepare_inputs_for_generation (line 1021) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1033) | def _reorder_cache(self, past_key_values, beam_idx):
  class XLMRobertaForMaskedLM (line 1045) | class XLMRobertaForMaskedLM(XLMRobertaPreTrainedModel):
    method __init__ (line 1050) | def __init__(self, config):
    method get_output_embeddings (line 1068) | def get_output_embeddings(self):
    method set_output_embeddings (line 1071) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1083) | def forward(
  class XLMRobertaLMHead (line 1144) | class XLMRobertaLMHead(nn.Module):
    method __init__ (line 1147) | def __init__(self, config):
    method forward (line 1156) | def forward(self, features, **kwargs):
    method _tie_weights (line 1166) | def _tie_weights(self):
  class XLMRobertaForSequenceClassification (line 1183) | class XLMRobertaForSequenceClassification(XLMRobertaPreTrainedModel):
    method __init__ (line 1186) | def __init__(self, config):
    method forward (line 1205) | def forward(
  class XLMRobertaForMultipleChoice (line 1285) | class XLMRobertaForMultipleChoice(XLMRobertaPreTrainedModel):
    method __init__ (line 1288) | def __init__(self, config):
    method forward (line 1306) | def forward(
  class XLMRobertaForTokenClassification (line 1382) | class XLMRobertaForTokenClassification(XLMRobertaPreTrainedModel):
    method __init__ (line 1386) | def __init__(self, config):
    method forward (line 1408) | def forward(
  class XLMRobertaClassificationHead (line 1464) | class XLMRobertaClassificationHead(nn.Module):
    method __init__ (line 1467) | def __init__(self, config):
    method forward (line 1476) | def forward(self, features, **kwargs):
  class XLMRobertaForQuestionAnswering (line 1494) | class XLMRobertaForQuestionAnswering(XLMRobertaPreTrainedModel):
    method __init__ (line 1498) | def __init__(self, config):
    method forward (line 1516) | def forward(
  function create_position_ids_from_input_ids (line 1592) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/xlm_roberta/tokenization_xlm_roberta.py
  class XLMRobertaTokenizer (line 63) | class XLMRobertaTokenizer(PreTrainedTokenizer):
    method __init__ (line 137) | def __init__(
    method __getstate__ (line 186) | def __getstate__(self):
    method __setstate__ (line 192) | def __setstate__(self, d):
    method build_inputs_with_special_tokens (line 202) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 228) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 256) | def create_token_type_ids_from_sequences(
    method vocab_size (line 282) | def vocab_size(self):
    method get_vocab (line 285) | def get_vocab(self):
    method _tokenize (line 290) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 293) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 302) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 308) | def convert_tokens_to_string(self, tokens):
    method save_vocabulary (line 313) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/xlm_roberta/tokenization_xlm_roberta_fast.py
  class XLMRobertaTokenizerFast (line 82) | class XLMRobertaTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 139) | def __init__(
    method build_inputs_with_special_tokens (line 171) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 197) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 222) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/xlm_roberta_xl/configuration_xlm_roberta_xl.py
  class XLMRobertaXLConfig (line 34) | class XLMRobertaXLConfig(PretrainedConfig):
    method __init__ (line 102) | def __init__(
  class XLMRobertaXLOnnxConfig (line 143) | class XLMRobertaXLOnnxConfig(OnnxConfig):
    method inputs (line 145) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/xlm_roberta_xl/convert_xlm_roberta_xl_original_pytorch_checkpoint_to_pytorch.py
  function convert_xlm_roberta_xl_checkpoint_to_pytorch (line 47) | def convert_xlm_roberta_xl_checkpoint_to_pytorch(

FILE: transformers/models/xlm_roberta_xl/modeling_xlm_roberta_xl.py
  class XLMRobertaXLEmbeddings (line 60) | class XLMRobertaXLEmbeddings(nn.Module):
    method __init__ (line 65) | def __init__(self, config):
    method forward (line 87) | def forward(
    method create_position_ids_from_inputs_embeds (line 128) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class XLMRobertaXLSelfAttention (line 147) | class XLMRobertaXLSelfAttention(nn.Module):
    method __init__ (line 148) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 174) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 179) | def forward(
  class XLMRobertaXLSelfOutput (line 281) | class XLMRobertaXLSelfOutput(nn.Module):
    method __init__ (line 282) | def __init__(self, config):
    method forward (line 287) | def forward(self, hidden_states, input_tensor):
  class XLMRobertaXLAttention (line 294) | class XLMRobertaXLAttention(nn.Module):
    method __init__ (line 295) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 302) | def prune_heads(self, heads):
    method forward (line 320) | def forward(
  class XLMRobertaXLIntermediate (line 346) | class XLMRobertaXLIntermediate(nn.Module):
    method __init__ (line 347) | def __init__(self, config):
    method forward (line 355) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class XLMRobertaXLOutput (line 361) | class XLMRobertaXLOutput(nn.Module):
    method __init__ (line 362) | def __init__(self, config):
    method forward (line 366) | def forward(self, hidden_states, input_tensor):
  class XLMRobertaXLLayer (line 372) | class XLMRobertaXLLayer(nn.Module):
    method __init__ (line 373) | def __init__(self, config):
    method forward (line 388) | def forward(
    method feed_forward_chunk (line 453) | def feed_forward_chunk(self, attention_output):
  class XLMRobertaXLEncoder (line 460) | class XLMRobertaXLEncoder(nn.Module):
    method __init__ (line 461) | def __init__(self, config):
    method forward (line 468) | def forward(
  class XLMRobertaXLPooler (line 561) | class XLMRobertaXLPooler(nn.Module):
    method __init__ (line 562) | def __init__(self, config):
    method forward (line 567) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class XLMRobertaXLPreTrainedModel (line 576) | class XLMRobertaXLPreTrainedModel(PreTrainedModel):
    method _init_weights (line 586) | def _init_weights(self, module):
    method update_keys_to_ignore (line 602) | def update_keys_to_ignore(self, config, del_keys_to_ignore):
  class XLMRobertaXLModel (line 671) | class XLMRobertaXLModel(XLMRobertaXLPreTrainedModel):
    method __init__ (line 685) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 697) | def get_input_embeddings(self):
    method set_input_embeddings (line 700) | def set_input_embeddings(self, value):
    method _prune_heads (line 703) | def _prune_heads(self, heads_to_prune):
    method forward (line 718) | def forward(
  class XLMRobertaXLForCausalLM (line 852) | class XLMRobertaXLForCausalLM(XLMRobertaXLPreTrainedModel):
    method __init__ (line 857) | def __init__(self, config):
    method get_output_embeddings (line 871) | def get_output_embeddings(self):
    method set_output_embeddings (line 874) | def set_output_embeddings(self, new_embeddings):
    method forward (line 879) | def forward(
    method prepare_inputs_for_generation (line 980) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 992) | def _reorder_cache(self, past_key_values, beam_idx):
  class XLMRobertaXLForMaskedLM (line 1002) | class XLMRobertaXLForMaskedLM(XLMRobertaXLPreTrainedModel):
    method __init__ (line 1007) | def __init__(self, config):
    method get_output_embeddings (line 1024) | def get_output_embeddings(self):
    method set_output_embeddings (line 1027) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1037) | def forward(
  class XLMRobertaXLLMHead (line 1095) | class XLMRobertaXLLMHead(nn.Module):
    method __init__ (line 1098) | def __init__(self, config):
    method forward (line 1107) | def forward(self, features, **kwargs):
    method _tie_weights (line 1117) | def _tie_weights(self):
  class XLMRobertaXLForSequenceClassification (line 1129) | class XLMRobertaXLForSequenceClassification(XLMRobertaXLPreTrainedModel):
    method __init__ (line 1132) | def __init__(self, config):
    method forward (line 1148) | def forward(
  class XLMRobertaXLForMultipleChoice (line 1225) | class XLMRobertaXLForMultipleChoice(XLMRobertaXLPreTrainedModel):
    method __init__ (line 1228) | def __init__(self, config):
    method forward (line 1245) | def forward(
  class XLMRobertaXLForTokenClassification (line 1318) | class XLMRobertaXLForTokenClassification(XLMRobertaXLPreTrainedModel):
    method __init__ (line 1322) | def __init__(self, config):
    method forward (line 1341) | def forward(
  class XLMRobertaXLClassificationHead (line 1403) | class XLMRobertaXLClassificationHead(nn.Module):
    method __init__ (line 1406) | def __init__(self, config):
    method forward (line 1415) | def forward(self, features, **kwargs):
  class XLMRobertaXLForQuestionAnswering (line 1432) | class XLMRobertaXLForQuestionAnswering(XLMRobertaXLPreTrainedModel):
    method __init__ (line 1436) | def __init__(self, config):
    method forward (line 1451) | def forward(
  function create_position_ids_from_input_ids (line 1527) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/xlnet/configuration_xlnet.py
  class XLNetConfig (line 32) | class XLNetConfig(PretrainedConfig):
    method __init__ (line 150) | def __init__(
    method max_position_embeddings (line 233) | def max_position_embeddings(self):
    method max_position_embeddings (line 238) | def max_position_embeddings(self, value):

FILE: transformers/models/xlnet/convert_xlnet_original_tf_checkpoint_to_pytorch.py
  function convert_xlnet_checkpoint_to_pytorch (line 49) | def convert_xlnet_checkpoint_to_pytorch(

FILE: transformers/models/xlnet/modeling_tf_xlnet.py
  class TFXLNetRelativeAttention (line 69) | class TFXLNetRelativeAttention(tf.keras.layers.Layer):
    method __init__ (line 70) | def __init__(self, config, **kwargs):
    method build (line 89) | def build(self, input_shape):
    method prune_heads (line 120) | def prune_heads(self, heads):
    method rel_shift (line 123) | def rel_shift(self, x, klen=-1):
    method rel_attn_core (line 135) | def rel_attn_core(
    method post_attention (line 179) | def post_attention(self, h, attn_vec, residual=True, training=False):
    method call (line 192) | def call(
  class TFXLNetFeedForward (line 332) | class TFXLNetFeedForward(tf.keras.layers.Layer):
    method __init__ (line 333) | def __init__(self, config, **kwargs):
    method call (line 348) | def call(self, inp, training=False):
  class TFXLNetLayer (line 359) | class TFXLNetLayer(tf.keras.layers.Layer):
    method __init__ (line 360) | def __init__(self, config, **kwargs):
    method call (line 366) | def call(
  class TFXLNetLMHead (line 403) | class TFXLNetLMHead(tf.keras.layers.Layer):
    method __init__ (line 404) | def __init__(self, config, input_embeddings, **kwargs):
    method build (line 411) | def build(self, input_shape):
    method get_output_embeddings (line 415) | def get_output_embeddings(self):
    method set_output_embeddings (line 418) | def set_output_embeddings(self, value):
    method get_bias (line 422) | def get_bias(self):
    method set_bias (line 425) | def set_bias(self, value):
    method call (line 429) | def call(self, hidden_states):
  class TFXLNetMainLayer (line 436) | class TFXLNetMainLayer(tf.keras.layers.Layer):
    method __init__ (line 439) | def __init__(self, config, **kwargs):
    method get_input_embeddings (line 467) | def get_input_embeddings(self):
    method set_input_embeddings (line 470) | def set_input_embeddings(self, value):
    method build (line 474) | def build(self, input_shape):
    method _prune_heads (line 481) | def _prune_heads(self, heads_to_prune):
    method create_mask (line 484) | def create_mask(self, qlen, mlen):
    method cache_mem (line 513) | def cache_mem(self, curr_out, prev_mem):
    method positional_embedding (line 535) | def positional_embedding(pos_seq, inv_freq, bsz=None):
    method relative_positional_encoding (line 545) | def relative_positional_encoding(self, qlen, klen, bsz=None):
    method call (line 586) | def call(
  class TFXLNetPreTrainedModel (line 791) | class TFXLNetPreTrainedModel(TFPreTrainedModel):
  class TFXLNetModelOutput (line 802) | class TFXLNetModelOutput(ModelOutput):
  class TFXLNetLMHeadModelOutput (line 836) | class TFXLNetLMHeadModelOutput(ModelOutput):
  class TFXLNetForSequenceClassificationOutput (line 873) | class TFXLNetForSequenceClassificationOutput(ModelOutput):
  class TFXLNetForTokenClassificationOutput (line 907) | class TFXLNetForTokenClassificationOutput(ModelOutput):
  class TFXLNetForMultipleChoiceOutput (line 941) | class TFXLNetForMultipleChoiceOutput(ModelOutput):
  class TFXLNetForQuestionAnsweringSimpleOutput (line 977) | class TFXLNetForQuestionAnsweringSimpleOutput(ModelOutput):
  class TFXLNetModel (line 1132) | class TFXLNetModel(TFXLNetPreTrainedModel):
    method __init__ (line 1133) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1144) | def call(
  class TFXLNetLMHeadModel (line 1187) | class TFXLNetLMHeadModel(TFXLNetPreTrainedModel, TFCausalLanguageModelin...
    method __init__ (line 1188) | def __init__(self, config, *inputs, **kwargs):
    method get_lm_head (line 1195) | def get_lm_head(self):
    method get_prefix_bias_name (line 1198) | def get_prefix_bias_name(self):
    method prepare_inputs_for_generation (line 1202) | def prepare_inputs_for_generation(self, inputs, past_key_values=None, ...
    method call (line 1244) | def call(
  class TFXLNetForSequenceClassification (line 1347) | class TFXLNetForSequenceClassification(TFXLNetPreTrainedModel, TFSequenc...
    method __init__ (line 1348) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1367) | def call(
  class TFXLNetForMultipleChoice (line 1434) | class TFXLNetForMultipleChoice(TFXLNetPreTrainedModel, TFMultipleChoiceL...
    method __init__ (line 1435) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1453) | def call(
  class TFXLNetForTokenClassification (line 1535) | class TFXLNetForTokenClassification(TFXLNetPreTrainedModel, TFTokenClass...
    method __init__ (line 1536) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1552) | def call(
  class TFXLNetForQuestionAnsweringSimple (line 1615) | class TFXLNetForQuestionAnsweringSimple(TFXLNetPreTrainedModel, TFQuesti...
    method __init__ (line 1616) | def __init__(self, config, *inputs, **kwargs):
    method call (line 1630) | def call(

FILE: transformers/models/xlnet/modeling_xlnet.py
  function build_tf_xlnet_to_pytorch_map (line 53) | def build_tf_xlnet_to_pytorch_map(model, config, tf_weights=None):
  function load_tf_weights_in_xlnet (line 136) | def load_tf_weights_in_xlnet(model, config, tf_path):
  class XLNetRelativeAttention (line 203) | class XLNetRelativeAttention(nn.Module):
    method __init__ (line 204) | def __init__(self, config):
    method prune_heads (line 232) | def prune_heads(self, heads):
    method rel_shift (line 236) | def rel_shift(x, klen=-1):
    method rel_shift_bnij (line 249) | def rel_shift_bnij(x, klen=-1):
    method rel_attn_core (line 263) | def rel_attn_core(
    method post_attention (line 315) | def post_attention(self, h, attn_vec, residual=True):
    method forward (line 327) | def forward(
  class XLNetFeedForward (line 463) | class XLNetFeedForward(nn.Module):
    method __init__ (line 464) | def __init__(self, config):
    method forward (line 475) | def forward(self, inp):
  class XLNetLayer (line 486) | class XLNetLayer(nn.Module):
    method __init__ (line 487) | def __init__(self, config):
    method forward (line 495) | def forward(
    method ff_chunk (line 531) | def ff_chunk(self, output_x):
  class XLNetPreTrainedModel (line 536) | class XLNetPreTrainedModel(PreTrainedModel):
    method _init_weights (line 546) | def _init_weights(self, module):
  class XLNetModelOutput (line 579) | class XLNetModelOutput(ModelOutput):
  class XLNetLMHeadModelOutput (line 613) | class XLNetLMHeadModelOutput(ModelOutput):
  class XLNetForSequenceClassificationOutput (line 650) | class XLNetForSequenceClassificationOutput(ModelOutput):
  class XLNetForTokenClassificationOutput (line 684) | class XLNetForTokenClassificationOutput(ModelOutput):
  class XLNetForMultipleChoiceOutput (line 718) | class XLNetForMultipleChoiceOutput(ModelOutput):
  class XLNetForQuestionAnsweringSimpleOutput (line 754) | class XLNetForQuestionAnsweringSimpleOutput(ModelOutput):
  class XLNetForQuestionAnsweringOutput (line 791) | class XLNetForQuestionAnsweringOutput(ModelOutput):
  class XLNetModel (line 931) | class XLNetModel(XLNetPreTrainedModel):
    method __init__ (line 932) | def __init__(self, config):
    method get_input_embeddings (line 952) | def get_input_embeddings(self):
    method set_input_embeddings (line 955) | def set_input_embeddings(self, new_embeddings):
    method _prune_heads (line 958) | def _prune_heads(self, heads_to_prune):
    method create_mask (line 961) | def create_mask(self, qlen, mlen):
    method cache_mem (line 989) | def cache_mem(self, curr_out, prev_mem):
    method positional_embedding (line 1011) | def positional_embedding(pos_seq, inv_freq, bsz=None):
    method relative_positional_encoding (line 1021) | def relative_positional_encoding(self, qlen, klen, bsz=None):
    method forward (line 1065) | def forward(
  class XLNetLMHeadModel (line 1294) | class XLNetLMHeadModel(XLNetPreTrainedModel):
    method __init__ (line 1297) | def __init__(self, config):
    method get_output_embeddings (line 1308) | def get_output_embeddings(self):
    method set_output_embeddings (line 1311) | def set_output_embeddings(self, new_embeddings):
    method prepare_inputs_for_generation (line 1314) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method forward (line 1358) | def forward(
    method _reorder_cache (line 1484) | def _reorder_cache(mems: List[torch.Tensor], beam_idx: torch.Tensor) -...
  class XLNetForSequenceClassification (line 1500) | class XLNetForSequenceClassification(XLNetPreTrainedModel):
    method __init__ (line 1501) | def __init__(self, config):
    method forward (line 1519) | def forward(
  class XLNetForTokenClassification (line 1609) | class XLNetForTokenClassification(XLNetPreTrainedModel):
    method __init__ (line 1610) | def __init__(self, config):
    method forward (line 1626) | def forward(
  class XLNetForMultipleChoice (line 1696) | class XLNetForMultipleChoice(XLNetPreTrainedModel):
    method __init__ (line 1697) | def __init__(self, config):
    method forward (line 1713) | def forward(
  class XLNetForQuestionAnsweringSimple (line 1799) | class XLNetForQuestionAnsweringSimple(XLNetPreTrainedModel):
    method __init__ (line 1800) | def __init__(self, config):
    method forward (line 1816) | def forward(
  class XLNetForQuestionAnswering (line 1909) | class XLNetForQuestionAnswering(XLNetPreTrainedModel):
    method __init__ (line 1910) | def __init__(self, config):
    method forward (line 1925) | def forward(

FILE: transformers/models/xlnet/tokenization_xlnet.py
  class XLNetTokenizer (line 53) | class XLNetTokenizer(PreTrainedTokenizer):
    method __init__ (line 133) | def __init__(
    method vocab_size (line 182) | def vocab_size(self):
    method get_vocab (line 185) | def get_vocab(self):
    method __getstate__ (line 190) | def __getstate__(self):
    method __setstate__ (line 195) | def __setstate__(self, d):
    method preprocess_text (line 205) | def preprocess_text(self, inputs):
    method _tokenize (line 220) | def _tokenize(self, text: str) -> List[str]:
    method _convert_token_to_id (line 240) | def _convert_token_to_id(self, token):
    method _convert_id_to_token (line 244) | def _convert_id_to_token(self, index):
    method convert_tokens_to_string (line 248) | def convert_tokens_to_string(self, tokens):
    method _decode (line 253) | def _decode(
    method build_inputs_with_special_tokens (line 298) | def build_inputs_with_special_tokens(
    method get_special_tokens_mask (line 323) | def get_special_tokens_mask(
    method create_token_type_ids_from_sequences (line 351) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 381) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/xlnet/tokenization_xlnet_fast.py
  class XLNetTokenizerFast (line 63) | class XLNetTokenizerFast(PreTrainedTokenizerFast):
    method __init__ (line 130) | def __init__(
    method build_inputs_with_special_tokens (line 174) | def build_inputs_with_special_tokens(
    method create_token_type_ids_from_sequences (line 199) | def create_token_type_ids_from_sequences(
    method save_vocabulary (line 229) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...

FILE: transformers/models/xmod/configuration_xmod.py
  class XmodConfig (line 40) | class XmodConfig(PretrainedConfig):
    method __init__ (line 124) | def __init__(
  class XmodOnnxConfig (line 180) | class XmodOnnxConfig(OnnxConfig):
    method inputs (line 182) | def inputs(self) -> Mapping[str, Mapping[int, str]]:

FILE: transformers/models/xmod/convert_xmod_original_pytorch_checkpoint_to_pytorch.py
  function convert_xmod_checkpoint_to_pytorch (line 41) | def convert_xmod_checkpoint_to_pytorch(

FILE: transformers/models/xmod/modeling_xmod.py
  class XmodEmbeddings (line 59) | class XmodEmbeddings(nn.Module):
    method __init__ (line 65) | def __init__(self, config):
    method forward (line 88) | def forward(
    method create_position_ids_from_inputs_embeds (line 128) | def create_position_ids_from_inputs_embeds(self, inputs_embeds):
  class XmodSelfAttention (line 147) | class XmodSelfAttention(nn.Module):
    method __init__ (line 148) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 174) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 179) | def forward(
  class XmodSelfOutput (line 281) | class XmodSelfOutput(nn.Module):
    method __init__ (line 283) | def __init__(self, config):
    method forward (line 289) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class XmodAttention (line 296) | class XmodAttention(nn.Module):
    method __init__ (line 297) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 305) | def prune_heads(self, heads):
    method forward (line 323) | def forward(
  class XmodIntermediate (line 353) | class XmodIntermediate(nn.Module):
    method __init__ (line 354) | def __init__(self, config):
    method forward (line 362) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class XmodAdapter (line 368) | class XmodAdapter(nn.Module):
    method __init__ (line 369) | def __init__(self, config):
    method forward (line 379) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class XmodOutput (line 386) | class XmodOutput(nn.Module):
    method __init__ (line 387) | def __init__(self, config):
    method forward (line 402) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
    method lang_adapter (line 409) | def lang_adapter(self, lang_ids: torch.Tensor, hidden_states: torch.Te...
  class XmodLayer (line 436) | class XmodLayer(nn.Module):
    method __init__ (line 437) | def __init__(self, config):
    method forward (line 452) | def forward(
    method feed_forward_chunk (line 527) | def feed_forward_chunk(self, attention_output):
  class XmodEncoder (line 531) | class XmodEncoder(nn.Module):
    method __init__ (line 532) | def __init__(self, config):
    method forward (line 541) | def forward(
  class XmodPooler (line 638) | class XmodPooler(nn.Module):
    method __init__ (line 639) | def __init__(self, config):
    method forward (line 644) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class XmodPreTrainedModel (line 653) | class XmodPreTrainedModel(PreTrainedModel):
    method _init_weights (line 664) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 681) | def _set_gradient_checkpointing(self, module, value=False):
    method update_keys_to_ignore (line 686) | def update_keys_to_ignore(self, config, del_keys_to_ignore):
    method set_default_language (line 695) | def set_default_language(self, language: str):
    method freeze_embeddings_and_language_adapters (line 708) | def freeze_embeddings_and_language_adapters(self):
  class XmodModel (line 798) | class XmodModel(XmodPreTrainedModel):
    method __init__ (line 817) | def __init__(self, config, add_pooling_layer=True):
    method get_input_embeddings (line 830) | def get_input_embeddings(self):
    method set_input_embeddings (line 834) | def set_input_embeddings(self, value):
    method _prune_heads (line 838) | def _prune_heads(self, heads_to_prune):
    method forward (line 847) | def forward(
  class XmodForCausalLM (line 991) | class XmodForCausalLM(XmodPreTrainedModel):
    method __init__ (line 997) | def __init__(self, config):
    method get_output_embeddings (line 1013) | def get_output_embeddings(self):
    method set_output_embeddings (line 1017) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1021) | def forward(
    method prepare_inputs_for_generation (line 1129) | def prepare_inputs_for_generation(self, input_ids, past_key_values=Non...
    method _reorder_cache (line 1142) | def _reorder_cache(self, past_key_values, beam_idx):
  class XmodForMaskedLM (line 1153) | class XmodForMaskedLM(XmodPreTrainedModel):
    method __init__ (line 1159) | def __init__(self, config):
    method get_output_embeddings (line 1178) | def get_output_embeddings(self):
    method set_output_embeddings (line 1182) | def set_output_embeddings(self, new_embeddings):
    method forward (line 1186) | def forward(
  class XmodLMHead (line 1247) | class XmodLMHead(nn.Module):
    method __init__ (line 1250) | def __init__(self, config):
    method forward (line 1259) | def forward(self, features, **kwargs):
    method _tie_weights (line 1269) | def _tie_weights(self):
  class XmodForSequenceClassification (line 1285) | class XmodForSequenceClassification(XmodPreTrainedModel):
    method __init__ (line 1289) | def __init__(self, config):
    method forward (line 1301) | def forward(
  class XmodForMultipleChoice (line 1380) | class XmodForMultipleChoice(XmodPreTrainedModel):
    method __init__ (line 1384) | def __init__(self, config):
    method forward (line 1395) | def forward(
  class XmodForTokenClassification (line 1471) | class XmodForTokenClassification(XmodPreTrainedModel):
    method __init__ (line 1476) | def __init__(self, config):
    method forward (line 1491) | def forward(
  class XmodClassificationHead (line 1547) | class XmodClassificationHead(nn.Module):
    method __init__ (line 1550) | def __init__(self, config):
    method forward (line 1559) | def forward(self, features, **kwargs):
  class XmodForQuestionAnswering (line 1576) | class XmodForQuestionAnswering(XmodPreTrainedModel):
    method __init__ (line 1581) | def __init__(self, config):
    method forward (line 1592) | def forward(
  function create_position_ids_from_input_ids (line 1670) | def create_position_ids_from_input_ids(input_ids, padding_idx, past_key_...

FILE: transformers/models/yolos/configuration_yolos.py
  class YolosConfig (line 35) | class YolosConfig(PretrainedConfig):
    method __init__ (line 108) | def __init__(
  class YolosOnnxConfig (line 162) | class YolosOnnxConfig(OnnxConfig):
    method inputs (line 166) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method atol_for_validation (line 174) | def atol_for_validation(self) -> float:
    method default_onnx_opset (line 178) | def default_onnx_opset(self) -> int:

FILE: transformers/models/yolos/convert_yolos_to_pytorch.py
  function get_yolos_config (line 35) | def get_yolos_config(yolos_name: str) -> YolosConfig:
  function read_in_q_k_v (line 71) | def read_in_q_k_v(state_dict: dict, config: YolosConfig, base_model: boo...
  function rename_key (line 89) | def rename_key(name: str) -> str:
  function convert_state_dict (line 126) | def convert_state_dict(orig_state_dict: dict, model: YolosForObjectDetec...
  function prepare_img (line 151) | def prepare_img() -> torch.Tensor:
  function convert_yolos_checkpoint (line 158) | def convert_yolos_checkpoint(

FILE: transformers/models/yolos/feature_extraction_yolos.py
  class YolosFeatureExtractor (line 26) | class YolosFeatureExtractor(YolosImageProcessor):
    method __init__ (line 27) | def __init__(self, *args, **kwargs) -> None:

FILE: transformers/models/yolos/image_processing_yolos.py
  class AnnotionFormat (line 83) | class AnnotionFormat(ExplicitEnum):
  function get_max_height_width (line 92) | def get_max_height_width(images: List[np.ndarray]) -> List[int]:
  function get_size_with_aspect_ratio (line 108) | def get_size_with_aspect_ratio(image_size, size, max_size=None) -> Tuple...
  function get_resize_output_image_size (line 140) | def get_resize_output_image_size(
  function get_numpy_to_framework_fn (line 164) | def get_numpy_to_framework_fn(arr) -> Callable:
  function safe_squeeze (line 189) | def safe_squeeze(arr: np.ndarray, axis: Optional[int] = None) -> np.ndar...
  function normalize_annotation (line 203) | def normalize_annotation(annotation: Dict, image_size: Tuple[int, int]) ...
  function max_across_indices (line 218) | def max_across_indices(values: Iterable[Any]) -> List[Any]:
  function make_pixel_mask (line 226) | def make_pixel_mask(image: np.ndarray, output_size: Tuple[int, int]) -> ...
  function convert_coco_poly_to_mask (line 243) | def convert_coco_poly_to_mask(segmentations, height: int, width: int) ->...
  function prepare_coco_detection_annotation (line 278) | def prepare_coco_detection_annotation(image, target, return_segmentation...
  function masks_to_boxes (line 331) | def masks_to_boxes(masks: np.ndarray) -> np.ndarray:
  function prepare_coco_panoptic_annotation (line 366) | def prepare_coco_panoptic_annotation(
  function get_segmentation_image (line 404) | def get_segmentation_image(
  function get_mask_area (line 430) | def get_mask_area(seg_img: np.ndarray, target_size: Tuple[int, int], n_c...
  function score_labels_from_class_probabilities (line 440) | def score_labels_from_class_probabilities(logits: np.ndarray) -> Tuple[n...
  function resize_annotation (line 449) | def resize_annotation(
  function binary_mask_to_rle (line 501) | def binary_mask_to_rle(mask):
  function convert_segmentation_to_rle (line 524) | def convert_segmentation_to_rle(segmentation):
  function remove_low_and_no_objects (line 546) | def remove_low_and_no_objects(masks, scores, labels, object_mask_thresho...
  function check_segment_validity (line 575) | def check_segment_validity(mask_labels, mask_probs, k, mask_threshold=0....
  function compute_segments (line 594) | def compute_segments(
  class YolosImageProcessor (line 654) | class YolosImageProcessor(BaseImageProcessor):
    method __init__ (line 691) | def __init__(
    method max_size (line 734) | def max_size(self):
    method from_dict (line 743) | def from_dict(cls, image_processor_dict: Dict[str, Any], **kwargs):
    method prepare_annotation (line 757) | def prepare_annotation(
    method prepare (line 783) | def prepare(self, image, target, return_segmentation_masks=False, mask...
    method convert_coco_poly_to_mask (line 793) | def convert_coco_poly_to_mask(self, *args, **kwargs):
    method prepare_coco_detection (line 800) | def prepare_coco_detection(self, *args, **kwargs):
    method prepare_coco_panoptic (line 807) | def prepare_coco_panoptic(self, *args, **kwargs):
    method resize (line 814) | def resize(
    method resize_annotation (line 848) | def resize_annotation(
    method rescale (line 862) | def rescale(
    method normalize (line 871) | def normalize(
    method normalize_annotation (line 884) | def normalize_annotation(self, annotation: Dict, image_size: Tuple[int...
    method _pad_image (line 892) | def _pad_image(
    method pad (line 913) | def pad(
    method preprocess (line 945) | def preprocess(
    method post_process (line 1151) | def post_process(self, outputs, target_sizes):
    method post_process_object_detection (line 1193) | def post_process_object_detection(

FILE: transformers/models/yolos/modeling_yolos.py
  class YolosObjectDetectionOutput (line 69) | class YolosObjectDetectionOutput(ModelOutput):
  class YolosEmbeddings (line 113) | class YolosEmbeddings(nn.Module):
    method __init__ (line 119) | def __init__(self, config: YolosConfig) -> None:
    method forward (line 134) | def forward(self, pixel_values: torch.Tensor) -> torch.Tensor:
  class InterpolateInitialPositionEmbeddings (line 156) | class InterpolateInitialPositionEmbeddings(nn.Module):
    method __init__ (line 157) | def __init__(self, config) -> None:
    method forward (line 161) | def forward(self, pos_embed, img_size=(800, 1344)) -> torch.Tensor:
  class InterpolateMidPositionEmbeddings (line 185) | class InterpolateMidPositionEmbeddings(nn.Module):
    method __init__ (line 186) | def __init__(self, config) -> None:
    method forward (line 190) | def forward(self, pos_embed, img_size=(800, 1344)) -> torch.Tensor:
  class YolosPatchEmbeddings (line 218) | class YolosPatchEmbeddings(nn.Module):
    method __init__ (line 225) | def __init__(self, config):
    method forward (line 240) | def forward(self, pixel_values: torch.Tensor) -> torch.Tensor:
  class YolosSelfAttention (line 252) | class YolosSelfAttention(nn.Module):
    method __init__ (line 253) | def __init__(self, config: YolosConfig) -> None:
    method transpose_for_scores (line 271) | def transpose_for_scores(self, x: torch.Tensor) -> torch.Tensor:
    method forward (line 276) | def forward(
  class YolosSelfOutput (line 313) | class YolosSelfOutput(nn.Module):
    method __init__ (line 319) | def __init__(self, config: YolosConfig) -> None:
    method forward (line 324) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class YolosAttention (line 332) | class YolosAttention(nn.Module):
    method __init__ (line 333) | def __init__(self, config: YolosConfig) -> None:
    method prune_heads (line 339) | def prune_heads(self, heads: Set[int]) -> None:
    method forward (line 357) | def forward(
  class YolosIntermediate (line 372) | class YolosIntermediate(nn.Module):
    method __init__ (line 373) | def __init__(self, config: YolosConfig) -> None:
    method forward (line 381) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class YolosOutput (line 389) | class YolosOutput(nn.Module):
    method __init__ (line 390) | def __init__(self, config: YolosConfig) -> None:
    method forward (line 395) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class YolosLayer (line 405) | class YolosLayer(nn.Module):
    method __init__ (line 408) | def __init__(self, config: YolosConfig) -> None:
    method forward (line 418) | def forward(
  class YolosEncoder (line 447) | class YolosEncoder(nn.Module):
    method __init__ (line 448) | def __init__(self, config: YolosConfig) -> None:
    method forward (line 472) | def forward(
  class YolosPreTrainedModel (line 531) | class YolosPreTrainedModel(PreTrainedModel):
    method _init_weights (line 542) | def _init_weights(self, module: Union[nn.Linear, nn.Conv2d, nn.LayerNo...
    method _set_gradient_checkpointing (line 554) | def _set_gradient_checkpointing(self, module: YolosEncoder, value: boo...
  class YolosModel (line 597) | class YolosModel(YolosPreTrainedModel):
    method __init__ (line 598) | def __init__(self, config: YolosConfig, add_pooling_layer: bool = True):
    method get_input_embeddings (line 611) | def get_input_embeddings(self) -> YolosPatchEmbeddings:
    method _prune_heads (line 614) | def _prune_heads(self, heads_to_prune: Dict[int, List[int]]) -> None:
    method forward (line 633) | def forward(
  class YolosPooler (line 684) | class YolosPooler(nn.Module):
    method __init__ (line 685) | def __init__(self, config: YolosConfig):
    method forward (line 690) | def forward(self, hidden_states):
  class YolosForObjectDetection (line 705) | class YolosForObjectDetection(YolosPreTrainedModel):
    method __init__ (line 706) | def __init__(self, config: YolosConfig):
    method _set_aux_loss (line 726) | def _set_aux_loss(self, outputs_class, outputs_coord):
    method forward (line 734) | def forward(
  function dice_loss (line 863) | def dice_loss(inputs, targets, num_boxes):
  function sigmoid_focal_loss (line 883) | def sigmoid_focal_loss(inputs, targets, num_boxes, alpha: float = 0.25, ...
  class YolosLoss (line 915) | class YolosLoss(nn.Module):
    method __init__ (line 940) | def __init__(self, matcher, num_classes, eos_coef, losses):
    method loss_labels (line 951) | def loss_labels(self, outputs, targets, indices, num_boxes):
    method loss_cardinality (line 973) | def loss_cardinality(self, outputs, targets, indices, num_boxes):
    method loss_boxes (line 988) | def loss_boxes(self, outputs, targets, indices, num_boxes):
    method loss_masks (line 1012) | def loss_masks(self, outputs, targets, indices, num_boxes):
    method _get_source_permutation_idx (line 1045) | def _get_source_permutation_idx(self, indices):
    method _get_target_permutation_idx (line 1051) | def _get_target_permutation_idx(self, indices):
    method get_loss (line 1057) | def get_loss(self, loss, outputs, targets, indices, num_boxes):
    method forward (line 1068) | def forward(self, outputs, targets):
  class YolosMLPPredictionHead (line 1114) | class YolosMLPPredictionHead(nn.Module):
    method __init__ (line 1123) | def __init__(self, input_dim, hidden_dim, output_dim, num_layers):
    method forward (line 1129) | def forward(self, x):
  class YolosHungarianMatcher (line 1136) | class YolosHungarianMatcher(nn.Module):
    method __init__ (line 1153) | def __init__(self, class_cost: float = 1, bbox_cost: float = 1, giou_c...
    method forward (line 1164) | def forward(self, outputs, targets):
  function _upcast (line 1215) | def _upcast(t: Tensor) -> Tensor:
  function box_area (line 1224) | def box_area(boxes: Tensor) -> Tensor:
  function box_iou (line 1241) | def box_iou(boxes1, boxes2):
  function generalized_box_iou (line 1258) | def generalized_box_iou(boxes1, boxes2):
  function _max_by_axis (line 1283) | def _max_by_axis(the_list):
  class NestedTensor (line 1293) | class NestedTensor(object):
    method __init__ (line 1294) | def __init__(self, tensors, mask: Optional[Tensor]):
    method to (line 1298) | def to(self, device):
    method decompose (line 1307) | def decompose(self):
    method __repr__ (line 1310) | def __repr__(self):
  function nested_tensor_from_tensor_list (line 1315) | def nested_tensor_from_tensor_list(tensor_list: List[Tensor]):

FILE: transformers/models/yoso/configuration_yoso.py
  class YosoConfig (line 29) | class YosoConfig(PretrainedConfig):
    method __init__ (line 99) | def __init__(

FILE: transformers/models/yoso/convert_yoso_pytorch_to_pytorch.py
  function rename_key (line 24) | def rename_key(orig_key):
  function convert_checkpoint_helper (line 62) | def convert_checkpoint_helper(max_position_embeddings, orig_state_dict):
  function convert_yoso_checkpoint (line 77) | def convert_yoso_checkpoint(checkpoint_path, yoso_config_file, pytorch_d...

FILE: transformers/models/yoso/modeling_yoso.py
  function load_cuda_kernels (line 53) | def load_cuda_kernels():
  function to_contiguous (line 76) | def to_contiguous(input_tensors):
  function normalize (line 90) | def normalize(input_tensors):
  function hashing (line 100) | def hashing(query, key, num_hash, hash_len):
  class YosoCumulation (line 119) | class YosoCumulation(torch.autograd.Function):
    method forward (line 121) | def forward(ctx, query_mask, key_mask, query, key, value, config):
    method backward (line 134) | def backward(ctx, grad):
  class YosoLSHCumulation (line 150) | class YosoLSHCumulation(torch.autograd.Function):
    method forward (line 152) | def forward(ctx, query_mask, key_mask, query, key, value, config):
    method backward (line 190) | def backward(ctx, grad):
  class YosoEmbeddings (line 240) | class YosoEmbeddings(nn.Module):
    method __init__ (line 243) | def __init__(self, config):
    method forward (line 263) | def forward(self, input_ids=None, token_type_ids=None, position_ids=No...
  class YosoSelfAttention (line 298) | class YosoSelfAttention(nn.Module):
    method __init__ (line 299) | def __init__(self, config, position_embedding_type=None):
    method transpose_for_scores (line 344) | def transpose_for_scores(self, layer):
    method forward (line 349) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class YosoSelfOutput (line 432) | class YosoSelfOutput(nn.Module):
    method __init__ (line 433) | def __init__(self, config):
    method forward (line 439) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class YosoAttention (line 446) | class YosoAttention(nn.Module):
    method __init__ (line 447) | def __init__(self, config, position_embedding_type=None):
    method prune_heads (line 453) | def prune_heads(self, heads):
    method forward (line 471) | def forward(self, hidden_states, attention_mask=None, output_attention...
  class YosoIntermediate (line 479) | class YosoIntermediate(nn.Module):
    method __init__ (line 480) | def __init__(self, config):
    method forward (line 488) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class YosoOutput (line 495) | class YosoOutput(nn.Module):
    method __init__ (line 496) | def __init__(self, config):
    method forward (line 502) | def forward(self, hidden_states: torch.Tensor, input_tensor: torch.Ten...
  class YosoLayer (line 509) | class YosoLayer(nn.Module):
    method __init__ (line 510) | def __init__(self, config):
    method forward (line 519) | def forward(self, hidden_states, attention_mask=None, output_attention...
    method feed_forward_chunk (line 532) | def feed_forward_chunk(self, attention_output):
  class YosoEncoder (line 538) | class YosoEncoder(nn.Module):
    method __init__ (line 539) | def __init__(self, config):
    method forward (line 545) | def forward(
  class YosoPredictionHeadTransform (line 594) | class YosoPredictionHeadTransform(nn.Module):
    method __init__ (line 595) | def __init__(self, config):
    method forward (line 604) | def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
  class YosoLMPredictionHead (line 612) | class YosoLMPredictionHead(nn.Module):
    method __init__ (line 613) | def __init__(self, config):
    method forward (line 626) | def forward(self, hidden_states):
  class YosoOnlyMLMHead (line 633) | class YosoOnlyMLMHead(nn.Module):
    method __init__ (line 634) | def __init__(self, config):
    method forward (line 638) | def forward(self, sequence_output: torch.Tensor) -> torch.Tensor:
  class YosoPreTrainedModel (line 643) | class YosoPreTrainedModel(PreTrainedModel):
    method _init_weights (line 654) | def _init_weights(self, module):
    method _set_gradient_checkpointing (line 670) | def _set_gradient_checkpointing(self, module, value=False):
  class YosoModel (line 740) | class YosoModel(YosoPreTrainedModel):
    method __init__ (line 741) | def __init__(self, config):
    method get_input_embeddings (line 751) | def get_input_embeddings(self):
    method set_input_embeddings (line 754) | def set_input_embeddings(self, value):
    method _prune_heads (line 757) | def _prune_heads(self, heads_to_prune):
    method forward (line 771) | def forward(
  class YosoForMaskedLM (line 851) | class YosoForMaskedLM(YosoPreTrainedModel):
    method __init__ (line 858) | def __init__(self, config):
    method get_output_embeddings (line 867) | def get_output_embeddings(self):
    method set_output_embeddings (line 870) | def set_output_embeddings(self, new_embeddings):
    method forward (line 879) | def forward(
  class YosoClassificationHead (line 932) | class YosoClassificationHead(nn.Module):
    method __init__ (line 935) | def __init__(self, config):
    method forward (line 943) | def forward(self, features, **kwargs):
  class YosoForSequenceClassification (line 958) | class YosoForSequenceClassification(YosoPreTrainedModel):
    method __init__ (line 959) | def __init__(self, config):
    method forward (line 974) | def forward(
  class YosoForMultipleChoice (line 1049) | class YosoForMultipleChoice(YosoPreTrainedModel):
    method __init__ (line 1050) | def __init__(self, config):
    method forward (line 1066) | def forward(
  class YosoForTokenClassification (line 1140) | class YosoForTokenClassification(YosoPreTrainedModel):
    method __init__ (line 1141) | def __init__(self, config):
    method forward (line 1158) | def forward(
  class YosoForQuestionAnswering (line 1225) | class YosoForQuestionAnswering(YosoPreTrainedModel):
    method __init__ (line 1226) | def __init__(self, config):
    method forward (line 1244) | def forward(

FILE: transformers/onnx/__main__.py
  function export_with_optimum (line 35) | def export_with_optimum(args):
  function export_with_transformers (line 69) | def export_with_transformers(args):
  function main (line 186) | def main():

FILE: transformers/onnx/config.py
  class PatchingSpec (line 48) | class PatchingSpec:
  class OnnxConfig (line 68) | class OnnxConfig(ABC):
    method __init__ (line 111) | def __init__(self, config: "PretrainedConfig", task: str = "default", ...
    method from_model_config (line 128) | def from_model_config(cls, config: "PretrainedConfig", task: str = "de...
    method inputs (line 142) | def inputs(self) -> Mapping[str, Mapping[int, str]]:
    method outputs (line 152) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method values_override (line 163) | def values_override(self) -> Optional[Mapping[str, Any]]:
    method default_batch_size (line 176) | def default_batch_size(self) -> int:
    method default_sequence_length (line 187) | def default_sequence_length(self) -> int:
    method default_num_choices (line 197) | def default_num_choices(self) -> int:
    method default_onnx_opset (line 207) | def default_onnx_opset(self) -> int:
    method atol_for_validation (line 217) | def atol_for_validation(self) -> float:
    method is_torch_support_available (line 227) | def is_torch_support_available(self) -> bool:
    method use_external_data_format (line 242) | def use_external_data_format(num_parameters: int) -> bool:
    method _generate_dummy_images (line 258) | def _generate_dummy_images(
    method _generate_dummy_audio (line 267) | def _generate_dummy_audio(
    method generate_dummy_inputs (line 280) | def generate_dummy_inputs(
    method generate_dummy_inputs_onnxruntime (line 400) | def generate_dummy_inputs_onnxruntime(self, reference_model_inputs: Ma...
    method patch_ops (line 414) | def patch_ops(self):
    method restore_ops (line 419) | def restore_ops(self):
    method flatten_output_collection_property (line 425) | def flatten_output_collection_property(cls, name: str, field: Iterable...
  class OnnxConfigWithPast (line 443) | class OnnxConfigWithPast(OnnxConfig, ABC):
    method __init__ (line 444) | def __init__(
    method with_past (line 455) | def with_past(cls, config: "PretrainedConfig", task: str = "default") ...
    method outputs (line 468) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method values_override (line 476) | def values_override(self) -> Optional[Mapping[str, Any]]:
    method num_layers (line 483) | def num_layers(self) -> int:
    method num_attention_heads (line 496) | def num_attention_heads(self) -> int:
    method generate_dummy_inputs (line 508) | def generate_dummy_inputs(
    method fill_with_past_key_values_ (line 550) | def fill_with_past_key_values_(
    method _flatten_past_key_values_ (line 575) | def _flatten_past_key_values_(self, flattened_output, name, idx, t):
    method flatten_output_collection_property (line 579) | def flatten_output_collection_property(self, name: str, field: Iterabl...
  class OnnxSeq2SeqConfigWithPast (line 590) | class OnnxSeq2SeqConfigWithPast(OnnxConfigWithPast):
    method outputs (line 592) | def outputs(self) -> Mapping[str, Mapping[int, str]]:
    method num_layers (line 609) | def num_layers(self) -> Tuple[int]:
    method num_attention_heads (line 625) | def num_attention_heads(self) -> Tuple[int]:
    method generate_dummy_inputs (line 640) | def generate_dummy_inputs(
    method fill_with_past_key_values_ (line 709) | def fill_with_past_key_values_(self, inputs_or_outputs: Mapping[str, M...
    method _flatten_past_key_values_ (line 737) | def _flatten_past_key_values_(self, flattened_output, name, idx, t):

FILE: transformers/onnx/convert.py
  function check_onnxruntime_requirements (line 54) | def check_onnxruntime_requirements(minimum_version: Version):
  function export_pytorch (line 83) | def export_pytorch(
  function export_tensorflow (line 219) | def export_tensorflow(
  function export (line 285) | def export(
  function validate_model_outputs (line 351) | def validate_model_outputs(
  function ensure_model_and_config_inputs_match (line 474) | def ensure_model_and_config_inputs_match(

FILE: transformers/onnx/features.py
  function supported_features_mapping (line 55) | def supported_features_mapping(
  class FeaturesManager (line 85) | class FeaturesManager:
    method get_supported_features_for_model_type (line 557) | def get_supported_features_for_model_type(
    method feature_to_task (line 583) | def feature_to_task(feature: str) -> str:
    method _validate_framework_choice (line 587) | def _validate_framework_choice(framework: str):
    method get_model_class_for_feature (line 602) | def get_model_class_for_feature(feature: str, framework: str = "pt") -...
    method determine_framework (line 629) | def determine_framework(model: str, framework: str = None) -> str:
    method get_model_from_feature (line 679) | def get_model_from_feature(
    method check_supported_model_or_raise (line 712) | def check_supported_model_or_raise(
    method get_config (line 736) | def get_config(model_type: str, feature: str) -> OnnxConfig:

FILE: transformers/onnx/utils.py
  class ParameterFormat (line 24) | class ParameterFormat(Enum):
    method size (line 28) | def size(self) -> int:
  function compute_effective_axis_dimension (line 38) | def compute_effective_axis_dimension(dimension: int, fixed_dimension: in...
  function compute_serialized_parameters_size (line 57) | def compute_serialized_parameters_size(num_parameters: int, dtype: Param...
  function get_preprocessor (line 71) | def get_preprocessor(model_name: str) -> Optional[Union["AutoTokenizer",...

FILE: transformers/optimization.py
  function _get_constant_lambda (line 35) | def _get_constant_lambda(_=None):
  function get_constant_schedule (line 39) | def get_constant_schedule(optimizer: Optimizer, last_epoch: int = -1):
  function get_reduce_on_plateau_schedule (line 56) | def get_reduce_on_plateau_schedule(optimizer: Optimizer):
  function _get_constant_schedule_with_warmup_lr_lambda (line 71) | def _get_constant_schedule_with_warmup_lr_lambda(current_step: int, *, n...
  function get_constant_schedule_with_warmup (line 77) | def get_constant_schedule_with_warmup(optimizer: Optimizer, num_warmup_s...
  function _get_linear_schedule_with_warmup_lr_lambda (line 98) | def _get_linear_schedule_with_warmup_lr_lambda(current_step: int, *, num...
  function get_linear_schedule_with_warmup (line 104) | def get_linear_schedule_with_warmup(optimizer, num_warmup_steps, num_tra...
  function _get_cosine_schedule_with_warmup_lr_lambda (line 131) | def _get_cosine_schedule_with_warmup_lr_lambda(
  function get_cosine_schedule_with_warmup (line 140) | def get_cosine_schedule_with_warmup(
  function _get_cosine_with_hard_restarts_schedule_with_warmup_lr_lambda (line 174) | def _get_cosine_with_hard_restarts_schedule_with_warmup_lr_lambda(
  function get_cosine_with_hard_restarts_schedule_with_warmup (line 185) | def get_cosine_with_hard_restarts_schedule_with_warmup(
  function _get_polynomial_decay_schedule_with_warmup_lr_lambda (line 218) | def _get_polynomial_decay_schedule_with_warmup_lr_lambda(
  function get_polynomial_decay_schedule_with_warmup (line 239) | def get_polynomial_decay_schedule_with_warmup(
  function _get_inverse_sqrt_schedule_lr_lambda (line 285) | def _get_inverse_sqrt_schedule_lr_lambda(current_step: int, *, num_warmu...
  function get_inverse_sqrt_schedule (line 293) | def get_inverse_sqrt_schedule(
  function get_scheduler (line 335) | def get_scheduler(
  class AdamW (line 378) | class AdamW(Optimizer):
    method __init__ (line 400) | def __init__(
    method step (line 430) | def step(self, closure: Callable = None):
  class Adafactor (line 492) | class Adafactor(Optimizer):
    method __init__ (line 577) | def __init__(
    method _get_lr (line 610) | def _get_lr(param_group, param_state):
    method _get_options (line 621) | def _get_options(param_group, param_shape):
    method _rms (line 627) | def _rms(tensor):
    method _approx_sq_grad (line 631) | def _approx_sq_grad(exp_avg_sq_row, exp_avg_sq_col):
    method step (line 639) | def step(self, closure=None):
  class AdafactorSchedule (line 733) | class AdafactorSchedule(LambdaLR):
    method __init__ (line 741) | def __init__(self, optimizer, initial_lr=0.0):
    method get_lr (line 751) | def get_lr(self):
  function get_adafactor_schedule (line 763) | def get_adafactor_schedule(optimizer, initial_lr=0.0):

FILE: transformers/optimization_tf.py
  class WarmUp (line 30) | class WarmUp(tf.keras.optimizers.schedules.LearningRateSchedule):
    method __init__ (line 48) | def __init__(
    method __call__ (line 63) | def __call__(self, step):
    method get_config (line 78) | def get_config(self):
  function create_optimizer (line 88) | def create_optimizer(
  class AdamWeightDecay (line 172) | class AdamWeightDecay(Adam):
    method __init__ (line 211) | def __init__(
    method from_config (line 230) | def from_config(cls, config):
    method _prepare_local (line 235) | def _prepare_local(self, var_device, var_dtype, apply_state):
    method _decay_weights_op (line 241) | def _decay_weights_op(self, var, learning_rate, apply_state):
    method apply_gradients (line 250) | def apply_gradients(self, grads_and_vars, name=None, **kwargs):
    method _get_lr (line 254) | def _get_lr(self, var_device, var_dtype, apply_state):
    method _resource_apply_dense (line 267) | def _resource_apply_dense(self, grad, var, apply_state=None):
    method _resource_apply_sparse (line 273) | def _resource_apply_sparse(self, grad, var, indices, apply_state=None):
    method get_config (line 279) | def get_config(self):
    method _do_use_weight_decay (line 284) | def _do_use_weight_decay(self, param_name):
  class GradientAccumulator (line 302) | class GradientAccumulator(object):
    method __init__ (line 313) | def __init__(self):
    method step (line 319) | def step(self):
    method gradients (line 332) | def gradients(self):
    method __call__ (line 338) | def __call__(self, gradients):
    method reset (line 364) | def reset(self):

FILE: transformers/pipelines/__init__.py
  function get_supported_tasks (line 421) | def get_supported_tasks() -> List[str]:
  function get_task (line 428) | def get_task(model: str, use_auth_token: Optional[str] = None) -> str:
  function check_task (line 445) | def check_task(task: str) -> Tuple[str, Dict, Any]:
  function clean_custom_task (line 490) | def clean_custom_task(task_info):
  function pipeline (line 506) | def pipeline(

FILE: transformers/pipelines/audio_classification.py
  function ffmpeg_read (line 30) | def ffmpeg_read(bpayload: bytes, sampling_rate: int) -> np.array:
  class AudioClassificationPipeline (line 67) | class AudioClassificationPipeline(Pipeline):
    method __init__ (line 93) | def __init__(self, *args, **kwargs):
    method __call__ (line 103) | def __call__(
    method _sanitize_parameters (line 132) | def _sanitize_parameters(self, top_k=None, **kwargs):
    method preprocess (line 141) | def preprocess(self, inputs):
    method _forward (line 164) | def _forward(self, model_inputs):
    method postprocess (line 168) | def postprocess(self, model_outputs, top_k=5):

FILE: transformers/pipelines/audio_utils.py
  function ffmpeg_read (line 10) | def ffmpeg_read(bpayload: bytes, sampling_rate: int) -> np.array:
  function ffmpeg_microphone (line 45) | def ffmpeg_microphone(
  function ffmpeg_microphone_live (line 98) | def ffmpeg_microphone_live(
  function chunk_bytes_iter (line 176) | def chunk_bytes_iter(iterator, chunk_len: int, stride: Tuple[int, int], ...
  function _ffmpeg_stream (line 211) | def _ffmpeg_stream(ffmpeg_command, buflen: int):

FILE: transformers/pipelines/automatic_speech_recognition.py
  function rescale_stride (line 36) | def rescale_stride(stride, ratio):
  function chunk_iter (line 56) | def chunk_iter(inputs, feature_extractor, chunk_len, stride_left, stride...
  function _fast_find_longest_common_sequence (line 85) | def _fast_find_longest_common_sequence(sequence_left, sequence_right):
  function _find_longest_common_sequence (line 105) | def _find_longest_common_sequence(sequences, tokenizer):
  class AutomaticSpeechRecognitionPipeline (line 130) | class AutomaticSpeechRecognitionPipeline(ChunkPipeline):
    method __init__ (line 196) | def __init__(
    method __call__ (line 225) | def __call__(
    method _sanitize_parameters (line 274) | def _sanitize_parameters(
    method preprocess (line 316) | def preprocess(self, inputs, chunk_length_s=0, stride_length_s=None, i...
    method _forward (line 419) | def _forward(self, model_inputs, return_timestamps=False, generate_kwa...
    method postprocess (line 480) | def postprocess(
  function _find_timestamp_sequence (line 582) | def _find_timestamp_sequence(sequences, tokenizer, feature_extractor, ma...

FILE: transformers/pipelines/base.py
  function no_collate_fn (line 68) | def no_collate_fn(items):
  function _pad (line 74) | def _pad(items, key, padding_value, padding_side):
  function pad_collate_fn (line 124) | def pad_collate_fn(tokenizer, feature_extractor):
  function infer_framework_load_model (line 187) | def infer_framework_load_model(
  function infer_framework_from_model (line 284) | def infer_framework_from_model(
  function get_framework (line 323) | def get_framework(model, revision: Optional[str] = None):
  function get_default_model_and_revision (line 357) | def get_default_model_and_revision(
  class PipelineException (line 401) | class PipelineException(Exception):
    method __init__ (line 411) | def __init__(self, task: str, model: str, reason: str):
  class ArgumentHandler (line 418) | class ArgumentHandler(ABC):
    method __call__ (line 424) | def __call__(self, *args, **kwargs):
  class PipelineDataFormat (line 428) | class PipelineDataFormat:
    method __init__ (line 450) | def __init__(
    method __iter__ (line 474) | def __iter__(self):
    method save (line 478) | def save(self, data: Union[dict, List[dict]]):
    method save_binary (line 487) | def save_binary(self, data: Union[dict, List[dict]]) -> str:
    method from_str (line 506) | def from_str(
  class CsvPipelineDataFormat (line 541) | class CsvPipelineDataFormat(PipelineDataFormat):
    method __init__ (line 553) | def __init__(
    method __iter__ (line 562) | def __iter__(self):
    method save (line 571) | def save(self, data: List[dict]):
  class JsonPipelineDataFormat (line 585) | class JsonPipelineDataFormat(PipelineDataFormat):
    method __init__ (line 597) | def __init__(
    method __iter__ (line 609) | def __iter__(self):
    method save (line 616) | def save(self, data: dict):
  class PipedPipelineDataFormat (line 627) | class PipedPipelineDataFormat(PipelineDataFormat):
    method __iter__ (line 641) | def __iter__(self):
    method save (line 656) | def save(self, data: dict):
    method save_binary (line 665) | def save_binary(self, data: Union[dict, List[dict]]) -> str:
  class _ScikitCompat (line 675) | class _ScikitCompat(ABC):
    method transform (line 681) | def transform(self, X):
    method predict (line 685) | def predict(self, X):
  class Pipeline (line 734) | class Pipeline(_ScikitCompat):
    method __init__ (line 753) | def __init__(
    method save_pretrained (line 824) | def save_pretrained(self, save_directory: str, safe_serialization: boo...
    method transform (line 871) | def transform(self, X):
    method predict (line 877) | def predict(self, X):
    method device_placement (line 884) | def device_placement(self):
    method ensure_tensor_on_device (line 909) | def ensure_tensor_on_device(self, **inputs):
    method _ensure_tensor_on_device (line 923) | def _ensure_tensor_on_device(self, inputs, device):
    method check_model_type (line 943) | def check_model_type(self, supported_models: Union[List[str], dict]):
    method _sanitize_parameters (line 967) | def _sanitize_parameters(self, **pipeline_parameters):
    method preprocess (line 980) | def preprocess(self, input_: Any, **preprocess_parameters: Dict) -> Di...
    method _forward (line 988) | def _forward(self, input_tensors: Dict[str, GenericTensor], **forward_...
    method postprocess (line 1001) | def postprocess(self, model_outputs: ModelOutput, **postprocess_parame...
    method get_inference_context (line 1009) | def get_inference_context(self):
    method forward (line 1017) | def forward(self, model_inputs, **forward_params):
    method get_iterator (line 1032) | def get_iterator(
    method __call__ (line 1057) | def __call__(self, inputs, *args, num_workers=None, batch_size=None, *...
    method run_multi (line 1122) | def run_multi(self, inputs, preprocess_params, forward_params, postpro...
    method run_single (line 1125) | def run_single(self, inputs, preprocess_params, forward_params, postpr...
    method iterate (line 1131) | def iterate(self, inputs, preprocess_params, forward_params, postproce...
  class ChunkPipeline (line 1138) | class ChunkPipeline(Pipeline):
    method run_single (line 1139) | def run_single(self, inputs, preprocess_params, forward_params, postpr...
    method get_iterator (line 1147) | def get_iterator(
  class PipelineRegistry (line 1170) | class PipelineRegistry:
    method __init__ (line 1171) | def __init__(self, supported_tasks: Dict[str, Any], task_aliases: Dict...
    method get_supported_tasks (line 1175) | def get_supported_tasks(self) -> List[str]:
    method check_task (line 1180) | def check_task(self, task: str) -> Tuple[str, Dict, Any]:
    method register_pipeline (line 1199) | def register_pipeline(
    method to_dict (line 1234) | def to_dict(self):

FILE: transformers/pipelines/conversational.py
  class Conversation (line 18) | class Conversation:
    method __init__ (line 56) | def __init__(
    method __eq__ (line 71) | def __eq__(self, other):
    method add_user_input (line 82) | def add_user_input(self, text: str, overwrite: bool = False):
    method mark_processed (line 106) | def mark_processed(self):
    method append_response (line 115) | def append_response(self, response: str):
    method iter_texts (line 124) | def iter_texts(self):
    method __repr__ (line 137) | def __repr__(self):
  class ConversationalPipeline (line 163) | class ConversationalPipeline(Pipeline):
    method __init__ (line 195) | def __init__(self, *args, **kwargs):
    method _sanitize_parameters (line 200) | def _sanitize_parameters(
    method __call__ (line 222) | def __call__(self, conversations: Union[Conversation, List[Conversatio...
    method preprocess (line 248) | def preprocess(self, conversation: Conversation, min_length_for_respon...
    method _forward (line 268) | def _forward(self, model_inputs, minimum_tokens=10, **generate_kwargs):
    method postprocess (line 287) | def postprocess(self, model_outputs, clean_up_tokenization_spaces=True):
    method _legacy_parse_and_tokenize (line 299) | def _legacy_parse_and_tokenize(self, conversation: Conversation) -> Dict:

FILE: transformers/pipelines/depth_estimation.py
  class DepthEstimationPipeline (line 23) | class DepthEstimationPipeline(Pipeline):
    method __init__ (line 48) | def __init__(self, *args, **kwargs):
    method __call__ (line 53) | def __call__(self, images: Union[str, List[str], "Image.Image", List["...
    method _sanitize_parameters (line 84) | def _sanitize_parameters(self, **kwargs):
    method preprocess (line 87) | def preprocess(self, image):
    method _forward (line 93) | def _forward(self, model_inputs):
    method postprocess (line 97) | def postprocess(self, model_outputs):

FILE: transformers/pipelines/document_question_answering.py
  function normalize_box (line 53) | def normalize_box(box, width, height):
  function apply_tesseract (line 62) | def apply_tesseract(image: "Image.Image", lang: Optional[str], tesseract...
  class ModelType (line 95) | class ModelType(ExplicitEnum):
  class DocumentQuestionAnsweringPipeline (line 102) | class DocumentQuestionAnsweringPipeline(ChunkPipeline):
    method __init__ (line 132) | def __init__(self, *args, **kwargs):
    method _sanitize_parameters (line 151) | def _sanitize_parameters(
    method __call__ (line 191) | def __call__(
    method preprocess (line 267) | def preprocess(
    method _forward (line 415) | def _forward(self, model_inputs):
    method postprocess (line 434) | def postprocess(self, model_outputs, top_k=1, **kwargs):
    method postprocess_encoder_decoder_single (line 443) | def postprocess_encoder_decoder_single(self, model_outputs, **kwargs):
    method postprocess_extractive_qa (line 459) | def postprocess_extractive_qa(

FILE: transformers/pipelines/feature_extraction.py
  class FeatureExtractionPipeline (line 7) | class FeatureExtractionPipeline(Pipeline):
    method _sanitize_parameters (line 60) | def _sanitize_parameters(self, truncation=None, tokenize_kwargs=None, ...
    method preprocess (line 79) | def preprocess(self, inputs, **tokenize_kwargs) -> Dict[str, GenericTe...
    method _forward (line 84) | def _forward(self, model_inputs):
    method postprocess (line 88) | def postprocess(self, model_outputs, return_tensors=False):
    method __call__ (line 97) | def __call__(self, *args, **kwargs):

FILE: transformers/pipelines/fill_mask.py
  class FillMaskPipeline (line 34) | class FillMaskPipeline(Pipeline):
    method get_masked_index (line 66) | def get_masked_index(self, input_ids: GenericTensor) -> np.ndarray:
    method _ensure_exactly_one_mask_token (line 75) | def _ensure_exactly_one_mask_token(self, input_ids: GenericTensor) -> ...
    method ensure_exactly_one_mask_token (line 85) | def ensure_exactly_one_mask_token(self, model_inputs: GenericTensor):
    method preprocess (line 93) | def preprocess(self, inputs, return_tensors=None, **preprocess_paramet...
    method _forward (line 100) | def _forward(self, model_inputs):
    method postprocess (line 105) | def postprocess(self, model_outputs, top_k=5, target_ids=None):
    method get_target_ids (line 160) | def get_target_ids(self, targets, top_k=None):
    method _sanitize_parameters (line 201) | def _sanitize_parameters(self, top_k=None, targets=None):
    method __call__ (line 217) | def __call__(self, inputs, *args, **kwargs):

FILE: transformers/pipelines/image_classification.py
  class ImageClassificationPipeline (line 32) | class ImageClassificationPipeline(Pipeline):
    method __init__ (line 56) | def __init__(self, *args, **kwargs):
    method _sanitize_parameters (line 65) | def _sanitize_parameters(self, top_k=None):
    method __call__ (line 71) | def __call__(self, images: Union[str, List[str], "Image.Image", List["...
    method preprocess (line 102) | def preprocess(self, image):
    method _forward (line 107) | def _forward(self, model_inputs):
    method postprocess (line 111) | def postprocess(self, model_outputs, top_k=5):

FILE: transformers/pipelines/image_segmentation.py
  class ImageSegmentationPipeline (line 31) | class ImageSegmentationPipeline(Pipeline):
    method __init__ (line 67) | def __init__(self, *args, **kwargs):
    method _sanitize_parameters (line 83) | def _sanitize_parameters(self, **kwargs):
    method __call__ (line 98) | def __call__(self, images, **kwargs) -> Union[Predictions, List[Predic...
    method preprocess (line 139) | def preprocess(self, image, subtask=None):
    method _forward (line 159) | def _forward(self, model_inputs):
    method postprocess (line 165) | def postprocess(

FILE: transformers/pipelines/image_to_text.py
  class ImageToTextPipeline (line 31) | class ImageToTextPipeline(Pipeline):
    method __init__ (line 54) | def __init__(self, *args, **kwargs):
    method _sanitize_parameters (line 61) | def _sanitize_parameters(self, max_new_tokens=None, generate_kwargs=No...
    method __call__ (line 81) | def __call__(self, images: Union[str, List[str], "Image.Image", List["...
    method preprocess (line 108) | def preprocess(self, image, prompt=None):
    method _forward (line 147) | def _forward(self, model_inputs, generate_kwargs=None):
    method postprocess (line 158) | def postprocess(self, model_outputs):

FILE: transformers/pipelines/mask_generation.py
  class MaskGenerationPipeline (line 23) | class MaskGenerationPipeline(ChunkPipeline):
    method __init__ (line 91) | def __init__(self, **kwargs):
    method _sanitize_parameters (line 101) | def _sanitize_parameters(self, **kwargs):
    method __call__ (line 133) | def __call__(self, image, *args, num_workers=None, batch_size=None, **...
    method preprocess (line 170) | def preprocess(
    method _forward (line 215) | def _forward(
    method postprocess (line 253) | def postprocess(

FILE: transformers/pipelines/object_detection.py
  class ObjectDetectionPipeline (line 24) | class ObjectDetectionPipeline(Pipeline):
    method __init__ (line 49) | def __init__(self, *args, **kwargs):
    method _sanitize_parameters (line 60) | def _sanitize_parameters(self, **kwargs):
    method __call__ (line 66) | def __call__(self, *args, **kwargs) -> Union[Predictions, List[Predict...
    method preprocess (line 97) | def preprocess(self, image):
    method _forward (line 106) | def _forward(self, model_inputs):
    method postprocess (line 114) | def postprocess(self, model_outputs, threshold=0.9):
    method _get_bounding_box (line 159) | def _get_bounding_box(self, box: "torch.Tensor") -> Dict[str, int]:

FILE: transformers/pipelines/pt_utils.py
  class PipelineDataset (line 8) | class PipelineDataset(Dataset):
    method __init__ (line 9) | def __init__(self, dataset, process, params):
    method __len__ (line 14) | def __len__(self):
    method __getitem__ (line 17) | def __getitem__(self, i):
  class PipelineIterator (line 23) | class PipelineIterator(IterableDataset):
    method __init__ (line 24) | def __init__(self, loader, infer, params, loader_batch_size=None):
    method __len__ (line 63) | def __len__(self):
    method __iter__ (line 66) | def __iter__(self):
    method loader_batch_item (line 70) | def loader_batch_item(self):
    method __next__ (line 117) | def __next__(self):
  class PipelineChunkIterator (line 151) | class PipelineChunkIterator(PipelineIterator):
    method __init__ (line 152) | def __init__(self, loader, infer, params, loader_batch_size=None):
    method __iter__ (line 172) | def __iter__(self):
    method __next__ (line 177) | def __next__(self):
  class PipelinePackIterator (line 196) | class PipelinePackIterator(PipelineIterator):
    method __iter__ (line 242) | def __iter__(self):
    method __next__ (line 246) | def __next__(self):
  class KeyDataset (line 296) | class KeyDataset(Dataset):
    method __init__ (line 297) | def __init__(self, dataset: Dataset, key: str):
    method __len__ (line 301) | def __len__(self):
    method __getitem__ (line 304) | def __getitem__(self, i):
  class KeyPairDataset (line 308) | class KeyPairDataset(Dataset):
    method __init__ (line 309) | def __init__(self, dataset: Dataset, key1: str, key2: str):
    method __len__ (line 314) | def __len__(self):
    method __getitem__ (line 317) | def __getitem__(self, i):

FILE: transformers/pipelines/question_answering.py
  function decode_spans (line 45) | def decode_spans(
  function select_starts_ends (line 95) | def select_starts_ends(
  class QuestionAnsweringArgumentHandler (line 149) | class QuestionAnsweringArgumentHandler(ArgumentHandler):
    method normalize (line 158) | def normalize(self, item):
    method __call__ (line 173) | def __call__(self, *args, **kwargs):
  class QuestionAnsweringPipeline (line 224) | class QuestionAnsweringPipeline(ChunkPipeline):
    method __init__ (line 252) | def __init__(
    method create_sample (line 276) | def create_sample(
    method _sanitize_parameters (line 297) | def _sanitize_parameters(
    method __call__ (line 340) | def __call__(self, *args, **kwargs):
    method preprocess (line 393) | def preprocess(self, example, padding="do_not_pad", doc_stride=None, m...
    method _forward (line 510) | def _forward(self, inputs):
    method postprocess (line 520) | def postprocess(
    method get_indices (line 605) | def get_indices(
    method span_to_answer (line 623) | def span_to_answer(self, text: str, start: int, end: int) -> Dict[str,...

FILE: transformers/pipelines/table_question_answering.py
  class TableQuestionAnsweringArgumentHandler (line 34) | class TableQuestionAnsweringArgumentHandler(ArgumentHandler):
    method __call__ (line 39) | def __call__(self, table=None, query=None, **kwargs):
  class TableQuestionAnsweringPipeline (line 88) | class TableQuestionAnsweringPipeline(Pipeline):
    method __init__ (line 121) | def __init__(self, args_parser=TableQuestionAnsweringArgumentHandler()...
    method batch_inference (line 141) | def batch_inference(self, **inputs):
    method sequential_inference (line 144) | def sequential_inference(self, **inputs):
    method __call__ (line 273) | def __call__(self, *args, **kwargs):
    method _sanitize_parameters (line 355) | def _sanitize_parameters(self, sequential=None, padding=None, truncati...
    method preprocess (line 367) | def preprocess(self, pipeline_input, sequential=None, padding=True, tr...
    method _forward (line 383) | def _forward(self, model_inputs, sequential=False):
    method postprocess (line 396) | def postprocess(self, model_outputs):

FILE: transformers/pipelines/text2text_generation.py
  class ReturnType (line 20) | class ReturnType(enum.Enum):
  class Text2TextGenerationPipeline (line 26) | class Text2TextGenerationPipeline(Pipeline):
    method __init__ (line 64) | def __init__(self, *args, **kwargs):
    method _sanitize_parameters (line 73) | def _sanitize_parameters(
    method check_inputs (line 109) | def check_inputs(self, input_length: int, min_length: int, max_length:...
    method _parse_and_tokenize (line 115) | def _parse_and_tokenize(self, *args, truncation):
    method __call__ (line 136) | def __call__(self, *args, **kwargs):
    method preprocess (line 174) | def preprocess(self, inputs, truncation=TruncationStrategy.DO_NOT_TRUN...
    method _forward (line 178) | def _forward(self, model_inputs, **generate_kwargs):
    method postprocess (line 195) | def postprocess(self, model_outputs, return_type=ReturnType.TEXT, clea...
  class SummarizationPipeline (line 213) | class SummarizationPipeline(Text2TextGenerationPipeline):
    method __call__ (line 241) | def __call__(self, *args, **kwargs):
    method check_inputs (line 267) | def check_inputs(self, input_length: int, min_length: int, max_length:...
  class TranslationPipeline (line 283) | class TranslationPipeline(Text2TextGenerationPipeline):
    method check_inputs (line 305) | def check_inputs(self, input_length: int, min_length: int, max_length:...
    method preprocess (line 313) | def preprocess(self, *args, truncation=TruncationStrategy.DO_NOT_TRUNC...
    method _sanitize_parameters (line 321) | def _sanitize_parameters(self, src_lang=None, tgt_lang=None, **kwargs):
    method __call__ (line 337) | def __call__(self, *args, **kwargs):

FILE: transformers/pipelines/text_classification.py
  function sigmoid (line 17) | def sigmoid(_outputs):
  function softmax (line 21) | def softmax(_outputs):
  class ClassificationFunction (line 27) | class ClassificationFunction(ExplicitEnum):
  class TextClassificationPipeline (line 48) | class TextClassificationPipeline(Pipeline):
    method __init__ (line 82) | def __init__(self, **kwargs):
    method _sanitize_parameters (line 91) | def _sanitize_parameters(self, return_all_scores=None, function_to_app...
    method __call__ (line 121) | def __call__(self, *args, **kwargs):
    method preprocess (line 164) | def preprocess(self, inputs, **tokenizer_kwargs) -> Dict[str, GenericT...
    method _forward (line 181) | def _forward(self, model_inputs):
    method postprocess (line 184) | def postprocess(self, model_outputs, function_to_apply=None, top_k=1, ...

FILE: transformers/pipelines/text_generation.py
  class ReturnType (line 14) | class ReturnType(enum.Enum):
  class TextGenerationPipeline (line 21) | class TextGenerationPipeline(Pipeline):
    method __init__ (line 63) | def __init__(self, *args, **kwargs):
    method _sanitize_parameters (line 90) | def _sanitize_parameters(
    method _parse_and_tokenize (line 150) | def _parse_and_tokenize(self, *args, **kwargs):
    method __call__ (line 160) | def __call__(self, text_inputs, **kwargs):
    method preprocess (line 203) | def preprocess(self, prompt_text, prefix="", handle_long_generation=No...
    method _forward (line 231) | def _forward(self, model_inputs, **generate_kwargs):
    method postprocess (line 271) | def postprocess(self, model_outputs, return_type=ReturnType.FULL_TEXT,...

FILE: transformers/pipelines/token_classification.py
  class TokenClassificationArgumentHandler (line 25) | class TokenClassificationArgumentHandler(ArgumentHandler):
    method __call__ (line 30) | def __call__(self, inputs: Union[str, List[str]], **kwargs):
  class AggregationStrategy (line 51) | class AggregationStrategy(ExplicitEnum):
  class TokenClassificationPipeline (line 96) | class TokenClassificationPipeline(ChunkPipeline):
    method __init__ (line 135) | def __init__(self, args_parser=TokenClassificationArgumentHandler(), *...
    method _sanitize_parameters (line 146) | def _sanitize_parameters(
    method __call__ (line 219) | def __call__(self, inputs: Union[str, List[str]], **kwargs):
    method preprocess (line 251) | def preprocess(self, sentence, offset_mapping=None, **preprocess_params):
    method _forward (line 277) | def _forward(self, model_inputs):
    method postprocess (line 298) | def postprocess(self, all_outputs, aggregation_strategy=AggregationStr...
    method aggregate_overlapping_entities (line 336) | def aggregate_overlapping_entities(self, entities):
    method gather_pre_entities (line 356) | def gather_pre_entities(
    method aggregate (line 418) | def aggregate(self, pre_entities: List[dict], aggregation_strategy: Ag...
    method aggregate_word (line 440) | def aggregate_word(self, entities: List[dict], aggregation_strategy: A...
    method aggregate_words (line 470) | def aggregate_words(self, entities: List[dict], aggregation_strategy: ...
    method group_sub_entities (line 497) | def group_sub_entities(self, entities: List[dict]) -> dict:
    method get_tag (line 518) | def get_tag(self, entity_name: str) -> Tuple[str, str]:
    method group_entities (line 532) | def group_entities(self, entities: List[dict]) -> List[dict]:

FILE: transformers/pipelines/video_classification.py
  class VideoClassificationPipeline (line 22) | class VideoClassificationPipeline(Pipeline):
    method __init__ (line 34) | def __init__(self, *args, **kwargs):
    method _sanitize_parameters (line 39) | def _sanitize_parameters(self, top_k=None, num_frames=None, frame_samp...
    method __call__ (line 51) | def __call__(self, videos: Union[str, List[str]], **kwargs):
    method preprocess (line 86) | def preprocess(self, video, num_frames=None, frame_sampling_rate=1):
    method _forward (line 106) | def _forward(self, model_inputs):
    method postprocess (line 110) | def postprocess(self, model_outputs, top_k=5):

FILE: transformers/pipelines/visual_question_answering.py
  class VisualQuestionAnsweringPipeline (line 19) | class VisualQuestionAnsweringPipeline(Pipeline):
    method __init__ (line 54) | def __init__(self, *args, **kwargs):
    method _sanitize_parameters (line 58) | def _sanitize_parameters(self, top_k=None, padding=None, truncation=No...
    method __call__ (line 68) | def __call__(self, image: Union["Image.Image", str], question: str = N...
    method preprocess (line 112) | def preprocess(self, inputs, padding=False, truncation=False):
    method _forward (line 121) | def _forward(self, model_inputs):
    method postprocess (line 125) | def postprocess(self, model_outputs, top_k=5):

FILE: transformers/pipelines/zero_shot_audio_classification.py
  class ZeroShotAudioClassificationPipeline (line 33) | class ZeroShotAudioClassificationPipeline(Pipeline):
    method __init__ (line 57) | def __init__(self, **kwargs):
    method __call__ (line 64) | def __call__(self, audios: Union[np.ndarray, bytes, str], **kwargs):
    method _sanitize_parameters (line 88) | def _sanitize_parameters(self, **kwargs):
    method preprocess (line 97) | def preprocess(self, audio, candidate_labels=None, hypothesis_template...
    method _forward (line 124) | def _forward(self, model_inputs):
    method postprocess (line 141) | def postprocess(self, model_outputs):

FILE: transformers/pipelines/zero_shot_classification.py
  class ZeroShotClassificationArgumentHandler (line 13) | class ZeroShotClassificationArgumentHandler(ArgumentHandler):
    method _parse_labels (line 19) | def _parse_labels(self, labels):
    method __call__ (line 24) | def __call__(self, sequences, labels, hypothesis_template):
  class ZeroShotClassificationPipeline (line 46) | class ZeroShotClassificationPipeline(ChunkPipeline):
    method __init__ (line 86) | def __init__(self, args_parser=ZeroShotClassificationArgumentHandler()...
    method entailment_id (line 96) | def entailment_id(self):
    method _parse_and_tokenize (line 102) | def _parse_and_tokenize(
    method _sanitize_parameters (line 144) | def _sanitize_parameters(self, **kwargs):
    method __call__ (line 162) | def __call__(
    method preprocess (line 207) | def preprocess(self, inputs, candidate_labels=None, hypothesis_templat...
    method _forward (line 220) | def _forward(self, inputs):
    method postprocess (line 234) | def postprocess(self, model_outputs, multi_label=False):

FILE: transformers/pipelines/zero_shot_image_classification.py
  class ZeroShotImageClassificationPipeline (line 31) | class ZeroShotImageClassificationPipeline(Pipeline):
    method __init__ (line 64) | def __init__(self, **kwargs):
    method __call__ (line 74) | def __call__(self, images: Union[str, List[str], "Image", List["Image"...
    method _sanitize_parameters (line 103) | def _sanitize_parameters(self, **kwargs):
    method preprocess (line 112) | def preprocess(self, image, candidate_labels=None, hypothesis_template...
    method _forward (line 121) | def _forward(self, model_inputs):
    method postprocess (line 138) | def postprocess(self, model_outputs):

FILE: transformers/pipelines/zero_shot_object_detection.py
  class ZeroShotObjectDetectionPipeline (line 23) | class ZeroShotObjectDetectionPipeline(ChunkPipeline):
    method __init__ (line 56) | def __init__(self, **kwargs):
    method __call__ (line 65) | def __call__(
    method _sanitize_parameters (line 134) | def _sanitize_parameters(self, **kwargs):
    method preprocess (line 142) | def preprocess(self, inputs):
    method _forward (line 160) | def _forward(self, model_inputs):
    method postprocess (line 170) | def postprocess(self, model_outputs, threshold=0.1, top_k=None):
    method _get_bounding_box (line 192) | def _get_bounding_box(self, box: "torch.Tensor") -> Dict[str, int]:

FILE: transformers/processing_utils.py
  class ProcessorMixin (line 40) | class ProcessorMixin(PushToHubMixin):
    method __init__ (line 52) | def __init__(self, *args, **kwargs):
    method __repr__ (line 86) | def __repr__(self):
    method save_pretrained (line 91) | def save_pretrained(self, save_directory, push_to_hub: bool = False, *...
    method from_pretrained (line 154) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method register_for_auto_class (line 188) | def register_for_auto_class(cls, auto_class="AutoProcessor"):
    method _get_arguments_from_pretrained (line 214) | def _get_arguments_from_pretrained(cls, pretrained_model_name_or_path,...
    method model_input_names (line 232) | def model_input_names(self):

FILE: transformers/pytorch_utils.py
  function softmax_backward_data (line 35) | def softmax_backward_data(parent, grad_output, output, dim, self):
  function prune_linear_layer (line 49) | def prune_linear_layer(layer: nn.Linear, index: torch.LongTensor, dim: i...
  class Conv1D (line 83) | class Conv1D(nn.Module):
    method __init__ (line 94) | def __init__(self, nf, nx):
    method forward (line 101) | def forward(self, x):
  function prune_conv1d_layer (line 108) | def prune_conv1d_layer(layer: Conv1D, index: torch.LongTensor, dim: int ...
  function prune_layer (line 141) | def prune_layer(
  function apply_chunking_to_forward (line 165) | def apply_chunking_to_forward(
  function find_pruneable_heads_and_indices (line 240) | def find_pruneable_heads_and_indices(
  function meshgrid (line 267) | def meshgrid(
  function id_tensor_storage (line 283) | def id_tensor_storage(tensor: torch.Tensor) -> Tuple[torch.device, int, ...

FILE: transformers/sagemaker/trainer_sm.py
  class SageMakerTrainer (line 23) | class SageMakerTrainer(Trainer):
    method __init__ (line 24) | def __init__(self, args=None, **kwargs):

FILE: transformers/sagemaker/training_args_sm.py
  function is_sagemaker_model_parallel_available (line 32) | def is_sagemaker_model_parallel_available():
  class SageMakerTrainingArguments (line 63) | class SageMakerTrainingArguments(TrainingArguments):
    method __post_init__ (line 69) | def __post_init__(self):
    method _setup_devices (line 78) | def _setup_devices(self) -> "torch.device":
    method world_size (line 124) | def world_size(self):
    method place_model_on_device (line 131) | def place_model_on_device(self):
    method _no_sync_in_gradient_accumulation (line 135) | def _no_sync_in_gradient_accumulation(self):

FILE: transformers/testing_utils.py
  function parse_flag_from_env (line 132) | def parse_flag_from_env(key, default=False):
  function parse_int_from_env (line 148) | def parse_int_from_env(key, default=None):
  function is_pt_tf_cross_test (line 171) | def is_pt_tf_cross_test(test_case):
  function is_pt_flax_cross_test (line 190) | def is_pt_flax_cross_test(test_case):
  function is_staging_test (line 209) | def is_staging_test(test_case):
  function is_pipeline_test (line 226) | def is_pipeline_test(test_case):
  function is_tool_test (line 242) | def is_tool_test(test_case):
  function slow (line 257) | def slow(test_case):
  function tooslow (line 267) | def tooslow(test_case):
  function custom_tokenizers (line 278) | def custom_tokenizers(test_case):
  function require_bs4 (line 288) | def require_bs4(test_case):
  function require_accelerate (line 295) | def require_accelerate(test_case):
  function require_safetensors (line 302) | def require_safetensors(test_case):
  function require_rjieba (line 309) | def require_rjieba(test_case):
  function require_jieba (line 316) | def require_jieba(test_case):
  function require_tf2onnx (line 323) | def require_tf2onnx(test_case):
  function require_onnx (line 327) | def require_onnx(test_case):
  function require_timm (line 331) | def require_timm(test_case):
  function require_natten (line 341) | def require_natten(test_case):
  function require_torch (line 351) | def require_torch(test_case):
  function require_torchvision (line 361) | def require_torchvision(test_case):
  function require_torch_or_tf (line 371) | def require_torch_or_tf(test_case):
  function require_intel_extension_for_pytorch (line 383) | def require_intel_extension_for_pytorch(test_case):
  function require_tensorflow_probability (line 398) | def require_tensorflow_probability(test_case):
  function require_torchaudio (line 410) | def require_torchaudio(test_case):
  function require_tf (line 417) | def require_tf(test_case):
  function require_flax (line 424) | def require_flax(test_case):
  function require_sentencepiece (line 431) | def require_sentencepiece(test_case):
  function require_scipy (line 438) | def require_scipy(test_case):
  function require_tokenizers (line 445) | def require_tokenizers(test_case):
  function require_tensorflow_text (line 452) | def require_tensorflow_text(test_case):
  function require_keras_nlp (line 460) | def require_keras_nlp(test_case):
  function require_pandas (line 467) | def require_pandas(test_case):
  function require_pytesseract (line 474) | def require_pytesseract(test_case):
  function require_pytorch_quantization (line 481) | def require_pytorch_quantization(test_case):
  function require_vision (line 491) | def require_vision(test_case):
  function require_ftfy (line 499) | def require_ftfy(test_case):
  function require_spacy (line 506) | def require_spacy(test_case):
  function require_decord (line 513) | def require_decord(test_case):
  function require_torch_multi_gpu (line 520) | def require_torch_multi_gpu(test_case):
  function require_torch_non_multi_gpu (line 535) | def require_torch_non_multi_gpu(test_case):
  function require_torch_up_to_2_gpus (line 547) | def require_torch_up_to_2_gpus(test_case):
  function require_torch_tpu (line 559) | def require_torch_tpu(test_case):
  function require_torch_neuroncore (line 566) | def require_torch_neuroncore(test_case):
  function require_torchdynamo (line 594) | def require_torchdynamo(test_case):
  function require_torch_tensorrt_fx (line 599) | def require_torch_tensorrt_fx(test_case):
  function require_torch_gpu (line 604) | def require_torch_gpu(test_case):
  function require_torch_bf16_gpu (line 609) | def require_torch_bf16_gpu(test_case):
  function require_torch_bf16_cpu (line 617) | def require_torch_bf16_cpu(test_case):
  function require_torch_tf32 (line 625) | def require_torch_tf32(test_case):
  function require_detectron2 (line 632) | def require_detectron2(test_case):
  function require_faiss (line 637) | def require_faiss(test_case):
  function require_optuna (line 642) | def require_optuna(test_case):
  function require_ray (line 652) | def require_ray(test_case):
  function require_sigopt (line 662) | def require_sigopt(test_case):
  function require_wandb (line 672) | def require_wandb(test_case):
  function require_clearml (line 682) | def require_clearml(test_case):
  function require_soundfile (line 692) | def require_soundfile(test_case):
  function require_deepspeed (line 702) | def require_deepspeed(test_case):
  function require_fairscale (line 709) | def require_fairscale(test_case):
  function require_apex (line 716) | def require_apex(test_case):
  function require_bitsandbytes (line 723) | def require_bitsandbytes(test_case):
  function require_optimum (line 730) | def require_optimum(test_case):
  function require_phonemizer (line 737) | def require_phonemizer(test_case):
  function require_pyctcdecode (line 744) | def require_pyctcdecode(test_case):
  function require_librosa (line 751) | def require_librosa(test_case):
  function cmd_exists (line 758) | def cmd_exists(cmd):
  function require_usr_bin_time (line 762) | def require_usr_bin_time(test_case):
  function require_sudachi (line 769) | def require_sudachi(test_case):
  function require_jumanpp (line 776) | def require_jumanpp(test_case):
  function require_cython (line 783) | def require_cython(test_case):
  function get_gpu_count (line 790) | def get_gpu_count():
  function get_tests_dir (line 810) | def get_tests_dir(append_path=None):
  function apply_print_resets (line 848) | def apply_print_resets(buf):
  function assert_screenout (line 852) | def assert_screenout(out, what):
  class CaptureStd (line 858) | class CaptureStd:
    method __init__ (line 907) | def __init__(self, out=True, err=True, replay=True):
    method __enter__ (line 924) | def __enter__(self):
    method __exit__ (line 935) | def __exit__(self, *exc):
    method __repr__ (line 950) | def __repr__(self):
  class CaptureStdout (line 965) | class CaptureStdout(CaptureStd):
    method __init__ (line 968) | def __init__(self, replay=True):
  class CaptureStderr (line 972) | class CaptureStderr(CaptureStd):
    method __init__ (line 975) | def __init__(self, replay=True):
  class CaptureLogger (line 979) | class CaptureLogger:
    method __init__ (line 1004) | def __init__(self, logger):
    method __enter__ (line 1010) | def __enter__(self):
    method __exit__ (line 1014) | def __exit__(self, *exc):
    method __repr__ (line 1018) | def __repr__(self):
  function LoggingLevel (line 1023) | def LoggingLevel(level):
  function ExtendSysPath (line 1045) | def ExtendSysPath(path: Union[str, os.PathLike]) -> Iterator[None]:
  class TestCasePlus (line 1065) | class TestCasePlus(unittest.TestCase):
    method setUp (line 1144) | def setUp(self):
    method test_file_path (line 1165) | def test_file_path(self):
    method test_file_path_str (line 1169) | def test_file_path_str(self):
    method test_file_dir (line 1173) | def test_file_dir(self):
    method test_file_dir_str (line 1177) | def test_file_dir_str(self):
    method tests_dir (line 1181) | def tests_dir(self):
    method tests_dir_str (line 1185) | def tests_dir_str(self):
    method examples_dir (line 1189) | def examples_dir(self):
    method examples_dir_str (line 1193) | def examples_dir_str(self):
    method repo_root_dir (line 1197) | def repo_root_dir(self):
    method repo_root_dir_str (line 1201) | def repo_root_dir_str(self):
    method src_dir (line 1205) | def src_dir(self):
    method src_dir_str (line 1209) | def src_dir_str(self):
    method get_env (line 1212) | def get_env(self):
    method get_auto_remove_tmp_dir (line 1232) | def get_auto_remove_tmp_dir(self, tmp_dir=None, before=None, after=None):
    method python_one_liner_max_rss (line 1300) | def python_one_liner_max_rss(self, one_liner_str):
    method tearDown (line 1333) | def tearDown(self):
  function mockenv (line 1348) | def mockenv(**kwargs):
  function mockenv_context (line 1361) | def mockenv_context(*remove, **update):
  function pytest_addoption_shared (line 1397) | def pytest_addoption_shared(parser):
  function pytest_terminal_summary_main (line 1416) | def pytest_terminal_summary_main(tr, id):
  class _RunOutput (line 1555) | class _RunOutput:
    method __init__ (line 1556) | def __init__(self, returncode, stdout, stderr):
  function _read_stream (line 1562) | async def _read_stream(stream, callback):
  function _stream_subprocess (line 1571) | async def _stream_subprocess(cmd, env=None, stdin=None, timeout=None, qu...
  function execute_subprocess_async (line 1612) | def execute_subprocess_async(cmd, env=None, stdin=None, timeout=180, qui...
  function pytest_xdist_worker_id (line 1634) | def pytest_xdist_worker_id():
  function get_torch_dist_unique_port (line 1644) | def get_torch_dist_unique_port():
  function nested_simplify (line 1656) | def nested_simplify(obj, decimals=3):
  function check_json_file_has_correct_format (line 1687) | def check_json_file_has_correct_format(file_path):
  function to_2tuple (line 1704) | def to_2tuple(x):
  class SubprocessCallException (line 1711) | class SubprocessCallException(Exception):
  function run_command (line 1715) | def run_command(command: List[str], return_stdout=False):
  class RequestCounter (line 1732) | class RequestCounter:
    method __enter__ (line 1737) | def __enter__(self):
    method __exit__ (line 1749) | def __exit__(self, *args, **kwargs):
    method new_request (line 1752) | def new_request(self, method, **kwargs):
  function is_flaky (line 1763) | def is_flaky(max_attempts: int = 5, wait_before_retry: Optional[float] =...
  function run_test_in_subprocess (line 1799) | def run_test_in_subprocess(test_case, target_func, inputs=None, timeout=...
  function preprocess_string (line 1852) | def preprocess_string(string, skip_cuda_tests):
  class HfDocTestParser (line 1879) | class HfDocTestParser(doctest.DocTestParser):
    method parse (line 1916) | def parse(self, string, name="<string>"):
  class HfDoctestModule (line 1925) | class HfDoctestModule(Module):
    method collect (line 1931) | def collect(self) -> Iterable[DoctestItem]:

FILE: transformers/tf_utils.py
  function shape_list (line 26) | def shape_list(tensor: Union[tf.Tensor, np.ndarray]) -> List[int]:
  function stable_softmax (line 49) | def stable_softmax(logits: tf.Tensor, axis: Optional[int] = None, name: ...
  function functional_layernorm (line 73) | def functional_layernorm(inputs, weight, bias, epsilon=1e-5, axis=-1):
  function flatten (line 105) | def flatten(input, start_dim=0, end_dim=-1):
  function invert_attention_mask (line 123) | def invert_attention_mask(encoder_attention_mask: tf.Tensor) -> tf.Tensor:
  function check_embeddings_within_bounds (line 151) | def check_embeddings_within_bounds(tensor: tf.Tensor, embed_dim: int, te...
  function save_attributes_to_hdf5_group (line 171) | def save_attributes_to_hdf5_group(group, name, data):
  function load_attributes_from_hdf5_group (line 218) | def load_attributes_from_hdf5_group(group, name):
  function expand_1d (line 246) | def expand_1d(data):

FILE: transformers/time_series_utils.py
  class AffineTransformed (line 34) | class AffineTransformed(TransformedDistribution):
    method __init__ (line 35) | def __init__(self, base_distribution: Distribution, loc=None, scale=No...
    method mean (line 42) | def mean(self):
    method variance (line 49) | def variance(self):
    method stddev (line 56) | def stddev(self):
  class ParameterProjection (line 63) | class ParameterProjection(nn.Module):
    method __init__ (line 64) | def __init__(
    method forward (line 72) | def forward(self, x: torch.Tensor) -> Tuple[torch.Tensor]:
  class LambdaLayer (line 78) | class LambdaLayer(nn.Module):
    method __init__ (line 79) | def __init__(self, function):
    method forward (line 83) | def forward(self, x, *args):
  class DistributionOutput (line 87) | class DistributionOutput:
    method __init__ (line 92) | def __init__(self, dim: int = 1) -> None:
    method _base_distribution (line 96) | def _base_distribution(self, distr_args):
    method distribution (line 102) | def distribution(
    method event_shape (line 115) | def event_shape(self) -> Tuple:
    method event_dim (line 122) | def event_dim(self) -> int:
    method value_in_support (line 130) | def value_in_support(self) -> float:
    method get_parameter_projection (line 137) | def get_parameter_projection(self, in_features: int) -> nn.Module:
    method domain_map (line 147) | def domain_map(self, *args: torch.Tensor):
    method squareplus (line 156) | def squareplus(x: torch.Tensor) -> torch.Tensor:
  class StudentTOutput (line 164) | class StudentTOutput(DistributionOutput):
    method domain_map (line 173) | def domain_map(cls, df: torch.Tensor, loc: torch.Tensor, scale: torch....
  class NormalOutput (line 179) | class NormalOutput(DistributionOutput):
    method domain_map (line 188) | def domain_map(cls, loc: torch.Tensor, scale: torch.Tensor):
  class NegativeBinomialOutput (line 193) | class NegativeBinomialOutput(DistributionOutput):
    method domain_map (line 202) | def domain_map(cls, total_count: torch.Tensor, logits: torch.Tensor):
    method _base_distribution (line 206) | def _base_distribution(self, distr_args) -> Distribution:
    method distribution (line 216) | def distribution(

FILE: transformers/tokenization_utils.py
  class Trie (line 52) | class Trie:
    method __init__ (line 58) | def __init__(self):
    method add (line 61) | def add(self, word: str):
    method split (line 90) | def split(self, text: str) -> List[str]:
    method cut_text (line 243) | def cut_text(self, text, offsets):
  function _is_whitespace (line 268) | def _is_whitespace(char):
  function _is_control (line 280) | def _is_control(char):
  function _is_punctuation (line 292) | def _is_punctuation(char):
  function _is_end_of_word (line 307) | def _is_end_of_word(text):
  function _is_start_of_word (line 313) | def _is_start_of_word(text):
  function _insert_one_token_to_ordered_list (line 319) | def _insert_one_token_to_ordered_list(token_list: List[str], new_token: ...
  class PreTrainedTokenizer (line 333) | class PreTrainedTokenizer(PreTrainedTokenizerBase):
    method __init__ (line 346) | def __init__(self, **kwargs):
    method is_fast (line 359) | def is_fast(self) -> bool:
    method vocab_size (line 363) | def vocab_size(self) -> int:
    method get_added_vocab (line 369) | def get_added_vocab(self) -> Dict[str, int]:
    method __len__ (line 378) | def __len__(self):
    method _add_tokens (line 384) | def _add_tokens(self, new_tokens: Union[List[str], List[AddedToken]], ...
    method _create_trie (line 449) | def _create_trie(self, unique_no_split_tokens):
    method num_special_tokens_to_add (line 458) | def num_special_tokens_to_add(self, pair: bool = False) -> int:
    method tokenize (line 481) | def tokenize(self, text: TextInput, **kwargs) -> List[str]:
    method _tokenize (line 551) | def _tokenize(self, text, **kwargs):
    method convert_tokens_to_ids (line 560) | def convert_tokens_to_ids(self, tokens: Union[str, List[str]]) -> Unio...
    method _convert_token_to_id_with_added_voc (line 582) | def _convert_token_to_id_with_added_voc(self, token):
    method _convert_token_to_id (line 590) | def _convert_token_to_id(self, token):
    method _encode_plus (line 593) | def _encode_plus(
    method _batch_encode_plus (line 671) | def _batch_encode_plus(
    method _batch_prepare_for_model (line 757) | def _batch_prepare_for_model(
    method prepare_for_tokenization (line 821) | def prepare_for_tokenization(
    method get_special_tokens_mask (line 845) | def get_special_tokens_mask(
    method convert_ids_to_tokens (line 876) | def convert_ids_to_tokens(self, ids: int, skip_special_tokens: bool = ...
    method convert_ids_to_tokens (line 880) | def convert_ids_to_tokens(self, ids: List[int], skip_special_tokens: b...
    method convert_ids_to_tokens (line 883) | def convert_ids_to_tokens(
    method _convert_id_to_token (line 915) | def _convert_id_to_token(self, index: int) -> str:
    method convert_tokens_to_string (line 918) | def convert_tokens_to_string(self, tokens: List[str]) -> str:
    method _decode (line 921) | def _decode(

FILE: transformers/tokenization_utils_base.py
  class AddedToken (line 80) | class AddedToken:
    method __getstate__ (line 92) | def __getstate__(self):
  class EncodingFast (line 96) | class EncodingFast:
  class TruncationStrategy (line 126) | class TruncationStrategy(ExplicitEnum):
  class CharSpan (line 138) | class CharSpan(NamedTuple):
  class TokenSpan (line 151) | class TokenSpan(NamedTuple):
  class BatchEncoding (line 164) | class BatchEncoding(UserDict):
    method __init__ (line 191) | def __init__(
    method n_sequences (line 214) | def n_sequences(self) -> Optional[int]:
    method is_fast (line 223) | def is_fast(self) -> bool:
    method __getitem__ (line 230) | def __getitem__(self, item: Union[int, str]) -> Union[Any, EncodingFast]:
    method __getattr__ (line 252) | def __getattr__(self, item: str):
    method __getstate__ (line 258) | def __getstate__(self):
    method __setstate__ (line 261) | def __setstate__(self, state):
    method keys (line 268) | def keys(self):
    method values (line 271) | def values(self):
    method items (line 274) | def items(self):
    method encodings (line 282) | def encodings(self) -> Optional[List[EncodingFast]]:
    method tokens (line 289) | def tokens(self, batch_index: int = 0) -> List[str]:
    method sequence_ids (line 307) | def sequence_ids(self, batch_index: int = 0) -> List[Optional[int]]:
    method words (line 331) | def words(self, batch_index: int = 0) -> List[Optional[int]]:
    method word_ids (line 355) | def word_ids(self, batch_index: int = 0) -> List[Optional[int]]:
    method token_to_sequence (line 374) | def token_to_sequence(self, batch_or_token_index: int, token_index: Op...
    method token_to_word (line 413) | def token_to_word(self, batch_or_token_index: int, token_index: Option...
    method word_to_tokens (line 451) | def word_to_tokens(
    method token_to_chars (line 504) | def token_to_chars(self, batch_or_token_index: int, token_index: Optio...
    method char_to_token (line 543) | def char_to_token(
    method word_to_chars (line 584) | def word_to_chars(
    method char_to_word (line 629) | def char_to_word(self, batch_or_char_index: int, char_index: Optional[...
    method convert_to_tensors (line 668) | def convert_to_tensors(
    method to (line 756) | def to(self, device: Union[str, "torch.device"]) -> "BatchEncoding":
  class SpecialTokensMixin (line 778) | class SpecialTokensMixin:
    method __init__ (line 816) | def __init__(self, verbose=True, **kwargs):
    method sanitize_special_tokens (line 846) | def sanitize_special_tokens(self) -> int:
    method add_special_tokens (line 858) | def add_special_tokens(
    method add_tokens (line 952) | def add_tokens(
    method _add_tokens (line 1001) | def _add_tokens(self, new_tokens: Union[List[str], List[AddedToken]], ...
    method bos_token (line 1005) | def bos_token(self) -> str:
    method eos_token (line 1016) | def eos_token(self) -> str:
    method unk_token (line 1027) | def unk_token(self) -> str:
    method sep_token (line 1038) | def sep_token(self) -> str:
    method pad_token (line 1050) | def pad_token(self) -> str:
    method cls_token (line 1061) | def cls_token(self) -> str:
    method mask_token (line 1073) | def mask_token(self) -> str:
    method additional_special_tokens (line 1085) | def additional_special_tokens(self) -> List[str]:
    method bos_token (line 1097) | def bos_token(self, value):
    method eos_token (line 1101) | def eos_token(self, value):
    method unk_token (line 1105) | def unk_token(self, value):
    method sep_token (line 1109) | def sep_token(self, value):
    method pad_token (line 1113) | def pad_token(self, value):
    method cls_token (line 1117) | def cls_token(self, value):
    method mask_token (line 1121) | def mask_token(self, value):
    method additional_special_tokens (line 1125) | def additional_special_tokens(self, value):
    method bos_token_id (line 1129) | def bos_token_id(self) -> Optional[int]:
    method eos_token_id (line 1139) | def eos_token_id(self) -> Optional[int]:
    method unk_token_id (line 1149) | def unk_token_id(self) -> Optional[int]:
    method sep_token_id (line 1158) | def sep_token_id(self) -> Optional[int]:
    method pad_token_id (line 1168) | def pad_token_id(self) -> Optional[int]:
    method pad_token_type_id (line 1177) | def pad_token_type_id(self) -> int:
    method cls_token_id (line 1184) | def cls_token_id(self) -> Optional[int]:
    method mask_token_id (line 1196) | def mask_token_id(self) -> Optional[int]:
    method additional_special_tokens_ids (line 1206) | def additional_special_tokens_ids(self) -> List[int]:
    method bos_token_id (line 1214) | def bos_token_id(self, value):
    method eos_token_id (line 1218) | def eos_token_id(self, value):
    method unk_token_id (line 1222) | def unk_token_id(self, value):
    method sep_token_id (line 1226) | def sep_token_id(self, value):
    method pad_token_id (line 1230) | def pad_token_id(self, value):
    method cls_token_id (line 1234) | def cls_token_id(self, value):
    method mask_token_id (line 1238) | def mask_token_id(self, value):
    method additional_special_tokens_ids (line 1242) | def additional_special_tokens_ids(self, values):
    method special_tokens_map (line 1246) | def special_tokens_map(self) -> Dict[str, Union[str, List[str]]]:
    method special_tokens_map_extended (line 1265) | def special_tokens_map_extended(self) -> Dict[str, Union[str, AddedTok...
    method all_special_tokens (line 1281) | def all_special_tokens(self) -> List[str]:
    method all_special_tokens_extended (line 1291) | def all_special_tokens_extended(self) -> List[Union[str, AddedToken]]:
    method all_special_ids (line 1307) | def all_special_ids(self) -> List[int]:
  class PreTrainedTokenizerBase (line 1494) | class PreTrainedTokenizerBase(SpecialTokensMixin, PushToHubMixin):
    method __init__ (line 1514) | def __init__(self, **kwargs):
    method max_len_single_sentence (line 1551) | def max_len_single_sentence(self) -> int:
    method max_len_sentences_pair (line 1558) | def max_len_sentences_pair(self) -> int:
    method max_len_single_sentence (line 1565) | def max_len_single_sentence(self, value) -> int:
    method max_len_sentences_pair (line 1579) | def max_len_sentences_pair(self, value) -> int:
    method _set_processor_class (line 1590) | def _set_processor_class(self, processor_class: str):
    method __repr__ (line 1594) | def __repr__(self) -> str:
    method __len__ (line 1602) | def __len__(self) -> int:
    method get_vocab (line 1605) | def get_vocab(self) -> Dict[str, int]:
    method from_pretrained (line 1618) | def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os....
    method _from_pretrained (line 1839) | def _from_pretrained(
    method _eventually_correct_t5_max_length (line 2071) | def _eventually_correct_t5_max_length(pretrained_model_name_or_path, m...
    method save_pretrained (line 2078) | def save_pretrained(
    method _save_pretrained (line 2223) | def _save_pretrained(
    method save_vocabulary (line 2257) | def save_vocabulary(self, save_directory: str, filename_prefix: Option...
    method tokenize (line 2275) | def tokenize(self, text: str, pair: Optional[str] = None, add_special_...
    method encode (line 2305) | def encode(
    method num_special_tokens_to_add (line 2346) | def num_special_tokens_to_add(self, pair: bool = False) -> int:
    method _get_padding_truncation_strategies (line 2349) | def _get_padding_truncation_strategies(
    method __call__ (line 2488) | def __call__(
    method _call_one (line 2576) | def _call_one(
    method encode_plus (line 2689) | def encode_plus(
    method _encode_plus (line 2761) | def _encode_plus(
    method batch_encode_plus (line 2785) | def batch_encode_plus(
    method _batch_encode_plus (line 2858) | def _batch_encode_plus(
    method pad (line 2887) | def pad(
    method create_token_type_ids_from_sequences (line 3060) | def create_token_type_ids_from_sequences(
    method build_inputs_with_special_tokens (line 3080) | def build_inputs_with_special_tokens(
    method prepare_for_model (line 3101) | def prepare_for_model(
    method truncate_sequences (line 3236) | def truncate_sequences(
    method _pad (line 3355) | def _pad(
    method convert_tokens_to_string (line 3432) | def convert_tokens_to_string(self, tokens: List[str]) -> str:
    method batch_decode (line 3445) | def batch_decode(
    method decode (line 3479) | def decode(
    method _decode (line 3516) | def _decode(
    method get_special_tokens_mask (line 3525) | def get_special_tokens_mask(
    method clean_up_tokenization (line 3557) | def clean_up_tokenization(out_string: str) -> str:
    method _eventual_warn_about_too_long_sequence (line 3581) | def _eventual_warn_about_too_long_sequence(self, ids: List[int], max_l...
    method _switch_to_input_mode (line 3601) | def _switch_to_input_mode(self):
    method _switch_to_target_mode (line 3607) | def _switch_to_target_mode(self):
    method as_target_tokenizer (line 3614) | def as_target_tokenizer(self):
    method register_for_auto_class (line 3631) | def register_for_auto_class(cls, auto_class="AutoTokenizer"):
    method prepare_seq2seq_batch (line 3656) | def prepare_seq2seq_batch(
  function get_fast_tokenizer_file (line 3779) | def get_fast_tokenizer_file(tokenization_files: List[str]) -> str:

FILE: transformers/tokenization_utils_fast.py
  class PreTrainedTokenizerFast (line 78) | class PreTrainedTokenizerFast(PreTrainedTokenizerBase):
    method __init__ (line 95) | def __init__(self, *args, **kwargs):
    method is_fast (line 139) | def is_fast(self) -> bool:
    method vocab_size (line 143) | def vocab_size(self) -> int:
    method get_vocab (line 149) | def get_vocab(self) -> Dict[str, int]:
    method vocab (line 153) | def vocab(self) -> Dict[str, int]:
    method get_added_vocab (line 156) | def get_added_vocab(self) -> Dict[str, int]:
    method __len__ (line 168) | def __len__(self) -> int:
    method backend_tokenizer (line 175) | def backend_tokenizer(self) -> TokenizerFast:
    method decoder (line 182) | def decoder(self) -> DecoderFast:
    method _convert_encoding (line 188) | def _convert_encoding(
    method convert_tokens_to_ids (line 235) | def convert_tokens_to_ids(self, tokens: Union[str, List[str]]) -> Unio...
    method _convert_token_to_id_with_added_voc (line 254) | def _convert_token_to_id_with_added_voc(self, token: str) -> int:
    method _convert_id_to_token (line 260) | def _convert_id_to_token(self, index: int) -> Optional[str]:
    method _add_tokens (line 263) | def _add_tokens(self, new_tokens: List[Union[str, AddedToken]], specia...
    method num_special_tokens_to_add (line 269) | def num_special_tokens_to_add(self, pair: bool = False) -> int:
    method convert_ids_to_tokens (line 290) | def convert_ids_to_tokens(
    method tokenize (line 316) | def tokenize(self, text: str, pair: Optional[str] = None, add_special_...
    method set_truncation_and_padding (line 319) | def set_truncation_and_padding(
    method _batch_encode_plus (line 390) | def _batch_encode_plus(
    method _encode_plus (line 475) | def _encode_plus(
    method convert_tokens_to_string (line 532) | def convert_tokens_to_string(self, tokens: List[str]) -> str:
    method _decode (line 535) | def _decode(
    method _save_pretrained (line 559) | def _save_pretrained(
    method train_new_from_iterator (line 607) | def train_new_from_iterator(

FILE: transformers/tools/agents.py
  class PreTool (line 59) | class PreTool:
  function get_remote_tools (line 76) | def get_remote_tools(organization="huggingface-tools"):
  function _setup_default_tools (line 91) | def _setup_default_tools():
  function resolve_tools (line 121) | def resolve_tools(code, toolbox, remote=False, cached_tools=None):
  function get_tool_creation_code (line 140) | def get_tool_creation_code(code, toolbox, remote=False):
  function clean_code_for_chat (line 156) | def clean_code_for_chat(result):
  function clean_code_for_run (line 174) | def clean_code_for_run(result):
  class Agent (line 190) | class Agent:
    method __init__ (line 208) | def __init__(self, chat_prompt_template=None, run_prompt_template=None...
    method toolbox (line 236) | def toolbox(self) -> Dict[str, Tool]:
    method format_prompt (line 240) | def format_prompt(self, task, chat_mode=False):
    method set_stream (line 253) | def set_stream(self, streamer):
    method chat (line 262) | def chat(self, task, *, return_code=False, remote=False, **kwargs):
    method prepare_for_new_chat (line 304) | def prepare_for_new_chat(self):
    method run (line 312) | def run(self, task, *, return_code=False, remote=False, **kwargs):
    method generate_one (line 349) | def generate_one(self, prompt, stop):
    method generate_many (line 353) | def generate_many(self, prompts, stop):
  class OpenAiAgent (line 358) | class OpenAiAgent(Agent):
    method __init__ (line 396) | def __init__(
    method generate_many (line 424) | def generate_many(self, prompts, stop):
    method generate_one (line 430) | def generate_one(self, prompt, stop):
    method _chat_generate (line 436) | def _chat_generate(self, prompt, stop):
    method _completion_generate (line 445) | def _completion_generate(self, prompts, stop):
  class AzureOpenAiAgent (line 456) | class AzureOpenAiAgent(Agent):
    method __init__ (line 504) | def __init__(
    method generate_many (line 550) | def generate_many(self, prompts, stop):
    method generate_one (line 556) | def generate_one(self, prompt, stop):
    method _chat_generate (line 562) | def _chat_generate(self, prompt, stop):
    method _completion_generate (line 571) | def _completion_generate(self, prompts, stop):
  class HfAgent (line 582) | class HfAgent(Agent):
    method __init__ (line 614) | def __init__(
    method generate_one (line 630) | def generate_one(self, prompt, stop):
  class LocalAgent (line 653) | class LocalAgent(Agent):
    method __init__ (line 689) | def __init__(self, model, tokenizer, chat_prompt_template=None, run_pr...
    method from_pretrained (line 699) | def from_pretrained(cls, pretrained_model_name_or_path, **kwargs):
    method _model_device (line 724) | def _model_device(self):
    method generate_one (line 730) | def generate_one(self, prompt, stop):
  class StopSequenceCriteria (line 746) | class StopSequenceCriteria(StoppingCriteria):
    method __init__ (line 757) | def __init__(self, stop_sequences, tokenizer):
    method __call__ (line 763) | def __call__(self, input_ids, scores, **kwargs) -> bool:

FILE: transformers/tools/base.py
  function get_repo_type (line 54) | def get_repo_type(repo_id, repo_type=None, **hub_kwargs):
  class Tool (line 80) | class Tool:
    method __init__ (line 108) | def __init__(self, *args, **kwargs):
    method __call__ (line 111) | def __call__(self, *args, **kwargs):
    method setup (line 114) | def setup(self):
    method save (line 121) | def save(self, output_dir):
    method from_hub (line 176) | def from_hub(
    method push_to_hub (line 285) | def push_to_hub(
    method from_gradio (line 330) | def from_gradio(gradio_tool):
  class RemoteTool (line 345) | class RemoteTool(Tool):
    method __init__ (line 360) | def __init__(self, endpoint_url=None, token=None, tool_class=None):
    method prepare_inputs (line 365) | def prepare_inputs(self, *args, **kwargs):
    method extract_outputs (line 408) | def extract_outputs(self, outputs):
    method __call__ (line 415) | def __call__(self, *args, **kwargs):
  class PipelineTool (line 427) | class PipelineTool(Tool):
    method __init__ (line 470) | def __init__(
    method setup (line 507) | def setup(self):
    method encode (line 533) | def encode(self, raw_inputs):
    method forward (line 539) | def forward(self, inputs):
    method decode (line 546) | def decode(self, outputs):
    method __call__ (line 552) | def __call__(self, *args, **kwargs):
  function launch_gradio_demo (line 563) | def launch_gradio_demo(tool_class: Tool):
  function get_default_device (line 591) | def get_default_device():
  function get_default_endpoints (line 617) | def get_default_endpoints():
  function supports_remote (line 624) | def supports_remote(task_or_repo_id):
  function load_tool (line 629) | def load_tool(task_or_repo_id, model_repo_id=None, remote=False, token=N...
  function add_description (line 683) | def add_description(description):
  class EndpointClient (line 697) | class EndpointClient:
    method __init__ (line 698) | def __init__(self, endpoint_url: str, token: Optional[str] = None):
    method encode_image (line 703) | def encode_image(image):
    method decode_image (line 710) | def decode_image(raw_image):
    method __call__ (line 722) | def __call__(

FILE: transformers/tools/document_question_answering.py
  class DocumentQuestionAnsweringTool (line 29) | class DocumentQuestionAnsweringTool(PipelineTool):
    method __init__ (line 43) | def __init__(self, *args, **kwargs):
    method encode (line 49) | def encode(self, document: "Image", question: str):
    method forward (line 59) | def forward(self, inputs):
    method decode (line 73) | def decode(self, outputs):

FILE: transformers/tools/evaluate_agent.py
  function classifier (line 22) | def classifier(text, labels):
  function translator (line 26) | def translator(text, src_lang, tgt_lang):
  function speaker (line 30) | def speaker(text):
  function transcriber (line 34) | def transcriber(audio):
  function image_generator (line 40) | def image_generator(prompt):
  function image_captioner (line 44) | def image_captioner(image):
  function image_transformer (line 50) | def image_transformer(image, prompt):
  function question_answerer (line 56) | def question_answerer(text, question):
  function image_qa (line 60) | def image_qa(image, question):
  function text_downloader (line 66) | def text_downloader(url):
  function summarizer (line 70) | def summarizer(text):
  function video_generator (line 74) | def video_generator(prompt, seconds=2):
  function document_qa (line 78) | def document_qa(image, question):
  function image_segmenter (line 82) | def image_segmenter(image, prompt):
  class Problem (line 104) | class Problem:
    method __init__ (line 120) | def __init__(self, task, inputs, answer):
  function get_theoretical_tools (line 400) | def get_theoretical_tools(agent_answer, theoretical_answer, code_answer):
  function evaluate_code (line 416) | def evaluate_code(code, inputs=None, state=None, verbose=False, return_i...
  function score_code (line 443) | def score_code(agent_answer, theoretical_answer, verbose: bool = False):
  function evaluate_one_result (line 462) | def evaluate_one_result(explanation, code, agent_answer, theoretical_ans...
  function evaluate_agent (line 505) | def evaluate_agent(agent, batch_size=8, verbose=False, return_errors=Fal...
  function evaluate_chat_agent (line 594) | def evaluate_chat_agent(agent, verbose=False, return_errors=False):

FILE: transformers/tools/image_captioning.py
  class ImageCaptioningTool (line 28) | class ImageCaptioningTool(PipelineTool):
    method __init__ (line 40) | def __init__(self, *args, **kwargs):
    method encode (line 44) | def encode(self, image: "Image"):
    method forward (line 47) | def forward(self, inputs):
    method decode (line 50) | def decode(self, outputs):

FILE: transformers/tools/image_question_answering.py
  class ImageQuestionAnsweringTool (line 30) | class ImageQuestionAnsweringTool(PipelineTool):
    method __init__ (line 44) | def __init__(self, *args, **kwargs):
    method encode (line 48) | def encode(self, image: "Image", question: str):
    method forward (line 51) | def forward(self, inputs):
    method decode (line 55) | def decode(self, outputs):

FILE: transformers/tools/image_segmentation.py
  class ImageSegmentationTool (line 29) | class ImageSegmentationTool(PipelineTool):
    method __init__ (line 42) | def __init__(self, *args, **kwargs):
    method encode (line 46) | def encode(self, image: "Image", label: str):
    method forward (line 50) | def forward(self, inputs):
    method decode (line 55) | def decode(self, outputs):

FILE: transformers/tools/prompts.py
  function download_prompt (line 33) | def download_prompt(prompt_or_repo_id, agent_name, mode="run"):

FILE: transformers/tools/python_interpreter.py
  class InterpretorError (line 23) | class InterpretorError(ValueError):
  function evaluate (line 32) | def evaluate(code: str, tools: Dict[str, Callable], state=None, chat_mod...
  function evaluate_ast (line 78) | def evaluate_ast(expression: ast.AST, state: Dict[str, Any], tools: Dict...
  function evaluate_assign (line 137) | def evaluate_assign(assign, state, tools):
  function evaluate_call (line 151) | def evaluate_call(call, state, tools):
  function evaluate_subscript (line 170) | def evaluate_subscript(subscript, state, tools):
  function evaluate_name (line 185) | def evaluate_name(name, state, tools):
  function evaluate_condition (line 194) | def evaluate_condition(condition, state, tools):
  function evaluate_if (line 226) | def evaluate_if(if_statement, state, tools):

FILE: transformers/tools/speech_to_text.py
  class SpeechToTextTool (line 21) | class SpeechToTextTool(PipelineTool):
    method encode (line 34) | def encode(self, audio):
    method forward (line 37) | def forward(self, inputs):
    method decode (line 40) | def decode(self, outputs):

FILE: transformers/tools/text_classification.py
  class TextClassificationTool (line 23) | class TextClassificationTool(PipelineTool):
    method setup (line 48) | def setup(self):
    method encode (line 58) | def encode(self, text, labels):
    method decode (line 67) | def decode(self, outputs):

FILE: transformers/tools/text_question_answering.py
  class TextQuestionAnsweringTool (line 26) | class TextQuestionAnsweringTool(PipelineTool):
    method encode (line 39) | def encode(self, text: str, question: str):
    method forward (line 43) | def forward(self, inputs):
    method decode (line 51) | def decode(self, outputs):

FILE: transformers/tools/text_summarization.py
  class TextSummarizationTool (line 21) | class TextSummarizationTool(PipelineTool):
    method encode (line 45) | def encode(self, text):
    method forward (line 48) | def forward(self, inputs):
    method decode (line 51) | def decode(self, outputs):

FILE: transformers/tools/text_to_speech.py
  class TextToSpeechTool (line 28) | class TextToSpeechTool(PipelineTool):
    method setup (line 42) | def setup(self):
    method encode (line 47) | def encode(self, text, speaker_embeddings=None):
    method forward (line 59) | def forward(self, inputs):
    method decode (line 63) | def decode(self, outputs):

FILE: transformers/tools/translation.py
  class TranslationTool (line 229) | class TranslationTool(PipelineTool):
    method encode (line 256) | def encode(self, text, src_lang, tgt_lang):
    method forward (line 267) | def forward(self, inputs):
    method decode (line 270) | def decode(self, outputs):

FILE: transformers/trainer.py
  class Trainer (line 240) | class Trainer:
    method __init__ (line 325) | def __init__(
    method add_callback (line 715) | def add_callback(self, callback):
    method pop_callback (line 726) | def pop_callback(self, callback):
    method remove_callback (line 742) | def remove_callback(self, callback):
    method _move_model_to_device (line 753) | def _move_model_to_device(self, model, device):
    method _set_signature_columns_if_needed (line 759) | def _set_signature_columns_if_needed(self):
    method _remove_unused_columns (line 767) | def _remove_unused_columns(self, dataset: "datasets.Dataset", descript...
    method _get_collator_with_removed_columns (line 793) | def _get_collator_with_removed_columns(
    method _get_train_sampler (line 811) | def _get_train_sampler(self) -> Optional[torch.utils.data.Sampler]:
    method get_train_dataloader (line 882) | def get_train_dataloader(self) -> DataLoader:
    method _get_eval_sampler (line 932) | def _get_eval_sampler(self, eval_dataset: Dataset) -> Optional[torch.u...
    method get_eval_dataloader (line 961) | def get_eval_dataloader(self, eval_dataset: Optional[Dataset] = None) ...
    method get_test_dataloader (line 1011) | def get_test_dataloader(self, test_dataset: Dataset) -> DataLoader:
    method create_optimizer_and_scheduler (line 1059) | def create_optimizer_and_scheduler(self, num_training_steps: int):
    method create_optimizer (line 1075) | def create_optimizer(self):
    method get_optimizer_cls_and_kwargs (line 1132) | def get_optimizer_cls_and_kwargs(args: TrainingArguments) -> Tuple[Any...
    method create_scheduler (line 1254) | def create_scheduler(self, num_training_steps: int, optimizer: torch.o...
    method num_examples (line 1271) | def num_examples(self, dataloader: DataLoader) -> int:
    method _hp_search_setup (line 1285) | def _hp_search_setup(self, trial: Union["optuna.Trial", Dict[str, Any]]):
    method _report_to_hp_search (line 1332) | def _report_to_hp_search(self, trial: Union["optuna.Trial", Dict[str, ...
    method _tune_save_checkpoint (line 1350) | def _tune_save_checkpoint(self):
    method call_model_init (line 1363) | def call_model_init(self, trial=None):
    method torch_jit_model_eval (line 1377) | def torch_jit_model_eval(self, model, dataloader, training=False):
    method ipex_optimize_model (line 1415) | def ipex_optimize_model(self, model, training=False, dtype=torch.float...
    method _wrap_model (line 1438) | def _wrap_model(self, model, training=True, dataloader=None):
    method train (line 1566) | def train(
    method _inner_training_loop (line 1652) | def _inner_training_loop(
    method _get_output_dir (line 2094) | def _get_output_dir(self, trial):
    method _load_from_checkpoint (line 2114) | def _load_from_checkpoint(self, resume_from_checkpoint, model=None):
    method _load_best_model (line 2187) | def _load_best_model(self):
    method _issue_warnings_after_load (line 2274) | def _issue_warnings_after_load(self, load_result):
    method _maybe_log_save_evaluate (line 2287) | def _maybe_log_save_evaluate(self, tr_loss, model, trial, epoch, ignor...
    method _load_rng_state (line 2335) | def _load_rng_state(self, checkpoint):
    method _save_checkpoint (line 2376) | def _save_checkpoint(self, model, trial, metrics=None):
    method _load_optimizer_and_scheduler (line 2492) | def _load_optimizer_and_scheduler(self, checkpoint):
    method hyperparameter_search (line 2561) | def hyperparameter_search(
    method log (line 2657) | def log(self, logs: Dict[str, float]) -> None:
    method _prepare_input (line 2674) | def _prepare_input(self, data: Union[torch.Tensor, Any]) -> Union[torc...
    method _prepare_inputs (line 2692) | def _prepare_inputs(self, inputs: Dict[str, Union[torch.Tensor, Any]])...
    method compute_loss_context_manager (line 2708) | def compute_loss_context_manager(self):
    method autocast_smart_context_manager (line 2714) | def autocast_smart_context_manager(self, cache_enabled: Optional[bool]...
    method training_step (line 2733) | def training_step(self, model: nn.Module, inputs: Dict[str, Union[torc...
    method compute_loss (line 2774) | def compute_loss(self, model, inputs, return_outputs=False):
    method is_local_process_zero (line 2810) | def is_local_process_zero(self) -> bool:
    method is_world_process_zero (line 2817) | def is_world_process_zero(self) -> bool:
    method save_model (line 2829) | def save_model(self, output_dir: Optional[str] = None, _internal_call:...
    method _save_tpu (line 2898) | def _save_tpu(self, output_dir: Optional[str] = None):
    method _save (line 2926) | def _save(self, output_dir: Optional[str] = None, state_dict=None):
    method store_flos (line 2960) | def store_flos(self):
    method _sorted_checkpoints (line 2971) | def _sorted_checkpoints(
    method _rotate_checkpoints (line 2995) | def _rotate_checkpoints(self, use_mtime=False, output_dir=None) -> None:
    method evaluate (line 3020) | def evaluate(
    method predict (line 3091) | def predict(
    method evaluation_loop (line 3153) | def evaluation_loop(
    method _nested_gather (line 3376) | def _nested_gather(self, tensors, name=None):
    method _pad_across_processes (line 3396) | def _pad_across_processes(self, tensor, pad_index=-100):
    method prediction_step (line 3430) | def prediction_step(
    method floating_point_ops (line 3536) | def floating_point_ops(self, inputs: Dict[str, Union[torch.Tensor, Any...
    method init_git_repo (line 3554) | def init_git_repo(self, at_init: bool = False):
    method create_model_card (line 3601) | def create_model_card(
    method _push_from_checkpoint (line 3657) | def _push_from_checkpoint(self, checkpoint_folder):
    method push_to_hub (line 3702) | def push_to_hub(self, commit_message: Optional[str] = "End of training...
    method prediction_loop (line 3762) | def prediction_loop(
    method _gather_and_numpify (line 3916) | def _gather_and_numpify(self, tensors, name):
    method _add_sm_patterns_to_gitignore (line 3932) | def _add_sm_patterns_to_gitignore(self) -> None:
    method create_accelerator_and_postprocess (line 3971) | def create_accelerator_and_postprocess(self):

FILE: transformers/trainer_callback.py
  class TrainerState (line 35) | class TrainerState:
    method __post_init__ (line 91) | def __post_init__(self):
    method save_to_json (line 95) | def save_to_json(self, json_path: str):
    method load_from_json (line 102) | def load_from_json(cls, json_path: str):
  class TrainerControl (line 110) | class TrainerControl:
    method _new_training (line 144) | def _new_training(self):
    method _new_epoch (line 148) | def _new_epoch(self):
    method _new_step (line 152) | def _new_step(self):
  class TrainerCallback (line 159) | class TrainerCallback:
    method on_init_end (line 209) | def on_init_end(self, args: TrainingArguments, state: TrainerState, co...
    method on_train_begin (line 215) | def on_train_begin(self, args: TrainingArguments, state: TrainerState,...
    method on_train_end (line 221) | def on_train_end(self, args: TrainingArguments, state: TrainerState, c...
    method on_epoch_begin (line 227) | def on_epoch_begin(self, args: TrainingArguments, state: TrainerState,...
    method on_epoch_end (line 233) | def on_epoch_end(self, args: TrainingArguments, state: TrainerState, c...
    method on_step_begin (line 239) | def on_step_begin(self, args: TrainingArguments, state: TrainerState, ...
    method on_substep_end (line 246) | def on_substep_end(self, args: TrainingArguments, state: TrainerState,...
    method on_step_end (line 252) | def on_step_end(self, args: TrainingArguments, state: TrainerState, co...
    method on_evaluate (line 259) | def on_evaluate(self, args: TrainingArguments, state: TrainerState, co...
    method on_predict (line 265) | def on_predict(self, args: TrainingArguments, state: TrainerState, con...
    method on_save (line 271) | def on_save(self, args: TrainingArguments, state: TrainerState, contro...
    method on_log (line 277) | def on_log(self, args: TrainingArguments, state: TrainerState, control...
    method on_prediction_step (line 283) | def on_prediction_step(self, args: TrainingArguments, state: TrainerSt...
  class CallbackHandler (line 290) | class CallbackHandler(TrainerCallback):
    method __init__ (line 293) | def __init__(self, callbacks, model, tokenizer, optimizer, lr_scheduler):
    method add_callback (line 312) | def add_callback(self, callback):
    method pop_callback (line 323) | def pop_callback(self, callback):
    method remove_callback (line 335) | def remove_callback(self, callback):
    method callback_list (line 345) | def callback_list(self):
    method on_init_end (line 348) | def on_init_end(self, args: TrainingArguments, state: TrainerState, co...
    method on_train_begin (line 351) | def on_train_begin(self, args: TrainingArguments, state: TrainerState,...
    method on_train_end (line 355) | def on_train_end(self, args: TrainingArguments, state: TrainerState, c...
    method on_epoch_begin (line 358) | def on_epoch_begin(self, args: TrainingArguments, state: TrainerState,...
    method on_epoch_end (line 362) | def on_epoch_end(self, args: TrainingArguments, state: TrainerState, c...
    method on_step_begin (line 365) | def on_step_begin(self, args: TrainingArguments, state: TrainerState, ...
    method on_substep_end (line 371) | def on_substep_end(self, args: TrainingArguments, state: TrainerState,...
    method on_step_end (line 374) | def on_step_end(self, args: TrainingArguments, state: TrainerState, co...
    method on_evaluate (line 377) | def on_evaluate(self, args: TrainingArguments, state: TrainerState, co...
    method on_predict (line 381) | def on_predict(self, args: TrainingArguments, state: TrainerState, con...
    method on_save (line 384) | def on_save(self, args: TrainingArguments, state: TrainerState, contro...
    method on_log (line 388) | def on_log(self, args: TrainingArguments, state: TrainerState, control...
    method on_prediction_step (line 392) | def on_prediction_step(self, args: TrainingArguments, state: TrainerSt...
    method call_event (line 395) | def call_event(self, event, args, state, control, **kwargs):
  class DefaultFlowCallback (line 415) | class DefaultFlowCallback(TrainerCallback):
    method on_step_end (line 420) | def on_step_end(self, args: TrainingArguments, state: TrainerState, co...
    method on_epoch_end (line 449) | def on_epoch_end(self, args: TrainingArguments, state: TrainerState, c...
  class ProgressCallback (line 465) | class ProgressCallback(TrainerCallback):
    method __init__ (line 470) | def __init__(self):
    method on_train_begin (line 474) | def on_train_begin(self, args, state, control, **kwargs):
    method on_step_end (line 479) | def on_step_end(self, args, state, control, **kwargs):
    method on_prediction_step (line 484) | def on_prediction_step(self, args, state, control, eval_dataloader=Non...
    method on_evaluate (line 490) | def on_evaluate(self, args, state, control, **kwargs):
    method on_predict (line 496) | def on_predict(self, args, state, control, **kwargs):
    method on_log (line 502) | def on_log(self, args, state, control, logs=None, **kwargs):
    method on_train_end (line 507) | def on_train_end(self, args, state, control, **kwargs):
  class PrinterCallback (line 513) | class PrinterCallback(TrainerCallback):
    method on_log (line 518) | def on_log(self, args, state, control, logs=None, **kwargs):
  class EarlyStoppingCallback (line 524) | class EarlyStoppingCallback(TrainerCallback):
    method __init__ (line 541) | def __init__(self, early_stopping_patience: int = 1, early_stopping_th...
    method check_metric_value (line 547) | def check_metric_value(self, args, state, control, metric_value):
    method on_train_begin (line 558) | def on_train_begin(self, args, state, control, **kwargs):
    method on_evaluate (line 567) | def on_evaluate(self, args, state, control, metrics, **kwargs):

FILE: transformers/trainer_pt_utils.py
  function atleast_1d (line 58) | def atleast_1d(tensor_or_array: Union[torch.Tensor, np.ndarray]):
  function torch_pad_and_concatenate (line 69) | def torch_pad_and_concatenate(tensor1, tensor2, padding_index=-100):
  function numpy_pad_and_concatenate (line 87) | def numpy_pad_and_concatenate(array1, array2, padding_index=-100):
  function nested_concat (line 105) | def nested_concat(tensors, new_tensors, padding_index=-100):
  function find_batch_size (line 127) | def find_batch_size(tensors):
  function nested_numpify (line 147) | def nested_numpify(tensors):
  function nested_detach (line 163) | def nested_detach(tensors):
  function nested_xla_mesh_reduce (line 172) | def nested_xla_mesh_reduce(tensors, name):
  function distributed_concat (line 189) | def distributed_concat(tensor: Any, num_total_examples: Optional[int] = ...
  function distributed_broadcast_scalars (line 208) | def distributed_broadcast_scalars(
  function reissue_pt_warnings (line 227) | def reissue_pt_warnings(caught_warnings):
  function torch_distributed_zero_first (line 236) | def torch_distributed_zero_first(local_rank: int):
  class DistributedSamplerWithLoop (line 250) | class DistributedSamplerWithLoop(DistributedSampler):
    method __init__ (line 264) | def __init__(self, dataset, batch_size, **kwargs):
    method __iter__ (line 268) | def __iter__(self):
  class SequentialDistributedSampler (line 278) | class SequentialDistributedSampler(Sampler):
    method __init__ (line 288) | def __init__(self, dataset, num_replicas=None, rank=None, batch_size=N...
    method __iter__ (line 313) | def __iter__(self):
    method __len__ (line 330) | def __len__(self):
  function get_tpu_sampler (line 334) | def get_tpu_sampler(dataset: torch.utils.data.Dataset, batch_size: int):
  function nested_new_like (line 340) | def nested_new_like(arrays, num_samples, padding_index=-100):
  function expand_like (line 347) | def expand_like(arrays, new_seq_length, padding_index=-100):
  function nested_truncate (line 354) | def nested_truncate(tensors, limit):
  class DistributedTensorGatherer (line 364) | class DistributedTensorGatherer:
    method __init__ (line 410) | def __init__(self, world_size, num_samples, make_multiple_of=None, pad...
    method add_arrays (line 424) | def add_arrays(self, arrays):
    method _nested_set_tensors (line 439) | def _nested_set_tensors(self, storage, arrays):
    method finalize (line 460) | def finalize(self):
  class LabelSmoother (line 473) | class LabelSmoother:
    method __call__ (line 487) | def __call__(self, model_output, labels, shift_labels=False):
  function get_length_grouped_indices (line 515) | def get_length_grouped_indices(lengths, batch_size, mega_batch_mult=None...
  class LengthGroupedSampler (line 550) | class LengthGroupedSampler(Sampler):
    method __init__ (line 556) | def __init__(
    method __len__ (line 588) | def __len__(self):
    method __iter__ (line 591) | def __iter__(self):
  class DistributedLengthGroupedSampler (line 596) | class DistributedLengthGroupedSampler(DistributedSampler):
    method __init__ (line 603) | def __init__(
    method __iter__ (line 663) | def __iter__(self) -> Iterator:
  class ShardSampler (line 684) | class ShardSampler(Sampler):
    method __init__ (line 693) | def __init__(
    method __iter__ (line 712) | def __iter__(self):
    method __len__ (line 726) | def __len__(self):
  class IterableDatasetShard (line 731) | class IterableDatasetShard(IterableDataset):
    method __init__ (line 772) | def __init__(
    method set_epoch (line 790) | def set_epoch(self, epoch):
    method __iter__ (line 795) | def __iter__(self):
    method __len__ (line 828) | def __len__(self):
  function _get_learning_rate (line 840) | def _get_learning_rate(self):
  function _secs2timedelta (line 863) | def _secs2timedelta(secs):
  function metrics_format (line 872) | def metrics_format(self, metrics: Dict[str, float]) -> Dict[str, float]:
  function log_metrics (line 898) | def log_metrics(self, split, metrics):
  function save_metrics (line 988) | def save_metrics(self, split, metrics, combined=True):
  function save_state (line 1026) | def save_state(self):
  function get_model_param_count (line 1039) | def get_model_param_count(model, trainable_only=False):
  function get_parameter_names (line 1056) | def get_parameter_names(model, forbidden_layer_types):
  function get_module_class_from_name (line 1072) | def get_module_class_from_name(module, name):
  function smp_forward_backward (line 1096) | def smp_forward_backward(model, inputs, gradient_accumulation_steps=1):
  function smp_forward_only (line 1104) | def smp_forward_only(model, inputs):
  function smp_gather (line 1107) | def smp_gather(tensor):
  function smp_nested_concat (line 1120) | def smp_nested_concat(tensor):

FILE: transformers/trainer_seq2seq.py
  class Seq2SeqTrainer (line 41) | class Seq2SeqTrainer(Trainer):
    method __init__ (line 42) | def __init__(
    method load_generation_config (line 77) | def load_generation_config(gen_config_arg: Union[str, GenerationConfig...
    method evaluate (line 112) | def evaluate(
    method predict (line 161) | def predict(
    method prediction_step (line 218) | def prediction_step(
    method _pad_tensors_to_max_len (line 317) | def _pad_tensors_to_max_len(self, tensor, max_length):

FILE: transformers/trainer_tf.py
  class TFTrainer (line 61) | class TFTrainer:
    method __init__ (line 94) | def __init__(
    method get_train_tfdataset (line 149) | def get_train_tfdataset(self) -> tf.data.Dataset:
    method get_eval_tfdataset (line 173) | def get_eval_tfdataset(self, eval_dataset: Optional[tf.data.Dataset] =...
    method get_test_tfdataset (line 206) | def get_test_tfdataset(self, test_dataset: tf.data.Dataset) -> tf.data...
    method create_optimizer_and_scheduler (line 231) | def create_optimizer_and_scheduler(self, num_training_steps: int):
    method setup_wandb (line 256) | def setup_wandb(self):
    method setup_comet (line 275) | def setup_comet(self):
    method prediction_loop (line 305) | def prediction_loop(
    method log (line 391) | def log(self, logs: Dict[str, float]) -> None:
    method evaluate (line 423) | def evaluate(self, eval_dataset: Optional[tf.data.Dataset] = None) -> ...
    method prediction_step (line 451) | def prediction_step(
    method distributed_prediction_steps (line 467) | def distributed_prediction_steps(self, batch):
    method train (line 475) | def train(self) -> None:
    method training_step (line 630) | def training_step(self, features, labels, nb_instances_in_global_batch):
    method apply_gradients (line 651) | def apply_gradients(self, features, labels, nb_instances_in_global_bat...
    method distributed_training_steps (line 705) | def distributed_training_steps(self, batch):
    method _compute_nb_instances (line 713) | def _compute_nb_instances(batch):
    method _get_step_inputs (line 723) | def _get_step_inputs(batch, nb_instances):
    method run_model (line 734) | def run_model(self, features, labels, training):
    method predict (line 764) | def predict(self, test_dataset: tf.data.Dataset) -> PredictionOutput:
    method save_model (line 790) | def save_model(self, output_dir: Optional[str] = None):

FILE: transformers/trainer_utils.py
  function seed_worker (line 50) | def seed_worker(_):
  function enable_full_determinism (line 58) | def enable_full_determinism(seed: int):
  function set_seed (line 83) | def set_seed(seed: int):
  class EvalPrediction (line 100) | class EvalPrediction:
    method __init__ (line 110) | def __init__(
    method __iter__ (line 120) | def __iter__(self):
    method __getitem__ (line 126) | def __getitem__(self, idx):
  class EvalLoopOutput (line 139) | class EvalLoopOutput(NamedTuple):
  class PredictionOutput (line 146) | class PredictionOutput(NamedTuple):
  class TrainOutput (line 152) | class TrainOutput(NamedTuple):
  function get_last_checkpoint (line 162) | def get_last_checkpoint(folder):
  class IntervalStrategy (line 174) | class IntervalStrategy(ExplicitEnum):
  class EvaluationStrategy (line 180) | class EvaluationStrategy(ExplicitEnum):
  class HubStrategy (line 186) | class HubStrategy(ExplicitEnum):
  class BestRun (line 193) | class BestRun(NamedTuple):
  function default_compute_objective (line 215) | def default_compute_objective(metrics: Dict[str, float]) -> float:
  function default_hp_space_optuna (line 240) | def default_hp_space_optuna(trial) -> Dict[str, float]:
  function default_hp_space_ray (line 252) | def default_hp_space_ray(trial) -> Dict[str, float]:
  function default_hp_space_sigopt (line 266) | def default_hp_space_sigopt(trial):
  function default_hp_space_wandb (line 279) | def default_hp_space_wandb(trial) -> Dict[str, float]:
  class HPSearchBackend (line 297) | class HPSearchBackend(ExplicitEnum):
  function is_main_process (line 312) | def is_main_process(local_rank):
  function total_processes_number (line 324) | def total_processes_number(local_rank):
  function speed_metrics (line 339) | def speed_metrics(split, start_time, num_samples=None, num_steps=None):
  class SchedulerType (line 364) | class SchedulerType(ExplicitEnum):
  class TrainerMemoryTracker (line 375) | class TrainerMemoryTracker:
    method __init__ (line 407) | def __init__(self, skip_memory_metrics=False):
    method derive_stage (line 433) | def derive_stage(self):
    method cpu_mem_used (line 443) | def cpu_mem_used(self):
    method peak_monitor_func (line 447) | def peak_monitor_func(self):
    method start (line 459) | def start(self):
    method stop (line 489) | def stop(self, stage):
    method update_metrics (line 533) | def update_metrics(self, stage, metrics):
    method stop_and_update_metrics (line 574) | def stop_and_update_metrics(self, metrics=None):
  function has_length (line 587) | def has_length(dataset):
  function denumpify_detensorize (line 598) | def denumpify_detensorize(metrics):
  function number_of_arguments (line 613) | def number_of_arguments(func):
  class ShardedDDPOption (line 623) | class ShardedDDPOption(ExplicitEnum):
  function find_executable_batch_size (line 631) | def find_executable_batch_size(
  class FSDPOption (line 662) | class FSDPOption(ExplicitEnum):
  class RemoveColumnsCollator (line 670) | class RemoveColumnsCollator:
    method __init__ (line 673) | def __init__(
    method _remove_columns (line 688) | def _remove_columns(self, feature: dict) -> dict:
    method __call__ (line 704) | def __call__(self, features: List[dict]):

FILE: transformers/training_args.py
  function default_logdir (line 102) | def default_logdir() -> str:
  function get_int_from_env (line 113) | def get_int_from_env(env_keys, default):
  function get_xla_device_type (line 122) | def get_xla_device_type(device: "torch.device") -> Optional[str]:
  class OptimizerNames (line 131) | class OptimizerNames(ExplicitEnum):
  class TrainingArguments (line 156) | class TrainingArguments:
    method __post_init__ (line 1168) | def __post_init__(self):
    method __str__ (line 1616) | def __str__(self):
    method train_batch_size (line 1632) | def train_batch_size(self) -> int:
    method eval_batch_size (line 1646) | def eval_batch_size(self) -> int:
    method ddp_timeout_delta (line 1660) | def ddp_timeout_delta(self) -> timedelta:
    method _setup_devices (line 1667) | def _setup_devices(self) -> "torch.device":
    method device (line 1759) | def device(self) -> "torch.device":
    method n_gpu (line 1767) | def n_gpu(self):
    method parallel_mode (line 1782) | def parallel_mode(self):
    method world_size (line 1809) | def world_size(self):
    method process_index (line 1826) | def process_index(self):
    method local_process_index (line 1842) | def local_process_index(self):
    method should_log (line 1858) | def should_log(self):
    method should_save (line 1871) | def should_save(self):
    method get_process_log_level (line 1883) | def get_process_log_level(self):
    method place_model_on_device (line 1906) | def place_model_on_device(self):
    method _no_sync_in_gradient_accumulation (line 1913) | def _no_sync_in_gradient_accumulation(self):
    method main_process_first (line 1922) | def main_process_first(self, local=True, desc="work"):
    method get_warmup_steps (line 1976) | def get_warmup_steps(self, num_training_steps: int):
    method to_dict (line 1985) | def to_dict(self):
    method to_json_string (line 2002) | def to_json_string(self):
    method to_sanitized_dict (line 2008) | def to_sanitized_dict(self) -> Dict[str, Any]:
    method set_training (line 2022) | def set_training(
    method set_evaluate (line 2097) | def set_evaluate(
    method set_testing (line 2158) | def set_testing(
    method set_save (line 2198) | def set_save(
    method set_logging (line 2247) | def set_logging(
    method set_push_to_hub (line 2321) | def set_push_to_hub(
    method set_optimizer (line 2386) | def set_optimizer(
    method set_lr_scheduler (line 2437) | def set_lr_scheduler(
    method set_dataloader (line 2482) | def set_dataloader(
  class ParallelMode (line 2540) | class ParallelMode(Enum):

FILE: transformers/training_args_seq2seq.py
  class Seq2SeqTrainingArguments (line 30) | class Seq2SeqTrainingArguments(TrainingArguments):
    method to_dict (line 87) | def to_dict(self):

FILE: transformers/training_args_tf.py
  class TFTrainingArguments (line 30) | class TFTrainingArguments(TrainingArguments):
    method _setup_strategy (line 188) | def _setup_strategy(self) -> Tuple["tf.distribute.Strategy", int]:
    method strategy (line 237) | def strategy(self) -> "tf.distribute.Strategy":
    method n_replicas (line 245) | def n_replicas(self) -> int:
    method should_log (line 253) | def should_log(self):
    method train_batch_size (line 260) | def train_batch_size(self) -> int:
    method eval_batch_size (line 273) | def eval_batch_size(self) -> int:
    method n_gpu (line 286) | def n_gpu(self) -> int:

FILE: transformers/utils/__init__.py
  function check_min_version (line 205) | def check_min_version(min_version):

FILE: transformers/utils/backbone_utils.py
  class BackboneType (line 23) | class BackboneType(enum.Enum):
  function verify_out_features_out_indices (line 28) | def verify_out_features_out_indices(
  function _align_output_features_output_indices (line 56) | def _align_output_features_output_indices(
  function get_aligned_output_features_output_indices (line 87) | def get_aligned_output_features_output_indices(
  class BackboneMixin (line 118) | class BackboneMixin:
    method _init_timm_backbone (line 121) | def _init_timm_backbone(self, config) -> None:
    method _init_transformers_backbone (line 142) | def _init_transformers_backbone(self, config) -> None:
    method _init_backbone (line 154) | def _init_backbone(self, config) -> None:
    method out_features (line 172) | def out_features(self):
    method out_features (line 176) | def out_features(self, out_features: List[str]):
    method out_indices (line 185) | def out_indices(self):
    method out_indices (line 189) | def out_indices(self, out_indices: Union[Tuple[int], List[int]]):
    method out_feature_channels (line 198) | def out_feature_channels(self):
    method channels (line 204) | def channels(self):
    method forward_with_filtered_kwargs (line 207) | def forward_with_filtered_kwargs(self, *args, **kwargs):
    method forward (line 212) | def forward(
    method to_dict (line 221) | def to_dict(self):
  class BackboneConfigMixin (line 232) | class BackboneConfigMixin:
    method out_features (line 238) | def out_features(self):
    method out_features (line 242) | def out_features(self, out_features: List[str]):
    method out_indices (line 251) | def out_indices(self):
    method out_indices (line 255) | def out_indices(self, out_indices: Union[Tuple[int], List[int]]):
    method to_dict (line 263) | def to_dict(self):

FILE: transformers/utils/bitsandbytes.py
  function set_module_quantized_tensor_to_device (line 22) | def set_module_quantized_tensor_to_device(module, tensor_name, device, v...
  function _replace_with_bnb_linear (line 112) | def _replace_with_bnb_linear(
  function replace_with_bnb_linear (line 169) | def replace_with_bnb_linear(model, modules_to_not_convert=None, current_...
  function replace_8bit_linear (line 212) | def replace_8bit_linear(*args, **kwargs):
  function set_module_8bit_tensor_to_device (line 221) | def set_module_8bit_tensor_to_device(*args, **kwargs):
  function get_keys_to_not_convert (line 229) | def get_keys_to_not_convert(model):

FILE: transformers/utils/doc.py
  function add_start_docstrings (line 23) | def add_start_docstrings(*docstr):
  function add_start_docstrings_to_model_forward (line 31) | def add_start_docstrings_to_model_forward(*docstr):
  function add_end_docstrings (line 53) | def add_end_docstrings(*docstr):
  function _get_indent (line 79) | def _get_indent(t):
  function _convert_output_args_doc (line 85) | def _convert_output_args_doc(output_args_doc):
  function _prepare_output_docstrings (line 111) | def _prepare_output_docstrings(output_type, config_class, min_indent=None):
  function filter_outputs_from_example (line 1045) | def filter_outputs_from_example(docstring, **kwargs):
  function add_code_sample_docstrings (line 1059) | def add_code_sample_docstrings(
  function replace_return_docstrings (line 1147) | def replace_return_docstrings(output_type=None, config_class=None):
  function copy_func (line 1169) | def copy_func(f):

FILE: transformers/utils/dummy_detectron2_objects.py
  class LayoutLMv2Model (line 8) | class LayoutLMv2Model:
    method __init__ (line 9) | def __init__(self, *args, **kwargs):
    method from_pretrained (line 13) | def from_pretrained(cls, *args, **kwargs):

FILE: transformers/utils/dummy_flax_objects.py
  class FlaxForcedBOSTokenLogitsProcessor (line 5) | class FlaxForcedBOSTokenLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 8) | def __init__(self, *args, **kwargs):
  class FlaxForcedEOSTokenLogitsProcessor (line 12) | class FlaxForcedEOSTokenLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 15) | def __init__(self, *args, **kwargs):
  class FlaxGenerationMixin (line 19) | class FlaxGenerationMixin(metaclass=DummyObject):
    method __init__ (line 22) | def __init__(self, *args, **kwargs):
  class FlaxLogitsProcessor (line 26) | class FlaxLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 29) | def __init__(self, *args, **kwargs):
  class FlaxLogitsProcessorList (line 33) | class FlaxLogitsProcessorList(metaclass=DummyObject):
    method __init__ (line 36) | def __init__(self, *args, **kwargs):
  class FlaxLogitsWarper (line 40) | class FlaxLogitsWarper(metaclass=DummyObject):
    method __init__ (line 43) | def __init__(self, *args, **kwargs):
  class FlaxMinLengthLogitsProcessor (line 47) | class FlaxMinLengthLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 50) | def __init__(self, *args, **kwargs):
  class FlaxTemperatureLogitsWarper (line 54) | class FlaxTemperatureLogitsWarper(metaclass=DummyObject):
    method __init__ (line 57) | def __init__(self, *args, **kwargs):
  class FlaxTopKLogitsWarper (line 61) | class FlaxTopKLogitsWarper(metaclass=DummyObject):
    method __init__ (line 64) | def __init__(self, *args, **kwargs):
  class FlaxTopPLogitsWarper (line 68) | class FlaxTopPLogitsWarper(metaclass=DummyObject):
    method __init__ (line 71) | def __init__(self, *args, **kwargs):
  class FlaxPreTrainedModel (line 75) | class FlaxPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 78) | def __init__(self, *args, **kwargs):
  class FlaxAlbertForMaskedLM (line 82) | class FlaxAlbertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 85) | def __init__(self, *args, **kwargs):
  class FlaxAlbertForMultipleChoice (line 89) | class FlaxAlbertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 92) | def __init__(self, *args, **kwargs):
  class FlaxAlbertForPreTraining (line 96) | class FlaxAlbertForPreTraining(metaclass=DummyObject):
    method __init__ (line 99) | def __init__(self, *args, **kwargs):
  class FlaxAlbertForQuestionAnswering (line 103) | class FlaxAlbertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 106) | def __init__(self, *args, **kwargs):
  class FlaxAlbertForSequenceClassification (line 110) | class FlaxAlbertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 113) | def __init__(self, *args, **kwargs):
  class FlaxAlbertForTokenClassification (line 117) | class FlaxAlbertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 120) | def __init__(self, *args, **kwargs):
  class FlaxAlbertModel (line 124) | class FlaxAlbertModel(metaclass=DummyObject):
    method __init__ (line 127) | def __init__(self, *args, **kwargs):
  class FlaxAlbertPreTrainedModel (line 131) | class FlaxAlbertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 134) | def __init__(self, *args, **kwargs):
  class FlaxAutoModel (line 177) | class FlaxAutoModel(metaclass=DummyObject):
    method __init__ (line 180) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForCausalLM (line 184) | class FlaxAutoModelForCausalLM(metaclass=DummyObject):
    method __init__ (line 187) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForImageClassification (line 191) | class FlaxAutoModelForImageClassification(metaclass=DummyObject):
    method __init__ (line 194) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForMaskedLM (line 198) | class FlaxAutoModelForMaskedLM(metaclass=DummyObject):
    method __init__ (line 201) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForMultipleChoice (line 205) | class FlaxAutoModelForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 208) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForNextSentencePrediction (line 212) | class FlaxAutoModelForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 215) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForPreTraining (line 219) | class FlaxAutoModelForPreTraining(metaclass=DummyObject):
    method __init__ (line 222) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForQuestionAnswering (line 226) | class FlaxAutoModelForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 229) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForSeq2SeqLM (line 233) | class FlaxAutoModelForSeq2SeqLM(metaclass=DummyObject):
    method __init__ (line 236) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForSequenceClassification (line 240) | class FlaxAutoModelForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 243) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForSpeechSeq2Seq (line 247) | class FlaxAutoModelForSpeechSeq2Seq(metaclass=DummyObject):
    method __init__ (line 250) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForTokenClassification (line 254) | class FlaxAutoModelForTokenClassification(metaclass=DummyObject):
    method __init__ (line 257) | def __init__(self, *args, **kwargs):
  class FlaxAutoModelForVision2Seq (line 261) | class FlaxAutoModelForVision2Seq(metaclass=DummyObject):
    method __init__ (line 264) | def __init__(self, *args, **kwargs):
  class FlaxBartDecoderPreTrainedModel (line 268) | class FlaxBartDecoderPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 271) | def __init__(self, *args, **kwargs):
  class FlaxBartForCausalLM (line 275) | class FlaxBartForCausalLM(metaclass=DummyObject):
    method __init__ (line 278) | def __init__(self, *args, **kwargs):
  class FlaxBartForConditionalGeneration (line 282) | class FlaxBartForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 285) | def __init__(self, *args, **kwargs):
  class FlaxBartForQuestionAnswering (line 289) | class FlaxBartForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 292) | def __init__(self, *args, **kwargs):
  class FlaxBartForSequenceClassification (line 296) | class FlaxBartForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 299) | def __init__(self, *args, **kwargs):
  class FlaxBartModel (line 303) | class FlaxBartModel(metaclass=DummyObject):
    method __init__ (line 306) | def __init__(self, *args, **kwargs):
  class FlaxBartPreTrainedModel (line 310) | class FlaxBartPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 313) | def __init__(self, *args, **kwargs):
  class FlaxBeitForImageClassification (line 317) | class FlaxBeitForImageClassification(metaclass=DummyObject):
    method __init__ (line 320) | def __init__(self, *args, **kwargs):
  class FlaxBeitForMaskedImageModeling (line 324) | class FlaxBeitForMaskedImageModeling(metaclass=DummyObject):
    method __init__ (line 327) | def __init__(self, *args, **kwargs):
  class FlaxBeitModel (line 331) | class FlaxBeitModel(metaclass=DummyObject):
    method __init__ (line 334) | def __init__(self, *args, **kwargs):
  class FlaxBeitPreTrainedModel (line 338) | class FlaxBeitPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 341) | def __init__(self, *args, **kwargs):
  class FlaxBertForCausalLM (line 345) | class FlaxBertForCausalLM(metaclass=DummyObject):
    method __init__ (line 348) | def __init__(self, *args, **kwargs):
  class FlaxBertForMaskedLM (line 352) | class FlaxBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 355) | def __init__(self, *args, **kwargs):
  class FlaxBertForMultipleChoice (line 359) | class FlaxBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 362) | def __init__(self, *args, **kwargs):
  class FlaxBertForNextSentencePrediction (line 366) | class FlaxBertForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 369) | def __init__(self, *args, **kwargs):
  class FlaxBertForPreTraining (line 373) | class FlaxBertForPreTraining(metaclass=DummyObject):
    method __init__ (line 376) | def __init__(self, *args, **kwargs):
  class FlaxBertForQuestionAnswering (line 380) | class FlaxBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 383) | def __init__(self, *args, **kwargs):
  class FlaxBertForSequenceClassification (line 387) | class FlaxBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 390) | def __init__(self, *args, **kwargs):
  class FlaxBertForTokenClassification (line 394) | class FlaxBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 397) | def __init__(self, *args, **kwargs):
  class FlaxBertModel (line 401) | class FlaxBertModel(metaclass=DummyObject):
    method __init__ (line 404) | def __init__(self, *args, **kwargs):
  class FlaxBertPreTrainedModel (line 408) | class FlaxBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 411) | def __init__(self, *args, **kwargs):
  class FlaxBigBirdForCausalLM (line 415) | class FlaxBigBirdForCausalLM(metaclass=DummyObject):
    method __init__ (line 418) | def __init__(self, *args, **kwargs):
  class FlaxBigBirdForMaskedLM (line 422) | class FlaxBigBirdForMaskedLM(metaclass=DummyObject):
    method __init__ (line 425) | def __init__(self, *args, **kwargs):
  class FlaxBigBirdForMultipleChoice (line 429) | class FlaxBigBirdForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 432) | def __init__(self, *args, **kwargs):
  class FlaxBigBirdForPreTraining (line 436) | class FlaxBigBirdForPreTraining(metaclass=DummyObject):
    method __init__ (line 439) | def __init__(self, *args, **kwargs):
  class FlaxBigBirdForQuestionAnswering (line 443) | class FlaxBigBirdForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 446) | def __init__(self, *args, **kwargs):
  class FlaxBigBirdForSequenceClassification (line 450) | class FlaxBigBirdForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 453) | def __init__(self, *args, **kwargs):
  class FlaxBigBirdForTokenClassification (line 457) | class FlaxBigBirdForTokenClassification(metaclass=DummyObject):
    method __init__ (line 460) | def __init__(self, *args, **kwargs):
  class FlaxBigBirdModel (line 464) | class FlaxBigBirdModel(metaclass=DummyObject):
    method __init__ (line 467) | def __init__(self, *args, **kwargs):
  class FlaxBigBirdPreTrainedModel (line 471) | class FlaxBigBirdPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 474) | def __init__(self, *args, **kwargs):
  class FlaxBlenderbotForConditionalGeneration (line 478) | class FlaxBlenderbotForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 481) | def __init__(self, *args, **kwargs):
  class FlaxBlenderbotModel (line 485) | class FlaxBlenderbotModel(metaclass=DummyObject):
    method __init__ (line 488) | def __init__(self, *args, **kwargs):
  class FlaxBlenderbotPreTrainedModel (line 492) | class FlaxBlenderbotPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 495) | def __init__(self, *args, **kwargs):
  class FlaxBlenderbotSmallForConditionalGeneration (line 499) | class FlaxBlenderbotSmallForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 502) | def __init__(self, *args, **kwargs):
  class FlaxBlenderbotSmallModel (line 506) | class FlaxBlenderbotSmallModel(metaclass=DummyObject):
    method __init__ (line 509) | def __init__(self, *args, **kwargs):
  class FlaxBlenderbotSmallPreTrainedModel (line 513) | class FlaxBlenderbotSmallPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 516) | def __init__(self, *args, **kwargs):
  class FlaxCLIPModel (line 520) | class FlaxCLIPModel(metaclass=DummyObject):
    method __init__ (line 523) | def __init__(self, *args, **kwargs):
  class FlaxCLIPPreTrainedModel (line 527) | class FlaxCLIPPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 530) | def __init__(self, *args, **kwargs):
  class FlaxCLIPTextModel (line 534) | class FlaxCLIPTextModel(metaclass=DummyObject):
    method __init__ (line 537) | def __init__(self, *args, **kwargs):
  class FlaxCLIPTextPreTrainedModel (line 541) | class FlaxCLIPTextPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 544) | def __init__(self, *args, **kwargs):
  class FlaxCLIPVisionModel (line 548) | class FlaxCLIPVisionModel(metaclass=DummyObject):
    method __init__ (line 551) | def __init__(self, *args, **kwargs):
  class FlaxCLIPVisionPreTrainedModel (line 555) | class FlaxCLIPVisionPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 558) | def __init__(self, *args, **kwargs):
  class FlaxDistilBertForMaskedLM (line 562) | class FlaxDistilBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 565) | def __init__(self, *args, **kwargs):
  class FlaxDistilBertForMultipleChoice (line 569) | class FlaxDistilBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 572) | def __init__(self, *args, **kwargs):
  class FlaxDistilBertForQuestionAnswering (line 576) | class FlaxDistilBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 579) | def __init__(self, *args, **kwargs):
  class FlaxDistilBertForSequenceClassification (line 583) | class FlaxDistilBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 586) | def __init__(self, *args, **kwargs):
  class FlaxDistilBertForTokenClassification (line 590) | class FlaxDistilBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 593) | def __init__(self, *args, **kwargs):
  class FlaxDistilBertModel (line 597) | class FlaxDistilBertModel(metaclass=DummyObject):
    method __init__ (line 600) | def __init__(self, *args, **kwargs):
  class FlaxDistilBertPreTrainedModel (line 604) | class FlaxDistilBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 607) | def __init__(self, *args, **kwargs):
  class FlaxElectraForCausalLM (line 611) | class FlaxElectraForCausalLM(metaclass=DummyObject):
    method __init__ (line 614) | def __init__(self, *args, **kwargs):
  class FlaxElectraForMaskedLM (line 618) | class FlaxElectraForMaskedLM(metaclass=DummyObject):
    method __init__ (line 621) | def __init__(self, *args, **kwargs):
  class FlaxElectraForMultipleChoice (line 625) | class FlaxElectraForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 628) | def __init__(self, *args, **kwargs):
  class FlaxElectraForPreTraining (line 632) | class FlaxElectraForPreTraining(metaclass=DummyObject):
    method __init__ (line 635) | def __init__(self, *args, **kwargs):
  class FlaxElectraForQuestionAnswering (line 639) | class FlaxElectraForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 642) | def __init__(self, *args, **kwargs):
  class FlaxElectraForSequenceClassification (line 646) | class FlaxElectraForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 649) | def __init__(self, *args, **kwargs):
  class FlaxElectraForTokenClassification (line 653) | class FlaxElectraForTokenClassification(metaclass=DummyObject):
    method __init__ (line 656) | def __init__(self, *args, **kwargs):
  class FlaxElectraModel (line 660) | class FlaxElectraModel(metaclass=DummyObject):
    method __init__ (line 663) | def __init__(self, *args, **kwargs):
  class FlaxElectraPreTrainedModel (line 667) | class FlaxElectraPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 670) | def __init__(self, *args, **kwargs):
  class FlaxEncoderDecoderModel (line 674) | class FlaxEncoderDecoderModel(metaclass=DummyObject):
    method __init__ (line 677) | def __init__(self, *args, **kwargs):
  class FlaxGPT2LMHeadModel (line 681) | class FlaxGPT2LMHeadModel(metaclass=DummyObject):
    method __init__ (line 684) | def __init__(self, *args, **kwargs):
  class FlaxGPT2Model (line 688) | class FlaxGPT2Model(metaclass=DummyObject):
    method __init__ (line 691) | def __init__(self, *args, **kwargs):
  class FlaxGPT2PreTrainedModel (line 695) | class FlaxGPT2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 698) | def __init__(self, *args, **kwargs):
  class FlaxGPTNeoForCausalLM (line 702) | class FlaxGPTNeoForCausalLM(metaclass=DummyObject):
    method __init__ (line 705) | def __init__(self, *args, **kwargs):
  class FlaxGPTNeoModel (line 709) | class FlaxGPTNeoModel(metaclass=DummyObject):
    method __init__ (line 712) | def __init__(self, *args, **kwargs):
  class FlaxGPTNeoPreTrainedModel (line 716) | class FlaxGPTNeoPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 719) | def __init__(self, *args, **kwargs):
  class FlaxGPTJForCausalLM (line 723) | class FlaxGPTJForCausalLM(metaclass=DummyObject):
    method __init__ (line 726) | def __init__(self, *args, **kwargs):
  class FlaxGPTJModel (line 730) | class FlaxGPTJModel(metaclass=DummyObject):
    method __init__ (line 733) | def __init__(self, *args, **kwargs):
  class FlaxGPTJPreTrainedModel (line 737) | class FlaxGPTJPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 740) | def __init__(self, *args, **kwargs):
  class FlaxLongT5ForConditionalGeneration (line 744) | class FlaxLongT5ForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 747) | def __init__(self, *args, **kwargs):
  class FlaxLongT5Model (line 751) | class FlaxLongT5Model(metaclass=DummyObject):
    method __init__ (line 754) | def __init__(self, *args, **kwargs):
  class FlaxLongT5PreTrainedModel (line 758) | class FlaxLongT5PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 761) | def __init__(self, *args, **kwargs):
  class FlaxMarianModel (line 765) | class FlaxMarianModel(metaclass=DummyObject):
    method __init__ (line 768) | def __init__(self, *args, **kwargs):
  class FlaxMarianMTModel (line 772) | class FlaxMarianMTModel(metaclass=DummyObject):
    method __init__ (line 775) | def __init__(self, *args, **kwargs):
  class FlaxMarianPreTrainedModel (line 779) | class FlaxMarianPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 782) | def __init__(self, *args, **kwargs):
  class FlaxMBartForConditionalGeneration (line 786) | class FlaxMBartForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 789) | def __init__(self, *args, **kwargs):
  class FlaxMBartForQuestionAnswering (line 793) | class FlaxMBartForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 796) | def __init__(self, *args, **kwargs):
  class FlaxMBartForSequenceClassification (line 800) | class FlaxMBartForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 803) | def __init__(self, *args, **kwargs):
  class FlaxMBartModel (line 807) | class FlaxMBartModel(metaclass=DummyObject):
    method __init__ (line 810) | def __init__(self, *args, **kwargs):
  class FlaxMBartPreTrainedModel (line 814) | class FlaxMBartPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 817) | def __init__(self, *args, **kwargs):
  class FlaxMT5EncoderModel (line 821) | class FlaxMT5EncoderModel(metaclass=DummyObject):
    method __init__ (line 824) | def __init__(self, *args, **kwargs):
  class FlaxMT5ForConditionalGeneration (line 828) | class FlaxMT5ForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 831) | def __init__(self, *args, **kwargs):
  class FlaxMT5Model (line 835) | class FlaxMT5Model(metaclass=DummyObject):
    method __init__ (line 838) | def __init__(self, *args, **kwargs):
  class FlaxOPTForCausalLM (line 842) | class FlaxOPTForCausalLM(metaclass=DummyObject):
    method __init__ (line 845) | def __init__(self, *args, **kwargs):
  class FlaxOPTModel (line 849) | class FlaxOPTModel(metaclass=DummyObject):
    method __init__ (line 852) | def __init__(self, *args, **kwargs):
  class FlaxOPTPreTrainedModel (line 856) | class FlaxOPTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 859) | def __init__(self, *args, **kwargs):
  class FlaxPegasusForConditionalGeneration (line 863) | class FlaxPegasusForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 866) | def __init__(self, *args, **kwargs):
  class FlaxPegasusModel (line 870) | class FlaxPegasusModel(metaclass=DummyObject):
    method __init__ (line 873) | def __init__(self, *args, **kwargs):
  class FlaxPegasusPreTrainedModel (line 877) | class FlaxPegasusPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 880) | def __init__(self, *args, **kwargs):
  class FlaxRegNetForImageClassification (line 884) | class FlaxRegNetForImageClassification(metaclass=DummyObject):
    method __init__ (line 887) | def __init__(self, *args, **kwargs):
  class FlaxRegNetModel (line 891) | class FlaxRegNetModel(metaclass=DummyObject):
    method __init__ (line 894) | def __init__(self, *args, **kwargs):
  class FlaxRegNetPreTrainedModel (line 898) | class FlaxRegNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 901) | def __init__(self, *args, **kwargs):
  class FlaxResNetForImageClassification (line 905) | class FlaxResNetForImageClassification(metaclass=DummyObject):
    method __init__ (line 908) | def __init__(self, *args, **kwargs):
  class FlaxResNetModel (line 912) | class FlaxResNetModel(metaclass=DummyObject):
    method __init__ (line 915) | def __init__(self, *args, **kwargs):
  class FlaxResNetPreTrainedModel (line 919) | class FlaxResNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 922) | def __init__(self, *args, **kwargs):
  class FlaxRobertaForCausalLM (line 926) | class FlaxRobertaForCausalLM(metaclass=DummyObject):
    method __init__ (line 929) | def __init__(self, *args, **kwargs):
  class FlaxRobertaForMaskedLM (line 933) | class FlaxRobertaForMaskedLM(metaclass=DummyObject):
    method __init__ (line 936) | def __init__(self, *args, **kwargs):
  class FlaxRobertaForMultipleChoice (line 940) | class FlaxRobertaForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 943) | def __init__(self, *args, **kwargs):
  class FlaxRobertaForQuestionAnswering (line 947) | class FlaxRobertaForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 950) | def __init__(self, *args, **kwargs):
  class FlaxRobertaForSequenceClassification (line 954) | class FlaxRobertaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 957) | def __init__(self, *args, **kwargs):
  class FlaxRobertaForTokenClassification (line 961) | class FlaxRobertaForTokenClassification(metaclass=DummyObject):
    method __init__ (line 964) | def __init__(self, *args, **kwargs):
  class FlaxRobertaModel (line 968) | class FlaxRobertaModel(metaclass=DummyObject):
    method __init__ (line 971) | def __init__(self, *args, **kwargs):
  class FlaxRobertaPreTrainedModel (line 975) | class FlaxRobertaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 978) | def __init__(self, *args, **kwargs):
  class FlaxRobertaPreLayerNormForCausalLM (line 982) | class FlaxRobertaPreLayerNormForCausalLM(metaclass=DummyObject):
    method __init__ (line 985) | def __init__(self, *args, **kwargs):
  class FlaxRobertaPreLayerNormForMaskedLM (line 989) | class FlaxRobertaPreLayerNormForMaskedLM(metaclass=DummyObject):
    method __init__ (line 992) | def __init__(self, *args, **kwargs):
  class FlaxRobertaPreLayerNormForMultipleChoice (line 996) | class FlaxRobertaPreLayerNormForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 999) | def __init__(self, *args, **kwargs):
  class FlaxRobertaPreLayerNormForQuestionAnswering (line 1003) | class FlaxRobertaPreLayerNormForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1006) | def __init__(self, *args, **kwargs):
  class FlaxRobertaPreLayerNormForSequenceClassification (line 1010) | class FlaxRobertaPreLayerNormForSequenceClassification(metaclass=DummyOb...
    method __init__ (line 1013) | def __init__(self, *args, **kwargs):
  class FlaxRobertaPreLayerNormForTokenClassification (line 1017) | class FlaxRobertaPreLayerNormForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1020) | def __init__(self, *args, **kwargs):
  class FlaxRobertaPreLayerNormModel (line 1024) | class FlaxRobertaPreLayerNormModel(metaclass=DummyObject):
    method __init__ (line 1027) | def __init__(self, *args, **kwargs):
  class FlaxRobertaPreLayerNormPreTrainedModel (line 1031) | class FlaxRobertaPreLayerNormPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1034) | def __init__(self, *args, **kwargs):
  class FlaxRoFormerForMaskedLM (line 1038) | class FlaxRoFormerForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1041) | def __init__(self, *args, **kwargs):
  class FlaxRoFormerForMultipleChoice (line 1045) | class FlaxRoFormerForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1048) | def __init__(self, *args, **kwargs):
  class FlaxRoFormerForQuestionAnswering (line 1052) | class FlaxRoFormerForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1055) | def __init__(self, *args, **kwargs):
  class FlaxRoFormerForSequenceClassification (line 1059) | class FlaxRoFormerForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1062) | def __init__(self, *args, **kwargs):
  class FlaxRoFormerForTokenClassification (line 1066) | class FlaxRoFormerForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1069) | def __init__(self, *args, **kwargs):
  class FlaxRoFormerModel (line 1073) | class FlaxRoFormerModel(metaclass=DummyObject):
    method __init__ (line 1076) | def __init__(self, *args, **kwargs):
  class FlaxRoFormerPreTrainedModel (line 1080) | class FlaxRoFormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1083) | def __init__(self, *args, **kwargs):
  class FlaxSpeechEncoderDecoderModel (line 1087) | class FlaxSpeechEncoderDecoderModel(metaclass=DummyObject):
    method __init__ (line 1090) | def __init__(self, *args, **kwargs):
  class FlaxT5EncoderModel (line 1094) | class FlaxT5EncoderModel(metaclass=DummyObject):
    method __init__ (line 1097) | def __init__(self, *args, **kwargs):
  class FlaxT5ForConditionalGeneration (line 1101) | class FlaxT5ForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1104) | def __init__(self, *args, **kwargs):
  class FlaxT5Model (line 1108) | class FlaxT5Model(metaclass=DummyObject):
    method __init__ (line 1111) | def __init__(self, *args, **kwargs):
  class FlaxT5PreTrainedModel (line 1115) | class FlaxT5PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1118) | def __init__(self, *args, **kwargs):
  class FlaxVisionEncoderDecoderModel (line 1122) | class FlaxVisionEncoderDecoderModel(metaclass=DummyObject):
    method __init__ (line 1125) | def __init__(self, *args, **kwargs):
  class FlaxVisionTextDualEncoderModel (line 1129) | class FlaxVisionTextDualEncoderModel(metaclass=DummyObject):
    method __init__ (line 1132) | def __init__(self, *args, **kwargs):
  class FlaxViTForImageClassification (line 1136) | class FlaxViTForImageClassification(metaclass=DummyObject):
    method __init__ (line 1139) | def __init__(self, *args, **kwargs):
  class FlaxViTModel (line 1143) | class FlaxViTModel(metaclass=DummyObject):
    method __init__ (line 1146) | def __init__(self, *args, **kwargs):
  class FlaxViTPreTrainedModel (line 1150) | class FlaxViTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1153) | def __init__(self, *args, **kwargs):
  class FlaxWav2Vec2ForCTC (line 1157) | class FlaxWav2Vec2ForCTC(metaclass=DummyObject):
    method __init__ (line 1160) | def __init__(self, *args, **kwargs):
  class FlaxWav2Vec2ForPreTraining (line 1164) | class FlaxWav2Vec2ForPreTraining(metaclass=DummyObject):
    method __init__ (line 1167) | def __init__(self, *args, **kwargs):
  class FlaxWav2Vec2Model (line 1171) | class FlaxWav2Vec2Model(metaclass=DummyObject):
    method __init__ (line 1174) | def __init__(self, *args, **kwargs):
  class FlaxWav2Vec2PreTrainedModel (line 1178) | class FlaxWav2Vec2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1181) | def __init__(self, *args, **kwargs):
  class FlaxWhisperForAudioClassification (line 1185) | class FlaxWhisperForAudioClassification(metaclass=DummyObject):
    method __init__ (line 1188) | def __init__(self, *args, **kwargs):
  class FlaxWhisperForConditionalGeneration (line 1192) | class FlaxWhisperForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1195) | def __init__(self, *args, **kwargs):
  class FlaxWhisperModel (line 1199) | class FlaxWhisperModel(metaclass=DummyObject):
    method __init__ (line 1202) | def __init__(self, *args, **kwargs):
  class FlaxWhisperPreTrainedModel (line 1206) | class FlaxWhisperPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1209) | def __init__(self, *args, **kwargs):
  class FlaxXGLMForCausalLM (line 1213) | class FlaxXGLMForCausalLM(metaclass=DummyObject):
    method __init__ (line 1216) | def __init__(self, *args, **kwargs):
  class FlaxXGLMModel (line 1220) | class FlaxXGLMModel(metaclass=DummyObject):
    method __init__ (line 1223) | def __init__(self, *args, **kwargs):
  class FlaxXGLMPreTrainedModel (line 1227) | class FlaxXGLMPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1230) | def __init__(self, *args, **kwargs):
  class FlaxXLMRobertaForCausalLM (line 1237) | class FlaxXLMRobertaForCausalLM(metaclass=DummyObject):
    method __init__ (line 1240) | def __init__(self, *args, **kwargs):
  class FlaxXLMRobertaForMaskedLM (line 1244) | class FlaxXLMRobertaForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1247) | def __init__(self, *args, **kwargs):
  class FlaxXLMRobertaForMultipleChoice (line 1251) | class FlaxXLMRobertaForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1254) | def __init__(self, *args, **kwargs):
  class FlaxXLMRobertaForQuestionAnswering (line 1258) | class FlaxXLMRobertaForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1261) | def __init__(self, *args, **kwargs):
  class FlaxXLMRobertaForSequenceClassification (line 1265) | class FlaxXLMRobertaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1268) | def __init__(self, *args, **kwargs):
  class FlaxXLMRobertaForTokenClassification (line 1272) | class FlaxXLMRobertaForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1275) | def __init__(self, *args, **kwargs):
  class FlaxXLMRobertaModel (line 1279) | class FlaxXLMRobertaModel(metaclass=DummyObject):
    method __init__ (line 1282) | def __init__(self, *args, **kwargs):
  class FlaxXLMRobertaPreTrainedModel (line 1286) | class FlaxXLMRobertaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1289) | def __init__(self, *args, **kwargs):

FILE: transformers/utils/dummy_keras_nlp_objects.py
  class TFGPT2Tokenizer (line 5) | class TFGPT2Tokenizer(metaclass=DummyObject):
    method __init__ (line 8) | def __init__(self, *args, **kwargs):

FILE: transformers/utils/dummy_pt_objects.py
  class PyTorchBenchmark (line 5) | class PyTorchBenchmark(metaclass=DummyObject):
    method __init__ (line 8) | def __init__(self, *args, **kwargs):
  class PyTorchBenchmarkArguments (line 12) | class PyTorchBenchmarkArguments(metaclass=DummyObject):
    method __init__ (line 15) | def __init__(self, *args, **kwargs):
  class GlueDataset (line 19) | class GlueDataset(metaclass=DummyObject):
    method __init__ (line 22) | def __init__(self, *args, **kwargs):
  class GlueDataTrainingArguments (line 26) | class GlueDataTrainingArguments(metaclass=DummyObject):
    method __init__ (line 29) | def __init__(self, *args, **kwargs):
  class LineByLineTextDataset (line 33) | class LineByLineTextDataset(metaclass=DummyObject):
    method __init__ (line 36) | def __init__(self, *args, **kwargs):
  class LineByLineWithRefDataset (line 40) | class LineByLineWithRefDataset(metaclass=DummyObject):
    method __init__ (line 43) | def __init__(self, *args, **kwargs):
  class LineByLineWithSOPTextDataset (line 47) | class LineByLineWithSOPTextDataset(metaclass=DummyObject):
    method __init__ (line 50) | def __init__(self, *args, **kwargs):
  class SquadDataset (line 54) | class SquadDataset(metaclass=DummyObject):
    method __init__ (line 57) | def __init__(self, *args, **kwargs):
  class SquadDataTrainingArguments (line 61) | class SquadDataTrainingArguments(metaclass=DummyObject):
    method __init__ (line 64) | def __init__(self, *args, **kwargs):
  class TextDataset (line 68) | class TextDataset(metaclass=DummyObject):
    method __init__ (line 71) | def __init__(self, *args, **kwargs):
  class TextDatasetForNextSentencePrediction (line 75) | class TextDatasetForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 78) | def __init__(self, *args, **kwargs):
  class BeamScorer (line 82) | class BeamScorer(metaclass=DummyObject):
    method __init__ (line 85) | def __init__(self, *args, **kwargs):
  class BeamSearchScorer (line 89) | class BeamSearchScorer(metaclass=DummyObject):
    method __init__ (line 92) | def __init__(self, *args, **kwargs):
  class ConstrainedBeamSearchScorer (line 96) | class ConstrainedBeamSearchScorer(metaclass=DummyObject):
    method __init__ (line 99) | def __init__(self, *args, **kwargs):
  class Constraint (line 103) | class Constraint(metaclass=DummyObject):
    method __init__ (line 106) | def __init__(self, *args, **kwargs):
  class ConstraintListState (line 110) | class ConstraintListState(metaclass=DummyObject):
    method __init__ (line 113) | def __init__(self, *args, **kwargs):
  class DisjunctiveConstraint (line 117) | class DisjunctiveConstraint(metaclass=DummyObject):
    method __init__ (line 120) | def __init__(self, *args, **kwargs):
  class ForcedBOSTokenLogitsProcessor (line 124) | class ForcedBOSTokenLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 127) | def __init__(self, *args, **kwargs):
  class ForcedEOSTokenLogitsProcessor (line 131) | class ForcedEOSTokenLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 134) | def __init__(self, *args, **kwargs):
  class GenerationMixin (line 138) | class GenerationMixin(metaclass=DummyObject):
    method __init__ (line 141) | def __init__(self, *args, **kwargs):
  class HammingDiversityLogitsProcessor (line 145) | class HammingDiversityLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 148) | def __init__(self, *args, **kwargs):
  class InfNanRemoveLogitsProcessor (line 152) | class InfNanRemoveLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 155) | def __init__(self, *args, **kwargs):
  class LogitsProcessor (line 159) | class LogitsProcessor(metaclass=DummyObject):
    method __init__ (line 162) | def __init__(self, *args, **kwargs):
  class LogitsProcessorList (line 166) | class LogitsProcessorList(metaclass=DummyObject):
    method __init__ (line 169) | def __init__(self, *args, **kwargs):
  class LogitsWarper (line 173) | class LogitsWarper(metaclass=DummyObject):
    method __init__ (line 176) | def __init__(self, *args, **kwargs):
  class MaxLengthCriteria (line 180) | class MaxLengthCriteria(metaclass=DummyObject):
    method __init__ (line 183) | def __init__(self, *args, **kwargs):
  class MaxTimeCriteria (line 187) | class MaxTimeCriteria(metaclass=DummyObject):
    method __init__ (line 190) | def __init__(self, *args, **kwargs):
  class MinLengthLogitsProcessor (line 194) | class MinLengthLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 197) | def __init__(self, *args, **kwargs):
  class MinNewTokensLengthLogitsProcessor (line 201) | class MinNewTokensLengthLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 204) | def __init__(self, *args, **kwargs):
  class NoBadWordsLogitsProcessor (line 208) | class NoBadWordsLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 211) | def __init__(self, *args, **kwargs):
  class NoRepeatNGramLogitsProcessor (line 215) | class NoRepeatNGramLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 218) | def __init__(self, *args, **kwargs):
  class PhrasalConstraint (line 222) | class PhrasalConstraint(metaclass=DummyObject):
    method __init__ (line 225) | def __init__(self, *args, **kwargs):
  class PrefixConstrainedLogitsProcessor (line 229) | class PrefixConstrainedLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 232) | def __init__(self, *args, **kwargs):
  class RepetitionPenaltyLogitsProcessor (line 236) | class RepetitionPenaltyLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 239) | def __init__(self, *args, **kwargs):
  class StoppingCriteria (line 243) | class StoppingCriteria(metaclass=DummyObject):
    method __init__ (line 246) | def __init__(self, *args, **kwargs):
  class StoppingCriteriaList (line 250) | class StoppingCriteriaList(metaclass=DummyObject):
    method __init__ (line 253) | def __init__(self, *args, **kwargs):
  class TemperatureLogitsWarper (line 257) | class TemperatureLogitsWarper(metaclass=DummyObject):
    method __init__ (line 260) | def __init__(self, *args, **kwargs):
  class TopKLogitsWarper (line 264) | class TopKLogitsWarper(metaclass=DummyObject):
    method __init__ (line 267) | def __init__(self, *args, **kwargs):
  class TopPLogitsWarper (line 271) | class TopPLogitsWarper(metaclass=DummyObject):
    method __init__ (line 274) | def __init__(self, *args, **kwargs):
  class TypicalLogitsWarper (line 278) | class TypicalLogitsWarper(metaclass=DummyObject):
    method __init__ (line 281) | def __init__(self, *args, **kwargs):
  function top_k_top_p_filtering (line 285) | def top_k_top_p_filtering(*args, **kwargs):
  class PreTrainedModel (line 289) | class PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 292) | def __init__(self, *args, **kwargs):
  class AlbertForMaskedLM (line 299) | class AlbertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 302) | def __init__(self, *args, **kwargs):
  class AlbertForMultipleChoice (line 306) | class AlbertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 309) | def __init__(self, *args, **kwargs):
  class AlbertForPreTraining (line 313) | class AlbertForPreTraining(metaclass=DummyObject):
    method __init__ (line 316) | def __init__(self, *args, **kwargs):
  class AlbertForQuestionAnswering (line 320) | class AlbertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 323) | def __init__(self, *args, **kwargs):
  class AlbertForSequenceClassification (line 327) | class AlbertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 330) | def __init__(self, *args, **kwargs):
  class AlbertForTokenClassification (line 334) | class AlbertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 337) | def __init__(self, *args, **kwargs):
  class AlbertModel (line 341) | class AlbertModel(metaclass=DummyObject):
    method __init__ (line 344) | def __init__(self, *args, **kwargs):
  class AlbertPreTrainedModel (line 348) | class AlbertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 351) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_albert (line 355) | def load_tf_weights_in_albert(*args, **kwargs):
  class AlignModel (line 362) | class AlignModel(metaclass=DummyObject):
    method __init__ (line 365) | def __init__(self, *args, **kwargs):
  class AlignPreTrainedModel (line 369) | class AlignPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 372) | def __init__(self, *args, **kwargs):
  class AlignTextModel (line 376) | class AlignTextModel(metaclass=DummyObject):
    method __init__ (line 379) | def __init__(self, *args, **kwargs):
  class AlignVisionModel (line 383) | class AlignVisionModel(metaclass=DummyObject):
    method __init__ (line 386) | def __init__(self, *args, **kwargs):
  class AltCLIPModel (line 393) | class AltCLIPModel(metaclass=DummyObject):
    method __init__ (line 396) | def __init__(self, *args, **kwargs):
  class AltCLIPPreTrainedModel (line 400) | class AltCLIPPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 403) | def __init__(self, *args, **kwargs):
  class AltCLIPTextModel (line 407) | class AltCLIPTextModel(metaclass=DummyObject):
    method __init__ (line 410) | def __init__(self, *args, **kwargs):
  class AltCLIPVisionModel (line 414) | class AltCLIPVisionModel(metaclass=DummyObject):
    method __init__ (line 417) | def __init__(self, *args, **kwargs):
  class ASTForAudioClassification (line 424) | class ASTForAudioClassification(metaclass=DummyObject):
    method __init__ (line 427) | def __init__(self, *args, **kwargs):
  class ASTModel (line 431) | class ASTModel(metaclass=DummyObject):
    method __init__ (line 434) | def __init__(self, *args, **kwargs):
  class ASTPreTrainedModel (line 438) | class ASTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 441) | def __init__(self, *args, **kwargs):
  class AutoBackbone (line 544) | class AutoBackbone(metaclass=DummyObject):
    method __init__ (line 547) | def __init__(self, *args, **kwargs):
  class AutoModel (line 551) | class AutoModel(metaclass=DummyObject):
    method __init__ (line 554) | def __init__(self, *args, **kwargs):
  class AutoModelForAudioClassification (line 558) | class AutoModelForAudioClassification(metaclass=DummyObject):
    method __init__ (line 561) | def __init__(self, *args, **kwargs):
  class AutoModelForAudioFrameClassification (line 565) | class AutoModelForAudioFrameClassification(metaclass=DummyObject):
    method __init__ (line 568) | def __init__(self, *args, **kwargs):
  class AutoModelForAudioXVector (line 572) | class AutoModelForAudioXVector(metaclass=DummyObject):
    method __init__ (line 575) | def __init__(self, *args, **kwargs):
  class AutoModelForCausalLM (line 579) | class AutoModelForCausalLM(metaclass=DummyObject):
    method __init__ (line 582) | def __init__(self, *args, **kwargs):
  class AutoModelForCTC (line 586) | class AutoModelForCTC(metaclass=DummyObject):
    method __init__ (line 589) | def __init__(self, *args, **kwargs):
  class AutoModelForDepthEstimation (line 593) | class AutoModelForDepthEstimation(metaclass=DummyObject):
    method __init__ (line 596) | def __init__(self, *args, **kwargs):
  class AutoModelForDocumentQuestionAnswering (line 600) | class AutoModelForDocumentQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 603) | def __init__(self, *args, **kwargs):
  class AutoModelForImageClassification (line 607) | class AutoModelForImageClassification(metaclass=DummyObject):
    method __init__ (line 610) | def __init__(self, *args, **kwargs):
  class AutoModelForImageSegmentation (line 614) | class AutoModelForImageSegmentation(metaclass=DummyObject):
    method __init__ (line 617) | def __init__(self, *args, **kwargs):
  class AutoModelForInstanceSegmentation (line 621) | class AutoModelForInstanceSegmentation(metaclass=DummyObject):
    method __init__ (line 624) | def __init__(self, *args, **kwargs):
  class AutoModelForMaskedImageModeling (line 628) | class AutoModelForMaskedImageModeling(metaclass=DummyObject):
    method __init__ (line 631) | def __init__(self, *args, **kwargs):
  class AutoModelForMaskedLM (line 635) | class AutoModelForMaskedLM(metaclass=DummyObject):
    method __init__ (line 638) | def __init__(self, *args, **kwargs):
  class AutoModelForMaskGeneration (line 642) | class AutoModelForMaskGeneration(metaclass=DummyObject):
    method __init__ (line 645) | def __init__(self, *args, **kwargs):
  class AutoModelForMultipleChoice (line 649) | class AutoModelForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 652) | def __init__(self, *args, **kwargs):
  class AutoModelForNextSentencePrediction (line 656) | class AutoModelForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 659) | def __init__(self, *args, **kwargs):
  class AutoModelForObjectDetection (line 663) | class AutoModelForObjectDetection(metaclass=DummyObject):
    method __init__ (line 666) | def __init__(self, *args, **kwargs):
  class AutoModelForPreTraining (line 670) | class AutoModelForPreTraining(metaclass=DummyObject):
    method __init__ (line 673) | def __init__(self, *args, **kwargs):
  class AutoModelForQuestionAnswering (line 677) | class AutoModelForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 680) | def __init__(self, *args, **kwargs):
  class AutoModelForSemanticSegmentation (line 684) | class AutoModelForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 687) | def __init__(self, *args, **kwargs):
  class AutoModelForSeq2SeqLM (line 691) | class AutoModelForSeq2SeqLM(metaclass=DummyObject):
    method __init__ (line 694) | def __init__(self, *args, **kwargs):
  class AutoModelForSequenceClassification (line 698) | class AutoModelForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 701) | def __init__(self, *args, **kwargs):
  class AutoModelForSpeechSeq2Seq (line 705) | class AutoModelForSpeechSeq2Seq(metaclass=DummyObject):
    method __init__ (line 708) | def __init__(self, *args, **kwargs):
  class AutoModelForTableQuestionAnswering (line 712) | class AutoModelForTableQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 715) | def __init__(self, *args, **kwargs):
  class AutoModelForTokenClassification (line 719) | class AutoModelForTokenClassification(metaclass=DummyObject):
    method __init__ (line 722) | def __init__(self, *args, **kwargs):
  class AutoModelForUniversalSegmentation (line 726) | class AutoModelForUniversalSegmentation(metaclass=DummyObject):
    method __init__ (line 729) | def __init__(self, *args, **kwargs):
  class AutoModelForVideoClassification (line 733) | class AutoModelForVideoClassification(metaclass=DummyObject):
    method __init__ (line 736) | def __init__(self, *args, **kwargs):
  class AutoModelForVision2Seq (line 740) | class AutoModelForVision2Seq(metaclass=DummyObject):
    method __init__ (line 743) | def __init__(self, *args, **kwargs):
  class AutoModelForVisualQuestionAnswering (line 747) | class AutoModelForVisualQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 750) | def __init__(self, *args, **kwargs):
  class AutoModelForZeroShotImageClassification (line 754) | class AutoModelForZeroShotImageClassification(metaclass=DummyObject):
    method __init__ (line 757) | def __init__(self, *args, **kwargs):
  class AutoModelForZeroShotObjectDetection (line 761) | class AutoModelForZeroShotObjectDetection(metaclass=DummyObject):
    method __init__ (line 764) | def __init__(self, *args, **kwargs):
  class AutoModelWithLMHead (line 768) | class AutoModelWithLMHead(metaclass=DummyObject):
    method __init__ (line 771) | def __init__(self, *args, **kwargs):
  class AutoformerForPrediction (line 778) | class AutoformerForPrediction(metaclass=DummyObject):
    method __init__ (line 781) | def __init__(self, *args, **kwargs):
  class AutoformerModel (line 785) | class AutoformerModel(metaclass=DummyObject):
    method __init__ (line 788) | def __init__(self, *args, **kwargs):
  class AutoformerPreTrainedModel (line 792) | class AutoformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 795) | def __init__(self, *args, **kwargs):
  class BartForCausalLM (line 802) | class BartForCausalLM(metaclass=DummyObject):
    method __init__ (line 805) | def __init__(self, *args, **kwargs):
  class BartForConditionalGeneration (line 809) | class BartForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 812) | def __init__(self, *args, **kwargs):
  class BartForQuestionAnswering (line 816) | class BartForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 819) | def __init__(self, *args, **kwargs):
  class BartForSequenceClassification (line 823) | class BartForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 826) | def __init__(self, *args, **kwargs):
  class BartModel (line 830) | class BartModel(metaclass=DummyObject):
    method __init__ (line 833) | def __init__(self, *args, **kwargs):
  class BartPretrainedModel (line 837) | class BartPretrainedModel(metaclass=DummyObject):
    method __init__ (line 840) | def __init__(self, *args, **kwargs):
  class PretrainedBartModel (line 844) | class PretrainedBartModel(metaclass=DummyObject):
    method __init__ (line 847) | def __init__(self, *args, **kwargs):
  class BeitForImageClassification (line 854) | class BeitForImageClassification(metaclass=DummyObject):
    method __init__ (line 857) | def __init__(self, *args, **kwargs):
  class BeitForMaskedImageModeling (line 861) | class BeitForMaskedImageModeling(metaclass=DummyObject):
    method __init__ (line 864) | def __init__(self, *args, **kwargs):
  class BeitForSemanticSegmentation (line 868) | class BeitForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 871) | def __init__(self, *args, **kwargs):
  class BeitModel (line 875) | class BeitModel(metaclass=DummyObject):
    method __init__ (line 878) | def __init__(self, *args, **kwargs):
  class BeitPreTrainedModel (line 882) | class BeitPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 885) | def __init__(self, *args, **kwargs):
  class BertForMaskedLM (line 892) | class BertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 895) | def __init__(self, *args, **kwargs):
  class BertForMultipleChoice (line 899) | class BertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 902) | def __init__(self, *args, **kwargs):
  class BertForNextSentencePrediction (line 906) | class BertForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 909) | def __init__(self, *args, **kwargs):
  class BertForPreTraining (line 913) | class BertForPreTraining(metaclass=DummyObject):
    method __init__ (line 916) | def __init__(self, *args, **kwargs):
  class BertForQuestionAnswering (line 920) | class BertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 923) | def __init__(self, *args, **kwargs):
  class BertForSequenceClassification (line 927) | class BertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 930) | def __init__(self, *args, **kwargs):
  class BertForTokenClassification (line 934) | class BertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 937) | def __init__(self, *args, **kwargs):
  class BertLayer (line 941) | class BertLayer(metaclass=DummyObject):
    method __init__ (line 944) | def __init__(self, *args, **kwargs):
  class BertLMHeadModel (line 948) | class BertLMHeadModel(metaclass=DummyObject):
    method __init__ (line 951) | def __init__(self, *args, **kwargs):
  class BertModel (line 955) | class BertModel(metaclass=DummyObject):
    method __init__ (line 958) | def __init__(self, *args, **kwargs):
  class BertPreTrainedModel (line 962) | class BertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 965) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_bert (line 969) | def load_tf_weights_in_bert(*args, **kwargs):
  class BertGenerationDecoder (line 973) | class BertGenerationDecoder(metaclass=DummyObject):
    method __init__ (line 976) | def __init__(self, *args, **kwargs):
  class BertGenerationEncoder (line 980) | class BertGenerationEncoder(metaclass=DummyObject):
    method __init__ (line 983) | def __init__(self, *args, **kwargs):
  class BertGenerationPreTrainedModel (line 987) | class BertGenerationPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 990) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_bert_generation (line 994) | def load_tf_weights_in_bert_generation(*args, **kwargs):
  class BigBirdForCausalLM (line 1001) | class BigBirdForCausalLM(metaclass=DummyObject):
    method __init__ (line 1004) | def __init__(self, *args, **kwargs):
  class BigBirdForMaskedLM (line 1008) | class BigBirdForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1011) | def __init__(self, *args, **kwargs):
  class BigBirdForMultipleChoice (line 1015) | class BigBirdForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1018) | def __init__(self, *args, **kwargs):
  class BigBirdForPreTraining (line 1022) | class BigBirdForPreTraining(metaclass=DummyObject):
    method __init__ (line 1025) | def __init__(self, *args, **kwargs):
  class BigBirdForQuestionAnswering (line 1029) | class BigBirdForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1032) | def __init__(self, *args, **kwargs):
  class BigBirdForSequenceClassification (line 1036) | class BigBirdForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1039) | def __init__(self, *args, **kwargs):
  class BigBirdForTokenClassification (line 1043) | class BigBirdForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1046) | def __init__(self, *args, **kwargs):
  class BigBirdLayer (line 1050) | class BigBirdLayer(metaclass=DummyObject):
    method __init__ (line 1053) | def __init__(self, *args, **kwargs):
  class BigBirdModel (line 1057) | class BigBirdModel(metaclass=DummyObject):
    method __init__ (line 1060) | def __init__(self, *args, **kwargs):
  class BigBirdPreTrainedModel (line 1064) | class BigBirdPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1067) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_big_bird (line 1071) | def load_tf_weights_in_big_bird(*args, **kwargs):
  class BigBirdPegasusForCausalLM (line 1078) | class BigBirdPegasusForCausalLM(metaclass=DummyObject):
    method __init__ (line 1081) | def __init__(self, *args, **kwargs):
  class BigBirdPegasusForConditionalGeneration (line 1085) | class BigBirdPegasusForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1088) | def __init__(self, *args, **kwargs):
  class BigBirdPegasusForQuestionAnswering (line 1092) | class BigBirdPegasusForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1095) | def __init__(self, *args, **kwargs):
  class BigBirdPegasusForSequenceClassification (line 1099) | class BigBirdPegasusForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1102) | def __init__(self, *args, **kwargs):
  class BigBirdPegasusModel (line 1106) | class BigBirdPegasusModel(metaclass=DummyObject):
    method __init__ (line 1109) | def __init__(self, *args, **kwargs):
  class BigBirdPegasusPreTrainedModel (line 1113) | class BigBirdPegasusPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1116) | def __init__(self, *args, **kwargs):
  class BioGptForCausalLM (line 1123) | class BioGptForCausalLM(metaclass=DummyObject):
    method __init__ (line 1126) | def __init__(self, *args, **kwargs):
  class BioGptForSequenceClassification (line 1130) | class BioGptForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1133) | def __init__(self, *args, **kwargs):
  class BioGptForTokenClassification (line 1137) | class BioGptForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1140) | def __init__(self, *args, **kwargs):
  class BioGptModel (line 1144) | class BioGptModel(metaclass=DummyObject):
    method __init__ (line 1147) | def __init__(self, *args, **kwargs):
  class BioGptPreTrainedModel (line 1151) | class BioGptPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1154) | def __init__(self, *args, **kwargs):
  class BitBackbone (line 1161) | class BitBackbone(metaclass=DummyObject):
    method __init__ (line 1164) | def __init__(self, *args, **kwargs):
  class BitForImageClassification (line 1168) | class BitForImageClassification(metaclass=DummyObject):
    method __init__ (line 1171) | def __init__(self, *args, **kwargs):
  class BitModel (line 1175) | class BitModel(metaclass=DummyObject):
    method __init__ (line 1178) | def __init__(self, *args, **kwargs):
  class BitPreTrainedModel (line 1182) | class BitPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1185) | def __init__(self, *args, **kwargs):
  class BlenderbotForCausalLM (line 1192) | class BlenderbotForCausalLM(metaclass=DummyObject):
    method __init__ (line 1195) | def __init__(self, *args, **kwargs):
  class BlenderbotForConditionalGeneration (line 1199) | class BlenderbotForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1202) | def __init__(self, *args, **kwargs):
  class BlenderbotModel (line 1206) | class BlenderbotModel(metaclass=DummyObject):
    method __init__ (line 1209) | def __init__(self, *args, **kwargs):
  class BlenderbotPreTrainedModel (line 1213) | class BlenderbotPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1216) | def __init__(self, *args, **kwargs):
  class BlenderbotSmallForCausalLM (line 1223) | class BlenderbotSmallForCausalLM(metaclass=DummyObject):
    method __init__ (line 1226) | def __init__(self, *args, **kwargs):
  class BlenderbotSmallForConditionalGeneration (line 1230) | class BlenderbotSmallForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1233) | def __init__(self, *args, **kwargs):
  class BlenderbotSmallModel (line 1237) | class BlenderbotSmallModel(metaclass=DummyObject):
    method __init__ (line 1240) | def __init__(self, *args, **kwargs):
  class BlenderbotSmallPreTrainedModel (line 1244) | class BlenderbotSmallPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1247) | def __init__(self, *args, **kwargs):
  class BlipForConditionalGeneration (line 1254) | class BlipForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1257) | def __init__(self, *args, **kwargs):
  class BlipForImageTextRetrieval (line 1261) | class BlipForImageTextRetrieval(metaclass=DummyObject):
    method __init__ (line 1264) | def __init__(self, *args, **kwargs):
  class BlipForQuestionAnswering (line 1268) | class BlipForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1271) | def __init__(self, *args, **kwargs):
  class BlipModel (line 1275) | class BlipModel(metaclass=DummyObject):
    method __init__ (line 1278) | def __init__(self, *args, **kwargs):
  class BlipPreTrainedModel (line 1282) | class BlipPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1285) | def __init__(self, *args, **kwargs):
  class BlipTextModel (line 1289) | class BlipTextModel(metaclass=DummyObject):
    method __init__ (line 1292) | def __init__(self, *args, **kwargs):
  class BlipVisionModel (line 1296) | class BlipVisionModel(metaclass=DummyObject):
    method __init__ (line 1299) | def __init__(self, *args, **kwargs):
  class Blip2ForConditionalGeneration (line 1306) | class Blip2ForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1309) | def __init__(self, *args, **kwargs):
  class Blip2Model (line 1313) | class Blip2Model(metaclass=DummyObject):
    method __init__ (line 1316) | def __init__(self, *args, **kwargs):
  class Blip2PreTrainedModel (line 1320) | class Blip2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1323) | def __init__(self, *args, **kwargs):
  class Blip2QFormerModel (line 1327) | class Blip2QFormerModel(metaclass=DummyObject):
    method __init__ (line 1330) | def __init__(self, *args, **kwargs):
  class Blip2VisionModel (line 1334) | class Blip2VisionModel(metaclass=DummyObject):
    method __init__ (line 1337) | def __init__(self, *args, **kwargs):
  class BloomForCausalLM (line 1344) | class BloomForCausalLM(metaclass=DummyObject):
    method __init__ (line 1347) | def __init__(self, *args, **kwargs):
  class BloomForQuestionAnswering (line 1351) | class BloomForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1354) | def __init__(self, *args, **kwargs):
  class BloomForSequenceClassification (line 1358) | class BloomForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1361) | def __init__(self, *args, **kwargs):
  class BloomForTokenClassification (line 1365) | class BloomForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1368) | def __init__(self, *args, **kwargs):
  class BloomModel (line 1372) | class BloomModel(metaclass=DummyObject):
    method __init__ (line 1375) | def __init__(self, *args, **kwargs):
  class BloomPreTrainedModel (line 1379) | class BloomPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1382) | def __init__(self, *args, **kwargs):
  class BridgeTowerForContrastiveLearning (line 1389) | class BridgeTowerForContrastiveLearning(metaclass=DummyObject):
    method __init__ (line 1392) | def __init__(self, *args, **kwargs):
  class BridgeTowerForImageAndTextRetrieval (line 1396) | class BridgeTowerForImageAndTextRetrieval(metaclass=DummyObject):
    method __init__ (line 1399) | def __init__(self, *args, **kwargs):
  class BridgeTowerForMaskedLM (line 1403) | class BridgeTowerForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1406) | def __init__(self, *args, **kwargs):
  class BridgeTowerModel (line 1410) | class BridgeTowerModel(metaclass=DummyObject):
    method __init__ (line 1413) | def __init__(self, *args, **kwargs):
  class BridgeTowerPreTrainedModel (line 1417) | class BridgeTowerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1420) | def __init__(self, *args, **kwargs):
  class CamembertForCausalLM (line 1427) | class CamembertForCausalLM(metaclass=DummyObject):
    method __init__ (line 1430) | def __init__(self, *args, **kwargs):
  class CamembertForMaskedLM (line 1434) | class CamembertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1437) | def __init__(self, *args, **kwargs):
  class CamembertForMultipleChoice (line 1441) | class CamembertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1444) | def __init__(self, *args, **kwargs):
  class CamembertForQuestionAnswering (line 1448) | class CamembertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1451) | def __init__(self, *args, **kwargs):
  class CamembertForSequenceClassification (line 1455) | class CamembertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1458) | def __init__(self, *args, **kwargs):
  class CamembertForTokenClassification (line 1462) | class CamembertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1465) | def __init__(self, *args, **kwargs):
  class CamembertModel (line 1469) | class CamembertModel(metaclass=DummyObject):
    method __init__ (line 1472) | def __init__(self, *args, **kwargs):
  class CamembertPreTrainedModel (line 1476) | class CamembertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1479) | def __init__(self, *args, **kwargs):
  class CanineForMultipleChoice (line 1486) | class CanineForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1489) | def __init__(self, *args, **kwargs):
  class CanineForQuestionAnswering (line 1493) | class CanineForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1496) | def __init__(self, *args, **kwargs):
  class CanineForSequenceClassification (line 1500) | class CanineForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1503) | def __init__(self, *args, **kwargs):
  class CanineForTokenClassification (line 1507) | class CanineForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1510) | def __init__(self, *args, **kwargs):
  class CanineLayer (line 1514) | class CanineLayer(metaclass=DummyObject):
    method __init__ (line 1517) | def __init__(self, *args, **kwargs):
  class CanineModel (line 1521) | class CanineModel(metaclass=DummyObject):
    method __init__ (line 1524) | def __init__(self, *args, **kwargs):
  class CaninePreTrainedModel (line 1528) | class CaninePreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1531) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_canine (line 1535) | def load_tf_weights_in_canine(*args, **kwargs):
  class ChineseCLIPModel (line 1542) | class ChineseCLIPModel(metaclass=DummyObject):
    method __init__ (line 1545) | def __init__(self, *args, **kwargs):
  class ChineseCLIPPreTrainedModel (line 1549) | class ChineseCLIPPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1552) | def __init__(self, *args, **kwargs):
  class ChineseCLIPTextModel (line 1556) | class ChineseCLIPTextModel(metaclass=DummyObject):
    method __init__ (line 1559) | def __init__(self, *args, **kwargs):
  class ChineseCLIPVisionModel (line 1563) | class ChineseCLIPVisionModel(metaclass=DummyObject):
    method __init__ (line 1566) | def __init__(self, *args, **kwargs):
  class ClapAudioModel (line 1573) | class ClapAudioModel(metaclass=DummyObject):
    method __init__ (line 1576) | def __init__(self, *args, **kwargs):
  class ClapAudioModelWithProjection (line 1580) | class ClapAudioModelWithProjection(metaclass=DummyObject):
    method __init__ (line 1583) | def __init__(self, *args, **kwargs):
  class ClapFeatureExtractor (line 1587) | class ClapFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 1590) | def __init__(self, *args, **kwargs):
  class ClapModel (line 1594) | class ClapModel(metaclass=DummyObject):
    method __init__ (line 1597) | def __init__(self, *args, **kwargs):
  class ClapPreTrainedModel (line 1601) | class ClapPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1604) | def __init__(self, *args, **kwargs):
  class ClapTextModel (line 1608) | class ClapTextModel(metaclass=DummyObject):
    method __init__ (line 1611) | def __init__(self, *args, **kwargs):
  class ClapTextModelWithProjection (line 1615) | class ClapTextModelWithProjection(metaclass=DummyObject):
    method __init__ (line 1618) | def __init__(self, *args, **kwargs):
  class CLIPModel (line 1625) | class CLIPModel(metaclass=DummyObject):
    method __init__ (line 1628) | def __init__(self, *args, **kwargs):
  class CLIPPreTrainedModel (line 1632) | class CLIPPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1635) | def __init__(self, *args, **kwargs):
  class CLIPTextModel (line 1639) | class CLIPTextModel(metaclass=DummyObject):
    method __init__ (line 1642) | def __init__(self, *args, **kwargs):
  class CLIPTextModelWithProjection (line 1646) | class CLIPTextModelWithProjection(metaclass=DummyObject):
    method __init__ (line 1649) | def __init__(self, *args, **kwargs):
  class CLIPVisionModel (line 1653) | class CLIPVisionModel(metaclass=DummyObject):
    method __init__ (line 1656) | def __init__(self, *args, **kwargs):
  class CLIPVisionModelWithProjection (line 1660) | class CLIPVisionModelWithProjection(metaclass=DummyObject):
    method __init__ (line 1663) | def __init__(self, *args, **kwargs):
  class CLIPSegForImageSegmentation (line 1670) | class CLIPSegForImageSegmentation(metaclass=DummyObject):
    method __init__ (line 1673) | def __init__(self, *args, **kwargs):
  class CLIPSegModel (line 1677) | class CLIPSegModel(metaclass=DummyObject):
    method __init__ (line 1680) | def __init__(self, *args, **kwargs):
  class CLIPSegPreTrainedModel (line 1684) | class CLIPSegPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1687) | def __init__(self, *args, **kwargs):
  class CLIPSegTextModel (line 1691) | class CLIPSegTextModel(metaclass=DummyObject):
    method __init__ (line 1694) | def __init__(self, *args, **kwargs):
  class CLIPSegVisionModel (line 1698) | class CLIPSegVisionModel(metaclass=DummyObject):
    method __init__ (line 1701) | def __init__(self, *args, **kwargs):
  class CodeGenForCausalLM (line 1708) | class CodeGenForCausalLM(metaclass=DummyObject):
    method __init__ (line 1711) | def __init__(self, *args, **kwargs):
  class CodeGenModel (line 1715) | class CodeGenModel(metaclass=DummyObject):
    method __init__ (line 1718) | def __init__(self, *args, **kwargs):
  class CodeGenPreTrainedModel (line 1722) | class CodeGenPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1725) | def __init__(self, *args, **kwargs):
  class ConditionalDetrForObjectDetection (line 1732) | class ConditionalDetrForObjectDetection(metaclass=DummyObject):
    method __init__ (line 1735) | def __init__(self, *args, **kwargs):
  class ConditionalDetrForSegmentation (line 1739) | class ConditionalDetrForSegmentation(metaclass=DummyObject):
    method __init__ (line 1742) | def __init__(self, *args, **kwargs):
  class ConditionalDetrModel (line 1746) | class ConditionalDetrModel(metaclass=DummyObject):
    method __init__ (line 1749) | def __init__(self, *args, **kwargs):
  class ConditionalDetrPreTrainedModel (line 1753) | class ConditionalDetrPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1756) | def __init__(self, *args, **kwargs):
  class ConvBertForMaskedLM (line 1763) | class ConvBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1766) | def __init__(self, *args, **kwargs):
  class ConvBertForMultipleChoice (line 1770) | class ConvBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1773) | def __init__(self, *args, **kwargs):
  class ConvBertForQuestionAnswering (line 1777) | class ConvBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1780) | def __init__(self, *args, **kwargs):
  class ConvBertForSequenceClassification (line 1784) | class ConvBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1787) | def __init__(self, *args, **kwargs):
  class ConvBertForTokenClassification (line 1791) | class ConvBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1794) | def __init__(self, *args, **kwargs):
  class ConvBertLayer (line 1798) | class ConvBertLayer(metaclass=DummyObject):
    method __init__ (line 1801) | def __init__(self, *args, **kwargs):
  class ConvBertModel (line 1805) | class ConvBertModel(metaclass=DummyObject):
    method __init__ (line 1808) | def __init__(self, *args, **kwargs):
  class ConvBertPreTrainedModel (line 1812) | class ConvBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1815) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_convbert (line 1819) | def load_tf_weights_in_convbert(*args, **kwargs):
  class ConvNextBackbone (line 1826) | class ConvNextBackbone(metaclass=DummyObject):
    method __init__ (line 1829) | def __init__(self, *args, **kwargs):
  class ConvNextForImageClassification (line 1833) | class ConvNextForImageClassification(metaclass=DummyObject):
    method __init__ (line 1836) | def __init__(self, *args, **kwargs):
  class ConvNextModel (line 1840) | class ConvNextModel(metaclass=DummyObject):
    method __init__ (line 1843) | def __init__(self, *args, **kwargs):
  class ConvNextPreTrainedModel (line 1847) | class ConvNextPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1850) | def __init__(self, *args, **kwargs):
  class ConvNextV2Backbone (line 1857) | class ConvNextV2Backbone(metaclass=DummyObject):
    method __init__ (line 1860) | def __init__(self, *args, **kwargs):
  class ConvNextV2ForImageClassification (line 1864) | class ConvNextV2ForImageClassification(metaclass=DummyObject):
    method __init__ (line 1867) | def __init__(self, *args, **kwargs):
  class ConvNextV2Model (line 1871) | class ConvNextV2Model(metaclass=DummyObject):
    method __init__ (line 1874) | def __init__(self, *args, **kwargs):
  class ConvNextV2PreTrainedModel (line 1878) | class ConvNextV2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1881) | def __init__(self, *args, **kwargs):
  class CpmAntForCausalLM (line 1888) | class CpmAntForCausalLM(metaclass=DummyObject):
    method __init__ (line 1891) | def __init__(self, *args, **kwargs):
  class CpmAntModel (line 1895) | class CpmAntModel(metaclass=DummyObject):
    method __init__ (line 1898) | def __init__(self, *args, **kwargs):
  class CpmAntPreTrainedModel (line 1902) | class CpmAntPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1905) | def __init__(self, *args, **kwargs):
  class CTRLForSequenceClassification (line 1912) | class CTRLForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1915) | def __init__(self, *args, **kwargs):
  class CTRLLMHeadModel (line 1919) | class CTRLLMHeadModel(metaclass=DummyObject):
    method __init__ (line 1922) | def __init__(self, *args, **kwargs):
  class CTRLModel (line 1926) | class CTRLModel(metaclass=DummyObject):
    method __init__ (line 1929) | def __init__(self, *args, **kwargs):
  class CTRLPreTrainedModel (line 1933) | class CTRLPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1936) | def __init__(self, *args, **kwargs):
  class CvtForImageClassification (line 1943) | class CvtForImageClassification(metaclass=DummyObject):
    method __init__ (line 1946) | def __init__(self, *args, **kwargs):
  class CvtModel (line 1950) | class CvtModel(metaclass=DummyObject):
    method __init__ (line 1953) | def __init__(self, *args, **kwargs):
  class CvtPreTrainedModel (line 1957) | class CvtPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1960) | def __init__(self, *args, **kwargs):
  class Data2VecAudioForAudioFrameClassification (line 1973) | class Data2VecAudioForAudioFrameClassification(metaclass=DummyObject):
    method __init__ (line 1976) | def __init__(self, *args, **kwargs):
  class Data2VecAudioForCTC (line 1980) | class Data2VecAudioForCTC(metaclass=DummyObject):
    method __init__ (line 1983) | def __init__(self, *args, **kwargs):
  class Data2VecAudioForSequenceClassification (line 1987) | class Data2VecAudioForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1990) | def __init__(self, *args, **kwargs):
  class Data2VecAudioForXVector (line 1994) | class Data2VecAudioForXVector(metaclass=DummyObject):
    method __init__ (line 1997) | def __init__(self, *args, **kwargs):
  class Data2VecAudioModel (line 2001) | class Data2VecAudioModel(metaclass=DummyObject):
    method __init__ (line 2004) | def __init__(self, *args, **kwargs):
  class Data2VecAudioPreTrainedModel (line 2008) | class Data2VecAudioPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2011) | def __init__(self, *args, **kwargs):
  class Data2VecTextForCausalLM (line 2015) | class Data2VecTextForCausalLM(metaclass=DummyObject):
    method __init__ (line 2018) | def __init__(self, *args, **kwargs):
  class Data2VecTextForMaskedLM (line 2022) | class Data2VecTextForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2025) | def __init__(self, *args, **kwargs):
  class Data2VecTextForMultipleChoice (line 2029) | class Data2VecTextForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2032) | def __init__(self, *args, **kwargs):
  class Data2VecTextForQuestionAnswering (line 2036) | class Data2VecTextForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2039) | def __init__(self, *args, **kwargs):
  class Data2VecTextForSequenceClassification (line 2043) | class Data2VecTextForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2046) | def __init__(self, *args, **kwargs):
  class Data2VecTextForTokenClassification (line 2050) | class Data2VecTextForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2053) | def __init__(self, *args, **kwargs):
  class Data2VecTextModel (line 2057) | class Data2VecTextModel(metaclass=DummyObject):
    method __init__ (line 2060) | def __init__(self, *args, **kwargs):
  class Data2VecTextPreTrainedModel (line 2064) | class Data2VecTextPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2067) | def __init__(self, *args, **kwargs):
  class Data2VecVisionForImageClassification (line 2071) | class Data2VecVisionForImageClassification(metaclass=DummyObject):
    method __init__ (line 2074) | def __init__(self, *args, **kwargs):
  class Data2VecVisionForSemanticSegmentation (line 2078) | class Data2VecVisionForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 2081) | def __init__(self, *args, **kwargs):
  class Data2VecVisionModel (line 2085) | class Data2VecVisionModel(metaclass=DummyObject):
    method __init__ (line 2088) | def __init__(self, *args, **kwargs):
  class Data2VecVisionPreTrainedModel (line 2092) | class Data2VecVisionPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2095) | def __init__(self, *args, **kwargs):
  class DebertaForMaskedLM (line 2102) | class DebertaForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2105) | def __init__(self, *args, **kwargs):
  class DebertaForQuestionAnswering (line 2109) | class DebertaForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2112) | def __init__(self, *args, **kwargs):
  class DebertaForSequenceClassification (line 2116) | class DebertaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2119) | def __init__(self, *args, **kwargs):
  class DebertaForTokenClassification (line 2123) | class DebertaForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2126) | def __init__(self, *args, **kwargs):
  class DebertaModel (line 2130) | class DebertaModel(metaclass=DummyObject):
    method __init__ (line 2133) | def __init__(self, *args, **kwargs):
  class DebertaPreTrainedModel (line 2137) | class DebertaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2140) | def __init__(self, *args, **kwargs):
  class DebertaV2ForMaskedLM (line 2147) | class DebertaV2ForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2150) | def __init__(self, *args, **kwargs):
  class DebertaV2ForMultipleChoice (line 2154) | class DebertaV2ForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2157) | def __init__(self, *args, **kwargs):
  class DebertaV2ForQuestionAnswering (line 2161) | class DebertaV2ForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2164) | def __init__(self, *args, **kwargs):
  class DebertaV2ForSequenceClassification (line 2168) | class DebertaV2ForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2171) | def __init__(self, *args, **kwargs):
  class DebertaV2ForTokenClassification (line 2175) | class DebertaV2ForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2178) | def __init__(self, *args, **kwargs):
  class DebertaV2Model (line 2182) | class DebertaV2Model(metaclass=DummyObject):
    method __init__ (line 2185) | def __init__(self, *args, **kwargs):
  class DebertaV2PreTrainedModel (line 2189) | class DebertaV2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2192) | def __init__(self, *args, **kwargs):
  class DecisionTransformerGPT2Model (line 2199) | class DecisionTransformerGPT2Model(metaclass=DummyObject):
    method __init__ (line 2202) | def __init__(self, *args, **kwargs):
  class DecisionTransformerGPT2PreTrainedModel (line 2206) | class DecisionTransformerGPT2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2209) | def __init__(self, *args, **kwargs):
  class DecisionTransformerModel (line 2213) | class DecisionTransformerModel(metaclass=DummyObject):
    method __init__ (line 2216) | def __init__(self, *args, **kwargs):
  class DecisionTransformerPreTrainedModel (line 2220) | class DecisionTransformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2223) | def __init__(self, *args, **kwargs):
  class DeformableDetrForObjectDetection (line 2230) | class DeformableDetrForObjectDetection(metaclass=DummyObject):
    method __init__ (line 2233) | def __init__(self, *args, **kwargs):
  class DeformableDetrModel (line 2237) | class DeformableDetrModel(metaclass=DummyObject):
    method __init__ (line 2240) | def __init__(self, *args, **kwargs):
  class DeformableDetrPreTrainedModel (line 2244) | class DeformableDetrPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2247) | def __init__(self, *args, **kwargs):
  class DeiTForImageClassification (line 2254) | class DeiTForImageClassification(metaclass=DummyObject):
    method __init__ (line 2257) | def __init__(self, *args, **kwargs):
  class DeiTForImageClassificationWithTeacher (line 2261) | class DeiTForImageClassificationWithTeacher(metaclass=DummyObject):
    method __init__ (line 2264) | def __init__(self, *args, **kwargs):
  class DeiTForMaskedImageModeling (line 2268) | class DeiTForMaskedImageModeling(metaclass=DummyObject):
    method __init__ (line 2271) | def __init__(self, *args, **kwargs):
  class DeiTModel (line 2275) | class DeiTModel(metaclass=DummyObject):
    method __init__ (line 2278) | def __init__(self, *args, **kwargs):
  class DeiTPreTrainedModel (line 2282) | class DeiTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2285) | def __init__(self, *args, **kwargs):
  class DetaForObjectDetection (line 2292) | class DetaForObjectDetection(metaclass=DummyObject):
    method __init__ (line 2295) | def __init__(self, *args, **kwargs):
  class DetaModel (line 2299) | class DetaModel(metaclass=DummyObject):
    method __init__ (line 2302) | def __init__(self, *args, **kwargs):
  class DetaPreTrainedModel (line 2306) | class DetaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2309) | def __init__(self, *args, **kwargs):
  class DetrForObjectDetection (line 2316) | class DetrForObjectDetection(metaclass=DummyObject):
    method __init__ (line 2319) | def __init__(self, *args, **kwargs):
  class DetrForSegmentation (line 2323) | class DetrForSegmentation(metaclass=DummyObject):
    method __init__ (line 2326) | def __init__(self, *args, **kwargs):
  class DetrModel (line 2330) | class DetrModel(metaclass=DummyObject):
    method __init__ (line 2333) | def __init__(self, *args, **kwargs):
  class DetrPreTrainedModel (line 2337) | class DetrPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2340) | def __init__(self, *args, **kwargs):
  class DinatBackbone (line 2347) | class DinatBackbone(metaclass=DummyObject):
    method __init__ (line 2350) | def __init__(self, *args, **kwargs):
  class DinatForImageClassification (line 2354) | class DinatForImageClassification(metaclass=DummyObject):
    method __init__ (line 2357) | def __init__(self, *args, **kwargs):
  class DinatModel (line 2361) | class DinatModel(metaclass=DummyObject):
    method __init__ (line 2364) | def __init__(self, *args, **kwargs):
  class DinatPreTrainedModel (line 2368) | class DinatPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2371) | def __init__(self, *args, **kwargs):
  class DistilBertForMaskedLM (line 2378) | class DistilBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2381) | def __init__(self, *args, **kwargs):
  class DistilBertForMultipleChoice (line 2385) | class DistilBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2388) | def __init__(self, *args, **kwargs):
  class DistilBertForQuestionAnswering (line 2392) | class DistilBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2395) | def __init__(self, *args, **kwargs):
  class DistilBertForSequenceClassification (line 2399) | class DistilBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2402) | def __init__(self, *args, **kwargs):
  class DistilBertForTokenClassification (line 2406) | class DistilBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2409) | def __init__(self, *args, **kwargs):
  class DistilBertModel (line 2413) | class DistilBertModel(metaclass=DummyObject):
    method __init__ (line 2416) | def __init__(self, *args, **kwargs):
  class DistilBertPreTrainedModel (line 2420) | class DistilBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2423) | def __init__(self, *args, **kwargs):
  class DonutSwinModel (line 2430) | class DonutSwinModel(metaclass=DummyObject):
    method __init__ (line 2433) | def __init__(self, *args, **kwargs):
  class DonutSwinPreTrainedModel (line 2437) | class DonutSwinPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2440) | def __init__(self, *args, **kwargs):
  class DPRContextEncoder (line 2453) | class DPRContextEncoder(metaclass=DummyObject):
    method __init__ (line 2456) | def __init__(self, *args, **kwargs):
  class DPRPretrainedContextEncoder (line 2460) | class DPRPretrainedContextEncoder(metaclass=DummyObject):
    method __init__ (line 2463) | def __init__(self, *args, **kwargs):
  class DPRPreTrainedModel (line 2467) | class DPRPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2470) | def __init__(self, *args, **kwargs):
  class DPRPretrainedQuestionEncoder (line 2474) | class DPRPretrainedQuestionEncoder(metaclass=DummyObject):
    method __init__ (line 2477) | def __init__(self, *args, **kwargs):
  class DPRPretrainedReader (line 2481) | class DPRPretrainedReader(metaclass=DummyObject):
    method __init__ (line 2484) | def __init__(self, *args, **kwargs):
  class DPRQuestionEncoder (line 2488) | class DPRQuestionEncoder(metaclass=DummyObject):
    method __init__ (line 2491) | def __init__(self, *args, **kwargs):
  class DPRReader (line 2495) | class DPRReader(metaclass=DummyObject):
    method __init__ (line 2498) | def __init__(self, *args, **kwargs):
  class DPTForDepthEstimation (line 2505) | class DPTForDepthEstimation(metaclass=DummyObject):
    method __init__ (line 2508) | def __init__(self, *args, **kwargs):
  class DPTForSemanticSegmentation (line 2512) | class DPTForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 2515) | def __init__(self, *args, **kwargs):
  class DPTModel (line 2519) | class DPTModel(metaclass=DummyObject):
    method __init__ (line 2522) | def __init__(self, *args, **kwargs):
  class DPTPreTrainedModel (line 2526) | class DPTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2529) | def __init__(self, *args, **kwargs):
  class EfficientFormerForImageClassification (line 2536) | class EfficientFormerForImageClassification(metaclass=DummyObject):
    method __init__ (line 2539) | def __init__(self, *args, **kwargs):
  class EfficientFormerForImageClassificationWithTeacher (line 2543) | class EfficientFormerForImageClassificationWithTeacher(metaclass=DummyOb...
    method __init__ (line 2546) | def __init__(self, *args, **kwargs):
  class EfficientFormerModel (line 2550) | class EfficientFormerModel(metaclass=DummyObject):
    method __init__ (line 2553) | def __init__(self, *args, **kwargs):
  class EfficientFormerPreTrainedModel (line 2557) | class EfficientFormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2560) | def __init__(self, *args, **kwargs):
  class EfficientNetForImageClassification (line 2567) | class EfficientNetForImageClassification(metaclass=DummyObject):
    method __init__ (line 2570) | def __init__(self, *args, **kwargs):
  class EfficientNetModel (line 2574) | class EfficientNetModel(metaclass=DummyObject):
    method __init__ (line 2577) | def __init__(self, *args, **kwargs):
  class EfficientNetPreTrainedModel (line 2581) | class EfficientNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2584) | def __init__(self, *args, **kwargs):
  class ElectraForCausalLM (line 2591) | class ElectraForCausalLM(metaclass=DummyObject):
    method __init__ (line 2594) | def __init__(self, *args, **kwargs):
  class ElectraForMaskedLM (line 2598) | class ElectraForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2601) | def __init__(self, *args, **kwargs):
  class ElectraForMultipleChoice (line 2605) | class ElectraForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2608) | def __init__(self, *args, **kwargs):
  class ElectraForPreTraining (line 2612) | class ElectraForPreTraining(metaclass=DummyObject):
    method __init__ (line 2615) | def __init__(self, *args, **kwargs):
  class ElectraForQuestionAnswering (line 2619) | class ElectraForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2622) | def __init__(self, *args, **kwargs):
  class ElectraForSequenceClassification (line 2626) | class ElectraForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2629) | def __init__(self, *args, **kwargs):
  class ElectraForTokenClassification (line 2633) | class ElectraForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2636) | def __init__(self, *args, **kwargs):
  class ElectraModel (line 2640) | class ElectraModel(metaclass=DummyObject):
    method __init__ (line 2643) | def __init__(self, *args, **kwargs):
  class ElectraPreTrainedModel (line 2647) | class ElectraPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2650) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_electra (line 2654) | def load_tf_weights_in_electra(*args, **kwargs):
  class EncoderDecoderModel (line 2658) | class EncoderDecoderModel(metaclass=DummyObject):
    method __init__ (line 2661) | def __init__(self, *args, **kwargs):
  class ErnieForCausalLM (line 2668) | class ErnieForCausalLM(metaclass=DummyObject):
    method __init__ (line 2671) | def __init__(self, *args, **kwargs):
  class ErnieForMaskedLM (line 2675) | class ErnieForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2678) | def __init__(self, *args, **kwargs):
  class ErnieForMultipleChoice (line 2682) | class ErnieForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2685) | def __init__(self, *args, **kwargs):
  class ErnieForNextSentencePrediction (line 2689) | class ErnieForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 2692) | def __init__(self, *args, **kwargs):
  class ErnieForPreTraining (line 2696) | class ErnieForPreTraining(metaclass=DummyObject):
    method __init__ (line 2699) | def __init__(self, *args, **kwargs):
  class ErnieForQuestionAnswering (line 2703) | class ErnieForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2706) | def __init__(self, *args, **kwargs):
  class ErnieForSequenceClassification (line 2710) | class ErnieForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2713) | def __init__(self, *args, **kwargs):
  class ErnieForTokenClassification (line 2717) | class ErnieForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2720) | def __init__(self, *args, **kwargs):
  class ErnieModel (line 2724) | class ErnieModel(metaclass=DummyObject):
    method __init__ (line 2727) | def __init__(self, *args, **kwargs):
  class ErniePreTrainedModel (line 2731) | class ErniePreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2734) | def __init__(self, *args, **kwargs):
  class ErnieMForInformationExtraction (line 2741) | class ErnieMForInformationExtraction(metaclass=DummyObject):
    method __init__ (line 2744) | def __init__(self, *args, **kwargs):
  class ErnieMForMultipleChoice (line 2748) | class ErnieMForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2751) | def __init__(self, *args, **kwargs):
  class ErnieMForQuestionAnswering (line 2755) | class ErnieMForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2758) | def __init__(self, *args, **kwargs):
  class ErnieMForSequenceClassification (line 2762) | class ErnieMForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2765) | def __init__(self, *args, **kwargs):
  class ErnieMForTokenClassification (line 2769) | class ErnieMForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2772) | def __init__(self, *args, **kwargs):
  class ErnieMModel (line 2776) | class ErnieMModel(metaclass=DummyObject):
    method __init__ (line 2779) | def __init__(self, *args, **kwargs):
  class ErnieMPreTrainedModel (line 2783) | class ErnieMPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2786) | def __init__(self, *args, **kwargs):
  class EsmFoldPreTrainedModel (line 2793) | class EsmFoldPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2796) | def __init__(self, *args, **kwargs):
  class EsmForMaskedLM (line 2800) | class EsmForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2803) | def __init__(self, *args, **kwargs):
  class EsmForProteinFolding (line 2807) | class EsmForProteinFolding(metaclass=DummyObject):
    method __init__ (line 2810) | def __init__(self, *args, **kwargs):
  class EsmForSequenceClassification (line 2814) | class EsmForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2817) | def __init__(self, *args, **kwargs):
  class EsmForTokenClassification (line 2821) | class EsmForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2824) | def __init__(self, *args, **kwargs):
  class EsmModel (line 2828) | class EsmModel(metaclass=DummyObject):
    method __init__ (line 2831) | def __init__(self, *args, **kwargs):
  class EsmPreTrainedModel (line 2835) | class EsmPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2838) | def __init__(self, *args, **kwargs):
  class FlaubertForMultipleChoice (line 2845) | class FlaubertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2848) | def __init__(self, *args, **kwargs):
  class FlaubertForQuestionAnswering (line 2852) | class FlaubertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2855) | def __init__(self, *args, **kwargs):
  class FlaubertForQuestionAnsweringSimple (line 2859) | class FlaubertForQuestionAnsweringSimple(metaclass=DummyObject):
    method __init__ (line 2862) | def __init__(self, *args, **kwargs):
  class FlaubertForSequenceClassification (line 2866) | class FlaubertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2869) | def __init__(self, *args, **kwargs):
  class FlaubertForTokenClassification (line 2873) | class FlaubertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2876) | def __init__(self, *args, **kwargs):
  class FlaubertModel (line 2880) | class FlaubertModel(metaclass=DummyObject):
    method __init__ (line 2883) | def __init__(self, *args, **kwargs):
  class FlaubertPreTrainedModel (line 2887) | class FlaubertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2890) | def __init__(self, *args, **kwargs):
  class FlaubertWithLMHeadModel (line 2894) | class FlaubertWithLMHeadModel(metaclass=DummyObject):
    method __init__ (line 2897) | def __init__(self, *args, **kwargs):
  class FlavaForPreTraining (line 2904) | class FlavaForPreTraining(metaclass=DummyObject):
    method __init__ (line 2907) | def __init__(self, *args, **kwargs):
  class FlavaImageCodebook (line 2911) | class FlavaImageCodebook(metaclass=DummyObject):
    method __init__ (line 2914) | def __init__(self, *args, **kwargs):
  class FlavaImageModel (line 2918) | class FlavaImageModel(metaclass=DummyObject):
    method __init__ (line 2921) | def __init__(self, *args, **kwargs):
  class FlavaModel (line 2925) | class FlavaModel(metaclass=DummyObject):
    method __init__ (line 2928) | def __init__(self, *args, **kwargs):
  class FlavaMultimodalModel (line 2932) | class FlavaMultimodalModel(metaclass=DummyObject):
    method __init__ (line 2935) | def __init__(self, *args, **kwargs):
  class FlavaPreTrainedModel (line 2939) | class FlavaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2942) | def __init__(self, *args, **kwargs):
  class FlavaTextModel (line 2946) | class FlavaTextModel(metaclass=DummyObject):
    method __init__ (line 2949) | def __init__(self, *args, **kwargs):
  class FNetForMaskedLM (line 2956) | class FNetForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2959) | def __init__(self, *args, **kwargs):
  class FNetForMultipleChoice (line 2963) | class FNetForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2966) | def __init__(self, *args, **kwargs):
  class FNetForNextSentencePrediction (line 2970) | class FNetForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 2973) | def __init__(self, *args, **kwargs):
  class FNetForPreTraining (line 2977) | class FNetForPreTraining(metaclass=DummyObject):
    method __init__ (line 2980) | def __init__(self, *args, **kwargs):
  class FNetForQuestionAnswering (line 2984) | class FNetForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2987) | def __init__(self, *args, **kwargs):
  class FNetForSequenceClassification (line 2991) | class FNetForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2994) | def __init__(self, *args, **kwargs):
  class FNetForTokenClassification (line 2998) | class FNetForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3001) | def __init__(self, *args, **kwargs):
  class FNetLayer (line 3005) | class FNetLayer(metaclass=DummyObject):
    method __init__ (line 3008) | def __init__(self, *args, **kwargs):
  class FNetModel (line 3012) | class FNetModel(metaclass=DummyObject):
    method __init__ (line 3015) | def __init__(self, *args, **kwargs):
  class FNetPreTrainedModel (line 3019) | class FNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3022) | def __init__(self, *args, **kwargs):
  class FocalNetBackbone (line 3029) | class FocalNetBackbone(metaclass=DummyObject):
    method __init__ (line 3032) | def __init__(self, *args, **kwargs):
  class FocalNetForImageClassification (line 3036) | class FocalNetForImageClassification(metaclass=DummyObject):
    method __init__ (line 3039) | def __init__(self, *args, **kwargs):
  class FocalNetForMaskedImageModeling (line 3043) | class FocalNetForMaskedImageModeling(metaclass=DummyObject):
    method __init__ (line 3046) | def __init__(self, *args, **kwargs):
  class FocalNetModel (line 3050) | class FocalNetModel(metaclass=DummyObject):
    method __init__ (line 3053) | def __init__(self, *args, **kwargs):
  class FocalNetPreTrainedModel (line 3057) | class FocalNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3060) | def __init__(self, *args, **kwargs):
  class FSMTForConditionalGeneration (line 3064) | class FSMTForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 3067) | def __init__(self, *args, **kwargs):
  class FSMTModel (line 3071) | class FSMTModel(metaclass=DummyObject):
    method __init__ (line 3074) | def __init__(self, *args, **kwargs):
  class PretrainedFSMTModel (line 3078) | class PretrainedFSMTModel(metaclass=DummyObject):
    method __init__ (line 3081) | def __init__(self, *args, **kwargs):
  class FunnelBaseModel (line 3088) | class FunnelBaseModel(metaclass=DummyObject):
    method __init__ (line 3091) | def __init__(self, *args, **kwargs):
  class FunnelForMaskedLM (line 3095) | class FunnelForMaskedLM(metaclass=DummyObject):
    method __init__ (line 3098) | def __init__(self, *args, **kwargs):
  class FunnelForMultipleChoice (line 3102) | class FunnelForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 3105) | def __init__(self, *args, **kwargs):
  class FunnelForPreTraining (line 3109) | class FunnelForPreTraining(metaclass=DummyObject):
    method __init__ (line 3112) | def __init__(self, *args, **kwargs):
  class FunnelForQuestionAnswering (line 3116) | class FunnelForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3119) | def __init__(self, *args, **kwargs):
  class FunnelForSequenceClassification (line 3123) | class FunnelForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3126) | def __init__(self, *args, **kwargs):
  class FunnelForTokenClassification (line 3130) | class FunnelForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3133) | def __init__(self, *args, **kwargs):
  class FunnelModel (line 3137) | class FunnelModel(metaclass=DummyObject):
    method __init__ (line 3140) | def __init__(self, *args, **kwargs):
  class FunnelPreTrainedModel (line 3144) | class FunnelPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3147) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_funnel (line 3151) | def load_tf_weights_in_funnel(*args, **kwargs):
  class GitForCausalLM (line 3158) | class GitForCausalLM(metaclass=DummyObject):
    method __init__ (line 3161) | def __init__(self, *args, **kwargs):
  class GitModel (line 3165) | class GitModel(metaclass=DummyObject):
    method __init__ (line 3168) | def __init__(self, *args, **kwargs):
  class GitPreTrainedModel (line 3172) | class GitPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3175) | def __init__(self, *args, **kwargs):
  class GitVisionModel (line 3179) | class GitVisionModel(metaclass=DummyObject):
    method __init__ (line 3182) | def __init__(self, *args, **kwargs):
  class GLPNForDepthEstimation (line 3189) | class GLPNForDepthEstimation(metaclass=DummyObject):
    method __init__ (line 3192) | def __init__(self, *args, **kwargs):
  class GLPNModel (line 3196) | class GLPNModel(metaclass=DummyObject):
    method __init__ (line 3199) | def __init__(self, *args, **kwargs):
  class GLPNPreTrainedModel (line 3203) | class GLPNPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3206) | def __init__(self, *args, **kwargs):
  class GPT2DoubleHeadsModel (line 3213) | class GPT2DoubleHeadsModel(metaclass=DummyObject):
    method __init__ (line 3216) | def __init__(self, *args, **kwargs):
  class GPT2ForQuestionAnswering (line 3220) | class GPT2ForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3223) | def __init__(self, *args, **kwargs):
  class GPT2ForSequenceClassification (line 3227) | class GPT2ForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3230) | def __init__(self, *args, **kwargs):
  class GPT2ForTokenClassification (line 3234) | class GPT2ForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3237) | def __init__(self, *args, **kwargs):
  class GPT2LMHeadModel (line 3241) | class GPT2LMHeadModel(metaclass=DummyObject):
    method __init__ (line 3244) | def __init__(self, *args, **kwargs):
  class GPT2Model (line 3248) | class GPT2Model(metaclass=DummyObject):
    method __init__ (line 3251) | def __init__(self, *args, **kwargs):
  class GPT2PreTrainedModel (line 3255) | class GPT2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3258) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_gpt2 (line 3262) | def load_tf_weights_in_gpt2(*args, **kwargs):
  class GPTBigCodeForCausalLM (line 3269) | class GPTBigCodeForCausalLM(metaclass=DummyObject):
    method __init__ (line 3272) | def __init__(self, *args, **kwargs):
  class GPTBigCodeForSequenceClassification (line 3276) | class GPTBigCodeForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3279) | def __init__(self, *args, **kwargs):
  class GPTBigCodeForTokenClassification (line 3283) | class GPTBigCodeForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3286) | def __init__(self, *args, **kwargs):
  class GPTBigCodeModel (line 3290) | class GPTBigCodeModel(metaclass=DummyObject):
    method __init__ (line 3293) | def __init__(self, *args, **kwargs):
  class GPTBigCodePreTrainedModel (line 3297) | class GPTBigCodePreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3300) | def __init__(self, *args, **kwargs):
  class GPTNeoForCausalLM (line 3307) | class GPTNeoForCausalLM(metaclass=DummyObject):
    method __init__ (line 3310) | def __init__(self, *args, **kwargs):
  class GPTNeoForQuestionAnswering (line 3314) | class GPTNeoForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3317) | def __init__(self, *args, **kwargs):
  class GPTNeoForSequenceClassification (line 3321) | class GPTNeoForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3324) | def __init__(self, *args, **kwargs):
  class GPTNeoForTokenClassification (line 3328) | class GPTNeoForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3331) | def __init__(self, *args, **kwargs):
  class GPTNeoModel (line 3335) | class GPTNeoModel(metaclass=DummyObject):
    method __init__ (line 3338) | def __init__(self, *args, **kwargs):
  class GPTNeoPreTrainedModel (line 3342) | class GPTNeoPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3345) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_gpt_neo (line 3349) | def load_tf_weights_in_gpt_neo(*args, **kwargs):
  class GPTNeoXForCausalLM (line 3356) | class GPTNeoXForCausalLM(metaclass=DummyObject):
    method __init__ (line 3359) | def __init__(self, *args, **kwargs):
  class GPTNeoXForQuestionAnswering (line 3363) | class GPTNeoXForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3366) | def __init__(self, *args, **kwargs):
  class GPTNeoXForSequenceClassification (line 3370) | class GPTNeoXForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3373) | def __init__(self, *args, **kwargs):
  class GPTNeoXForTokenClassification (line 3377) | class GPTNeoXForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3380) | def __init__(self, *args, **kwargs):
  class GPTNeoXLayer (line 3384) | class GPTNeoXLayer(metaclass=DummyObject):
    method __init__ (line 3387) | def __init__(self, *args, **kwargs):
  class GPTNeoXModel (line 3391) | class GPTNeoXModel(metaclass=DummyObject):
    method __init__ (line 3394) | def __init__(self, *args, **kwargs):
  class GPTNeoXPreTrainedModel (line 3398) | class GPTNeoXPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3401) | def __init__(self, *args, **kwargs):
  class GPTNeoXJapaneseForCausalLM (line 3408) | class GPTNeoXJapaneseForCausalLM(metaclass=DummyObject):
    method __init__ (line 3411) | def __init__(self, *args, **kwargs):
  class GPTNeoXJapaneseLayer (line 3415) | class GPTNeoXJapaneseLayer(metaclass=DummyObject):
    method __init__ (line 3418) | def __init__(self, *args, **kwargs):
  class GPTNeoXJapaneseModel (line 3422) | class GPTNeoXJapaneseModel(metaclass=DummyObject):
    method __init__ (line 3425) | def __init__(self, *args, **kwargs):
  class GPTNeoXJapanesePreTrainedModel (line 3429) | class GPTNeoXJapanesePreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3432) | def __init__(self, *args, **kwargs):
  class GPTJForCausalLM (line 3439) | class GPTJForCausalLM(metaclass=DummyObject):
    method __init__ (line 3442) | def __init__(self, *args, **kwargs):
  class GPTJForQuestionAnswering (line 3446) | class GPTJForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3449) | def __init__(self, *args, **kwargs):
  class GPTJForSequenceClassification (line 3453) | class GPTJForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3456) | def __init__(self, *args, **kwargs):
  class GPTJModel (line 3460) | class GPTJModel(metaclass=DummyObject):
    method __init__ (line 3463) | def __init__(self, *args, **kwargs):
  class GPTJPreTrainedModel (line 3467) | class GPTJPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3470) | def __init__(self, *args, **kwargs):
  class GPTSanJapaneseForConditionalGeneration (line 3477) | class GPTSanJapaneseForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 3480) | def __init__(self, *args, **kwargs):
  class GPTSanJapaneseModel (line 3484) | class GPTSanJapaneseModel(metaclass=DummyObject):
    method __init__ (line 3487) | def __init__(self, *args, **kwargs):
  class GPTSanJapanesePreTrainedModel (line 3491) | class GPTSanJapanesePreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3494) | def __init__(self, *args, **kwargs):
  class GraphormerForGraphClassification (line 3501) | class GraphormerForGraphClassification(metaclass=DummyObject):
    method __init__ (line 3504) | def __init__(self, *args, **kwargs):
  class GraphormerModel (line 3508) | class GraphormerModel(metaclass=DummyObject):
    method __init__ (line 3511) | def __init__(self, *args, **kwargs):
  class GraphormerPreTrainedModel (line 3515) | class GraphormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3518) | def __init__(self, *args, **kwargs):
  class GroupViTModel (line 3525) | class GroupViTModel(metaclass=DummyObject):
    method __init__ (line 3528) | def __init__(self, *args, **kwargs):
  class GroupViTPreTrainedModel (line 3532) | class GroupViTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3535) | def __init__(self, *args, **kwargs):
  class GroupViTTextModel (line 3539) | class GroupViTTextModel(metaclass=DummyObject):
    method __init__ (line 3542) | def __init__(self, *args, **kwargs):
  class GroupViTVisionModel (line 3546) | class GroupViTVisionModel(metaclass=DummyObject):
    method __init__ (line 3549) | def __init__(self, *args, **kwargs):
  class HubertForCTC (line 3556) | class HubertForCTC(metaclass=DummyObject):
    method __init__ (line 3559) | def __init__(self, *args, **kwargs):
  class HubertForSequenceClassification (line 3563) | class HubertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3566) | def __init__(self, *args, **kwargs):
  class HubertModel (line 3570) | class HubertModel(metaclass=DummyObject):
    method __init__ (line 3573) | def __init__(self, *args, **kwargs):
  class HubertPreTrainedModel (line 3577) | class HubertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3580) | def __init__(self, *args, **kwargs):
  class IBertForMaskedLM (line 3587) | class IBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 3590) | def __init__(self, *args, **kwargs):
  class IBertForMultipleChoice (line 3594) | class IBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 3597) | def __init__(self, *args, **kwargs):
  class IBertForQuestionAnswering (line 3601) | class IBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3604) | def __init__(self, *args, **kwargs):
  class IBertForSequenceClassification (line 3608) | class IBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3611) | def __init__(self, *args, **kwargs):
  class IBertForTokenClassification (line 3615) | class IBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3618) | def __init__(self, *args, **kwargs):
  class IBertModel (line 3622) | class IBertModel(metaclass=DummyObject):
    method __init__ (line 3625) | def __init__(self, *args, **kwargs):
  class IBertPreTrainedModel (line 3629) | class IBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3632) | def __init__(self, *args, **kwargs):
  class ImageGPTForCausalImageModeling (line 3639) | class ImageGPTForCausalImageModeling(metaclass=DummyObject):
    method __init__ (line 3642) | def __init__(self, *args, **kwargs):
  class ImageGPTForImageClassification (line 3646) | class ImageGPTForImageClassification(metaclass=DummyObject):
    method __init__ (line 3649) | def __init__(self, *args, **kwargs):
  class ImageGPTModel (line 3653) | class ImageGPTModel(metaclass=DummyObject):
    method __init__ (line 3656) | def __init__(self, *args, **kwargs):
  class ImageGPTPreTrainedModel (line 3660) | class ImageGPTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3663) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_imagegpt (line 3667) | def load_tf_weights_in_imagegpt(*args, **kwargs):
  class InformerForPrediction (line 3674) | class InformerForPrediction(metaclass=DummyObject):
    method __init__ (line 3677) | def __init__(self, *args, **kwargs):
  class InformerModel (line 3681) | class InformerModel(metaclass=DummyObject):
    method __init__ (line 3684) | def __init__(self, *args, **kwargs):
  class InformerPreTrainedModel (line 3688) | class InformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3691) | def __init__(self, *args, **kwargs):
  class JukeboxModel (line 3698) | class JukeboxModel(metaclass=DummyObject):
    method __init__ (line 3701) | def __init__(self, *args, **kwargs):
  class JukeboxPreTrainedModel (line 3705) | class JukeboxPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3708) | def __init__(self, *args, **kwargs):
  class JukeboxPrior (line 3712) | class JukeboxPrior(metaclass=DummyObject):
    method __init__ (line 3715) | def __init__(self, *args, **kwargs):
  class JukeboxVQVAE (line 3719) | class JukeboxVQVAE(metaclass=DummyObject):
    method __init__ (line 3722) | def __init__(self, *args, **kwargs):
  class LayoutLMForMaskedLM (line 3729) | class LayoutLMForMaskedLM(metaclass=DummyObject):
    method __init__ (line 3732) | def __init__(self, *args, **kwargs):
  class LayoutLMForQuestionAnswering (line 3736) | class LayoutLMForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3739) | def __init__(self, *args, **kwargs):
  class LayoutLMForSequenceClassification (line 3743) | class LayoutLMForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3746) | def __init__(self, *args, **kwargs):
  class LayoutLMForTokenClassification (line 3750) | class LayoutLMForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3753) | def __init__(self, *args, **kwargs):
  class LayoutLMModel (line 3757) | class LayoutLMModel(metaclass=DummyObject):
    method __init__ (line 3760) | def __init__(self, *args, **kwargs):
  class LayoutLMPreTrainedModel (line 3764) | class LayoutLMPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3767) | def __init__(self, *args, **kwargs):
  class LayoutLMv2ForQuestionAnswering (line 3774) | class LayoutLMv2ForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3777) | def __init__(self, *args, **kwargs):
  class LayoutLMv2ForSequenceClassification (line 3781) | class LayoutLMv2ForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3784) | def __init__(self, *args, **kwargs):
  class LayoutLMv2ForTokenClassification (line 3788) | class LayoutLMv2ForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3791) | def __init__(self, *args, **kwargs):
  class LayoutLMv2Model (line 3795) | class LayoutLMv2Model(metaclass=DummyObject):
    method __init__ (line 3798) | def __init__(self, *args, **kwargs):
  class LayoutLMv2PreTrainedModel (line 3802) | class LayoutLMv2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3805) | def __init__(self, *args, **kwargs):
  class LayoutLMv3ForQuestionAnswering (line 3812) | class LayoutLMv3ForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3815) | def __init__(self, *args, **kwargs):
  class LayoutLMv3ForSequenceClassification (line 3819) | class LayoutLMv3ForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3822) | def __init__(self, *args, **kwargs):
  class LayoutLMv3ForTokenClassification (line 3826) | class LayoutLMv3ForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3829) | def __init__(self, *args, **kwargs):
  class LayoutLMv3Model (line 3833) | class LayoutLMv3Model(metaclass=DummyObject):
    method __init__ (line 3836) | def __init__(self, *args, **kwargs):
  class LayoutLMv3PreTrainedModel (line 3840) | class LayoutLMv3PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3843) | def __init__(self, *args, **kwargs):
  class LEDForConditionalGeneration (line 3850) | class LEDForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 3853) | def __init__(self, *args, **kwargs):
  class LEDForQuestionAnswering (line 3857) | class LEDForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3860) | def __init__(self, *args, **kwargs):
  class LEDForSequenceClassification (line 3864) | class LEDForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3867) | def __init__(self, *args, **kwargs):
  class LEDModel (line 3871) | class LEDModel(metaclass=DummyObject):
    method __init__ (line 3874) | def __init__(self, *args, **kwargs):
  class LEDPreTrainedModel (line 3878) | class LEDPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3881) | def __init__(self, *args, **kwargs):
  class LevitForImageClassification (line 3888) | class LevitForImageClassification(metaclass=DummyObject):
    method __init__ (line 3891) | def __init__(self, *args, **kwargs):
  class LevitForImageClassificationWithTeacher (line 3895) | class LevitForImageClassificationWithTeacher(metaclass=DummyObject):
    method __init__ (line 3898) | def __init__(self, *args, **kwargs):
  class LevitModel (line 3902) | class LevitModel(metaclass=DummyObject):
    method __init__ (line 3905) | def __init__(self, *args, **kwargs):
  class LevitPreTrainedModel (line 3909) | class LevitPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3912) | def __init__(self, *args, **kwargs):
  class LiltForQuestionAnswering (line 3919) | class LiltForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 3922) | def __init__(self, *args, **kwargs):
  class LiltForSequenceClassification (line 3926) | class LiltForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3929) | def __init__(self, *args, **kwargs):
  class LiltForTokenClassification (line 3933) | class LiltForTokenClassification(metaclass=DummyObject):
    method __init__ (line 3936) | def __init__(self, *args, **kwargs):
  class LiltModel (line 3940) | class LiltModel(metaclass=DummyObject):
    method __init__ (line 3943) | def __init__(self, *args, **kwargs):
  class LiltPreTrainedModel (line 3947) | class LiltPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3950) | def __init__(self, *args, **kwargs):
  class LlamaForCausalLM (line 3954) | class LlamaForCausalLM(metaclass=DummyObject):
    method __init__ (line 3957) | def __init__(self, *args, **kwargs):
  class LlamaForSequenceClassification (line 3961) | class LlamaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 3964) | def __init__(self, *args, **kwargs):
  class LlamaModel (line 3968) | class LlamaModel(metaclass=DummyObject):
    method __init__ (line 3971) | def __init__(self, *args, **kwargs):
  class LlamaPreTrainedModel (line 3975) | class LlamaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 3978) | def __init__(self, *args, **kwargs):
  class LongformerForMaskedLM (line 3985) | class LongformerForMaskedLM(metaclass=DummyObject):
    method __init__ (line 3988) | def __init__(self, *args, **kwargs):
  class LongformerForMultipleChoice (line 3992) | class LongformerForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 3995) | def __init__(self, *args, **kwargs):
  class LongformerForQuestionAnswering (line 3999) | class LongformerForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4002) | def __init__(self, *args, **kwargs):
  class LongformerForSequenceClassification (line 4006) | class LongformerForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 4009) | def __init__(self, *args, **kwargs):
  class LongformerForTokenClassification (line 4013) | class LongformerForTokenClassification(metaclass=DummyObject):
    method __init__ (line 4016) | def __init__(self, *args, **kwargs):
  class LongformerModel (line 4020) | class LongformerModel(metaclass=DummyObject):
    method __init__ (line 4023) | def __init__(self, *args, **kwargs):
  class LongformerPreTrainedModel (line 4027) | class LongformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4030) | def __init__(self, *args, **kwargs):
  class LongformerSelfAttention (line 4034) | class LongformerSelfAttention(metaclass=DummyObject):
    method __init__ (line 4037) | def __init__(self, *args, **kwargs):
  class LongT5EncoderModel (line 4044) | class LongT5EncoderModel(metaclass=DummyObject):
    method __init__ (line 4047) | def __init__(self, *args, **kwargs):
  class LongT5ForConditionalGeneration (line 4051) | class LongT5ForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 4054) | def __init__(self, *args, **kwargs):
  class LongT5Model (line 4058) | class LongT5Model(metaclass=DummyObject):
    method __init__ (line 4061) | def __init__(self, *args, **kwargs):
  class LongT5PreTrainedModel (line 4065) | class LongT5PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4068) | def __init__(self, *args, **kwargs):
  class LukeForEntityClassification (line 4075) | class LukeForEntityClassification(metaclass=DummyObject):
    method __init__ (line 4078) | def __init__(self, *args, **kwargs):
  class LukeForEntityPairClassification (line 4082) | class LukeForEntityPairClassification(metaclass=DummyObject):
    method __init__ (line 4085) | def __init__(self, *args, **kwargs):
  class LukeForEntitySpanClassification (line 4089) | class LukeForEntitySpanClassification(metaclass=DummyObject):
    method __init__ (line 4092) | def __init__(self, *args, **kwargs):
  class LukeForMaskedLM (line 4096) | class LukeForMaskedLM(metaclass=DummyObject):
    method __init__ (line 4099) | def __init__(self, *args, **kwargs):
  class LukeForMultipleChoice (line 4103) | class LukeForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 4106) | def __init__(self, *args, **kwargs):
  class LukeForQuestionAnswering (line 4110) | class LukeForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4113) | def __init__(self, *args, **kwargs):
  class LukeForSequenceClassification (line 4117) | class LukeForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 4120) | def __init__(self, *args, **kwargs):
  class LukeForTokenClassification (line 4124) | class LukeForTokenClassification(metaclass=DummyObject):
    method __init__ (line 4127) | def __init__(self, *args, **kwargs):
  class LukeModel (line 4131) | class LukeModel(metaclass=DummyObject):
    method __init__ (line 4134) | def __init__(self, *args, **kwargs):
  class LukePreTrainedModel (line 4138) | class LukePreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4141) | def __init__(self, *args, **kwargs):
  class LxmertEncoder (line 4145) | class LxmertEncoder(metaclass=DummyObject):
    method __init__ (line 4148) | def __init__(self, *args, **kwargs):
  class LxmertForPreTraining (line 4152) | class LxmertForPreTraining(metaclass=DummyObject):
    method __init__ (line 4155) | def __init__(self, *args, **kwargs):
  class LxmertForQuestionAnswering (line 4159) | class LxmertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4162) | def __init__(self, *args, **kwargs):
  class LxmertModel (line 4166) | class LxmertModel(metaclass=DummyObject):
    method __init__ (line 4169) | def __init__(self, *args, **kwargs):
  class LxmertPreTrainedModel (line 4173) | class LxmertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4176) | def __init__(self, *args, **kwargs):
  class LxmertVisualFeatureEncoder (line 4180) | class LxmertVisualFeatureEncoder(metaclass=DummyObject):
    method __init__ (line 4183) | def __init__(self, *args, **kwargs):
  class LxmertXLayer (line 4187) | class LxmertXLayer(metaclass=DummyObject):
    method __init__ (line 4190) | def __init__(self, *args, **kwargs):
  class M2M100ForConditionalGeneration (line 4197) | class M2M100ForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 4200) | def __init__(self, *args, **kwargs):
  class M2M100Model (line 4204) | class M2M100Model(metaclass=DummyObject):
    method __init__ (line 4207) | def __init__(self, *args, **kwargs):
  class M2M100PreTrainedModel (line 4211) | class M2M100PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4214) | def __init__(self, *args, **kwargs):
  class MarianForCausalLM (line 4218) | class MarianForCausalLM(metaclass=DummyObject):
    method __init__ (line 4221) | def __init__(self, *args, **kwargs):
  class MarianModel (line 4225) | class MarianModel(metaclass=DummyObject):
    method __init__ (line 4228) | def __init__(self, *args, **kwargs):
  class MarianMTModel (line 4232) | class MarianMTModel(metaclass=DummyObject):
    method __init__ (line 4235) | def __init__(self, *args, **kwargs):
  class MarkupLMForQuestionAnswering (line 4242) | class MarkupLMForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4245) | def __init__(self, *args, **kwargs):
  class MarkupLMForSequenceClassification (line 4249) | class MarkupLMForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 4252) | def __init__(self, *args, **kwargs):
  class MarkupLMForTokenClassification (line 4256) | class MarkupLMForTokenClassification(metaclass=DummyObject):
    method __init__ (line 4259) | def __init__(self, *args, **kwargs):
  class MarkupLMModel (line 4263) | class MarkupLMModel(metaclass=DummyObject):
    method __init__ (line 4266) | def __init__(self, *args, **kwargs):
  class MarkupLMPreTrainedModel (line 4270) | class MarkupLMPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4273) | def __init__(self, *args, **kwargs):
  class Mask2FormerForUniversalSegmentation (line 4280) | class Mask2FormerForUniversalSegmentation(metaclass=DummyObject):
    method __init__ (line 4283) | def __init__(self, *args, **kwargs):
  class Mask2FormerModel (line 4287) | class Mask2FormerModel(metaclass=DummyObject):
    method __init__ (line 4290) | def __init__(self, *args, **kwargs):
  class Mask2FormerPreTrainedModel (line 4294) | class Mask2FormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4297) | def __init__(self, *args, **kwargs):
  class MaskFormerForInstanceSegmentation (line 4304) | class MaskFormerForInstanceSegmentation(metaclass=DummyObject):
    method __init__ (line 4307) | def __init__(self, *args, **kwargs):
  class MaskFormerModel (line 4311) | class MaskFormerModel(metaclass=DummyObject):
    method __init__ (line 4314) | def __init__(self, *args, **kwargs):
  class MaskFormerPreTrainedModel (line 4318) | class MaskFormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4321) | def __init__(self, *args, **kwargs):
  class MaskFormerSwinBackbone (line 4325) | class MaskFormerSwinBackbone(metaclass=DummyObject):
    method __init__ (line 4328) | def __init__(self, *args, **kwargs):
  class MBartForCausalLM (line 4332) | class MBartForCausalLM(metaclass=DummyObject):
    method __init__ (line 4335) | def __init__(self, *args, **kwargs):
  class MBartForConditionalGeneration (line 4339) | class MBartForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 4342) | def __init__(self, *args, **kwargs):
  class MBartForQuestionAnswering (line 4346) | class MBartForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4349) | def __init__(self, *args, **kwargs):
  class MBartForSequenceClassification (line 4353) | class MBartForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 4356) | def __init__(self, *args, **kwargs):
  class MBartModel (line 4360) | class MBartModel(metaclass=DummyObject):
    method __init__ (line 4363) | def __init__(self, *args, **kwargs):
  class MBartPreTrainedModel (line 4367) | class MBartPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4370) | def __init__(self, *args, **kwargs):
  class MCTCTForCTC (line 4377) | class MCTCTForCTC(metaclass=DummyObject):
    method __init__ (line 4380) | def __init__(self, *args, **kwargs):
  class MCTCTModel (line 4384) | class MCTCTModel(metaclass=DummyObject):
    method __init__ (line 4387) | def __init__(self, *args, **kwargs):
  class MCTCTPreTrainedModel (line 4391) | class MCTCTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4394) | def __init__(self, *args, **kwargs):
  class MegaForCausalLM (line 4401) | class MegaForCausalLM(metaclass=DummyObject):
    method __init__ (line 4404) | def __init__(self, *args, **kwargs):
  class MegaForMaskedLM (line 4408) | class MegaForMaskedLM(metaclass=DummyObject):
    method __init__ (line 4411) | def __init__(self, *args, **kwargs):
  class MegaForMultipleChoice (line 4415) | class MegaForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 4418) | def __init__(self, *args, **kwargs):
  class MegaForQuestionAnswering (line 4422) | class MegaForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4425) | def __init__(self, *args, **kwargs):
  class MegaForSequenceClassification (line 4429) | class MegaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 4432) | def __init__(self, *args, **kwargs):
  class MegaForTokenClassification (line 4436) | class MegaForTokenClassification(metaclass=DummyObject):
    method __init__ (line 4439) | def __init__(self, *args, **kwargs):
  class MegaModel (line 4443) | class MegaModel(metaclass=DummyObject):
    method __init__ (line 4446) | def __init__(self, *args, **kwargs):
  class MegaPreTrainedModel (line 4450) | class MegaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4453) | def __init__(self, *args, **kwargs):
  class MegatronBertForCausalLM (line 4460) | class MegatronBertForCausalLM(metaclass=DummyObject):
    method __init__ (line 4463) | def __init__(self, *args, **kwargs):
  class MegatronBertForMaskedLM (line 4467) | class MegatronBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 4470) | def __init__(self, *args, **kwargs):
  class MegatronBertForMultipleChoice (line 4474) | class MegatronBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 4477) | def __init__(self, *args, **kwargs):
  class MegatronBertForNextSentencePrediction (line 4481) | class MegatronBertForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 4484) | def __init__(self, *args, **kwargs):
  class MegatronBertForPreTraining (line 4488) | class MegatronBertForPreTraining(metaclass=DummyObject):
    method __init__ (line 4491) | def __init__(self, *args, **kwargs):
  class MegatronBertForQuestionAnswering (line 4495) | class MegatronBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4498) | def __init__(self, *args, **kwargs):
  class MegatronBertForSequenceClassification (line 4502) | class MegatronBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 4505) | def __init__(self, *args, **kwargs):
  class MegatronBertForTokenClassification (line 4509) | class MegatronBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 4512) | def __init__(self, *args, **kwargs):
  class MegatronBertModel (line 4516) | class MegatronBertModel(metaclass=DummyObject):
    method __init__ (line 4519) | def __init__(self, *args, **kwargs):
  class MegatronBertPreTrainedModel (line 4523) | class MegatronBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4526) | def __init__(self, *args, **kwargs):
  class MgpstrForSceneTextRecognition (line 4533) | class MgpstrForSceneTextRecognition(metaclass=DummyObject):
    method __init__ (line 4536) | def __init__(self, *args, **kwargs):
  class MgpstrModel (line 4540) | class MgpstrModel(metaclass=DummyObject):
    method __init__ (line 4543) | def __init__(self, *args, **kwargs):
  class MgpstrPreTrainedModel (line 4547) | class MgpstrPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4550) | def __init__(self, *args, **kwargs):
  class MMBTForClassification (line 4554) | class MMBTForClassification(metaclass=DummyObject):
    method __init__ (line 4557) | def __init__(self, *args, **kwargs):
  class MMBTModel (line 4561) | class MMBTModel(metaclass=DummyObject):
    method __init__ (line 4564) | def __init__(self, *args, **kwargs):
  class ModalEmbeddings (line 4568) | class ModalEmbeddings(metaclass=DummyObject):
    method __init__ (line 4571) | def __init__(self, *args, **kwargs):
  class MobileBertForMaskedLM (line 4578) | class MobileBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 4581) | def __init__(self, *args, **kwargs):
  class MobileBertForMultipleChoice (line 4585) | class MobileBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 4588) | def __init__(self, *args, **kwargs):
  class MobileBertForNextSentencePrediction (line 4592) | class MobileBertForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 4595) | def __init__(self, *args, **kwargs):
  class MobileBertForPreTraining (line 4599) | class MobileBertForPreTraining(metaclass=DummyObject):
    method __init__ (line 4602) | def __init__(self, *args, **kwargs):
  class MobileBertForQuestionAnswering (line 4606) | class MobileBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4609) | def __init__(self, *args, **kwargs):
  class MobileBertForSequenceClassification (line 4613) | class MobileBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 4616) | def __init__(self, *args, **kwargs):
  class MobileBertForTokenClassification (line 4620) | class MobileBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 4623) | def __init__(self, *args, **kwargs):
  class MobileBertLayer (line 4627) | class MobileBertLayer(metaclass=DummyObject):
    method __init__ (line 4630) | def __init__(self, *args, **kwargs):
  class MobileBertModel (line 4634) | class MobileBertModel(metaclass=DummyObject):
    method __init__ (line 4637) | def __init__(self, *args, **kwargs):
  class MobileBertPreTrainedModel (line 4641) | class MobileBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4644) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_mobilebert (line 4648) | def load_tf_weights_in_mobilebert(*args, **kwargs):
  class MobileNetV1ForImageClassification (line 4655) | class MobileNetV1ForImageClassification(metaclass=DummyObject):
    method __init__ (line 4658) | def __init__(self, *args, **kwargs):
  class MobileNetV1Model (line 4662) | class MobileNetV1Model(metaclass=DummyObject):
    method __init__ (line 4665) | def __init__(self, *args, **kwargs):
  class MobileNetV1PreTrainedModel (line 4669) | class MobileNetV1PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4672) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_mobilenet_v1 (line 4676) | def load_tf_weights_in_mobilenet_v1(*args, **kwargs):
  class MobileNetV2ForImageClassification (line 4683) | class MobileNetV2ForImageClassification(metaclass=DummyObject):
    method __init__ (line 4686) | def __init__(self, *args, **kwargs):
  class MobileNetV2ForSemanticSegmentation (line 4690) | class MobileNetV2ForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 4693) | def __init__(self, *args, **kwargs):
  class MobileNetV2Model (line 4697) | class MobileNetV2Model(metaclass=DummyObject):
    method __init__ (line 4700) | def __init__(self, *args, **kwargs):
  class MobileNetV2PreTrainedModel (line 4704) | class MobileNetV2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4707) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_mobilenet_v2 (line 4711) | def load_tf_weights_in_mobilenet_v2(*args, **kwargs):
  class MobileViTForImageClassification (line 4718) | class MobileViTForImageClassification(metaclass=DummyObject):
    method __init__ (line 4721) | def __init__(self, *args, **kwargs):
  class MobileViTForSemanticSegmentation (line 4725) | class MobileViTForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 4728) | def __init__(self, *args, **kwargs):
  class MobileViTModel (line 4732) | class MobileViTModel(metaclass=DummyObject):
    method __init__ (line 4735) | def __init__(self, *args, **kwargs):
  class MobileViTPreTrainedModel (line 4739) | class MobileViTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4742) | def __init__(self, *args, **kwargs):
  class MobileViTV2ForImageClassification (line 4749) | class MobileViTV2ForImageClassification(metaclass=DummyObject):
    method __init__ (line 4752) | def __init__(self, *args, **kwargs):
  class MobileViTV2ForSemanticSegmentation (line 4756) | class MobileViTV2ForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 4759) | def __init__(self, *args, **kwargs):
  class MobileViTV2Model (line 4763) | class MobileViTV2Model(metaclass=DummyObject):
    method __init__ (line 4766) | def __init__(self, *args, **kwargs):
  class MobileViTV2PreTrainedModel (line 4770) | class MobileViTV2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4773) | def __init__(self, *args, **kwargs):
  class MPNetForMaskedLM (line 4780) | class MPNetForMaskedLM(metaclass=DummyObject):
    method __init__ (line 4783) | def __init__(self, *args, **kwargs):
  class MPNetForMultipleChoice (line 4787) | class MPNetForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 4790) | def __init__(self, *args, **kwargs):
  class MPNetForQuestionAnswering (line 4794) | class MPNetForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4797) | def __init__(self, *args, **kwargs):
  class MPNetForSequenceClassification (line 4801) | class MPNetForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 4804) | def __init__(self, *args, **kwargs):
  class MPNetForTokenClassification (line 4808) | class MPNetForTokenClassification(metaclass=DummyObject):
    method __init__ (line 4811) | def __init__(self, *args, **kwargs):
  class MPNetLayer (line 4815) | class MPNetLayer(metaclass=DummyObject):
    method __init__ (line 4818) | def __init__(self, *args, **kwargs):
  class MPNetModel (line 4822) | class MPNetModel(metaclass=DummyObject):
    method __init__ (line 4825) | def __init__(self, *args, **kwargs):
  class MPNetPreTrainedModel (line 4829) | class MPNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4832) | def __init__(self, *args, **kwargs):
  class MT5EncoderModel (line 4836) | class MT5EncoderModel(metaclass=DummyObject):
    method __init__ (line 4839) | def __init__(self, *args, **kwargs):
  class MT5ForConditionalGeneration (line 4843) | class MT5ForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 4846) | def __init__(self, *args, **kwargs):
  class MT5Model (line 4850) | class MT5Model(metaclass=DummyObject):
    method __init__ (line 4853) | def __init__(self, *args, **kwargs):
  class MT5PreTrainedModel (line 4857) | class MT5PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4860) | def __init__(self, *args, **kwargs):
  class MvpForCausalLM (line 4867) | class MvpForCausalLM(metaclass=DummyObject):
    method __init__ (line 4870) | def __init__(self, *args, **kwargs):
  class MvpForConditionalGeneration (line 4874) | class MvpForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 4877) | def __init__(self, *args, **kwargs):
  class MvpForQuestionAnswering (line 4881) | class MvpForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4884) | def __init__(self, *args, **kwargs):
  class MvpForSequenceClassification (line 4888) | class MvpForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 4891) | def __init__(self, *args, **kwargs):
  class MvpModel (line 4895) | class MvpModel(metaclass=DummyObject):
    method __init__ (line 4898) | def __init__(self, *args, **kwargs):
  class MvpPreTrainedModel (line 4902) | class MvpPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4905) | def __init__(self, *args, **kwargs):
  class NatBackbone (line 4912) | class NatBackbone(metaclass=DummyObject):
    method __init__ (line 4915) | def __init__(self, *args, **kwargs):
  class NatForImageClassification (line 4919) | class NatForImageClassification(metaclass=DummyObject):
    method __init__ (line 4922) | def __init__(self, *args, **kwargs):
  class NatModel (line 4926) | class NatModel(metaclass=DummyObject):
    method __init__ (line 4929) | def __init__(self, *args, **kwargs):
  class NatPreTrainedModel (line 4933) | class NatPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 4936) | def __init__(self, *args, **kwargs):
  class NezhaForMaskedLM (line 4943) | class NezhaForMaskedLM(metaclass=DummyObject):
    method __init__ (line 4946) | def __init__(self, *args, **kwargs):
  class NezhaForMultipleChoice (line 4950) | class NezhaForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 4953) | def __init__(self, *args, **kwargs):
  class NezhaForNextSentencePrediction (line 4957) | class NezhaForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 4960) | def __init__(self, *args, **kwargs):
  class NezhaForPreTraining (line 4964) | class NezhaForPreTraining(metaclass=DummyObject):
    method __init__ (line 4967) | def __init__(self, *args, **kwargs):
  class NezhaForQuestionAnswering (line 4971) | class NezhaForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 4974) | def __init__(self, *args, **kwargs):
  class NezhaForSequenceClassification (line 4978) | class NezhaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 4981) | def __init__(self, *args, **kwargs):
  class NezhaForTokenClassification (line 4985) | class NezhaForTokenClassification(metaclass=DummyObject):
    method __init__ (line 4988) | def __init__(self, *args, **kwargs):
  class NezhaModel (line 4992) | class NezhaModel(metaclass=DummyObject):
    method __init__ (line 4995) | def __init__(self, *args, **kwargs):
  class NezhaPreTrainedModel (line 4999) | class NezhaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5002) | def __init__(self, *args, **kwargs):
  class NllbMoeForConditionalGeneration (line 5009) | class NllbMoeForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 5012) | def __init__(self, *args, **kwargs):
  class NllbMoeModel (line 5016) | class NllbMoeModel(metaclass=DummyObject):
    method __init__ (line 5019) | def __init__(self, *args, **kwargs):
  class NllbMoePreTrainedModel (line 5023) | class NllbMoePreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5026) | def __init__(self, *args, **kwargs):
  class NllbMoeSparseMLP (line 5030) | class NllbMoeSparseMLP(metaclass=DummyObject):
    method __init__ (line 5033) | def __init__(self, *args, **kwargs):
  class NllbMoeTop2Router (line 5037) | class NllbMoeTop2Router(metaclass=DummyObject):
    method __init__ (line 5040) | def __init__(self, *args, **kwargs):
  class NystromformerForMaskedLM (line 5047) | class NystromformerForMaskedLM(metaclass=DummyObject):
    method __init__ (line 5050) | def __init__(self, *args, **kwargs):
  class NystromformerForMultipleChoice (line 5054) | class NystromformerForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 5057) | def __init__(self, *args, **kwargs):
  class NystromformerForQuestionAnswering (line 5061) | class NystromformerForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 5064) | def __init__(self, *args, **kwargs):
  class NystromformerForSequenceClassification (line 5068) | class NystromformerForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5071) | def __init__(self, *args, **kwargs):
  class NystromformerForTokenClassification (line 5075) | class NystromformerForTokenClassification(metaclass=DummyObject):
    method __init__ (line 5078) | def __init__(self, *args, **kwargs):
  class NystromformerLayer (line 5082) | class NystromformerLayer(metaclass=DummyObject):
    method __init__ (line 5085) | def __init__(self, *args, **kwargs):
  class NystromformerModel (line 5089) | class NystromformerModel(metaclass=DummyObject):
    method __init__ (line 5092) | def __init__(self, *args, **kwargs):
  class NystromformerPreTrainedModel (line 5096) | class NystromformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5099) | def __init__(self, *args, **kwargs):
  class OneFormerForUniversalSegmentation (line 5106) | class OneFormerForUniversalSegmentation(metaclass=DummyObject):
    method __init__ (line 5109) | def __init__(self, *args, **kwargs):
  class OneFormerModel (line 5113) | class OneFormerModel(metaclass=DummyObject):
    method __init__ (line 5116) | def __init__(self, *args, **kwargs):
  class OneFormerPreTrainedModel (line 5120) | class OneFormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5123) | def __init__(self, *args, **kwargs):
  class OpenLlamaForCausalLM (line 5127) | class OpenLlamaForCausalLM(metaclass=DummyObject):
    method __init__ (line 5130) | def __init__(self, *args, **kwargs):
  class OpenLlamaForSequenceClassification (line 5134) | class OpenLlamaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5137) | def __init__(self, *args, **kwargs):
  class OpenLlamaModel (line 5141) | class OpenLlamaModel(metaclass=DummyObject):
    method __init__ (line 5144) | def __init__(self, *args, **kwargs):
  class OpenLlamaPreTrainedModel (line 5148) | class OpenLlamaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5151) | def __init__(self, *args, **kwargs):
  class OpenAIGPTDoubleHeadsModel (line 5158) | class OpenAIGPTDoubleHeadsModel(metaclass=DummyObject):
    method __init__ (line 5161) | def __init__(self, *args, **kwargs):
  class OpenAIGPTForSequenceClassification (line 5165) | class OpenAIGPTForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5168) | def __init__(self, *args, **kwargs):
  class OpenAIGPTLMHeadModel (line 5172) | class OpenAIGPTLMHeadModel(metaclass=DummyObject):
    method __init__ (line 5175) | def __init__(self, *args, **kwargs):
  class OpenAIGPTModel (line 5179) | class OpenAIGPTModel(metaclass=DummyObject):
    method __init__ (line 5182) | def __init__(self, *args, **kwargs):
  class OpenAIGPTPreTrainedModel (line 5186) | class OpenAIGPTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5189) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_openai_gpt (line 5193) | def load_tf_weights_in_openai_gpt(*args, **kwargs):
  class OPTForCausalLM (line 5200) | class OPTForCausalLM(metaclass=DummyObject):
    method __init__ (line 5203) | def __init__(self, *args, **kwargs):
  class OPTForQuestionAnswering (line 5207) | class OPTForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 5210) | def __init__(self, *args, **kwargs):
  class OPTForSequenceClassification (line 5214) | class OPTForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5217) | def __init__(self, *args, **kwargs):
  class OPTModel (line 5221) | class OPTModel(metaclass=DummyObject):
    method __init__ (line 5224) | def __init__(self, *args, **kwargs):
  class OPTPreTrainedModel (line 5228) | class OPTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5231) | def __init__(self, *args, **kwargs):
  class OwlViTForObjectDetection (line 5238) | class OwlViTForObjectDetection(metaclass=DummyObject):
    method __init__ (line 5241) | def __init__(self, *args, **kwargs):
  class OwlViTModel (line 5245) | class OwlViTModel(metaclass=DummyObject):
    method __init__ (line 5248) | def __init__(self, *args, **kwargs):
  class OwlViTPreTrainedModel (line 5252) | class OwlViTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5255) | def __init__(self, *args, **kwargs):
  class OwlViTTextModel (line 5259) | class OwlViTTextModel(metaclass=DummyObject):
    method __init__ (line 5262) | def __init__(self, *args, **kwargs):
  class OwlViTVisionModel (line 5266) | class OwlViTVisionModel(metaclass=DummyObject):
    method __init__ (line 5269) | def __init__(self, *args, **kwargs):
  class PegasusForCausalLM (line 5273) | class PegasusForCausalLM(metaclass=DummyObject):
    method __init__ (line 5276) | def __init__(self, *args, **kwargs):
  class PegasusForConditionalGeneration (line 5280) | class PegasusForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 5283) | def __init__(self, *args, **kwargs):
  class PegasusModel (line 5287) | class PegasusModel(metaclass=DummyObject):
    method __init__ (line 5290) | def __init__(self, *args, **kwargs):
  class PegasusPreTrainedModel (line 5294) | class PegasusPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5297) | def __init__(self, *args, **kwargs):
  class PegasusXForConditionalGeneration (line 5304) | class PegasusXForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 5307) | def __init__(self, *args, **kwargs):
  class PegasusXModel (line 5311) | class PegasusXModel(metaclass=DummyObject):
    method __init__ (line 5314) | def __init__(self, *args, **kwargs):
  class PegasusXPreTrainedModel (line 5318) | class PegasusXPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5321) | def __init__(self, *args, **kwargs):
  class PerceiverForImageClassificationConvProcessing (line 5328) | class PerceiverForImageClassificationConvProcessing(metaclass=DummyObject):
    method __init__ (line 5331) | def __init__(self, *args, **kwargs):
  class PerceiverForImageClassificationFourier (line 5335) | class PerceiverForImageClassificationFourier(metaclass=DummyObject):
    method __init__ (line 5338) | def __init__(self, *args, **kwargs):
  class PerceiverForImageClassificationLearned (line 5342) | class PerceiverForImageClassificationLearned(metaclass=DummyObject):
    method __init__ (line 5345) | def __init__(self, *args, **kwargs):
  class PerceiverForMaskedLM (line 5349) | class PerceiverForMaskedLM(metaclass=DummyObject):
    method __init__ (line 5352) | def __init__(self, *args, **kwargs):
  class PerceiverForMultimodalAutoencoding (line 5356) | class PerceiverForMultimodalAutoencoding(metaclass=DummyObject):
    method __init__ (line 5359) | def __init__(self, *args, **kwargs):
  class PerceiverForOpticalFlow (line 5363) | class PerceiverForOpticalFlow(metaclass=DummyObject):
    method __init__ (line 5366) | def __init__(self, *args, **kwargs):
  class PerceiverForSequenceClassification (line 5370) | class PerceiverForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5373) | def __init__(self, *args, **kwargs):
  class PerceiverLayer (line 5377) | class PerceiverLayer(metaclass=DummyObject):
    method __init__ (line 5380) | def __init__(self, *args, **kwargs):
  class PerceiverModel (line 5384) | class PerceiverModel(metaclass=DummyObject):
    method __init__ (line 5387) | def __init__(self, *args, **kwargs):
  class PerceiverPreTrainedModel (line 5391) | class PerceiverPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5394) | def __init__(self, *args, **kwargs):
  class Pix2StructForConditionalGeneration (line 5401) | class Pix2StructForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 5404) | def __init__(self, *args, **kwargs):
  class Pix2StructPreTrainedModel (line 5408) | class Pix2StructPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5411) | def __init__(self, *args, **kwargs):
  class Pix2StructTextModel (line 5415) | class Pix2StructTextModel(metaclass=DummyObject):
    method __init__ (line 5418) | def __init__(self, *args, **kwargs):
  class Pix2StructVisionModel (line 5422) | class Pix2StructVisionModel(metaclass=DummyObject):
    method __init__ (line 5425) | def __init__(self, *args, **kwargs):
  class PLBartForCausalLM (line 5432) | class PLBartForCausalLM(metaclass=DummyObject):
    method __init__ (line 5435) | def __init__(self, *args, **kwargs):
  class PLBartForConditionalGeneration (line 5439) | class PLBartForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 5442) | def __init__(self, *args, **kwargs):
  class PLBartForSequenceClassification (line 5446) | class PLBartForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5449) | def __init__(self, *args, **kwargs):
  class PLBartModel (line 5453) | class PLBartModel(metaclass=DummyObject):
    method __init__ (line 5456) | def __init__(self, *args, **kwargs):
  class PLBartPreTrainedModel (line 5460) | class PLBartPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5463) | def __init__(self, *args, **kwargs):
  class PoolFormerForImageClassification (line 5470) | class PoolFormerForImageClassification(metaclass=DummyObject):
    method __init__ (line 5473) | def __init__(self, *args, **kwargs):
  class PoolFormerModel (line 5477) | class PoolFormerModel(metaclass=DummyObject):
    method __init__ (line 5480) | def __init__(self, *args, **kwargs):
  class PoolFormerPreTrainedModel (line 5484) | class PoolFormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5487) | def __init__(self, *args, **kwargs):
  class ProphetNetDecoder (line 5494) | class ProphetNetDecoder(metaclass=DummyObject):
    method __init__ (line 5497) | def __init__(self, *args, **kwargs):
  class ProphetNetEncoder (line 5501) | class ProphetNetEncoder(metaclass=DummyObject):
    method __init__ (line 5504) | def __init__(self, *args, **kwargs):
  class ProphetNetForCausalLM (line 5508) | class ProphetNetForCausalLM(metaclass=DummyObject):
    method __init__ (line 5511) | def __init__(self, *args, **kwargs):
  class ProphetNetForConditionalGeneration (line 5515) | class ProphetNetForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 5518) | def __init__(self, *args, **kwargs):
  class ProphetNetModel (line 5522) | class ProphetNetModel(metaclass=DummyObject):
    method __init__ (line 5525) | def __init__(self, *args, **kwargs):
  class ProphetNetPreTrainedModel (line 5529) | class ProphetNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5532) | def __init__(self, *args, **kwargs):
  class QDQBertForMaskedLM (line 5539) | class QDQBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 5542) | def __init__(self, *args, **kwargs):
  class QDQBertForMultipleChoice (line 5546) | class QDQBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 5549) | def __init__(self, *args, **kwargs):
  class QDQBertForNextSentencePrediction (line 5553) | class QDQBertForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 5556) | def __init__(self, *args, **kwargs):
  class QDQBertForQuestionAnswering (line 5560) | class QDQBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 5563) | def __init__(self, *args, **kwargs):
  class QDQBertForSequenceClassification (line 5567) | class QDQBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5570) | def __init__(self, *args, **kwargs):
  class QDQBertForTokenClassification (line 5574) | class QDQBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 5577) | def __init__(self, *args, **kwargs):
  class QDQBertLayer (line 5581) | class QDQBertLayer(metaclass=DummyObject):
    method __init__ (line 5584) | def __init__(self, *args, **kwargs):
  class QDQBertLMHeadModel (line 5588) | class QDQBertLMHeadModel(metaclass=DummyObject):
    method __init__ (line 5591) | def __init__(self, *args, **kwargs):
  class QDQBertModel (line 5595) | class QDQBertModel(metaclass=DummyObject):
    method __init__ (line 5598) | def __init__(self, *args, **kwargs):
  class QDQBertPreTrainedModel (line 5602) | class QDQBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5605) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_qdqbert (line 5609) | def load_tf_weights_in_qdqbert(*args, **kwargs):
  class RagModel (line 5613) | class RagModel(metaclass=DummyObject):
    method __init__ (line 5616) | def __init__(self, *args, **kwargs):
  class RagPreTrainedModel (line 5620) | class RagPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5623) | def __init__(self, *args, **kwargs):
  class RagSequenceForGeneration (line 5627) | class RagSequenceForGeneration(metaclass=DummyObject):
    method __init__ (line 5630) | def __init__(self, *args, **kwargs):
  class RagTokenForGeneration (line 5634) | class RagTokenForGeneration(metaclass=DummyObject):
    method __init__ (line 5637) | def __init__(self, *args, **kwargs):
  class RealmEmbedder (line 5644) | class RealmEmbedder(metaclass=DummyObject):
    method __init__ (line 5647) | def __init__(self, *args, **kwargs):
  class RealmForOpenQA (line 5651) | class RealmForOpenQA(metaclass=DummyObject):
    method __init__ (line 5654) | def __init__(self, *args, **kwargs):
  class RealmKnowledgeAugEncoder (line 5658) | class RealmKnowledgeAugEncoder(metaclass=DummyObject):
    method __init__ (line 5661) | def __init__(self, *args, **kwargs):
  class RealmPreTrainedModel (line 5665) | class RealmPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5668) | def __init__(self, *args, **kwargs):
  class RealmReader (line 5672) | class RealmReader(metaclass=DummyObject):
    method __init__ (line 5675) | def __init__(self, *args, **kwargs):
  class RealmRetriever (line 5679) | class RealmRetriever(metaclass=DummyObject):
    method __init__ (line 5682) | def __init__(self, *args, **kwargs):
  class RealmScorer (line 5686) | class RealmScorer(metaclass=DummyObject):
    method __init__ (line 5689) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_realm (line 5693) | def load_tf_weights_in_realm(*args, **kwargs):
  class ReformerAttention (line 5700) | class ReformerAttention(metaclass=DummyObject):
    method __init__ (line 5703) | def __init__(self, *args, **kwargs):
  class ReformerForMaskedLM (line 5707) | class ReformerForMaskedLM(metaclass=DummyObject):
    method __init__ (line 5710) | def __init__(self, *args, **kwargs):
  class ReformerForQuestionAnswering (line 5714) | class ReformerForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 5717) | def __init__(self, *args, **kwargs):
  class ReformerForSequenceClassification (line 5721) | class ReformerForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5724) | def __init__(self, *args, **kwargs):
  class ReformerLayer (line 5728) | class ReformerLayer(metaclass=DummyObject):
    method __init__ (line 5731) | def __init__(self, *args, **kwargs):
  class ReformerModel (line 5735) | class ReformerModel(metaclass=DummyObject):
    method __init__ (line 5738) | def __init__(self, *args, **kwargs):
  class ReformerModelWithLMHead (line 5742) | class ReformerModelWithLMHead(metaclass=DummyObject):
    method __init__ (line 5745) | def __init__(self, *args, **kwargs):
  class ReformerPreTrainedModel (line 5749) | class ReformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5752) | def __init__(self, *args, **kwargs):
  class RegNetForImageClassification (line 5759) | class RegNetForImageClassification(metaclass=DummyObject):
    method __init__ (line 5762) | def __init__(self, *args, **kwargs):
  class RegNetModel (line 5766) | class RegNetModel(metaclass=DummyObject):
    method __init__ (line 5769) | def __init__(self, *args, **kwargs):
  class RegNetPreTrainedModel (line 5773) | class RegNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5776) | def __init__(self, *args, **kwargs):
  class RemBertForCausalLM (line 5783) | class RemBertForCausalLM(metaclass=DummyObject):
    method __init__ (line 5786) | def __init__(self, *args, **kwargs):
  class RemBertForMaskedLM (line 5790) | class RemBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 5793) | def __init__(self, *args, **kwargs):
  class RemBertForMultipleChoice (line 5797) | class RemBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 5800) | def __init__(self, *args, **kwargs):
  class RemBertForQuestionAnswering (line 5804) | class RemBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 5807) | def __init__(self, *args, **kwargs):
  class RemBertForSequenceClassification (line 5811) | class RemBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5814) | def __init__(self, *args, **kwargs):
  class RemBertForTokenClassification (line 5818) | class RemBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 5821) | def __init__(self, *args, **kwargs):
  class RemBertLayer (line 5825) | class RemBertLayer(metaclass=DummyObject):
    method __init__ (line 5828) | def __init__(self, *args, **kwargs):
  class RemBertModel (line 5832) | class RemBertModel(metaclass=DummyObject):
    method __init__ (line 5835) | def __init__(self, *args, **kwargs):
  class RemBertPreTrainedModel (line 5839) | class RemBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5842) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_rembert (line 5846) | def load_tf_weights_in_rembert(*args, **kwargs):
  class ResNetBackbone (line 5853) | class ResNetBackbone(metaclass=DummyObject):
    method __init__ (line 5856) | def __init__(self, *args, **kwargs):
  class ResNetForImageClassification (line 5860) | class ResNetForImageClassification(metaclass=DummyObject):
    method __init__ (line 5863) | def __init__(self, *args, **kwargs):
  class ResNetModel (line 5867) | class ResNetModel(metaclass=DummyObject):
    method __init__ (line 5870) | def __init__(self, *args, **kwargs):
  class ResNetPreTrainedModel (line 5874) | class ResNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5877) | def __init__(self, *args, **kwargs):
  class RetriBertModel (line 5884) | class RetriBertModel(metaclass=DummyObject):
    method __init__ (line 5887) | def __init__(self, *args, **kwargs):
  class RetriBertPreTrainedModel (line 5891) | class RetriBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5894) | def __init__(self, *args, **kwargs):
  class RobertaForCausalLM (line 5901) | class RobertaForCausalLM(metaclass=DummyObject):
    method __init__ (line 5904) | def __init__(self, *args, **kwargs):
  class RobertaForMaskedLM (line 5908) | class RobertaForMaskedLM(metaclass=DummyObject):
    method __init__ (line 5911) | def __init__(self, *args, **kwargs):
  class RobertaForMultipleChoice (line 5915) | class RobertaForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 5918) | def __init__(self, *args, **kwargs):
  class RobertaForQuestionAnswering (line 5922) | class RobertaForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 5925) | def __init__(self, *args, **kwargs):
  class RobertaForSequenceClassification (line 5929) | class RobertaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5932) | def __init__(self, *args, **kwargs):
  class RobertaForTokenClassification (line 5936) | class RobertaForTokenClassification(metaclass=DummyObject):
    method __init__ (line 5939) | def __init__(self, *args, **kwargs):
  class RobertaModel (line 5943) | class RobertaModel(metaclass=DummyObject):
    method __init__ (line 5946) | def __init__(self, *args, **kwargs):
  class RobertaPreTrainedModel (line 5950) | class RobertaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 5953) | def __init__(self, *args, **kwargs):
  class RobertaPreLayerNormForCausalLM (line 5960) | class RobertaPreLayerNormForCausalLM(metaclass=DummyObject):
    method __init__ (line 5963) | def __init__(self, *args, **kwargs):
  class RobertaPreLayerNormForMaskedLM (line 5967) | class RobertaPreLayerNormForMaskedLM(metaclass=DummyObject):
    method __init__ (line 5970) | def __init__(self, *args, **kwargs):
  class RobertaPreLayerNormForMultipleChoice (line 5974) | class RobertaPreLayerNormForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 5977) | def __init__(self, *args, **kwargs):
  class RobertaPreLayerNormForQuestionAnswering (line 5981) | class RobertaPreLayerNormForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 5984) | def __init__(self, *args, **kwargs):
  class RobertaPreLayerNormForSequenceClassification (line 5988) | class RobertaPreLayerNormForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 5991) | def __init__(self, *args, **kwargs):
  class RobertaPreLayerNormForTokenClassification (line 5995) | class RobertaPreLayerNormForTokenClassification(metaclass=DummyObject):
    method __init__ (line 5998) | def __init__(self, *args, **kwargs):
  class RobertaPreLayerNormModel (line 6002) | class RobertaPreLayerNormModel(metaclass=DummyObject):
    method __init__ (line 6005) | def __init__(self, *args, **kwargs):
  class RobertaPreLayerNormPreTrainedModel (line 6009) | class RobertaPreLayerNormPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6012) | def __init__(self, *args, **kwargs):
  class RoCBertForCausalLM (line 6019) | class RoCBertForCausalLM(metaclass=DummyObject):
    method __init__ (line 6022) | def __init__(self, *args, **kwargs):
  class RoCBertForMaskedLM (line 6026) | class RoCBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 6029) | def __init__(self, *args, **kwargs):
  class RoCBertForMultipleChoice (line 6033) | class RoCBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 6036) | def __init__(self, *args, **kwargs):
  class RoCBertForPreTraining (line 6040) | class RoCBertForPreTraining(metaclass=DummyObject):
    method __init__ (line 6043) | def __init__(self, *args, **kwargs):
  class RoCBertForQuestionAnswering (line 6047) | class RoCBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 6050) | def __init__(self, *args, **kwargs):
  class RoCBertForSequenceClassification (line 6054) | class RoCBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 6057) | def __init__(self, *args, **kwargs):
  class RoCBertForTokenClassification (line 6061) | class RoCBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 6064) | def __init__(self, *args, **kwargs):
  class RoCBertLayer (line 6068) | class RoCBertLayer(metaclass=DummyObject):
    method __init__ (line 6071) | def __init__(self, *args, **kwargs):
  class RoCBertModel (line 6075) | class RoCBertModel(metaclass=DummyObject):
    method __init__ (line 6078) | def __init__(self, *args, **kwargs):
  class RoCBertPreTrainedModel (line 6082) | class RoCBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6085) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_roc_bert (line 6089) | def load_tf_weights_in_roc_bert(*args, **kwargs):
  class RoFormerForCausalLM (line 6096) | class RoFormerForCausalLM(metaclass=DummyObject):
    method __init__ (line 6099) | def __init__(self, *args, **kwargs):
  class RoFormerForMaskedLM (line 6103) | class RoFormerForMaskedLM(metaclass=DummyObject):
    method __init__ (line 6106) | def __init__(self, *args, **kwargs):
  class RoFormerForMultipleChoice (line 6110) | class RoFormerForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 6113) | def __init__(self, *args, **kwargs):
  class RoFormerForQuestionAnswering (line 6117) | class RoFormerForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 6120) | def __init__(self, *args, **kwargs):
  class RoFormerForSequenceClassification (line 6124) | class RoFormerForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 6127) | def __init__(self, *args, **kwargs):
  class RoFormerForTokenClassification (line 6131) | class RoFormerForTokenClassification(metaclass=DummyObject):
    method __init__ (line 6134) | def __init__(self, *args, **kwargs):
  class RoFormerLayer (line 6138) | class RoFormerLayer(metaclass=DummyObject):
    method __init__ (line 6141) | def __init__(self, *args, **kwargs):
  class RoFormerModel (line 6145) | class RoFormerModel(metaclass=DummyObject):
    method __init__ (line 6148) | def __init__(self, *args, **kwargs):
  class RoFormerPreTrainedModel (line 6152) | class RoFormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6155) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_roformer (line 6159) | def load_tf_weights_in_roformer(*args, **kwargs):
  class RwkvForCausalLM (line 6166) | class RwkvForCausalLM(metaclass=DummyObject):
    method __init__ (line 6169) | def __init__(self, *args, **kwargs):
  class RwkvModel (line 6173) | class RwkvModel(metaclass=DummyObject):
    method __init__ (line 6176) | def __init__(self, *args, **kwargs):
  class RwkvPreTrainedModel (line 6180) | class RwkvPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6183) | def __init__(self, *args, **kwargs):
  class SamModel (line 6190) | class SamModel(metaclass=DummyObject):
    method __init__ (line 6193) | def __init__(self, *args, **kwargs):
  class SamPreTrainedModel (line 6197) | class SamPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6200) | def __init__(self, *args, **kwargs):
  class SegformerDecodeHead (line 6207) | class SegformerDecodeHead(metaclass=DummyObject):
    method __init__ (line 6210) | def __init__(self, *args, **kwargs):
  class SegformerForImageClassification (line 6214) | class SegformerForImageClassification(metaclass=DummyObject):
    method __init__ (line 6217) | def __init__(self, *args, **kwargs):
  class SegformerForSemanticSegmentation (line 6221) | class SegformerForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 6224) | def __init__(self, *args, **kwargs):
  class SegformerLayer (line 6228) | class SegformerLayer(metaclass=DummyObject):
    method __init__ (line 6231) | def __init__(self, *args, **kwargs):
  class SegformerModel (line 6235) | class SegformerModel(metaclass=DummyObject):
    method __init__ (line 6238) | def __init__(self, *args, **kwargs):
  class SegformerPreTrainedModel (line 6242) | class SegformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6245) | def __init__(self, *args, **kwargs):
  class SEWForCTC (line 6252) | class SEWForCTC(metaclass=DummyObject):
    method __init__ (line 6255) | def __init__(self, *args, **kwargs):
  class SEWForSequenceClassification (line 6259) | class SEWForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 6262) | def __init__(self, *args, **kwargs):
  class SEWModel (line 6266) | class SEWModel(metaclass=DummyObject):
    method __init__ (line 6269) | def __init__(self, *args, **kwargs):
  class SEWPreTrainedModel (line 6273) | class SEWPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6276) | def __init__(self, *args, **kwargs):
  class SEWDForCTC (line 6283) | class SEWDForCTC(metaclass=DummyObject):
    method __init__ (line 6286) | def __init__(self, *args, **kwargs):
  class SEWDForSequenceClassification (line 6290) | class SEWDForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 6293) | def __init__(self, *args, **kwargs):
  class SEWDModel (line 6297) | class SEWDModel(metaclass=DummyObject):
    method __init__ (line 6300) | def __init__(self, *args, **kwargs):
  class SEWDPreTrainedModel (line 6304) | class SEWDPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6307) | def __init__(self, *args, **kwargs):
  class SpeechEncoderDecoderModel (line 6311) | class SpeechEncoderDecoderModel(metaclass=DummyObject):
    method __init__ (line 6314) | def __init__(self, *args, **kwargs):
  class Speech2TextForConditionalGeneration (line 6321) | class Speech2TextForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 6324) | def __init__(self, *args, **kwargs):
  class Speech2TextModel (line 6328) | class Speech2TextModel(metaclass=DummyObject):
    method __init__ (line 6331) | def __init__(self, *args, **kwargs):
  class Speech2TextPreTrainedModel (line 6335) | class Speech2TextPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6338) | def __init__(self, *args, **kwargs):
  class Speech2Text2ForCausalLM (line 6342) | class Speech2Text2ForCausalLM(metaclass=DummyObject):
    method __init__ (line 6345) | def __init__(self, *args, **kwargs):
  class Speech2Text2PreTrainedModel (line 6349) | class Speech2Text2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6352) | def __init__(self, *args, **kwargs):
  class SpeechT5ForSpeechToSpeech (line 6359) | class SpeechT5ForSpeechToSpeech(metaclass=DummyObject):
    method __init__ (line 6362) | def __init__(self, *args, **kwargs):
  class SpeechT5ForSpeechToText (line 6366) | class SpeechT5ForSpeechToText(metaclass=DummyObject):
    method __init__ (line 6369) | def __init__(self, *args, **kwargs):
  class SpeechT5ForTextToSpeech (line 6373) | class SpeechT5ForTextToSpeech(metaclass=DummyObject):
    method __init__ (line 6376) | def __init__(self, *args, **kwargs):
  class SpeechT5HifiGan (line 6380) | class SpeechT5HifiGan(metaclass=DummyObject):
    method __init__ (line 6383) | def __init__(self, *args, **kwargs):
  class SpeechT5Model (line 6387) | class SpeechT5Model(metaclass=DummyObject):
    method __init__ (line 6390) | def __init__(self, *args, **kwargs):
  class SpeechT5PreTrainedModel (line 6394) | class SpeechT5PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6397) | def __init__(self, *args, **kwargs):
  class SplinterForPreTraining (line 6404) | class SplinterForPreTraining(metaclass=DummyObject):
    method __init__ (line 6407) | def __init__(self, *args, **kwargs):
  class SplinterForQuestionAnswering (line 6411) | class SplinterForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 6414) | def __init__(self, *args, **kwargs):
  class SplinterLayer (line 6418) | class SplinterLayer(metaclass=DummyObject):
    method __init__ (line 6421) | def __init__(self, *args, **kwargs):
  class SplinterModel (line 6425) | class SplinterModel(metaclass=DummyObject):
    method __init__ (line 6428) | def __init__(self, *args, **kwargs):
  class SplinterPreTrainedModel (line 6432) | class SplinterPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6435) | def __init__(self, *args, **kwargs):
  class SqueezeBertForMaskedLM (line 6442) | class SqueezeBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 6445) | def __init__(self, *args, **kwargs):
  class SqueezeBertForMultipleChoice (line 6449) | class SqueezeBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 6452) | def __init__(self, *args, **kwargs):
  class SqueezeBertForQuestionAnswering (line 6456) | class SqueezeBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 6459) | def __init__(self, *args, **kwargs):
  class SqueezeBertForSequenceClassification (line 6463) | class SqueezeBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 6466) | def __init__(self, *args, **kwargs):
  class SqueezeBertForTokenClassification (line 6470) | class SqueezeBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 6473) | def __init__(self, *args, **kwargs):
  class SqueezeBertModel (line 6477) | class SqueezeBertModel(metaclass=DummyObject):
    method __init__ (line 6480) | def __init__(self, *args, **kwargs):
  class SqueezeBertModule (line 6484) | class SqueezeBertModule(metaclass=DummyObject):
    method __init__ (line 6487) | def __init__(self, *args, **kwargs):
  class SqueezeBertPreTrainedModel (line 6491) | class SqueezeBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6494) | def __init__(self, *args, **kwargs):
  class SwiftFormerForImageClassification (line 6501) | class SwiftFormerForImageClassification(metaclass=DummyObject):
    method __init__ (line 6504) | def __init__(self, *args, **kwargs):
  class SwiftFormerModel (line 6508) | class SwiftFormerModel(metaclass=DummyObject):
    method __init__ (line 6511) | def __init__(self, *args, **kwargs):
  class SwiftFormerPreTrainedModel (line 6515) | class SwiftFormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6518) | def __init__(self, *args, **kwargs):
  class SwinBackbone (line 6525) | class SwinBackbone(metaclass=DummyObject):
    method __init__ (line 6528) | def __init__(self, *args, **kwargs):
  class SwinForImageClassification (line 6532) | class SwinForImageClassification(metaclass=DummyObject):
    method __init__ (line 6535) | def __init__(self, *args, **kwargs):
  class SwinForMaskedImageModeling (line 6539) | class SwinForMaskedImageModeling(metaclass=DummyObject):
    method __init__ (line 6542) | def __init__(self, *args, **kwargs):
  class SwinModel (line 6546) | class SwinModel(metaclass=DummyObject):
    method __init__ (line 6549) | def __init__(self, *args, **kwargs):
  class SwinPreTrainedModel (line 6553) | class SwinPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6556) | def __init__(self, *args, **kwargs):
  class Swin2SRForImageSuperResolution (line 6563) | class Swin2SRForImageSuperResolution(metaclass=DummyObject):
    method __init__ (line 6566) | def __init__(self, *args, **kwargs):
  class Swin2SRModel (line 6570) | class Swin2SRModel(metaclass=DummyObject):
    method __init__ (line 6573) | def __init__(self, *args, **kwargs):
  class Swin2SRPreTrainedModel (line 6577) | class Swin2SRPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6580) | def __init__(self, *args, **kwargs):
  class Swinv2ForImageClassification (line 6587) | class Swinv2ForImageClassification(metaclass=DummyObject):
    method __init__ (line 6590) | def __init__(self, *args, **kwargs):
  class Swinv2ForMaskedImageModeling (line 6594) | class Swinv2ForMaskedImageModeling(metaclass=DummyObject):
    method __init__ (line 6597) | def __init__(self, *args, **kwargs):
  class Swinv2Model (line 6601) | class Swinv2Model(metaclass=DummyObject):
    method __init__ (line 6604) | def __init__(self, *args, **kwargs):
  class Swinv2PreTrainedModel (line 6608) | class Swinv2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6611) | def __init__(self, *args, **kwargs):
  class SwitchTransformersEncoderModel (line 6618) | class SwitchTransformersEncoderModel(metaclass=DummyObject):
    method __init__ (line 6621) | def __init__(self, *args, **kwargs):
  class SwitchTransformersForConditionalGeneration (line 6625) | class SwitchTransformersForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 6628) | def __init__(self, *args, **kwargs):
  class SwitchTransformersModel (line 6632) | class SwitchTransformersModel(metaclass=DummyObject):
    method __init__ (line 6635) | def __init__(self, *args, **kwargs):
  class SwitchTransformersPreTrainedModel (line 6639) | class SwitchTransformersPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6642) | def __init__(self, *args, **kwargs):
  class SwitchTransformersSparseMLP (line 6646) | class SwitchTransformersSparseMLP(metaclass=DummyObject):
    method __init__ (line 6649) | def __init__(self, *args, **kwargs):
  class SwitchTransformersTop1Router (line 6653) | class SwitchTransformersTop1Router(metaclass=DummyObject):
    method __init__ (line 6656) | def __init__(self, *args, **kwargs):
  class T5EncoderModel (line 6663) | class T5EncoderModel(metaclass=DummyObject):
    method __init__ (line 6666) | def __init__(self, *args, **kwargs):
  class T5ForConditionalGeneration (line 6670) | class T5ForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 6673) | def __init__(self, *args, **kwargs):
  class T5Model (line 6677) | class T5Model(metaclass=DummyObject):
    method __init__ (line 6680) | def __init__(self, *args, **kwargs):
  class T5PreTrainedModel (line 6684) | class T5PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6687) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_t5 (line 6691) | def load_tf_weights_in_t5(*args, **kwargs):
  class TableTransformerForObjectDetection (line 6698) | class TableTransformerForObjectDetection(metaclass=DummyObject):
    method __init__ (line 6701) | def __init__(self, *args, **kwargs):
  class TableTransformerModel (line 6705) | class TableTransformerModel(metaclass=DummyObject):
    method __init__ (line 6708) | def __init__(self, *args, **kwargs):
  class TableTransformerPreTrainedModel (line 6712) | class TableTransformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6715) | def __init__(self, *args, **kwargs):
  class TapasForMaskedLM (line 6722) | class TapasForMaskedLM(metaclass=DummyObject):
    method __init__ (line 6725) | def __init__(self, *args, **kwargs):
  class TapasForQuestionAnswering (line 6729) | class TapasForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 6732) | def __init__(self, *args, **kwargs):
  class TapasForSequenceClassification (line 6736) | class TapasForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 6739) | def __init__(self, *args, **kwargs):
  class TapasModel (line 6743) | class TapasModel(metaclass=DummyObject):
    method __init__ (line 6746) | def __init__(self, *args, **kwargs):
  class TapasPreTrainedModel (line 6750) | class TapasPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6753) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_tapas (line 6757) | def load_tf_weights_in_tapas(*args, **kwargs):
  class TimeSeriesTransformerForPrediction (line 6764) | class TimeSeriesTransformerForPrediction(metaclass=DummyObject):
    method __init__ (line 6767) | def __init__(self, *args, **kwargs):
  class TimeSeriesTransformerModel (line 6771) | class TimeSeriesTransformerModel(metaclass=DummyObject):
    method __init__ (line 6774) | def __init__(self, *args, **kwargs):
  class TimeSeriesTransformerPreTrainedModel (line 6778) | class TimeSeriesTransformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6781) | def __init__(self, *args, **kwargs):
  class TimesformerForVideoClassification (line 6788) | class TimesformerForVideoClassification(metaclass=DummyObject):
    method __init__ (line 6791) | def __init__(self, *args, **kwargs):
  class TimesformerModel (line 6795) | class TimesformerModel(metaclass=DummyObject):
    method __init__ (line 6798) | def __init__(self, *args, **kwargs):
  class TimesformerPreTrainedModel (line 6802) | class TimesformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6805) | def __init__(self, *args, **kwargs):
  class TimmBackbone (line 6809) | class TimmBackbone(metaclass=DummyObject):
    method __init__ (line 6812) | def __init__(self, *args, **kwargs):
  class TrajectoryTransformerModel (line 6819) | class TrajectoryTransformerModel(metaclass=DummyObject):
    method __init__ (line 6822) | def __init__(self, *args, **kwargs):
  class TrajectoryTransformerPreTrainedModel (line 6826) | class TrajectoryTransformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6829) | def __init__(self, *args, **kwargs):
  class AdaptiveEmbedding (line 6836) | class AdaptiveEmbedding(metaclass=DummyObject):
    method __init__ (line 6839) | def __init__(self, *args, **kwargs):
  class TransfoXLForSequenceClassification (line 6843) | class TransfoXLForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 6846) | def __init__(self, *args, **kwargs):
  class TransfoXLLMHeadModel (line 6850) | class TransfoXLLMHeadModel(metaclass=DummyObject):
    method __init__ (line 6853) | def __init__(self, *args, **kwargs):
  class TransfoXLModel (line 6857) | class TransfoXLModel(metaclass=DummyObject):
    method __init__ (line 6860) | def __init__(self, *args, **kwargs):
  class TransfoXLPreTrainedModel (line 6864) | class TransfoXLPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6867) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_transfo_xl (line 6871) | def load_tf_weights_in_transfo_xl(*args, **kwargs):
  class TrOCRForCausalLM (line 6878) | class TrOCRForCausalLM(metaclass=DummyObject):
    method __init__ (line 6881) | def __init__(self, *args, **kwargs):
  class TrOCRPreTrainedModel (line 6885) | class TrOCRPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6888) | def __init__(self, *args, **kwargs):
  class TvltForAudioVisualClassification (line 6895) | class TvltForAudioVisualClassification(metaclass=DummyObject):
    method __init__ (line 6898) | def __init__(self, *args, **kwargs):
  class TvltForPreTraining (line 6902) | class TvltForPreTraining(metaclass=DummyObject):
    method __init__ (line 6905) | def __init__(self, *args, **kwargs):
  class TvltModel (line 6909) | class TvltModel(metaclass=DummyObject):
    method __init__ (line 6912) | def __init__(self, *args, **kwargs):
  class TvltPreTrainedModel (line 6916) | class TvltPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6919) | def __init__(self, *args, **kwargs):
  class UniSpeechForCTC (line 6926) | class UniSpeechForCTC(metaclass=DummyObject):
    method __init__ (line 6929) | def __init__(self, *args, **kwargs):
  class UniSpeechForPreTraining (line 6933) | class UniSpeechForPreTraining(metaclass=DummyObject):
    method __init__ (line 6936) | def __init__(self, *args, **kwargs):
  class UniSpeechForSequenceClassification (line 6940) | class UniSpeechForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 6943) | def __init__(self, *args, **kwargs):
  class UniSpeechModel (line 6947) | class UniSpeechModel(metaclass=DummyObject):
    method __init__ (line 6950) | def __init__(self, *args, **kwargs):
  class UniSpeechPreTrainedModel (line 6954) | class UniSpeechPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 6957) | def __init__(self, *args, **kwargs):
  class UniSpeechSatForAudioFrameClassification (line 6964) | class UniSpeechSatForAudioFrameClassification(metaclass=DummyObject):
    method __init__ (line 6967) | def __init__(self, *args, **kwargs):
  class UniSpeechSatForCTC (line 6971) | class UniSpeechSatForCTC(metaclass=DummyObject):
    method __init__ (line 6974) | def __init__(self, *args, **kwargs):
  class UniSpeechSatForPreTraining (line 6978) | class UniSpeechSatForPreTraining(metaclass=DummyObject):
    method __init__ (line 6981) | def __init__(self, *args, **kwargs):
  class UniSpeechSatForSequenceClassification (line 6985) | class UniSpeechSatForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 6988) | def __init__(self, *args, **kwargs):
  class UniSpeechSatForXVector (line 6992) | class UniSpeechSatForXVector(metaclass=DummyObject):
    method __init__ (line 6995) | def __init__(self, *args, **kwargs):
  class UniSpeechSatModel (line 6999) | class UniSpeechSatModel(metaclass=DummyObject):
    method __init__ (line 7002) | def __init__(self, *args, **kwargs):
  class UniSpeechSatPreTrainedModel (line 7006) | class UniSpeechSatPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7009) | def __init__(self, *args, **kwargs):
  class UperNetForSemanticSegmentation (line 7013) | class UperNetForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 7016) | def __init__(self, *args, **kwargs):
  class UperNetPreTrainedModel (line 7020) | class UperNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7023) | def __init__(self, *args, **kwargs):
  class VanForImageClassification (line 7030) | class VanForImageClassification(metaclass=DummyObject):
    method __init__ (line 7033) | def __init__(self, *args, **kwargs):
  class VanModel (line 7037) | class VanModel(metaclass=DummyObject):
    method __init__ (line 7040) | def __init__(self, *args, **kwargs):
  class VanPreTrainedModel (line 7044) | class VanPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7047) | def __init__(self, *args, **kwargs):
  class VideoMAEForPreTraining (line 7054) | class VideoMAEForPreTraining(metaclass=DummyObject):
    method __init__ (line 7057) | def __init__(self, *args, **kwargs):
  class VideoMAEForVideoClassification (line 7061) | class VideoMAEForVideoClassification(metaclass=DummyObject):
    method __init__ (line 7064) | def __init__(self, *args, **kwargs):
  class VideoMAEModel (line 7068) | class VideoMAEModel(metaclass=DummyObject):
    method __init__ (line 7071) | def __init__(self, *args, **kwargs):
  class VideoMAEPreTrainedModel (line 7075) | class VideoMAEPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7078) | def __init__(self, *args, **kwargs):
  class ViltForImageAndTextRetrieval (line 7085) | class ViltForImageAndTextRetrieval(metaclass=DummyObject):
    method __init__ (line 7088) | def __init__(self, *args, **kwargs):
  class ViltForImagesAndTextClassification (line 7092) | class ViltForImagesAndTextClassification(metaclass=DummyObject):
    method __init__ (line 7095) | def __init__(self, *args, **kwargs):
  class ViltForMaskedLM (line 7099) | class ViltForMaskedLM(metaclass=DummyObject):
    method __init__ (line 7102) | def __init__(self, *args, **kwargs):
  class ViltForQuestionAnswering (line 7106) | class ViltForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 7109) | def __init__(self, *args, **kwargs):
  class ViltForTokenClassification (line 7113) | class ViltForTokenClassification(metaclass=DummyObject):
    method __init__ (line 7116) | def __init__(self, *args, **kwargs):
  class ViltLayer (line 7120) | class ViltLayer(metaclass=DummyObject):
    method __init__ (line 7123) | def __init__(self, *args, **kwargs):
  class ViltModel (line 7127) | class ViltModel(metaclass=DummyObject):
    method __init__ (line 7130) | def __init__(self, *args, **kwargs):
  class ViltPreTrainedModel (line 7134) | class ViltPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7137) | def __init__(self, *args, **kwargs):
  class VisionEncoderDecoderModel (line 7141) | class VisionEncoderDecoderModel(metaclass=DummyObject):
    method __init__ (line 7144) | def __init__(self, *args, **kwargs):
  class VisionTextDualEncoderModel (line 7148) | class VisionTextDualEncoderModel(metaclass=DummyObject):
    method __init__ (line 7151) | def __init__(self, *args, **kwargs):
  class VisualBertForMultipleChoice (line 7158) | class VisualBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 7161) | def __init__(self, *args, **kwargs):
  class VisualBertForPreTraining (line 7165) | class VisualBertForPreTraining(metaclass=DummyObject):
    method __init__ (line 7168) | def __init__(self, *args, **kwargs):
  class VisualBertForQuestionAnswering (line 7172) | class VisualBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 7175) | def __init__(self, *args, **kwargs):
  class VisualBertForRegionToPhraseAlignment (line 7179) | class VisualBertForRegionToPhraseAlignment(metaclass=DummyObject):
    method __init__ (line 7182) | def __init__(self, *args, **kwargs):
  class VisualBertForVisualReasoning (line 7186) | class VisualBertForVisualReasoning(metaclass=DummyObject):
    method __init__ (line 7189) | def __init__(self, *args, **kwargs):
  class VisualBertLayer (line 7193) | class VisualBertLayer(metaclass=DummyObject):
    method __init__ (line 7196) | def __init__(self, *args, **kwargs):
  class VisualBertModel (line 7200) | class VisualBertModel(metaclass=DummyObject):
    method __init__ (line 7203) | def __init__(self, *args, **kwargs):
  class VisualBertPreTrainedModel (line 7207) | class VisualBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7210) | def __init__(self, *args, **kwargs):
  class ViTForImageClassification (line 7217) | class ViTForImageClassification(metaclass=DummyObject):
    method __init__ (line 7220) | def __init__(self, *args, **kwargs):
  class ViTForMaskedImageModeling (line 7224) | class ViTForMaskedImageModeling(metaclass=DummyObject):
    method __init__ (line 7227) | def __init__(self, *args, **kwargs):
  class ViTModel (line 7231) | class ViTModel(metaclass=DummyObject):
    method __init__ (line 7234) | def __init__(self, *args, **kwargs):
  class ViTPreTrainedModel (line 7238) | class ViTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7241) | def __init__(self, *args, **kwargs):
  class ViTHybridForImageClassification (line 7248) | class ViTHybridForImageClassification(metaclass=DummyObject):
    method __init__ (line 7251) | def __init__(self, *args, **kwargs):
  class ViTHybridModel (line 7255) | class ViTHybridModel(metaclass=DummyObject):
    method __init__ (line 7258) | def __init__(self, *args, **kwargs):
  class ViTHybridPreTrainedModel (line 7262) | class ViTHybridPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7265) | def __init__(self, *args, **kwargs):
  class ViTMAEForPreTraining (line 7272) | class ViTMAEForPreTraining(metaclass=DummyObject):
    method __init__ (line 7275) | def __init__(self, *args, **kwargs):
  class ViTMAELayer (line 7279) | class ViTMAELayer(metaclass=DummyObject):
    method __init__ (line 7282) | def __init__(self, *args, **kwargs):
  class ViTMAEModel (line 7286) | class ViTMAEModel(metaclass=DummyObject):
    method __init__ (line 7289) | def __init__(self, *args, **kwargs):
  class ViTMAEPreTrainedModel (line 7293) | class ViTMAEPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7296) | def __init__(self, *args, **kwargs):
  class ViTMSNForImageClassification (line 7303) | class ViTMSNForImageClassification(metaclass=DummyObject):
    method __init__ (line 7306) | def __init__(self, *args, **kwargs):
  class ViTMSNModel (line 7310) | class ViTMSNModel(metaclass=DummyObject):
    method __init__ (line 7313) | def __init__(self, *args, **kwargs):
  class ViTMSNPreTrainedModel (line 7317) | class ViTMSNPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7320) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ForAudioFrameClassification (line 7327) | class Wav2Vec2ForAudioFrameClassification(metaclass=DummyObject):
    method __init__ (line 7330) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ForCTC (line 7334) | class Wav2Vec2ForCTC(metaclass=DummyObject):
    method __init__ (line 7337) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ForMaskedLM (line 7341) | class Wav2Vec2ForMaskedLM(metaclass=DummyObject):
    method __init__ (line 7344) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ForPreTraining (line 7348) | class Wav2Vec2ForPreTraining(metaclass=DummyObject):
    method __init__ (line 7351) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ForSequenceClassification (line 7355) | class Wav2Vec2ForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 7358) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ForXVector (line 7362) | class Wav2Vec2ForXVector(metaclass=DummyObject):
    method __init__ (line 7365) | def __init__(self, *args, **kwargs):
  class Wav2Vec2Model (line 7369) | class Wav2Vec2Model(metaclass=DummyObject):
    method __init__ (line 7372) | def __init__(self, *args, **kwargs):
  class Wav2Vec2PreTrainedModel (line 7376) | class Wav2Vec2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7379) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ConformerForAudioFrameClassification (line 7386) | class Wav2Vec2ConformerForAudioFrameClassification(metaclass=DummyObject):
    method __init__ (line 7389) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ConformerForCTC (line 7393) | class Wav2Vec2ConformerForCTC(metaclass=DummyObject):
    method __init__ (line 7396) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ConformerForPreTraining (line 7400) | class Wav2Vec2ConformerForPreTraining(metaclass=DummyObject):
    method __init__ (line 7403) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ConformerForSequenceClassification (line 7407) | class Wav2Vec2ConformerForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 7410) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ConformerForXVector (line 7414) | class Wav2Vec2ConformerForXVector(metaclass=DummyObject):
    method __init__ (line 7417) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ConformerModel (line 7421) | class Wav2Vec2ConformerModel(metaclass=DummyObject):
    method __init__ (line 7424) | def __init__(self, *args, **kwargs):
  class Wav2Vec2ConformerPreTrainedModel (line 7428) | class Wav2Vec2ConformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7431) | def __init__(self, *args, **kwargs):
  class WavLMForAudioFrameClassification (line 7438) | class WavLMForAudioFrameClassification(metaclass=DummyObject):
    method __init__ (line 7441) | def __init__(self, *args, **kwargs):
  class WavLMForCTC (line 7445) | class WavLMForCTC(metaclass=DummyObject):
    method __init__ (line 7448) | def __init__(self, *args, **kwargs):
  class WavLMForSequenceClassification (line 7452) | class WavLMForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 7455) | def __init__(self, *args, **kwargs):
  class WavLMForXVector (line 7459) | class WavLMForXVector(metaclass=DummyObject):
    method __init__ (line 7462) | def __init__(self, *args, **kwargs):
  class WavLMModel (line 7466) | class WavLMModel(metaclass=DummyObject):
    method __init__ (line 7469) | def __init__(self, *args, **kwargs):
  class WavLMPreTrainedModel (line 7473) | class WavLMPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7476) | def __init__(self, *args, **kwargs):
  class WhisperForAudioClassification (line 7483) | class WhisperForAudioClassification(metaclass=DummyObject):
    method __init__ (line 7486) | def __init__(self, *args, **kwargs):
  class WhisperForConditionalGeneration (line 7490) | class WhisperForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 7493) | def __init__(self, *args, **kwargs):
  class WhisperModel (line 7497) | class WhisperModel(metaclass=DummyObject):
    method __init__ (line 7500) | def __init__(self, *args, **kwargs):
  class WhisperPreTrainedModel (line 7504) | class WhisperPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7507) | def __init__(self, *args, **kwargs):
  class XCLIPModel (line 7514) | class XCLIPModel(metaclass=DummyObject):
    method __init__ (line 7517) | def __init__(self, *args, **kwargs):
  class XCLIPPreTrainedModel (line 7521) | class XCLIPPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7524) | def __init__(self, *args, **kwargs):
  class XCLIPTextModel (line 7528) | class XCLIPTextModel(metaclass=DummyObject):
    method __init__ (line 7531) | def __init__(self, *args, **kwargs):
  class XCLIPVisionModel (line 7535) | class XCLIPVisionModel(metaclass=DummyObject):
    method __init__ (line 7538) | def __init__(self, *args, **kwargs):
  class XGLMForCausalLM (line 7545) | class XGLMForCausalLM(metaclass=DummyObject):
    method __init__ (line 7548) | def __init__(self, *args, **kwargs):
  class XGLMModel (line 7552) | class XGLMModel(metaclass=DummyObject):
    method __init__ (line 7555) | def __init__(self, *args, **kwargs):
  class XGLMPreTrainedModel (line 7559) | class XGLMPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7562) | def __init__(self, *args, **kwargs):
  class XLMForMultipleChoice (line 7569) | class XLMForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 7572) | def __init__(self, *args, **kwargs):
  class XLMForQuestionAnswering (line 7576) | class XLMForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 7579) | def __init__(self, *args, **kwargs):
  class XLMForQuestionAnsweringSimple (line 7583) | class XLMForQuestionAnsweringSimple(metaclass=DummyObject):
    method __init__ (line 7586) | def __init__(self, *args, **kwargs):
  class XLMForSequenceClassification (line 7590) | class XLMForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 7593) | def __init__(self, *args, **kwargs):
  class XLMForTokenClassification (line 7597) | class XLMForTokenClassification(metaclass=DummyObject):
    method __init__ (line 7600) | def __init__(self, *args, **kwargs):
  class XLMModel (line 7604) | class XLMModel(metaclass=DummyObject):
    method __init__ (line 7607) | def __init__(self, *args, **kwargs):
  class XLMPreTrainedModel (line 7611) | class XLMPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7614) | def __init__(self, *args, **kwargs):
  class XLMWithLMHeadModel (line 7618) | class XLMWithLMHeadModel(metaclass=DummyObject):
    method __init__ (line 7621) | def __init__(self, *args, **kwargs):
  class XLMProphetNetDecoder (line 7628) | class XLMProphetNetDecoder(metaclass=DummyObject):
    method __init__ (line 7631) | def __init__(self, *args, **kwargs):
  class XLMProphetNetEncoder (line 7635) | class XLMProphetNetEncoder(metaclass=DummyObject):
    method __init__ (line 7638) | def __init__(self, *args, **kwargs):
  class XLMProphetNetForCausalLM (line 7642) | class XLMProphetNetForCausalLM(metaclass=DummyObject):
    method __init__ (line 7645) | def __init__(self, *args, **kwargs):
  class XLMProphetNetForConditionalGeneration (line 7649) | class XLMProphetNetForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 7652) | def __init__(self, *args, **kwargs):
  class XLMProphetNetModel (line 7656) | class XLMProphetNetModel(metaclass=DummyObject):
    method __init__ (line 7659) | def __init__(self, *args, **kwargs):
  class XLMProphetNetPreTrainedModel (line 7663) | class XLMProphetNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7666) | def __init__(self, *args, **kwargs):
  class XLMRobertaForCausalLM (line 7673) | class XLMRobertaForCausalLM(metaclass=DummyObject):
    method __init__ (line 7676) | def __init__(self, *args, **kwargs):
  class XLMRobertaForMaskedLM (line 7680) | class XLMRobertaForMaskedLM(metaclass=DummyObject):
    method __init__ (line 7683) | def __init__(self, *args, **kwargs):
  class XLMRobertaForMultipleChoice (line 7687) | class XLMRobertaForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 7690) | def __init__(self, *args, **kwargs):
  class XLMRobertaForQuestionAnswering (line 7694) | class XLMRobertaForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 7697) | def __init__(self, *args, **kwargs):
  class XLMRobertaForSequenceClassification (line 7701) | class XLMRobertaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 7704) | def __init__(self, *args, **kwargs):
  class XLMRobertaForTokenClassification (line 7708) | class XLMRobertaForTokenClassification(metaclass=DummyObject):
    method __init__ (line 7711) | def __init__(self, *args, **kwargs):
  class XLMRobertaModel (line 7715) | class XLMRobertaModel(metaclass=DummyObject):
    method __init__ (line 7718) | def __init__(self, *args, **kwargs):
  class XLMRobertaPreTrainedModel (line 7722) | class XLMRobertaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7725) | def __init__(self, *args, **kwargs):
  class XLMRobertaXLForCausalLM (line 7732) | class XLMRobertaXLForCausalLM(metaclass=DummyObject):
    method __init__ (line 7735) | def __init__(self, *args, **kwargs):
  class XLMRobertaXLForMaskedLM (line 7739) | class XLMRobertaXLForMaskedLM(metaclass=DummyObject):
    method __init__ (line 7742) | def __init__(self, *args, **kwargs):
  class XLMRobertaXLForMultipleChoice (line 7746) | class XLMRobertaXLForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 7749) | def __init__(self, *args, **kwargs):
  class XLMRobertaXLForQuestionAnswering (line 7753) | class XLMRobertaXLForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 7756) | def __init__(self, *args, **kwargs):
  class XLMRobertaXLForSequenceClassification (line 7760) | class XLMRobertaXLForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 7763) | def __init__(self, *args, **kwargs):
  class XLMRobertaXLForTokenClassification (line 7767) | class XLMRobertaXLForTokenClassification(metaclass=DummyObject):
    method __init__ (line 7770) | def __init__(self, *args, **kwargs):
  class XLMRobertaXLModel (line 7774) | class XLMRobertaXLModel(metaclass=DummyObject):
    method __init__ (line 7777) | def __init__(self, *args, **kwargs):
  class XLMRobertaXLPreTrainedModel (line 7781) | class XLMRobertaXLPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7784) | def __init__(self, *args, **kwargs):
  class XLNetForMultipleChoice (line 7791) | class XLNetForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 7794) | def __init__(self, *args, **kwargs):
  class XLNetForQuestionAnswering (line 7798) | class XLNetForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 7801) | def __init__(self, *args, **kwargs):
  class XLNetForQuestionAnsweringSimple (line 7805) | class XLNetForQuestionAnsweringSimple(metaclass=DummyObject):
    method __init__ (line 7808) | def __init__(self, *args, **kwargs):
  class XLNetForSequenceClassification (line 7812) | class XLNetForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 7815) | def __init__(self, *args, **kwargs):
  class XLNetForTokenClassification (line 7819) | class XLNetForTokenClassification(metaclass=DummyObject):
    method __init__ (line 7822) | def __init__(self, *args, **kwargs):
  class XLNetLMHeadModel (line 7826) | class XLNetLMHeadModel(metaclass=DummyObject):
    method __init__ (line 7829) | def __init__(self, *args, **kwargs):
  class XLNetModel (line 7833) | class XLNetModel(metaclass=DummyObject):
    method __init__ (line 7836) | def __init__(self, *args, **kwargs):
  class XLNetPreTrainedModel (line 7840) | class XLNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7843) | def __init__(self, *args, **kwargs):
  function load_tf_weights_in_xlnet (line 7847) | def load_tf_weights_in_xlnet(*args, **kwargs):
  class XmodForCausalLM (line 7854) | class XmodForCausalLM(metaclass=DummyObject):
    method __init__ (line 7857) | def __init__(self, *args, **kwargs):
  class XmodForMaskedLM (line 7861) | class XmodForMaskedLM(metaclass=DummyObject):
    method __init__ (line 7864) | def __init__(self, *args, **kwargs):
  class XmodForMultipleChoice (line 7868) | class XmodForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 7871) | def __init__(self, *args, **kwargs):
  class XmodForQuestionAnswering (line 7875) | class XmodForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 7878) | def __init__(self, *args, **kwargs):
  class XmodForSequenceClassification (line 7882) | class XmodForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 7885) | def __init__(self, *args, **kwargs):
  class XmodForTokenClassification (line 7889) | class XmodForTokenClassification(metaclass=DummyObject):
    method __init__ (line 7892) | def __init__(self, *args, **kwargs):
  class XmodModel (line 7896) | class XmodModel(metaclass=DummyObject):
    method __init__ (line 7899) | def __init__(self, *args, **kwargs):
  class XmodPreTrainedModel (line 7903) | class XmodPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7906) | def __init__(self, *args, **kwargs):
  class YolosForObjectDetection (line 7913) | class YolosForObjectDetection(metaclass=DummyObject):
    method __init__ (line 7916) | def __init__(self, *args, **kwargs):
  class YolosModel (line 7920) | class YolosModel(metaclass=DummyObject):
    method __init__ (line 7923) | def __init__(self, *args, **kwargs):
  class YolosPreTrainedModel (line 7927) | class YolosPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7930) | def __init__(self, *args, **kwargs):
  class YosoForMaskedLM (line 7937) | class YosoForMaskedLM(metaclass=DummyObject):
    method __init__ (line 7940) | def __init__(self, *args, **kwargs):
  class YosoForMultipleChoice (line 7944) | class YosoForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 7947) | def __init__(self, *args, **kwargs):
  class YosoForQuestionAnswering (line 7951) | class YosoForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 7954) | def __init__(self, *args, **kwargs):
  class YosoForSequenceClassification (line 7958) | class YosoForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 7961) | def __init__(self, *args, **kwargs):
  class YosoForTokenClassification (line 7965) | class YosoForTokenClassification(metaclass=DummyObject):
    method __init__ (line 7968) | def __init__(self, *args, **kwargs):
  class YosoLayer (line 7972) | class YosoLayer(metaclass=DummyObject):
    method __init__ (line 7975) | def __init__(self, *args, **kwargs):
  class YosoModel (line 7979) | class YosoModel(metaclass=DummyObject):
    method __init__ (line 7982) | def __init__(self, *args, **kwargs):
  class YosoPreTrainedModel (line 7986) | class YosoPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 7989) | def __init__(self, *args, **kwargs):
  class Adafactor (line 7993) | class Adafactor(metaclass=DummyObject):
    method __init__ (line 7996) | def __init__(self, *args, **kwargs):
  class AdamW (line 8000) | class AdamW(metaclass=DummyObject):
    method __init__ (line 8003) | def __init__(self, *args, **kwargs):
  function get_constant_schedule (line 8007) | def get_constant_schedule(*args, **kwargs):
  function get_constant_schedule_with_warmup (line 8011) | def get_constant_schedule_with_warmup(*args, **kwargs):
  function get_cosine_schedule_with_warmup (line 8015) | def get_cosine_schedule_with_warmup(*args, **kwargs):
  function get_cosine_with_hard_restarts_schedule_with_warmup (line 8019) | def get_cosine_with_hard_restarts_schedule_with_warmup(*args, **kwargs):
  function get_inverse_sqrt_schedule (line 8023) | def get_inverse_sqrt_schedule(*args, **kwargs):
  function get_linear_schedule_with_warmup (line 8027) | def get_linear_schedule_with_warmup(*args, **kwargs):
  function get_polynomial_decay_schedule_with_warmup (line 8031) | def get_polynomial_decay_schedule_with_warmup(*args, **kwargs):
  function get_scheduler (line 8035) | def get_scheduler(*args, **kwargs):
  class Conv1D (line 8039) | class Conv1D(metaclass=DummyObject):
    method __init__ (line 8042) | def __init__(self, *args, **kwargs):
  function apply_chunking_to_forward (line 8046) | def apply_chunking_to_forward(*args, **kwargs):
  function prune_layer (line 8050) | def prune_layer(*args, **kwargs):
  class Trainer (line 8054) | class Trainer(metaclass=DummyObject):
    method __init__ (line 8057) | def __init__(self, *args, **kwargs):
  function torch_distributed_zero_first (line 8061) | def torch_distributed_zero_first(*args, **kwargs):
  class Seq2SeqTrainer (line 8065) | class Seq2SeqTrainer(metaclass=DummyObject):
    method __init__ (line 8068) | def __init__(self, *args, **kwargs):

FILE: transformers/utils/dummy_sentencepiece_and_tokenizers_objects.py
  function convert_slow_tokenizer (line 8) | def convert_slow_tokenizer(*args, **kwargs):

FILE: transformers/utils/dummy_sentencepiece_objects.py
  class AlbertTokenizer (line 5) | class AlbertTokenizer(metaclass=DummyObject):
    method __init__ (line 8) | def __init__(self, *args, **kwargs):
  class BarthezTokenizer (line 12) | class BarthezTokenizer(metaclass=DummyObject):
    method __init__ (line 15) | def __init__(self, *args, **kwargs):
  class BartphoTokenizer (line 19) | class BartphoTokenizer(metaclass=DummyObject):
    method __init__ (line 22) | def __init__(self, *args, **kwargs):
  class BertGenerationTokenizer (line 26) | class BertGenerationTokenizer(metaclass=DummyObject):
    method __init__ (line 29) | def __init__(self, *args, **kwargs):
  class BigBirdTokenizer (line 33) | class BigBirdTokenizer(metaclass=DummyObject):
    method __init__ (line 36) | def __init__(self, *args, **kwargs):
  class CamembertTokenizer (line 40) | class CamembertTokenizer(metaclass=DummyObject):
    method __init__ (line 43) | def __init__(self, *args, **kwargs):
  class CpmTokenizer (line 47) | class CpmTokenizer(metaclass=DummyObject):
    method __init__ (line 50) | def __init__(self, *args, **kwargs):
  class DebertaV2Tokenizer (line 54) | class DebertaV2Tokenizer(metaclass=DummyObject):
    method __init__ (line 57) | def __init__(self, *args, **kwargs):
  class ErnieMTokenizer (line 61) | class ErnieMTokenizer(metaclass=DummyObject):
    method __init__ (line 64) | def __init__(self, *args, **kwargs):
  class FNetTokenizer (line 68) | class FNetTokenizer(metaclass=DummyObject):
    method __init__ (line 71) | def __init__(self, *args, **kwargs):
  class GPTSw3Tokenizer (line 75) | class GPTSw3Tokenizer(metaclass=DummyObject):
    method __init__ (line 78) | def __init__(self, *args, **kwargs):
  class LayoutXLMTokenizer (line 82) | class LayoutXLMTokenizer(metaclass=DummyObject):
    method __init__ (line 85) | def __init__(self, *args, **kwargs):
  class LlamaTokenizer (line 89) | class LlamaTokenizer(metaclass=DummyObject):
    method __init__ (line 92) | def __init__(self, *args, **kwargs):
  class M2M100Tokenizer (line 96) | class M2M100Tokenizer(metaclass=DummyObject):
    method __init__ (line 99) | def __init__(self, *args, **kwargs):
  class MarianTokenizer (line 103) | class MarianTokenizer(metaclass=DummyObject):
    method __init__ (line 106) | def __init__(self, *args, **kwargs):
  class MBart50Tokenizer (line 110) | class MBart50Tokenizer(metaclass=DummyObject):
    method __init__ (line 113) | def __init__(self, *args, **kwargs):
  class MBartTokenizer (line 117) | class MBartTokenizer(metaclass=DummyObject):
    method __init__ (line 120) | def __init__(self, *args, **kwargs):
  class MLukeTokenizer (line 124) | class MLukeTokenizer(metaclass=DummyObject):
    method __init__ (line 127) | def __init__(self, *args, **kwargs):
  class MT5Tokenizer (line 131) | class MT5Tokenizer(metaclass=DummyObject):
    method __init__ (line 134) | def __init__(self, *args, **kwargs):
  class NllbTokenizer (line 138) | class NllbTokenizer(metaclass=DummyObject):
    method __init__ (line 141) | def __init__(self, *args, **kwargs):
  class PegasusTokenizer (line 145) | class PegasusTokenizer(metaclass=DummyObject):
    method __init__ (line 148) | def __init__(self, *args, **kwargs):
  class PLBartTokenizer (line 152) | class PLBartTokenizer(metaclass=DummyObject):
    method __init__ (line 155) | def __init__(self, *args, **kwargs):
  class ReformerTokenizer (line 159) | class ReformerTokenizer(metaclass=DummyObject):
    method __init__ (line 162) | def __init__(self, *args, **kwargs):
  class RemBertTokenizer (line 166) | class RemBertTokenizer(metaclass=DummyObject):
    method __init__ (line 169) | def __init__(self, *args, **kwargs):
  class Speech2TextTokenizer (line 173) | class Speech2TextTokenizer(metaclass=DummyObject):
    method __init__ (line 176) | def __init__(self, *args, **kwargs):
  class SpeechT5Tokenizer (line 180) | class SpeechT5Tokenizer(metaclass=DummyObject):
    method __init__ (line 183) | def __init__(self, *args, **kwargs):
  class T5Tokenizer (line 187) | class T5Tokenizer(metaclass=DummyObject):
    method __init__ (line 190) | def __init__(self, *args, **kwargs):
  class XGLMTokenizer (line 194) | class XGLMTokenizer(metaclass=DummyObject):
    method __init__ (line 197) | def __init__(self, *args, **kwargs):
  class XLMProphetNetTokenizer (line 201) | class XLMProphetNetTokenizer(metaclass=DummyObject):
    method __init__ (line 204) | def __init__(self, *args, **kwargs):
  class XLMRobertaTokenizer (line 208) | class XLMRobertaTokenizer(metaclass=DummyObject):
    method __init__ (line 211) | def __init__(self, *args, **kwargs):
  class XLNetTokenizer (line 215) | class XLNetTokenizer(metaclass=DummyObject):
    method __init__ (line 218) | def __init__(self, *args, **kwargs):

FILE: transformers/utils/dummy_speech_objects.py
  class ASTFeatureExtractor (line 5) | class ASTFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 8) | def __init__(self, *args, **kwargs):
  class MCTCTFeatureExtractor (line 12) | class MCTCTFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 15) | def __init__(self, *args, **kwargs):
  class Speech2TextFeatureExtractor (line 19) | class Speech2TextFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 22) | def __init__(self, *args, **kwargs):
  class SpeechT5FeatureExtractor (line 26) | class SpeechT5FeatureExtractor(metaclass=DummyObject):
    method __init__ (line 29) | def __init__(self, *args, **kwargs):
  class TvltFeatureExtractor (line 33) | class TvltFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 36) | def __init__(self, *args, **kwargs):

FILE: transformers/utils/dummy_tensorflow_text_objects.py
  class TFBertTokenizer (line 5) | class TFBertTokenizer(metaclass=DummyObject):
    method __init__ (line 8) | def __init__(self, *args, **kwargs):

FILE: transformers/utils/dummy_tf_objects.py
  class TensorFlowBenchmarkArguments (line 5) | class TensorFlowBenchmarkArguments(metaclass=DummyObject):
    method __init__ (line 8) | def __init__(self, *args, **kwargs):
  class TensorFlowBenchmark (line 12) | class TensorFlowBenchmark(metaclass=DummyObject):
    method __init__ (line 15) | def __init__(self, *args, **kwargs):
  class TFForcedBOSTokenLogitsProcessor (line 19) | class TFForcedBOSTokenLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 22) | def __init__(self, *args, **kwargs):
  class TFForcedEOSTokenLogitsProcessor (line 26) | class TFForcedEOSTokenLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 29) | def __init__(self, *args, **kwargs):
  class TFGenerationMixin (line 33) | class TFGenerationMixin(metaclass=DummyObject):
    method __init__ (line 36) | def __init__(self, *args, **kwargs):
  class TFLogitsProcessor (line 40) | class TFLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 43) | def __init__(self, *args, **kwargs):
  class TFLogitsProcessorList (line 47) | class TFLogitsProcessorList(metaclass=DummyObject):
    method __init__ (line 50) | def __init__(self, *args, **kwargs):
  class TFLogitsWarper (line 54) | class TFLogitsWarper(metaclass=DummyObject):
    method __init__ (line 57) | def __init__(self, *args, **kwargs):
  class TFMinLengthLogitsProcessor (line 61) | class TFMinLengthLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 64) | def __init__(self, *args, **kwargs):
  class TFNoBadWordsLogitsProcessor (line 68) | class TFNoBadWordsLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 71) | def __init__(self, *args, **kwargs):
  class TFNoRepeatNGramLogitsProcessor (line 75) | class TFNoRepeatNGramLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 78) | def __init__(self, *args, **kwargs):
  class TFRepetitionPenaltyLogitsProcessor (line 82) | class TFRepetitionPenaltyLogitsProcessor(metaclass=DummyObject):
    method __init__ (line 85) | def __init__(self, *args, **kwargs):
  class TFTemperatureLogitsWarper (line 89) | class TFTemperatureLogitsWarper(metaclass=DummyObject):
    method __init__ (line 92) | def __init__(self, *args, **kwargs):
  class TFTopKLogitsWarper (line 96) | class TFTopKLogitsWarper(metaclass=DummyObject):
    method __init__ (line 99) | def __init__(self, *args, **kwargs):
  class TFTopPLogitsWarper (line 103) | class TFTopPLogitsWarper(metaclass=DummyObject):
    method __init__ (line 106) | def __init__(self, *args, **kwargs):
  function tf_top_k_top_p_filtering (line 110) | def tf_top_k_top_p_filtering(*args, **kwargs):
  class KerasMetricCallback (line 114) | class KerasMetricCallback(metaclass=DummyObject):
    method __init__ (line 117) | def __init__(self, *args, **kwargs):
  class PushToHubCallback (line 121) | class PushToHubCallback(metaclass=DummyObject):
    method __init__ (line 124) | def __init__(self, *args, **kwargs):
  class TFPreTrainedModel (line 128) | class TFPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 131) | def __init__(self, *args, **kwargs):
  class TFSequenceSummary (line 135) | class TFSequenceSummary(metaclass=DummyObject):
    method __init__ (line 138) | def __init__(self, *args, **kwargs):
  class TFSharedEmbeddings (line 142) | class TFSharedEmbeddings(metaclass=DummyObject):
    method __init__ (line 145) | def __init__(self, *args, **kwargs):
  function shape_list (line 149) | def shape_list(*args, **kwargs):
  class TFAlbertForMaskedLM (line 156) | class TFAlbertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 159) | def __init__(self, *args, **kwargs):
  class TFAlbertForMultipleChoice (line 163) | class TFAlbertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 166) | def __init__(self, *args, **kwargs):
  class TFAlbertForPreTraining (line 170) | class TFAlbertForPreTraining(metaclass=DummyObject):
    method __init__ (line 173) | def __init__(self, *args, **kwargs):
  class TFAlbertForQuestionAnswering (line 177) | class TFAlbertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 180) | def __init__(self, *args, **kwargs):
  class TFAlbertForSequenceClassification (line 184) | class TFAlbertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 187) | def __init__(self, *args, **kwargs):
  class TFAlbertForTokenClassification (line 191) | class TFAlbertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 194) | def __init__(self, *args, **kwargs):
  class TFAlbertMainLayer (line 198) | class TFAlbertMainLayer(metaclass=DummyObject):
    method __init__ (line 201) | def __init__(self, *args, **kwargs):
  class TFAlbertModel (line 205) | class TFAlbertModel(metaclass=DummyObject):
    method __init__ (line 208) | def __init__(self, *args, **kwargs):
  class TFAlbertPreTrainedModel (line 212) | class TFAlbertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 215) | def __init__(self, *args, **kwargs):
  class TFAutoModel (line 276) | class TFAutoModel(metaclass=DummyObject):
    method __init__ (line 279) | def __init__(self, *args, **kwargs):
  class TFAutoModelForCausalLM (line 283) | class TFAutoModelForCausalLM(metaclass=DummyObject):
    method __init__ (line 286) | def __init__(self, *args, **kwargs):
  class TFAutoModelForDocumentQuestionAnswering (line 290) | class TFAutoModelForDocumentQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 293) | def __init__(self, *args, **kwargs):
  class TFAutoModelForImageClassification (line 297) | class TFAutoModelForImageClassification(metaclass=DummyObject):
    method __init__ (line 300) | def __init__(self, *args, **kwargs):
  class TFAutoModelForMaskedLM (line 304) | class TFAutoModelForMaskedLM(metaclass=DummyObject):
    method __init__ (line 307) | def __init__(self, *args, **kwargs):
  class TFAutoModelForMultipleChoice (line 311) | class TFAutoModelForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 314) | def __init__(self, *args, **kwargs):
  class TFAutoModelForNextSentencePrediction (line 318) | class TFAutoModelForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 321) | def __init__(self, *args, **kwargs):
  class TFAutoModelForPreTraining (line 325) | class TFAutoModelForPreTraining(metaclass=DummyObject):
    method __init__ (line 328) | def __init__(self, *args, **kwargs):
  class TFAutoModelForQuestionAnswering (line 332) | class TFAutoModelForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 335) | def __init__(self, *args, **kwargs):
  class TFAutoModelForSemanticSegmentation (line 339) | class TFAutoModelForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 342) | def __init__(self, *args, **kwargs):
  class TFAutoModelForSeq2SeqLM (line 346) | class TFAutoModelForSeq2SeqLM(metaclass=DummyObject):
    method __init__ (line 349) | def __init__(self, *args, **kwargs):
  class TFAutoModelForSequenceClassification (line 353) | class TFAutoModelForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 356) | def __init__(self, *args, **kwargs):
  class TFAutoModelForSpeechSeq2Seq (line 360) | class TFAutoModelForSpeechSeq2Seq(metaclass=DummyObject):
    method __init__ (line 363) | def __init__(self, *args, **kwargs):
  class TFAutoModelForTableQuestionAnswering (line 367) | class TFAutoModelForTableQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 370) | def __init__(self, *args, **kwargs):
  class TFAutoModelForTokenClassification (line 374) | class TFAutoModelForTokenClassification(metaclass=DummyObject):
    method __init__ (line 377) | def __init__(self, *args, **kwargs):
  class TFAutoModelForVision2Seq (line 381) | class TFAutoModelForVision2Seq(metaclass=DummyObject):
    method __init__ (line 384) | def __init__(self, *args, **kwargs):
  class TFAutoModelForZeroShotImageClassification (line 388) | class TFAutoModelForZeroShotImageClassification(metaclass=DummyObject):
    method __init__ (line 391) | def __init__(self, *args, **kwargs):
  class TFAutoModelWithLMHead (line 395) | class TFAutoModelWithLMHead(metaclass=DummyObject):
    method __init__ (line 398) | def __init__(self, *args, **kwargs):
  class TFBartForConditionalGeneration (line 402) | class TFBartForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 405) | def __init__(self, *args, **kwargs):
  class TFBartForSequenceClassification (line 409) | class TFBartForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 412) | def __init__(self, *args, **kwargs):
  class TFBartModel (line 416) | class TFBartModel(metaclass=DummyObject):
    method __init__ (line 419) | def __init__(self, *args, **kwargs):
  class TFBartPretrainedModel (line 423) | class TFBartPretrainedModel(metaclass=DummyObject):
    method __init__ (line 426) | def __init__(self, *args, **kwargs):
  class TFBertEmbeddings (line 433) | class TFBertEmbeddings(metaclass=DummyObject):
    method __init__ (line 436) | def __init__(self, *args, **kwargs):
  class TFBertForMaskedLM (line 440) | class TFBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 443) | def __init__(self, *args, **kwargs):
  class TFBertForMultipleChoice (line 447) | class TFBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 450) | def __init__(self, *args, **kwargs):
  class TFBertForNextSentencePrediction (line 454) | class TFBertForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 457) | def __init__(self, *args, **kwargs):
  class TFBertForPreTraining (line 461) | class TFBertForPreTraining(metaclass=DummyObject):
    method __init__ (line 464) | def __init__(self, *args, **kwargs):
  class TFBertForQuestionAnswering (line 468) | class TFBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 471) | def __init__(self, *args, **kwargs):
  class TFBertForSequenceClassification (line 475) | class TFBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 478) | def __init__(self, *args, **kwargs):
  class TFBertForTokenClassification (line 482) | class TFBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 485) | def __init__(self, *args, **kwargs):
  class TFBertLMHeadModel (line 489) | class TFBertLMHeadModel(metaclass=DummyObject):
    method __init__ (line 492) | def __init__(self, *args, **kwargs):
  class TFBertMainLayer (line 496) | class TFBertMainLayer(metaclass=DummyObject):
    method __init__ (line 499) | def __init__(self, *args, **kwargs):
  class TFBertModel (line 503) | class TFBertModel(metaclass=DummyObject):
    method __init__ (line 506) | def __init__(self, *args, **kwargs):
  class TFBertPreTrainedModel (line 510) | class TFBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 513) | def __init__(self, *args, **kwargs):
  class TFBlenderbotForConditionalGeneration (line 517) | class TFBlenderbotForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 520) | def __init__(self, *args, **kwargs):
  class TFBlenderbotModel (line 524) | class TFBlenderbotModel(metaclass=DummyObject):
    method __init__ (line 527) | def __init__(self, *args, **kwargs):
  class TFBlenderbotPreTrainedModel (line 531) | class TFBlenderbotPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 534) | def __init__(self, *args, **kwargs):
  class TFBlenderbotSmallForConditionalGeneration (line 538) | class TFBlenderbotSmallForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 541) | def __init__(self, *args, **kwargs):
  class TFBlenderbotSmallModel (line 545) | class TFBlenderbotSmallModel(metaclass=DummyObject):
    method __init__ (line 548) | def __init__(self, *args, **kwargs):
  class TFBlenderbotSmallPreTrainedModel (line 552) | class TFBlenderbotSmallPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 555) | def __init__(self, *args, **kwargs):
  class TFBlipForConditionalGeneration (line 562) | class TFBlipForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 565) | def __init__(self, *args, **kwargs):
  class TFBlipForImageTextRetrieval (line 569) | class TFBlipForImageTextRetrieval(metaclass=DummyObject):
    method __init__ (line 572) | def __init__(self, *args, **kwargs):
  class TFBlipForQuestionAnswering (line 576) | class TFBlipForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 579) | def __init__(self, *args, **kwargs):
  class TFBlipModel (line 583) | class TFBlipModel(metaclass=DummyObject):
    method __init__ (line 586) | def __init__(self, *args, **kwargs):
  class TFBlipPreTrainedModel (line 590) | class TFBlipPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 593) | def __init__(self, *args, **kwargs):
  class TFBlipTextModel (line 597) | class TFBlipTextModel(metaclass=DummyObject):
    method __init__ (line 600) | def __init__(self, *args, **kwargs):
  class TFBlipVisionModel (line 604) | class TFBlipVisionModel(metaclass=DummyObject):
    method __init__ (line 607) | def __init__(self, *args, **kwargs):
  class TFCamembertForCausalLM (line 614) | class TFCamembertForCausalLM(metaclass=DummyObject):
    method __init__ (line 617) | def __init__(self, *args, **kwargs):
  class TFCamembertForMaskedLM (line 621) | class TFCamembertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 624) | def __init__(self, *args, **kwargs):
  class TFCamembertForMultipleChoice (line 628) | class TFCamembertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 631) | def __init__(self, *args, **kwargs):
  class TFCamembertForQuestionAnswering (line 635) | class TFCamembertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 638) | def __init__(self, *args, **kwargs):
  class TFCamembertForSequenceClassification (line 642) | class TFCamembertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 645) | def __init__(self, *args, **kwargs):
  class TFCamembertForTokenClassification (line 649) | class TFCamembertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 652) | def __init__(self, *args, **kwargs):
  class TFCamembertModel (line 656) | class TFCamembertModel(metaclass=DummyObject):
    method __init__ (line 659) | def __init__(self, *args, **kwargs):
  class TFCamembertPreTrainedModel (line 663) | class TFCamembertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 666) | def __init__(self, *args, **kwargs):
  class TFCLIPModel (line 673) | class TFCLIPModel(metaclass=DummyObject):
    method __init__ (line 676) | def __init__(self, *args, **kwargs):
  class TFCLIPPreTrainedModel (line 680) | class TFCLIPPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 683) | def __init__(self, *args, **kwargs):
  class TFCLIPTextModel (line 687) | class TFCLIPTextModel(metaclass=DummyObject):
    method __init__ (line 690) | def __init__(self, *args, **kwargs):
  class TFCLIPVisionModel (line 694) | class TFCLIPVisionModel(metaclass=DummyObject):
    method __init__ (line 697) | def __init__(self, *args, **kwargs):
  class TFConvBertForMaskedLM (line 704) | class TFConvBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 707) | def __init__(self, *args, **kwargs):
  class TFConvBertForMultipleChoice (line 711) | class TFConvBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 714) | def __init__(self, *args, **kwargs):
  class TFConvBertForQuestionAnswering (line 718) | class TFConvBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 721) | def __init__(self, *args, **kwargs):
  class TFConvBertForSequenceClassification (line 725) | class TFConvBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 728) | def __init__(self, *args, **kwargs):
  class TFConvBertForTokenClassification (line 732) | class TFConvBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 735) | def __init__(self, *args, **kwargs):
  class TFConvBertLayer (line 739) | class TFConvBertLayer(metaclass=DummyObject):
    method __init__ (line 742) | def __init__(self, *args, **kwargs):
  class TFConvBertModel (line 746) | class TFConvBertModel(metaclass=DummyObject):
    method __init__ (line 749) | def __init__(self, *args, **kwargs):
  class TFConvBertPreTrainedModel (line 753) | class TFConvBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 756) | def __init__(self, *args, **kwargs):
  class TFConvNextForImageClassification (line 760) | class TFConvNextForImageClassification(metaclass=DummyObject):
    method __init__ (line 763) | def __init__(self, *args, **kwargs):
  class TFConvNextModel (line 767) | class TFConvNextModel(metaclass=DummyObject):
    method __init__ (line 770) | def __init__(self, *args, **kwargs):
  class TFConvNextPreTrainedModel (line 774) | class TFConvNextPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 777) | def __init__(self, *args, **kwargs):
  class TFCTRLForSequenceClassification (line 784) | class TFCTRLForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 787) | def __init__(self, *args, **kwargs):
  class TFCTRLLMHeadModel (line 791) | class TFCTRLLMHeadModel(metaclass=DummyObject):
    method __init__ (line 794) | def __init__(self, *args, **kwargs):
  class TFCTRLModel (line 798) | class TFCTRLModel(metaclass=DummyObject):
    method __init__ (line 801) | def __init__(self, *args, **kwargs):
  class TFCTRLPreTrainedModel (line 805) | class TFCTRLPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 808) | def __init__(self, *args, **kwargs):
  class TFCvtForImageClassification (line 815) | class TFCvtForImageClassification(metaclass=DummyObject):
    method __init__ (line 818) | def __init__(self, *args, **kwargs):
  class TFCvtModel (line 822) | class TFCvtModel(metaclass=DummyObject):
    method __init__ (line 825) | def __init__(self, *args, **kwargs):
  class TFCvtPreTrainedModel (line 829) | class TFCvtPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 832) | def __init__(self, *args, **kwargs):
  class TFData2VecVisionForImageClassification (line 836) | class TFData2VecVisionForImageClassification(metaclass=DummyObject):
    method __init__ (line 839) | def __init__(self, *args, **kwargs):
  class TFData2VecVisionForSemanticSegmentation (line 843) | class TFData2VecVisionForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 846) | def __init__(self, *args, **kwargs):
  class TFData2VecVisionModel (line 850) | class TFData2VecVisionModel(metaclass=DummyObject):
    method __init__ (line 853) | def __init__(self, *args, **kwargs):
  class TFData2VecVisionPreTrainedModel (line 857) | class TFData2VecVisionPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 860) | def __init__(self, *args, **kwargs):
  class TFDebertaForMaskedLM (line 867) | class TFDebertaForMaskedLM(metaclass=DummyObject):
    method __init__ (line 870) | def __init__(self, *args, **kwargs):
  class TFDebertaForQuestionAnswering (line 874) | class TFDebertaForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 877) | def __init__(self, *args, **kwargs):
  class TFDebertaForSequenceClassification (line 881) | class TFDebertaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 884) | def __init__(self, *args, **kwargs):
  class TFDebertaForTokenClassification (line 888) | class TFDebertaForTokenClassification(metaclass=DummyObject):
    method __init__ (line 891) | def __init__(self, *args, **kwargs):
  class TFDebertaModel (line 895) | class TFDebertaModel(metaclass=DummyObject):
    method __init__ (line 898) | def __init__(self, *args, **kwargs):
  class TFDebertaPreTrainedModel (line 902) | class TFDebertaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 905) | def __init__(self, *args, **kwargs):
  class TFDebertaV2ForMaskedLM (line 912) | class TFDebertaV2ForMaskedLM(metaclass=DummyObject):
    method __init__ (line 915) | def __init__(self, *args, **kwargs):
  class TFDebertaV2ForQuestionAnswering (line 919) | class TFDebertaV2ForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 922) | def __init__(self, *args, **kwargs):
  class TFDebertaV2ForSequenceClassification (line 926) | class TFDebertaV2ForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 929) | def __init__(self, *args, **kwargs):
  class TFDebertaV2ForTokenClassification (line 933) | class TFDebertaV2ForTokenClassification(metaclass=DummyObject):
    method __init__ (line 936) | def __init__(self, *args, **kwargs):
  class TFDebertaV2Model (line 940) | class TFDebertaV2Model(metaclass=DummyObject):
    method __init__ (line 943) | def __init__(self, *args, **kwargs):
  class TFDebertaV2PreTrainedModel (line 947) | class TFDebertaV2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 950) | def __init__(self, *args, **kwargs):
  class TFDeiTForImageClassification (line 957) | class TFDeiTForImageClassification(metaclass=DummyObject):
    method __init__ (line 960) | def __init__(self, *args, **kwargs):
  class TFDeiTForImageClassificationWithTeacher (line 964) | class TFDeiTForImageClassificationWithTeacher(metaclass=DummyObject):
    method __init__ (line 967) | def __init__(self, *args, **kwargs):
  class TFDeiTForMaskedImageModeling (line 971) | class TFDeiTForMaskedImageModeling(metaclass=DummyObject):
    method __init__ (line 974) | def __init__(self, *args, **kwargs):
  class TFDeiTModel (line 978) | class TFDeiTModel(metaclass=DummyObject):
    method __init__ (line 981) | def __init__(self, *args, **kwargs):
  class TFDeiTPreTrainedModel (line 985) | class TFDeiTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 988) | def __init__(self, *args, **kwargs):
  class TFDistilBertForMaskedLM (line 995) | class TFDistilBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 998) | def __init__(self, *args, **kwargs):
  class TFDistilBertForMultipleChoice (line 1002) | class TFDistilBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1005) | def __init__(self, *args, **kwargs):
  class TFDistilBertForQuestionAnswering (line 1009) | class TFDistilBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1012) | def __init__(self, *args, **kwargs):
  class TFDistilBertForSequenceClassification (line 1016) | class TFDistilBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1019) | def __init__(self, *args, **kwargs):
  class TFDistilBertForTokenClassification (line 1023) | class TFDistilBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1026) | def __init__(self, *args, **kwargs):
  class TFDistilBertMainLayer (line 1030) | class TFDistilBertMainLayer(metaclass=DummyObject):
    method __init__ (line 1033) | def __init__(self, *args, **kwargs):
  class TFDistilBertModel (line 1037) | class TFDistilBertModel(metaclass=DummyObject):
    method __init__ (line 1040) | def __init__(self, *args, **kwargs):
  class TFDistilBertPreTrainedModel (line 1044) | class TFDistilBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1047) | def __init__(self, *args, **kwargs):
  class TFDPRContextEncoder (line 1060) | class TFDPRContextEncoder(metaclass=DummyObject):
    method __init__ (line 1063) | def __init__(self, *args, **kwargs):
  class TFDPRPretrainedContextEncoder (line 1067) | class TFDPRPretrainedContextEncoder(metaclass=DummyObject):
    method __init__ (line 1070) | def __init__(self, *args, **kwargs):
  class TFDPRPretrainedQuestionEncoder (line 1074) | class TFDPRPretrainedQuestionEncoder(metaclass=DummyObject):
    method __init__ (line 1077) | def __init__(self, *args, **kwargs):
  class TFDPRPretrainedReader (line 1081) | class TFDPRPretrainedReader(metaclass=DummyObject):
    method __init__ (line 1084) | def __init__(self, *args, **kwargs):
  class TFDPRQuestionEncoder (line 1088) | class TFDPRQuestionEncoder(metaclass=DummyObject):
    method __init__ (line 1091) | def __init__(self, *args, **kwargs):
  class TFDPRReader (line 1095) | class TFDPRReader(metaclass=DummyObject):
    method __init__ (line 1098) | def __init__(self, *args, **kwargs):
  class TFEfficientFormerForImageClassification (line 1105) | class TFEfficientFormerForImageClassification(metaclass=DummyObject):
    method __init__ (line 1108) | def __init__(self, *args, **kwargs):
  class TFEfficientFormerForImageClassificationWithTeacher (line 1112) | class TFEfficientFormerForImageClassificationWithTeacher(metaclass=Dummy...
    method __init__ (line 1115) | def __init__(self, *args, **kwargs):
  class TFEfficientFormerModel (line 1119) | class TFEfficientFormerModel(metaclass=DummyObject):
    method __init__ (line 1122) | def __init__(self, *args, **kwargs):
  class TFEfficientFormerPreTrainedModel (line 1126) | class TFEfficientFormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1129) | def __init__(self, *args, **kwargs):
  class TFElectraForMaskedLM (line 1136) | class TFElectraForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1139) | def __init__(self, *args, **kwargs):
  class TFElectraForMultipleChoice (line 1143) | class TFElectraForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1146) | def __init__(self, *args, **kwargs):
  class TFElectraForPreTraining (line 1150) | class TFElectraForPreTraining(metaclass=DummyObject):
    method __init__ (line 1153) | def __init__(self, *args, **kwargs):
  class TFElectraForQuestionAnswering (line 1157) | class TFElectraForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1160) | def __init__(self, *args, **kwargs):
  class TFElectraForSequenceClassification (line 1164) | class TFElectraForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1167) | def __init__(self, *args, **kwargs):
  class TFElectraForTokenClassification (line 1171) | class TFElectraForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1174) | def __init__(self, *args, **kwargs):
  class TFElectraModel (line 1178) | class TFElectraModel(metaclass=DummyObject):
    method __init__ (line 1181) | def __init__(self, *args, **kwargs):
  class TFElectraPreTrainedModel (line 1185) | class TFElectraPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1188) | def __init__(self, *args, **kwargs):
  class TFEncoderDecoderModel (line 1192) | class TFEncoderDecoderModel(metaclass=DummyObject):
    method __init__ (line 1195) | def __init__(self, *args, **kwargs):
  class TFEsmForMaskedLM (line 1202) | class TFEsmForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1205) | def __init__(self, *args, **kwargs):
  class TFEsmForSequenceClassification (line 1209) | class TFEsmForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1212) | def __init__(self, *args, **kwargs):
  class TFEsmForTokenClassification (line 1216) | class TFEsmForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1219) | def __init__(self, *args, **kwargs):
  class TFEsmModel (line 1223) | class TFEsmModel(metaclass=DummyObject):
    method __init__ (line 1226) | def __init__(self, *args, **kwargs):
  class TFEsmPreTrainedModel (line 1230) | class TFEsmPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1233) | def __init__(self, *args, **kwargs):
  class TFFlaubertForMultipleChoice (line 1240) | class TFFlaubertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1243) | def __init__(self, *args, **kwargs):
  class TFFlaubertForQuestionAnsweringSimple (line 1247) | class TFFlaubertForQuestionAnsweringSimple(metaclass=DummyObject):
    method __init__ (line 1250) | def __init__(self, *args, **kwargs):
  class TFFlaubertForSequenceClassification (line 1254) | class TFFlaubertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1257) | def __init__(self, *args, **kwargs):
  class TFFlaubertForTokenClassification (line 1261) | class TFFlaubertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1264) | def __init__(self, *args, **kwargs):
  class TFFlaubertModel (line 1268) | class TFFlaubertModel(metaclass=DummyObject):
    method __init__ (line 1271) | def __init__(self, *args, **kwargs):
  class TFFlaubertPreTrainedModel (line 1275) | class TFFlaubertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1278) | def __init__(self, *args, **kwargs):
  class TFFlaubertWithLMHeadModel (line 1282) | class TFFlaubertWithLMHeadModel(metaclass=DummyObject):
    method __init__ (line 1285) | def __init__(self, *args, **kwargs):
  class TFFunnelBaseModel (line 1292) | class TFFunnelBaseModel(metaclass=DummyObject):
    method __init__ (line 1295) | def __init__(self, *args, **kwargs):
  class TFFunnelForMaskedLM (line 1299) | class TFFunnelForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1302) | def __init__(self, *args, **kwargs):
  class TFFunnelForMultipleChoice (line 1306) | class TFFunnelForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1309) | def __init__(self, *args, **kwargs):
  class TFFunnelForPreTraining (line 1313) | class TFFunnelForPreTraining(metaclass=DummyObject):
    method __init__ (line 1316) | def __init__(self, *args, **kwargs):
  class TFFunnelForQuestionAnswering (line 1320) | class TFFunnelForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1323) | def __init__(self, *args, **kwargs):
  class TFFunnelForSequenceClassification (line 1327) | class TFFunnelForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1330) | def __init__(self, *args, **kwargs):
  class TFFunnelForTokenClassification (line 1334) | class TFFunnelForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1337) | def __init__(self, *args, **kwargs):
  class TFFunnelModel (line 1341) | class TFFunnelModel(metaclass=DummyObject):
    method __init__ (line 1344) | def __init__(self, *args, **kwargs):
  class TFFunnelPreTrainedModel (line 1348) | class TFFunnelPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1351) | def __init__(self, *args, **kwargs):
  class TFGPT2DoubleHeadsModel (line 1358) | class TFGPT2DoubleHeadsModel(metaclass=DummyObject):
    method __init__ (line 1361) | def __init__(self, *args, **kwargs):
  class TFGPT2ForSequenceClassification (line 1365) | class TFGPT2ForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1368) | def __init__(self, *args, **kwargs):
  class TFGPT2LMHeadModel (line 1372) | class TFGPT2LMHeadModel(metaclass=DummyObject):
    method __init__ (line 1375) | def __init__(self, *args, **kwargs):
  class TFGPT2MainLayer (line 1379) | class TFGPT2MainLayer(metaclass=DummyObject):
    method __init__ (line 1382) | def __init__(self, *args, **kwargs):
  class TFGPT2Model (line 1386) | class TFGPT2Model(metaclass=DummyObject):
    method __init__ (line 1389) | def __init__(self, *args, **kwargs):
  class TFGPT2PreTrainedModel (line 1393) | class TFGPT2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1396) | def __init__(self, *args, **kwargs):
  class TFGPTJForCausalLM (line 1400) | class TFGPTJForCausalLM(metaclass=DummyObject):
    method __init__ (line 1403) | def __init__(self, *args, **kwargs):
  class TFGPTJForQuestionAnswering (line 1407) | class TFGPTJForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1410) | def __init__(self, *args, **kwargs):
  class TFGPTJForSequenceClassification (line 1414) | class TFGPTJForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1417) | def __init__(self, *args, **kwargs):
  class TFGPTJModel (line 1421) | class TFGPTJModel(metaclass=DummyObject):
    method __init__ (line 1424) | def __init__(self, *args, **kwargs):
  class TFGPTJPreTrainedModel (line 1428) | class TFGPTJPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1431) | def __init__(self, *args, **kwargs):
  class TFGroupViTModel (line 1438) | class TFGroupViTModel(metaclass=DummyObject):
    method __init__ (line 1441) | def __init__(self, *args, **kwargs):
  class TFGroupViTPreTrainedModel (line 1445) | class TFGroupViTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1448) | def __init__(self, *args, **kwargs):
  class TFGroupViTTextModel (line 1452) | class TFGroupViTTextModel(metaclass=DummyObject):
    method __init__ (line 1455) | def __init__(self, *args, **kwargs):
  class TFGroupViTVisionModel (line 1459) | class TFGroupViTVisionModel(metaclass=DummyObject):
    method __init__ (line 1462) | def __init__(self, *args, **kwargs):
  class TFHubertForCTC (line 1469) | class TFHubertForCTC(metaclass=DummyObject):
    method __init__ (line 1472) | def __init__(self, *args, **kwargs):
  class TFHubertModel (line 1476) | class TFHubertModel(metaclass=DummyObject):
    method __init__ (line 1479) | def __init__(self, *args, **kwargs):
  class TFHubertPreTrainedModel (line 1483) | class TFHubertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1486) | def __init__(self, *args, **kwargs):
  class TFLayoutLMForMaskedLM (line 1493) | class TFLayoutLMForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1496) | def __init__(self, *args, **kwargs):
  class TFLayoutLMForQuestionAnswering (line 1500) | class TFLayoutLMForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1503) | def __init__(self, *args, **kwargs):
  class TFLayoutLMForSequenceClassification (line 1507) | class TFLayoutLMForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1510) | def __init__(self, *args, **kwargs):
  class TFLayoutLMForTokenClassification (line 1514) | class TFLayoutLMForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1517) | def __init__(self, *args, **kwargs):
  class TFLayoutLMMainLayer (line 1521) | class TFLayoutLMMainLayer(metaclass=DummyObject):
    method __init__ (line 1524) | def __init__(self, *args, **kwargs):
  class TFLayoutLMModel (line 1528) | class TFLayoutLMModel(metaclass=DummyObject):
    method __init__ (line 1531) | def __init__(self, *args, **kwargs):
  class TFLayoutLMPreTrainedModel (line 1535) | class TFLayoutLMPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1538) | def __init__(self, *args, **kwargs):
  class TFLayoutLMv3ForQuestionAnswering (line 1545) | class TFLayoutLMv3ForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1548) | def __init__(self, *args, **kwargs):
  class TFLayoutLMv3ForSequenceClassification (line 1552) | class TFLayoutLMv3ForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1555) | def __init__(self, *args, **kwargs):
  class TFLayoutLMv3ForTokenClassification (line 1559) | class TFLayoutLMv3ForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1562) | def __init__(self, *args, **kwargs):
  class TFLayoutLMv3Model (line 1566) | class TFLayoutLMv3Model(metaclass=DummyObject):
    method __init__ (line 1569) | def __init__(self, *args, **kwargs):
  class TFLayoutLMv3PreTrainedModel (line 1573) | class TFLayoutLMv3PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1576) | def __init__(self, *args, **kwargs):
  class TFLEDForConditionalGeneration (line 1580) | class TFLEDForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1583) | def __init__(self, *args, **kwargs):
  class TFLEDModel (line 1587) | class TFLEDModel(metaclass=DummyObject):
    method __init__ (line 1590) | def __init__(self, *args, **kwargs):
  class TFLEDPreTrainedModel (line 1594) | class TFLEDPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1597) | def __init__(self, *args, **kwargs):
  class TFLongformerForMaskedLM (line 1604) | class TFLongformerForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1607) | def __init__(self, *args, **kwargs):
  class TFLongformerForMultipleChoice (line 1611) | class TFLongformerForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1614) | def __init__(self, *args, **kwargs):
  class TFLongformerForQuestionAnswering (line 1618) | class TFLongformerForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1621) | def __init__(self, *args, **kwargs):
  class TFLongformerForSequenceClassification (line 1625) | class TFLongformerForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1628) | def __init__(self, *args, **kwargs):
  class TFLongformerForTokenClassification (line 1632) | class TFLongformerForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1635) | def __init__(self, *args, **kwargs):
  class TFLongformerModel (line 1639) | class TFLongformerModel(metaclass=DummyObject):
    method __init__ (line 1642) | def __init__(self, *args, **kwargs):
  class TFLongformerPreTrainedModel (line 1646) | class TFLongformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1649) | def __init__(self, *args, **kwargs):
  class TFLongformerSelfAttention (line 1653) | class TFLongformerSelfAttention(metaclass=DummyObject):
    method __init__ (line 1656) | def __init__(self, *args, **kwargs):
  class TFLxmertForPreTraining (line 1663) | class TFLxmertForPreTraining(metaclass=DummyObject):
    method __init__ (line 1666) | def __init__(self, *args, **kwargs):
  class TFLxmertMainLayer (line 1670) | class TFLxmertMainLayer(metaclass=DummyObject):
    method __init__ (line 1673) | def __init__(self, *args, **kwargs):
  class TFLxmertModel (line 1677) | class TFLxmertModel(metaclass=DummyObject):
    method __init__ (line 1680) | def __init__(self, *args, **kwargs):
  class TFLxmertPreTrainedModel (line 1684) | class TFLxmertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1687) | def __init__(self, *args, **kwargs):
  class TFLxmertVisualFeatureEncoder (line 1691) | class TFLxmertVisualFeatureEncoder(metaclass=DummyObject):
    method __init__ (line 1694) | def __init__(self, *args, **kwargs):
  class TFMarianModel (line 1698) | class TFMarianModel(metaclass=DummyObject):
    method __init__ (line 1701) | def __init__(self, *args, **kwargs):
  class TFMarianMTModel (line 1705) | class TFMarianMTModel(metaclass=DummyObject):
    method __init__ (line 1708) | def __init__(self, *args, **kwargs):
  class TFMarianPreTrainedModel (line 1712) | class TFMarianPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1715) | def __init__(self, *args, **kwargs):
  class TFMBartForConditionalGeneration (line 1719) | class TFMBartForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1722) | def __init__(self, *args, **kwargs):
  class TFMBartModel (line 1726) | class TFMBartModel(metaclass=DummyObject):
    method __init__ (line 1729) | def __init__(self, *args, **kwargs):
  class TFMBartPreTrainedModel (line 1733) | class TFMBartPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1736) | def __init__(self, *args, **kwargs):
  class TFMobileBertForMaskedLM (line 1743) | class TFMobileBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1746) | def __init__(self, *args, **kwargs):
  class TFMobileBertForMultipleChoice (line 1750) | class TFMobileBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1753) | def __init__(self, *args, **kwargs):
  class TFMobileBertForNextSentencePrediction (line 1757) | class TFMobileBertForNextSentencePrediction(metaclass=DummyObject):
    method __init__ (line 1760) | def __init__(self, *args, **kwargs):
  class TFMobileBertForPreTraining (line 1764) | class TFMobileBertForPreTraining(metaclass=DummyObject):
    method __init__ (line 1767) | def __init__(self, *args, **kwargs):
  class TFMobileBertForQuestionAnswering (line 1771) | class TFMobileBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1774) | def __init__(self, *args, **kwargs):
  class TFMobileBertForSequenceClassification (line 1778) | class TFMobileBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1781) | def __init__(self, *args, **kwargs):
  class TFMobileBertForTokenClassification (line 1785) | class TFMobileBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1788) | def __init__(self, *args, **kwargs):
  class TFMobileBertMainLayer (line 1792) | class TFMobileBertMainLayer(metaclass=DummyObject):
    method __init__ (line 1795) | def __init__(self, *args, **kwargs):
  class TFMobileBertModel (line 1799) | class TFMobileBertModel(metaclass=DummyObject):
    method __init__ (line 1802) | def __init__(self, *args, **kwargs):
  class TFMobileBertPreTrainedModel (line 1806) | class TFMobileBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1809) | def __init__(self, *args, **kwargs):
  class TFMobileViTForImageClassification (line 1816) | class TFMobileViTForImageClassification(metaclass=DummyObject):
    method __init__ (line 1819) | def __init__(self, *args, **kwargs):
  class TFMobileViTForSemanticSegmentation (line 1823) | class TFMobileViTForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 1826) | def __init__(self, *args, **kwargs):
  class TFMobileViTModel (line 1830) | class TFMobileViTModel(metaclass=DummyObject):
    method __init__ (line 1833) | def __init__(self, *args, **kwargs):
  class TFMobileViTPreTrainedModel (line 1837) | class TFMobileViTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1840) | def __init__(self, *args, **kwargs):
  class TFMPNetForMaskedLM (line 1847) | class TFMPNetForMaskedLM(metaclass=DummyObject):
    method __init__ (line 1850) | def __init__(self, *args, **kwargs):
  class TFMPNetForMultipleChoice (line 1854) | class TFMPNetForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 1857) | def __init__(self, *args, **kwargs):
  class TFMPNetForQuestionAnswering (line 1861) | class TFMPNetForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 1864) | def __init__(self, *args, **kwargs):
  class TFMPNetForSequenceClassification (line 1868) | class TFMPNetForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1871) | def __init__(self, *args, **kwargs):
  class TFMPNetForTokenClassification (line 1875) | class TFMPNetForTokenClassification(metaclass=DummyObject):
    method __init__ (line 1878) | def __init__(self, *args, **kwargs):
  class TFMPNetMainLayer (line 1882) | class TFMPNetMainLayer(metaclass=DummyObject):
    method __init__ (line 1885) | def __init__(self, *args, **kwargs):
  class TFMPNetModel (line 1889) | class TFMPNetModel(metaclass=DummyObject):
    method __init__ (line 1892) | def __init__(self, *args, **kwargs):
  class TFMPNetPreTrainedModel (line 1896) | class TFMPNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1899) | def __init__(self, *args, **kwargs):
  class TFMT5EncoderModel (line 1903) | class TFMT5EncoderModel(metaclass=DummyObject):
    method __init__ (line 1906) | def __init__(self, *args, **kwargs):
  class TFMT5ForConditionalGeneration (line 1910) | class TFMT5ForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1913) | def __init__(self, *args, **kwargs):
  class TFMT5Model (line 1917) | class TFMT5Model(metaclass=DummyObject):
    method __init__ (line 1920) | def __init__(self, *args, **kwargs):
  class TFOpenAIGPTDoubleHeadsModel (line 1927) | class TFOpenAIGPTDoubleHeadsModel(metaclass=DummyObject):
    method __init__ (line 1930) | def __init__(self, *args, **kwargs):
  class TFOpenAIGPTForSequenceClassification (line 1934) | class TFOpenAIGPTForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 1937) | def __init__(self, *args, **kwargs):
  class TFOpenAIGPTLMHeadModel (line 1941) | class TFOpenAIGPTLMHeadModel(metaclass=DummyObject):
    method __init__ (line 1944) | def __init__(self, *args, **kwargs):
  class TFOpenAIGPTMainLayer (line 1948) | class TFOpenAIGPTMainLayer(metaclass=DummyObject):
    method __init__ (line 1951) | def __init__(self, *args, **kwargs):
  class TFOpenAIGPTModel (line 1955) | class TFOpenAIGPTModel(metaclass=DummyObject):
    method __init__ (line 1958) | def __init__(self, *args, **kwargs):
  class TFOpenAIGPTPreTrainedModel (line 1962) | class TFOpenAIGPTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1965) | def __init__(self, *args, **kwargs):
  class TFOPTForCausalLM (line 1969) | class TFOPTForCausalLM(metaclass=DummyObject):
    method __init__ (line 1972) | def __init__(self, *args, **kwargs):
  class TFOPTModel (line 1976) | class TFOPTModel(metaclass=DummyObject):
    method __init__ (line 1979) | def __init__(self, *args, **kwargs):
  class TFOPTPreTrainedModel (line 1983) | class TFOPTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 1986) | def __init__(self, *args, **kwargs):
  class TFPegasusForConditionalGeneration (line 1990) | class TFPegasusForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 1993) | def __init__(self, *args, **kwargs):
  class TFPegasusModel (line 1997) | class TFPegasusModel(metaclass=DummyObject):
    method __init__ (line 2000) | def __init__(self, *args, **kwargs):
  class TFPegasusPreTrainedModel (line 2004) | class TFPegasusPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2007) | def __init__(self, *args, **kwargs):
  class TFRagModel (line 2011) | class TFRagModel(metaclass=DummyObject):
    method __init__ (line 2014) | def __init__(self, *args, **kwargs):
  class TFRagPreTrainedModel (line 2018) | class TFRagPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2021) | def __init__(self, *args, **kwargs):
  class TFRagSequenceForGeneration (line 2025) | class TFRagSequenceForGeneration(metaclass=DummyObject):
    method __init__ (line 2028) | def __init__(self, *args, **kwargs):
  class TFRagTokenForGeneration (line 2032) | class TFRagTokenForGeneration(metaclass=DummyObject):
    method __init__ (line 2035) | def __init__(self, *args, **kwargs):
  class TFRegNetForImageClassification (line 2042) | class TFRegNetForImageClassification(metaclass=DummyObject):
    method __init__ (line 2045) | def __init__(self, *args, **kwargs):
  class TFRegNetModel (line 2049) | class TFRegNetModel(metaclass=DummyObject):
    method __init__ (line 2052) | def __init__(self, *args, **kwargs):
  class TFRegNetPreTrainedModel (line 2056) | class TFRegNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2059) | def __init__(self, *args, **kwargs):
  class TFRemBertForCausalLM (line 2066) | class TFRemBertForCausalLM(metaclass=DummyObject):
    method __init__ (line 2069) | def __init__(self, *args, **kwargs):
  class TFRemBertForMaskedLM (line 2073) | class TFRemBertForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2076) | def __init__(self, *args, **kwargs):
  class TFRemBertForMultipleChoice (line 2080) | class TFRemBertForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2083) | def __init__(self, *args, **kwargs):
  class TFRemBertForQuestionAnswering (line 2087) | class TFRemBertForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2090) | def __init__(self, *args, **kwargs):
  class TFRemBertForSequenceClassification (line 2094) | class TFRemBertForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2097) | def __init__(self, *args, **kwargs):
  class TFRemBertForTokenClassification (line 2101) | class TFRemBertForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2104) | def __init__(self, *args, **kwargs):
  class TFRemBertLayer (line 2108) | class TFRemBertLayer(metaclass=DummyObject):
    method __init__ (line 2111) | def __init__(self, *args, **kwargs):
  class TFRemBertModel (line 2115) | class TFRemBertModel(metaclass=DummyObject):
    method __init__ (line 2118) | def __init__(self, *args, **kwargs):
  class TFRemBertPreTrainedModel (line 2122) | class TFRemBertPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2125) | def __init__(self, *args, **kwargs):
  class TFResNetForImageClassification (line 2132) | class TFResNetForImageClassification(metaclass=DummyObject):
    method __init__ (line 2135) | def __init__(self, *args, **kwargs):
  class TFResNetModel (line 2139) | class TFResNetModel(metaclass=DummyObject):
    method __init__ (line 2142) | def __init__(self, *args, **kwargs):
  class TFResNetPreTrainedModel (line 2146) | class TFResNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2149) | def __init__(self, *args, **kwargs):
  class TFRobertaForCausalLM (line 2156) | class TFRobertaForCausalLM(metaclass=DummyObject):
    method __init__ (line 2159) | def __init__(self, *args, **kwargs):
  class TFRobertaForMaskedLM (line 2163) | class TFRobertaForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2166) | def __init__(self, *args, **kwargs):
  class TFRobertaForMultipleChoice (line 2170) | class TFRobertaForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2173) | def __init__(self, *args, **kwargs):
  class TFRobertaForQuestionAnswering (line 2177) | class TFRobertaForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2180) | def __init__(self, *args, **kwargs):
  class TFRobertaForSequenceClassification (line 2184) | class TFRobertaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2187) | def __init__(self, *args, **kwargs):
  class TFRobertaForTokenClassification (line 2191) | class TFRobertaForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2194) | def __init__(self, *args, **kwargs):
  class TFRobertaMainLayer (line 2198) | class TFRobertaMainLayer(metaclass=DummyObject):
    method __init__ (line 2201) | def __init__(self, *args, **kwargs):
  class TFRobertaModel (line 2205) | class TFRobertaModel(metaclass=DummyObject):
    method __init__ (line 2208) | def __init__(self, *args, **kwargs):
  class TFRobertaPreTrainedModel (line 2212) | class TFRobertaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2215) | def __init__(self, *args, **kwargs):
  class TFRobertaPreLayerNormForCausalLM (line 2222) | class TFRobertaPreLayerNormForCausalLM(metaclass=DummyObject):
    method __init__ (line 2225) | def __init__(self, *args, **kwargs):
  class TFRobertaPreLayerNormForMaskedLM (line 2229) | class TFRobertaPreLayerNormForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2232) | def __init__(self, *args, **kwargs):
  class TFRobertaPreLayerNormForMultipleChoice (line 2236) | class TFRobertaPreLayerNormForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2239) | def __init__(self, *args, **kwargs):
  class TFRobertaPreLayerNormForQuestionAnswering (line 2243) | class TFRobertaPreLayerNormForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2246) | def __init__(self, *args, **kwargs):
  class TFRobertaPreLayerNormForSequenceClassification (line 2250) | class TFRobertaPreLayerNormForSequenceClassification(metaclass=DummyObje...
    method __init__ (line 2253) | def __init__(self, *args, **kwargs):
  class TFRobertaPreLayerNormForTokenClassification (line 2257) | class TFRobertaPreLayerNormForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2260) | def __init__(self, *args, **kwargs):
  class TFRobertaPreLayerNormMainLayer (line 2264) | class TFRobertaPreLayerNormMainLayer(metaclass=DummyObject):
    method __init__ (line 2267) | def __init__(self, *args, **kwargs):
  class TFRobertaPreLayerNormModel (line 2271) | class TFRobertaPreLayerNormModel(metaclass=DummyObject):
    method __init__ (line 2274) | def __init__(self, *args, **kwargs):
  class TFRobertaPreLayerNormPreTrainedModel (line 2278) | class TFRobertaPreLayerNormPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2281) | def __init__(self, *args, **kwargs):
  class TFRoFormerForCausalLM (line 2288) | class TFRoFormerForCausalLM(metaclass=DummyObject):
    method __init__ (line 2291) | def __init__(self, *args, **kwargs):
  class TFRoFormerForMaskedLM (line 2295) | class TFRoFormerForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2298) | def __init__(self, *args, **kwargs):
  class TFRoFormerForMultipleChoice (line 2302) | class TFRoFormerForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2305) | def __init__(self, *args, **kwargs):
  class TFRoFormerForQuestionAnswering (line 2309) | class TFRoFormerForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2312) | def __init__(self, *args, **kwargs):
  class TFRoFormerForSequenceClassification (line 2316) | class TFRoFormerForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2319) | def __init__(self, *args, **kwargs):
  class TFRoFormerForTokenClassification (line 2323) | class TFRoFormerForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2326) | def __init__(self, *args, **kwargs):
  class TFRoFormerLayer (line 2330) | class TFRoFormerLayer(metaclass=DummyObject):
    method __init__ (line 2333) | def __init__(self, *args, **kwargs):
  class TFRoFormerModel (line 2337) | class TFRoFormerModel(metaclass=DummyObject):
    method __init__ (line 2340) | def __init__(self, *args, **kwargs):
  class TFRoFormerPreTrainedModel (line 2344) | class TFRoFormerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2347) | def __init__(self, *args, **kwargs):
  class TFSamModel (line 2354) | class TFSamModel(metaclass=DummyObject):
    method __init__ (line 2357) | def __init__(self, *args, **kwargs):
  class TFSamPreTrainedModel (line 2361) | class TFSamPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2364) | def __init__(self, *args, **kwargs):
  class TFSegformerDecodeHead (line 2371) | class TFSegformerDecodeHead(metaclass=DummyObject):
    method __init__ (line 2374) | def __init__(self, *args, **kwargs):
  class TFSegformerForImageClassification (line 2378) | class TFSegformerForImageClassification(metaclass=DummyObject):
    method __init__ (line 2381) | def __init__(self, *args, **kwargs):
  class TFSegformerForSemanticSegmentation (line 2385) | class TFSegformerForSemanticSegmentation(metaclass=DummyObject):
    method __init__ (line 2388) | def __init__(self, *args, **kwargs):
  class TFSegformerModel (line 2392) | class TFSegformerModel(metaclass=DummyObject):
    method __init__ (line 2395) | def __init__(self, *args, **kwargs):
  class TFSegformerPreTrainedModel (line 2399) | class TFSegformerPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2402) | def __init__(self, *args, **kwargs):
  class TFSpeech2TextForConditionalGeneration (line 2409) | class TFSpeech2TextForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 2412) | def __init__(self, *args, **kwargs):
  class TFSpeech2TextModel (line 2416) | class TFSpeech2TextModel(metaclass=DummyObject):
    method __init__ (line 2419) | def __init__(self, *args, **kwargs):
  class TFSpeech2TextPreTrainedModel (line 2423) | class TFSpeech2TextPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2426) | def __init__(self, *args, **kwargs):
  class TFSwinForImageClassification (line 2433) | class TFSwinForImageClassification(metaclass=DummyObject):
    method __init__ (line 2436) | def __init__(self, *args, **kwargs):
  class TFSwinForMaskedImageModeling (line 2440) | class TFSwinForMaskedImageModeling(metaclass=DummyObject):
    method __init__ (line 2443) | def __init__(self, *args, **kwargs):
  class TFSwinModel (line 2447) | class TFSwinModel(metaclass=DummyObject):
    method __init__ (line 2450) | def __init__(self, *args, **kwargs):
  class TFSwinPreTrainedModel (line 2454) | class TFSwinPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2457) | def __init__(self, *args, **kwargs):
  class TFT5EncoderModel (line 2464) | class TFT5EncoderModel(metaclass=DummyObject):
    method __init__ (line 2467) | def __init__(self, *args, **kwargs):
  class TFT5ForConditionalGeneration (line 2471) | class TFT5ForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 2474) | def __init__(self, *args, **kwargs):
  class TFT5Model (line 2478) | class TFT5Model(metaclass=DummyObject):
    method __init__ (line 2481) | def __init__(self, *args, **kwargs):
  class TFT5PreTrainedModel (line 2485) | class TFT5PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2488) | def __init__(self, *args, **kwargs):
  class TFTapasForMaskedLM (line 2495) | class TFTapasForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2498) | def __init__(self, *args, **kwargs):
  class TFTapasForQuestionAnswering (line 2502) | class TFTapasForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2505) | def __init__(self, *args, **kwargs):
  class TFTapasForSequenceClassification (line 2509) | class TFTapasForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2512) | def __init__(self, *args, **kwargs):
  class TFTapasModel (line 2516) | class TFTapasModel(metaclass=DummyObject):
    method __init__ (line 2519) | def __init__(self, *args, **kwargs):
  class TFTapasPreTrainedModel (line 2523) | class TFTapasPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2526) | def __init__(self, *args, **kwargs):
  class TFAdaptiveEmbedding (line 2533) | class TFAdaptiveEmbedding(metaclass=DummyObject):
    method __init__ (line 2536) | def __init__(self, *args, **kwargs):
  class TFTransfoXLForSequenceClassification (line 2540) | class TFTransfoXLForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2543) | def __init__(self, *args, **kwargs):
  class TFTransfoXLLMHeadModel (line 2547) | class TFTransfoXLLMHeadModel(metaclass=DummyObject):
    method __init__ (line 2550) | def __init__(self, *args, **kwargs):
  class TFTransfoXLMainLayer (line 2554) | class TFTransfoXLMainLayer(metaclass=DummyObject):
    method __init__ (line 2557) | def __init__(self, *args, **kwargs):
  class TFTransfoXLModel (line 2561) | class TFTransfoXLModel(metaclass=DummyObject):
    method __init__ (line 2564) | def __init__(self, *args, **kwargs):
  class TFTransfoXLPreTrainedModel (line 2568) | class TFTransfoXLPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2571) | def __init__(self, *args, **kwargs):
  class TFVisionEncoderDecoderModel (line 2575) | class TFVisionEncoderDecoderModel(metaclass=DummyObject):
    method __init__ (line 2578) | def __init__(self, *args, **kwargs):
  class TFVisionTextDualEncoderModel (line 2582) | class TFVisionTextDualEncoderModel(metaclass=DummyObject):
    method __init__ (line 2585) | def __init__(self, *args, **kwargs):
  class TFViTForImageClassification (line 2589) | class TFViTForImageClassification(metaclass=DummyObject):
    method __init__ (line 2592) | def __init__(self, *args, **kwargs):
  class TFViTModel (line 2596) | class TFViTModel(metaclass=DummyObject):
    method __init__ (line 2599) | def __init__(self, *args, **kwargs):
  class TFViTPreTrainedModel (line 2603) | class TFViTPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2606) | def __init__(self, *args, **kwargs):
  class TFViTMAEForPreTraining (line 2610) | class TFViTMAEForPreTraining(metaclass=DummyObject):
    method __init__ (line 2613) | def __init__(self, *args, **kwargs):
  class TFViTMAEModel (line 2617) | class TFViTMAEModel(metaclass=DummyObject):
    method __init__ (line 2620) | def __init__(self, *args, **kwargs):
  class TFViTMAEPreTrainedModel (line 2624) | class TFViTMAEPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2627) | def __init__(self, *args, **kwargs):
  class TFWav2Vec2ForCTC (line 2634) | class TFWav2Vec2ForCTC(metaclass=DummyObject):
    method __init__ (line 2637) | def __init__(self, *args, **kwargs):
  class TFWav2Vec2ForSequenceClassification (line 2641) | class TFWav2Vec2ForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2644) | def __init__(self, *args, **kwargs):
  class TFWav2Vec2Model (line 2648) | class TFWav2Vec2Model(metaclass=DummyObject):
    method __init__ (line 2651) | def __init__(self, *args, **kwargs):
  class TFWav2Vec2PreTrainedModel (line 2655) | class TFWav2Vec2PreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2658) | def __init__(self, *args, **kwargs):
  class TFWhisperForConditionalGeneration (line 2665) | class TFWhisperForConditionalGeneration(metaclass=DummyObject):
    method __init__ (line 2668) | def __init__(self, *args, **kwargs):
  class TFWhisperModel (line 2672) | class TFWhisperModel(metaclass=DummyObject):
    method __init__ (line 2675) | def __init__(self, *args, **kwargs):
  class TFWhisperPreTrainedModel (line 2679) | class TFWhisperPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2682) | def __init__(self, *args, **kwargs):
  class TFXGLMForCausalLM (line 2689) | class TFXGLMForCausalLM(metaclass=DummyObject):
    method __init__ (line 2692) | def __init__(self, *args, **kwargs):
  class TFXGLMModel (line 2696) | class TFXGLMModel(metaclass=DummyObject):
    method __init__ (line 2699) | def __init__(self, *args, **kwargs):
  class TFXGLMPreTrainedModel (line 2703) | class TFXGLMPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2706) | def __init__(self, *args, **kwargs):
  class TFXLMForMultipleChoice (line 2713) | class TFXLMForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2716) | def __init__(self, *args, **kwargs):
  class TFXLMForQuestionAnsweringSimple (line 2720) | class TFXLMForQuestionAnsweringSimple(metaclass=DummyObject):
    method __init__ (line 2723) | def __init__(self, *args, **kwargs):
  class TFXLMForSequenceClassification (line 2727) | class TFXLMForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2730) | def __init__(self, *args, **kwargs):
  class TFXLMForTokenClassification (line 2734) | class TFXLMForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2737) | def __init__(self, *args, **kwargs):
  class TFXLMMainLayer (line 2741) | class TFXLMMainLayer(metaclass=DummyObject):
    method __init__ (line 2744) | def __init__(self, *args, **kwargs):
  class TFXLMModel (line 2748) | class TFXLMModel(metaclass=DummyObject):
    method __init__ (line 2751) | def __init__(self, *args, **kwargs):
  class TFXLMPreTrainedModel (line 2755) | class TFXLMPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2758) | def __init__(self, *args, **kwargs):
  class TFXLMWithLMHeadModel (line 2762) | class TFXLMWithLMHeadModel(metaclass=DummyObject):
    method __init__ (line 2765) | def __init__(self, *args, **kwargs):
  class TFXLMRobertaForCausalLM (line 2772) | class TFXLMRobertaForCausalLM(metaclass=DummyObject):
    method __init__ (line 2775) | def __init__(self, *args, **kwargs):
  class TFXLMRobertaForMaskedLM (line 2779) | class TFXLMRobertaForMaskedLM(metaclass=DummyObject):
    method __init__ (line 2782) | def __init__(self, *args, **kwargs):
  class TFXLMRobertaForMultipleChoice (line 2786) | class TFXLMRobertaForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2789) | def __init__(self, *args, **kwargs):
  class TFXLMRobertaForQuestionAnswering (line 2793) | class TFXLMRobertaForQuestionAnswering(metaclass=DummyObject):
    method __init__ (line 2796) | def __init__(self, *args, **kwargs):
  class TFXLMRobertaForSequenceClassification (line 2800) | class TFXLMRobertaForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2803) | def __init__(self, *args, **kwargs):
  class TFXLMRobertaForTokenClassification (line 2807) | class TFXLMRobertaForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2810) | def __init__(self, *args, **kwargs):
  class TFXLMRobertaModel (line 2814) | class TFXLMRobertaModel(metaclass=DummyObject):
    method __init__ (line 2817) | def __init__(self, *args, **kwargs):
  class TFXLMRobertaPreTrainedModel (line 2821) | class TFXLMRobertaPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2824) | def __init__(self, *args, **kwargs):
  class TFXLNetForMultipleChoice (line 2831) | class TFXLNetForMultipleChoice(metaclass=DummyObject):
    method __init__ (line 2834) | def __init__(self, *args, **kwargs):
  class TFXLNetForQuestionAnsweringSimple (line 2838) | class TFXLNetForQuestionAnsweringSimple(metaclass=DummyObject):
    method __init__ (line 2841) | def __init__(self, *args, **kwargs):
  class TFXLNetForSequenceClassification (line 2845) | class TFXLNetForSequenceClassification(metaclass=DummyObject):
    method __init__ (line 2848) | def __init__(self, *args, **kwargs):
  class TFXLNetForTokenClassification (line 2852) | class TFXLNetForTokenClassification(metaclass=DummyObject):
    method __init__ (line 2855) | def __init__(self, *args, **kwargs):
  class TFXLNetLMHeadModel (line 2859) | class TFXLNetLMHeadModel(metaclass=DummyObject):
    method __init__ (line 2862) | def __init__(self, *args, **kwargs):
  class TFXLNetMainLayer (line 2866) | class TFXLNetMainLayer(metaclass=DummyObject):
    method __init__ (line 2869) | def __init__(self, *args, **kwargs):
  class TFXLNetModel (line 2873) | class TFXLNetModel(metaclass=DummyObject):
    method __init__ (line 2876) | def __init__(self, *args, **kwargs):
  class TFXLNetPreTrainedModel (line 2880) | class TFXLNetPreTrainedModel(metaclass=DummyObject):
    method __init__ (line 2883) | def __init__(self, *args, **kwargs):
  class AdamWeightDecay (line 2887) | class AdamWeightDecay(metaclass=DummyObject):
    method __init__ (line 2890) | def __init__(self, *args, **kwargs):
  class GradientAccumulator (line 2894) | class GradientAccumulator(metaclass=DummyObject):
    method __init__ (line 2897) | def __init__(self, *args, **kwargs):
  class WarmUp (line 2901) | class WarmUp(metaclass=DummyObject):
    method __init__ (line 2904) | def __init__(self, *args, **kwargs):
  function create_optimizer (line 2908) | def create_optimizer(*args, **kwargs):
  class TFTrainer (line 2912) | class TFTrainer(metaclass=DummyObject):
    method __init__ (line 2915) | def __init__(self, *args, **kwargs):

FILE: transformers/utils/dummy_tokenizers_objects.py
  class AlbertTokenizerFast (line 5) | class AlbertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 8) | def __init__(self, *args, **kwargs):
  class BartTokenizerFast (line 12) | class BartTokenizerFast(metaclass=DummyObject):
    method __init__ (line 15) | def __init__(self, *args, **kwargs):
  class BarthezTokenizerFast (line 19) | class BarthezTokenizerFast(metaclass=DummyObject):
    method __init__ (line 22) | def __init__(self, *args, **kwargs):
  class BertTokenizerFast (line 26) | class BertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 29) | def __init__(self, *args, **kwargs):
  class BigBirdTokenizerFast (line 33) | class BigBirdTokenizerFast(metaclass=DummyObject):
    method __init__ (line 36) | def __init__(self, *args, **kwargs):
  class BlenderbotTokenizerFast (line 40) | class BlenderbotTokenizerFast(metaclass=DummyObject):
    method __init__ (line 43) | def __init__(self, *args, **kwargs):
  class BlenderbotSmallTokenizerFast (line 47) | class BlenderbotSmallTokenizerFast(metaclass=DummyObject):
    method __init__ (line 50) | def __init__(self, *args, **kwargs):
  class BloomTokenizerFast (line 54) | class BloomTokenizerFast(metaclass=DummyObject):
    method __init__ (line 57) | def __init__(self, *args, **kwargs):
  class CamembertTokenizerFast (line 61) | class CamembertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 64) | def __init__(self, *args, **kwargs):
  class CLIPTokenizerFast (line 68) | class CLIPTokenizerFast(metaclass=DummyObject):
    method __init__ (line 71) | def __init__(self, *args, **kwargs):
  class CodeGenTokenizerFast (line 75) | class CodeGenTokenizerFast(metaclass=DummyObject):
    method __init__ (line 78) | def __init__(self, *args, **kwargs):
  class ConvBertTokenizerFast (line 82) | class ConvBertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 85) | def __init__(self, *args, **kwargs):
  class CpmTokenizerFast (line 89) | class CpmTokenizerFast(metaclass=DummyObject):
    method __init__ (line 92) | def __init__(self, *args, **kwargs):
  class DebertaTokenizerFast (line 96) | class DebertaTokenizerFast(metaclass=DummyObject):
    method __init__ (line 99) | def __init__(self, *args, **kwargs):
  class DebertaV2TokenizerFast (line 103) | class DebertaV2TokenizerFast(metaclass=DummyObject):
    method __init__ (line 106) | def __init__(self, *args, **kwargs):
  class DistilBertTokenizerFast (line 110) | class DistilBertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 113) | def __init__(self, *args, **kwargs):
  class DPRContextEncoderTokenizerFast (line 117) | class DPRContextEncoderTokenizerFast(metaclass=DummyObject):
    method __init__ (line 120) | def __init__(self, *args, **kwargs):
  class DPRQuestionEncoderTokenizerFast (line 124) | class DPRQuestionEncoderTokenizerFast(metaclass=DummyObject):
    method __init__ (line 127) | def __init__(self, *args, **kwargs):
  class DPRReaderTokenizerFast (line 131) | class DPRReaderTokenizerFast(metaclass=DummyObject):
    method __init__ (line 134) | def __init__(self, *args, **kwargs):
  class ElectraTokenizerFast (line 138) | class ElectraTokenizerFast(metaclass=DummyObject):
    method __init__ (line 141) | def __init__(self, *args, **kwargs):
  class FNetTokenizerFast (line 145) | class FNetTokenizerFast(metaclass=DummyObject):
    method __init__ (line 148) | def __init__(self, *args, **kwargs):
  class FunnelTokenizerFast (line 152) | class FunnelTokenizerFast(metaclass=DummyObject):
    method __init__ (line 155) | def __init__(self, *args, **kwargs):
  class GPT2TokenizerFast (line 159) | class GPT2TokenizerFast(metaclass=DummyObject):
    method __init__ (line 162) | def __init__(self, *args, **kwargs):
  class GPTNeoXTokenizerFast (line 166) | class GPTNeoXTokenizerFast(metaclass=DummyObject):
    method __init__ (line 169) | def __init__(self, *args, **kwargs):
  class GPTNeoXJapaneseTokenizer (line 173) | class GPTNeoXJapaneseTokenizer(metaclass=DummyObject):
    method __init__ (line 176) | def __init__(self, *args, **kwargs):
  class HerbertTokenizerFast (line 180) | class HerbertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 183) | def __init__(self, *args, **kwargs):
  class LayoutLMTokenizerFast (line 187) | class LayoutLMTokenizerFast(metaclass=DummyObject):
    method __init__ (line 190) | def __init__(self, *args, **kwargs):
  class LayoutLMv2TokenizerFast (line 194) | class LayoutLMv2TokenizerFast(metaclass=DummyObject):
    method __init__ (line 197) | def __init__(self, *args, **kwargs):
  class LayoutLMv3TokenizerFast (line 201) | class LayoutLMv3TokenizerFast(metaclass=DummyObject):
    method __init__ (line 204) | def __init__(self, *args, **kwargs):
  class LayoutXLMTokenizerFast (line 208) | class LayoutXLMTokenizerFast(metaclass=DummyObject):
    method __init__ (line 211) | def __init__(self, *args, **kwargs):
  class LEDTokenizerFast (line 215) | class LEDTokenizerFast(metaclass=DummyObject):
    method __init__ (line 218) | def __init__(self, *args, **kwargs):
  class LlamaTokenizerFast (line 222) | class LlamaTokenizerFast(metaclass=DummyObject):
    method __init__ (line 225) | def __init__(self, *args, **kwargs):
  class LongformerTokenizerFast (line 229) | class LongformerTokenizerFast(metaclass=DummyObject):
    method __init__ (line 232) | def __init__(self, *args, **kwargs):
  class LxmertTokenizerFast (line 236) | class LxmertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 239) | def __init__(self, *args, **kwargs):
  class MarkupLMTokenizerFast (line 243) | class MarkupLMTokenizerFast(metaclass=DummyObject):
    method __init__ (line 246) | def __init__(self, *args, **kwargs):
  class MBartTokenizerFast (line 250) | class MBartTokenizerFast(metaclass=DummyObject):
    method __init__ (line 253) | def __init__(self, *args, **kwargs):
  class MBart50TokenizerFast (line 257) | class MBart50TokenizerFast(metaclass=DummyObject):
    method __init__ (line 260) | def __init__(self, *args, **kwargs):
  class MobileBertTokenizerFast (line 264) | class MobileBertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 267) | def __init__(self, *args, **kwargs):
  class MPNetTokenizerFast (line 271) | class MPNetTokenizerFast(metaclass=DummyObject):
    method __init__ (line 274) | def __init__(self, *args, **kwargs):
  class MT5TokenizerFast (line 278) | class MT5TokenizerFast(metaclass=DummyObject):
    method __init__ (line 281) | def __init__(self, *args, **kwargs):
  class MvpTokenizerFast (line 285) | class MvpTokenizerFast(metaclass=DummyObject):
    method __init__ (line 288) | def __init__(self, *args, **kwargs):
  class NllbTokenizerFast (line 292) | class NllbTokenizerFast(metaclass=DummyObject):
    method __init__ (line 295) | def __init__(self, *args, **kwargs):
  class OpenAIGPTTokenizerFast (line 299) | class OpenAIGPTTokenizerFast(metaclass=DummyObject):
    method __init__ (line 302) | def __init__(self, *args, **kwargs):
  class PegasusTokenizerFast (line 306) | class PegasusTokenizerFast(metaclass=DummyObject):
    method __init__ (line 309) | def __init__(self, *args, **kwargs):
  class RealmTokenizerFast (line 313) | class RealmTokenizerFast(metaclass=DummyObject):
    method __init__ (line 316) | def __init__(self, *args, **kwargs):
  class ReformerTokenizerFast (line 320) | class ReformerTokenizerFast(metaclass=DummyObject):
    method __init__ (line 323) | def __init__(self, *args, **kwargs):
  class RemBertTokenizerFast (line 327) | class RemBertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 330) | def __init__(self, *args, **kwargs):
  class RetriBertTokenizerFast (line 334) | class RetriBertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 337) | def __init__(self, *args, **kwargs):
  class RobertaTokenizerFast (line 341) | class RobertaTokenizerFast(metaclass=DummyObject):
    method __init__ (line 344) | def __init__(self, *args, **kwargs):
  class RoFormerTokenizerFast (line 348) | class RoFormerTokenizerFast(metaclass=DummyObject):
    method __init__ (line 351) | def __init__(self, *args, **kwargs):
  class SplinterTokenizerFast (line 355) | class SplinterTokenizerFast(metaclass=DummyObject):
    method __init__ (line 358) | def __init__(self, *args, **kwargs):
  class SqueezeBertTokenizerFast (line 362) | class SqueezeBertTokenizerFast(metaclass=DummyObject):
    method __init__ (line 365) | def __init__(self, *args, **kwargs):
  class T5TokenizerFast (line 369) | class T5TokenizerFast(metaclass=DummyObject):
    method __init__ (line 372) | def __init__(self, *args, **kwargs):
  class WhisperTokenizerFast (line 376) | class WhisperTokenizerFast(metaclass=DummyObject):
    method __init__ (line 379) | def __init__(self, *args, **kwargs):
  class XGLMTokenizerFast (line 383) | class XGLMTokenizerFast(metaclass=DummyObject):
    method __init__ (line 386) | def __init__(self, *args, **kwargs):
  class XLMRobertaTokenizerFast (line 390) | class XLMRobertaTokenizerFast(metaclass=DummyObject):
    method __init__ (line 393) | def __init__(self, *args, **kwargs):
  class XLNetTokenizerFast (line 397) | class XLNetTokenizerFast(metaclass=DummyObject):
    method __init__ (line 400) | def __init__(self, *args, **kwargs):
  class PreTrainedTokenizerFast (line 404) | class PreTrainedTokenizerFast(metaclass=DummyObject):
    method __init__ (line 407) | def __init__(self, *args, **kwargs):

FILE: transformers/utils/dummy_vision_objects.py
  class ImageProcessingMixin (line 5) | class ImageProcessingMixin(metaclass=DummyObject):
    method __init__ (line 8) | def __init__(self, *args, **kwargs):
  class ImageFeatureExtractionMixin (line 12) | class ImageFeatureExtractionMixin(metaclass=DummyObject):
    method __init__ (line 15) | def __init__(self, *args, **kwargs):
  class BeitFeatureExtractor (line 19) | class BeitFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 22) | def __init__(self, *args, **kwargs):
  class BeitImageProcessor (line 26) | class BeitImageProcessor(metaclass=DummyObject):
    method __init__ (line 29) | def __init__(self, *args, **kwargs):
  class BitImageProcessor (line 33) | class BitImageProcessor(metaclass=DummyObject):
    method __init__ (line 36) | def __init__(self, *args, **kwargs):
  class BlipImageProcessor (line 40) | class BlipImageProcessor(metaclass=DummyObject):
    method __init__ (line 43) | def __init__(self, *args, **kwargs):
  class BridgeTowerImageProcessor (line 47) | class BridgeTowerImageProcessor(metaclass=DummyObject):
    method __init__ (line 50) | def __init__(self, *args, **kwargs):
  class ChineseCLIPFeatureExtractor (line 54) | class ChineseCLIPFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 57) | def __init__(self, *args, **kwargs):
  class ChineseCLIPImageProcessor (line 61) | class ChineseCLIPImageProcessor(metaclass=DummyObject):
    method __init__ (line 64) | def __init__(self, *args, **kwargs):
  class CLIPFeatureExtractor (line 68) | class CLIPFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 71) | def __init__(self, *args, **kwargs):
  class CLIPImageProcessor (line 75) | class CLIPImageProcessor(metaclass=DummyObject):
    method __init__ (line 78) | def __init__(self, *args, **kwargs):
  class ConditionalDetrFeatureExtractor (line 82) | class ConditionalDetrFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 85) | def __init__(self, *args, **kwargs):
  class ConditionalDetrImageProcessor (line 89) | class ConditionalDetrImageProcessor(metaclass=DummyObject):
    method __init__ (line 92) | def __init__(self, *args, **kwargs):
  class ConvNextFeatureExtractor (line 96) | class ConvNextFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 99) | def __init__(self, *args, **kwargs):
  class ConvNextImageProcessor (line 103) | class ConvNextImageProcessor(metaclass=DummyObject):
    method __init__ (line 106) | def __init__(self, *args, **kwargs):
  class DeformableDetrFeatureExtractor (line 110) | class DeformableDetrFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 113) | def __init__(self, *args, **kwargs):
  class DeformableDetrImageProcessor (line 117) | class DeformableDetrImageProcessor(metaclass=DummyObject):
    method __init__ (line 120) | def __init__(self, *args, **kwargs):
  class DeiTFeatureExtractor (line 124) | class DeiTFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 127) | def __init__(self, *args, **kwargs):
  class DeiTImageProcessor (line 131) | class DeiTImageProcessor(metaclass=DummyObject):
    method __init__ (line 134) | def __init__(self, *args, **kwargs):
  class DetaImageProcessor (line 138) | class DetaImageProcessor(metaclass=DummyObject):
    method __init__ (line 141) | def __init__(self, *args, **kwargs):
  class DetrFeatureExtractor (line 145) | class DetrFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 148) | def __init__(self, *args, **kwargs):
  class DetrImageProcessor (line 152) | class DetrImageProcessor(metaclass=DummyObject):
    method __init__ (line 155) | def __init__(self, *args, **kwargs):
  class DonutFeatureExtractor (line 159) | class DonutFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 162) | def __init__(self, *args, **kwargs):
  class DonutImageProcessor (line 166) | class DonutImageProcessor(metaclass=DummyObject):
    method __init__ (line 169) | def __init__(self, *args, **kwargs):
  class DPTFeatureExtractor (line 173) | class DPTFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 176) | def __init__(self, *args, **kwargs):
  class DPTImageProcessor (line 180) | class DPTImageProcessor(metaclass=DummyObject):
    method __init__ (line 183) | def __init__(self, *args, **kwargs):
  class EfficientFormerImageProcessor (line 187) | class EfficientFormerImageProcessor(metaclass=DummyObject):
    method __init__ (line 190) | def __init__(self, *args, **kwargs):
  class EfficientNetImageProcessor (line 194) | class EfficientNetImageProcessor(metaclass=DummyObject):
    method __init__ (line 197) | def __init__(self, *args, **kwargs):
  class FlavaFeatureExtractor (line 201) | class FlavaFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 204) | def __init__(self, *args, **kwargs):
  class FlavaImageProcessor (line 208) | class FlavaImageProcessor(metaclass=DummyObject):
    method __init__ (line 211) | def __init__(self, *args, **kwargs):
  class FlavaProcessor (line 215) | class FlavaProcessor(metaclass=DummyObject):
    method __init__ (line 218) | def __init__(self, *args, **kwargs):
  class GLPNFeatureExtractor (line 222) | class GLPNFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 225) | def __init__(self, *args, **kwargs):
  class GLPNImageProcessor (line 229) | class GLPNImageProcessor(metaclass=DummyObject):
    method __init__ (line 232) | def __init__(self, *args, **kwargs):
  class ImageGPTFeatureExtractor (line 236) | class ImageGPTFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 239) | def __init__(self, *args, **kwargs):
  class ImageGPTImageProcessor (line 243) | class ImageGPTImageProcessor(metaclass=DummyObject):
    method __init__ (line 246) | def __init__(self, *args, **kwargs):
  class LayoutLMv2FeatureExtractor (line 250) | class LayoutLMv2FeatureExtractor(metaclass=DummyObject):
    method __init__ (line 253) | def __init__(self, *args, **kwargs):
  class LayoutLMv2ImageProcessor (line 257) | class LayoutLMv2ImageProcessor(metaclass=DummyObject):
    method __init__ (line 260) | def __init__(self, *args, **kwargs):
  class LayoutLMv3FeatureExtractor (line 264) | class LayoutLMv3FeatureExtractor(metaclass=DummyObject):
    method __init__ (line 267) | def __init__(self, *args, **kwargs):
  class LayoutLMv3ImageProcessor (line 271) | class LayoutLMv3ImageProcessor(metaclass=DummyObject):
    method __init__ (line 274) | def __init__(self, *args, **kwargs):
  class LevitFeatureExtractor (line 278) | class LevitFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 281) | def __init__(self, *args, **kwargs):
  class LevitImageProcessor (line 285) | class LevitImageProcessor(metaclass=DummyObject):
    method __init__ (line 288) | def __init__(self, *args, **kwargs):
  class Mask2FormerImageProcessor (line 292) | class Mask2FormerImageProcessor(metaclass=DummyObject):
    method __init__ (line 295) | def __init__(self, *args, **kwargs):
  class MaskFormerFeatureExtractor (line 299) | class MaskFormerFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 302) | def __init__(self, *args, **kwargs):
  class MaskFormerImageProcessor (line 306) | class MaskFormerImageProcessor(metaclass=DummyObject):
    method __init__ (line 309) | def __init__(self, *args, **kwargs):
  class MobileNetV1FeatureExtractor (line 313) | class MobileNetV1FeatureExtractor(metaclass=DummyObject):
    method __init__ (line 316) | def __init__(self, *args, **kwargs):
  class MobileNetV1ImageProcessor (line 320) | class MobileNetV1ImageProcessor(metaclass=DummyObject):
    method __init__ (line 323) | def __init__(self, *args, **kwargs):
  class MobileNetV2FeatureExtractor (line 327) | class MobileNetV2FeatureExtractor(metaclass=DummyObject):
    method __init__ (line 330) | def __init__(self, *args, **kwargs):
  class MobileNetV2ImageProcessor (line 334) | class MobileNetV2ImageProcessor(metaclass=DummyObject):
    method __init__ (line 337) | def __init__(self, *args, **kwargs):
  class MobileViTFeatureExtractor (line 341) | class MobileViTFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 344) | def __init__(self, *args, **kwargs):
  class MobileViTImageProcessor (line 348) | class MobileViTImageProcessor(metaclass=DummyObject):
    method __init__ (line 351) | def __init__(self, *args, **kwargs):
  class OneFormerImageProcessor (line 355) | class OneFormerImageProcessor(metaclass=DummyObject):
    method __init__ (line 358) | def __init__(self, *args, **kwargs):
  class OwlViTFeatureExtractor (line 362) | class OwlViTFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 365) | def __init__(self, *args, **kwargs):
  class OwlViTImageProcessor (line 369) | class OwlViTImageProcessor(metaclass=DummyObject):
    method __init__ (line 372) | def __init__(self, *args, **kwargs):
  class PerceiverFeatureExtractor (line 376) | class PerceiverFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 379) | def __init__(self, *args, **kwargs):
  class PerceiverImageProcessor (line 383) | class PerceiverImageProcessor(metaclass=DummyObject):
    method __init__ (line 386) | def __init__(self, *args, **kwargs):
  class Pix2StructImageProcessor (line 390) | class Pix2StructImageProcessor(metaclass=DummyObject):
    method __init__ (line 393) | def __init__(self, *args, **kwargs):
  class PoolFormerFeatureExtractor (line 397) | class PoolFormerFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 400) | def __init__(self, *args, **kwargs):
  class PoolFormerImageProcessor (line 404) | class PoolFormerImageProcessor(metaclass=DummyObject):
    method __init__ (line 407) | def __init__(self, *args, **kwargs):
  class SamImageProcessor (line 411) | class SamImageProcessor(metaclass=DummyObject):
    method __init__ (line 414) | def __init__(self, *args, **kwargs):
  class SegformerFeatureExtractor (line 418) | class SegformerFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 421) | def __init__(self, *args, **kwargs):
  class SegformerImageProcessor (line 425) | class SegformerImageProcessor(metaclass=DummyObject):
    method __init__ (line 428) | def __init__(self, *args, **kwargs):
  class Swin2SRImageProcessor (line 432) | class Swin2SRImageProcessor(metaclass=DummyObject):
    method __init__ (line 435) | def __init__(self, *args, **kwargs):
  class TvltImageProcessor (line 439) | class TvltImageProcessor(metaclass=DummyObject):
    method __init__ (line 442) | def __init__(self, *args, **kwargs):
  class VideoMAEFeatureExtractor (line 446) | class VideoMAEFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 449) | def __init__(self, *args, **kwargs):
  class VideoMAEImageProcessor (line 453) | class VideoMAEImageProcessor(metaclass=DummyObject):
    method __init__ (line 456) | def __init__(self, *args, **kwargs):
  class ViltFeatureExtractor (line 460) | class ViltFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 463) | def __init__(self, *args, **kwargs):
  class ViltImageProcessor (line 467) | class ViltImageProcessor(metaclass=DummyObject):
    method __init__ (line 470) | def __init__(self, *args, **kwargs):
  class ViltProcessor (line 474) | class ViltProcessor(metaclass=DummyObject):
    method __init__ (line 477) | def __init__(self, *args, **kwargs):
  class ViTFeatureExtractor (line 481) | class ViTFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 484) | def __init__(self, *args, **kwargs):
  class ViTImageProcessor (line 488) | class ViTImageProcessor(metaclass=DummyObject):
    method __init__ (line 491) | def __init__(self, *args, **kwargs):
  class ViTHybridImageProcessor (line 495) | class ViTHybridImageProcessor(metaclass=DummyObject):
    method __init__ (line 498) | def __init__(self, *args, **kwargs):
  class YolosFeatureExtractor (line 502) | class YolosFeatureExtractor(metaclass=DummyObject):
    method __init__ (line 505) | def __init__(self, *args, **kwargs):
  class YolosImageProcessor (line 509) | class YolosImageProcessor(metaclass=DummyObject):
    method __init__ (line 512) | def __init__(self, *args, **kwargs):

FILE: transformers/utils/fx.py
  function _generate_supported_model_class_names (line 73) | def _generate_supported_model_class_names(
  function torch_nn_embedding (line 184) | def torch_nn_embedding(self, input):
  function torch_nn_functional_embedding (line 188) | def torch_nn_functional_embedding(
  function torch_nn_layernorm (line 194) | def torch_nn_layernorm(self, input):
  function torch_nn_groupnorm (line 198) | def torch_nn_groupnorm(self, input):
  function torch_nn_linear (line 202) | def torch_nn_linear(self, input):
  function torch_relu (line 206) | def torch_relu(x):
  function torch_nn_relu (line 210) | def torch_nn_relu(self, x):
  function torch_nn_functional_relu (line 214) | def torch_nn_functional_relu(x, inplace=False):
  function torch_where (line 220) | def torch_where(condition, x, y):
  function torch_abs (line 226) | def torch_abs(input, *, out=None):
  function torch_arange (line 232) | def torch_arange(*args, **kwargs):
  function torch_full (line 253) | def torch_full(*args, **kwargs):
  function torch_cat (line 262) | def torch_cat(tensors, dim=None, axis=None, *, out=None):
  function torch_stack (line 276) | def torch_stack(tensors, dim=None, axis=None, *, out=None):
  function torch_add (line 288) | def torch_add(input, other, *, alpha=1, out=None):
  function torch_mul (line 302) | def torch_mul(input, other, *, out=None):
  function torch_tensor_mul (line 306) | def torch_tensor_mul(self, other):
  function torch_matmul (line 310) | def torch_matmul(input, other, *, out=None):
  function torch_bmm (line 346) | def torch_bmm(input, mat2, *, out=None):
  function torch_baddbmm (line 354) | def torch_baddbmm(input, batch1, batch2, *, beta=1, alpha=1, out=None):
  function torch_tensor_baddbmm (line 360) | def torch_tensor_baddbmm(self, batch1, batch2, *, beta=1, alpha=1, out=N...
  function torch_einsum (line 364) | def torch_einsum(equation, *operands):
  function torch_tensor_repeat (line 370) | def torch_tensor_repeat(self, *sizes):
  function torch_repeat_interleave (line 377) | def torch_repeat_interleave(*args, dim=None, output_size=None):
  function torch_index_select (line 397) | def torch_index_select(input, dim, index, *, out=None):
  function torch_tensor_index_select (line 403) | def torch_tensor_index_select(self, dim, index):
  function torch_gather (line 407) | def torch_gather(input, dim, index, *, sparse_grad=False, out=None):
  function torch_tensor_gather (line 413) | def torch_tensor_gather(self, dim, index):
  function torch_roll (line 417) | def torch_roll(input, shifts, dims=None):
  function torch_flip (line 421) | def torch_flip(input, dims):
  function torch_tensor_flip (line 425) | def torch_tensor_flip(self, dims):
  function torch_nn_conv1d (line 429) | def torch_nn_conv1d(self, input):
  function torch_nn_conv2d (line 447) | def torch_nn_conv2d(self, input):
  function torch_squeeze (line 468) | def torch_squeeze(input, dim=None):
  function torch_tensor_squeeze (line 485) | def torch_tensor_squeeze(self, dim=None):
  function torch_unsqueeze (line 489) | def torch_unsqueeze(input, dim):
  function torch_tensor_unsqueeze (line 497) | def torch_tensor_unsqueeze(self, dim):
  function torch_unique_consecutive (line 501) | def torch_unique_consecutive(input, **kwargs):
  function torch_nn_functional_one_hot (line 509) | def torch_nn_functional_one_hot(tensor, num_classes=-1):
  function torch_nn_mseloss (line 516) | def torch_nn_mseloss(self, input, target):
  function torch_nn_crossentropyloss (line 524) | def torch_nn_crossentropyloss(self, input, target):
  function torch_nn_bcewithlogitsloss (line 532) | def torch_nn_bcewithlogitsloss(self, input, target):
  function operator_getitem (line 540) | def operator_getitem(a, b):
  class HFProxy (line 606) | class HFProxy(Proxy):
    method install_metadata (line 611) | def install_metadata(self, metadata):
    method shape (line 615) | def shape(self):
    method device (line 619) | def device(self):
    method __len__ (line 624) | def __len__(self):
    method __bool__ (line 629) | def __bool__(self):
    method __getattr__ (line 634) | def __getattr__(self, k):
    method __setitem__ (line 641) | def __setitem__(self, indices, values):
    method __contains__ (line 644) | def __contains__(self, key):
  class HFAttribute (line 650) | class HFAttribute(HFProxy):
    method __init__ (line 651) | def __init__(self, root, attr: str):
    method node (line 661) | def node(self):
    method __call__ (line 668) | def __call__(self, *args, **kwargs):
  class MetaDeviceAttribute (line 672) | class MetaDeviceAttribute(HFAttribute):
  function _proxies_to_metas (line 676) | def _proxies_to_metas(v):
  function _gen_constructor_wrapper (line 687) | def _gen_constructor_wrapper(target):
  function _generate_random_int (line 708) | def _generate_random_int(low: int = 10, high: int = 20, forbidden_values...
  class HFTracer (line 717) | class HFTracer(Tracer):
    method __init__ (line 740) | def __init__(self, autowrap_modules=(math,), autowrap_functions=()):
    method _generate_dummy_input (line 749) | def _generate_dummy_input(
    method create_proxy (line 876) | def create_proxy(self, kind, target, args, kwargs, name=None, type_exp...
    method _module_getattr (line 944) | def _module_getattr(self, attr, attr_val, parameter_proxy_cache):
    method getattr (line 982) | def getattr(self, attr: str, attr_val: Any, parameter_proxy_cache: Dic...
    method call_module (line 985) | def call_module(self, m, forward, args, kwargs):
    method proxy (line 989) | def proxy(self, node):
    method trace (line 992) | def trace(
    method _stateless_mod_instanciation_depends_on_proxies (line 1122) | def _stateless_mod_instanciation_depends_on_proxies(self, mod: nn.Modu...
    method _insert_module_as_submodule (line 1129) | def _insert_module_as_submodule(self, mod: nn.Module) -> str:
    method path_of_module (line 1153) | def path_of_module(self, mod: nn.Module) -> str:
    method is_leaf_module (line 1170) | def is_leaf_module(self, m: torch.nn.Module, module_qualified_name: st...
    method keys (line 1176) | def keys(self, obj: "Proxy") -> Any:
  function get_concrete_args (line 1187) | def get_concrete_args(model: nn.Module, input_names: List[str]):
  function check_if_model_is_supported (line 1201) | def check_if_model_is_supported(model: PreTrainedModel):
  function symbolic_trace (line 1209) | def symbolic_trace(

FILE: transformers/utils/generic.py
  class cached_property (line 36) | class cached_property(property):
    method __get__ (line 45) | def __get__(self, obj, objtype=None):
  function strtobool (line 60) | def strtobool(val):
  function is_tensor (line 74) | def is_tensor(x):
  function _is_numpy (line 101) | def _is_numpy(x):
  function is_numpy_array (line 105) | def is_numpy_array(x):
  function _is_torch (line 112) | def _is_torch(x):
  function is_torch_tensor (line 118) | def is_torch_tensor(x):
  function _is_torch_device (line 125) | def _is_torch_device(x):
  function is_torch_device (line 131) | def is_torch_device(x):
  function _is_torch_dtype (line 138) | def _is_torch_dtype(x):
  function is_torch_dtype (line 149) | def is_torch_dtype(x):
  function _is_tensorflow (line 156) | def _is_tensorflow(x):
  function is_tf_tensor (line 162) | def is_tf_tensor(x):
  function _is_tf_symbolic_tensor (line 169) | def _is_tf_symbolic_tensor(x):
  function is_tf_symbolic_tensor (line 178) | def is_tf_symbolic_tensor(x):
  function _is_jax (line 186) | def _is_jax(x):
  function is_jax_tensor (line 192) | def is_jax_tensor(x):
  function to_py_obj (line 199) | def to_py_obj(obj):
  function to_numpy (line 219) | def to_numpy(obj):
  class ModelOutput (line 237) | class ModelOutput(OrderedDict):
    method __post_init__ (line 251) | def __post_init__(self):
    method __delitem__ (line 303) | def __delitem__(self, *args, **kwargs):
    method setdefault (line 306) | def setdefault(self, *args, **kwargs):
    method pop (line 309) | def pop(self, *args, **kwargs):
    method update (line 312) | def update(self, *args, **kwargs):
    method __getitem__ (line 315) | def __getitem__(self, k):
    method __setattr__ (line 322) | def __setattr__(self, name, value):
    method __setitem__ (line 328) | def __setitem__(self, key, value):
    method to_tuple (line 334) | def to_tuple(self) -> Tuple[Any]:
  class ExplicitEnum (line 341) | class ExplicitEnum(str, Enum):
    method _missing_ (line 347) | def _missing_(cls, value):
  class PaddingStrategy (line 353) | class PaddingStrategy(ExplicitEnum):
  class TensorType (line 364) | class TensorType(ExplicitEnum):
  class ContextManagers (line 376) | class ContextManagers:
    method __init__ (line 382) | def __init__(self, context_managers: List[ContextManager]):
    method __enter__ (line 386) | def __enter__(self):
    method __exit__ (line 390) | def __exit__(self, *args, **kwargs):
  function can_return_loss (line 394) | def can_return_loss(model_class):
  function find_labels (line 416) | def find_labels(model_class):
  function flatten_dict (line 438) | def flatten_dict(d: MutableMapping, parent_key: str = "", delimiter: str...
  function working_or_temp_dir (line 453) | def working_or_temp_dir(working_dir, use_temp_dir: bool = False):
  function transpose (line 461) | def transpose(array, axes=None):
  function reshape (line 480) | def reshape(array, newshape):
  function squeeze (line 499) | def squeeze(array, axis=None):
  function expand_dims (line 518) | def expand_dims(array, axis):
  function tensor_size (line 537) | def tensor_size(array):
  function add_model_info_to_auto_map (line 555) | def add_model_info_to_auto_map(auto_map, repo_id):
  function infer_framework (line 568) | def infer_framework(model_class):

FILE: transformers/utils/hp_naming.py
  class TrialShortNamer (line 19) | class TrialShortNamer:
    method set_defaults (line 25) | def set_defaults(cls, prefix, defaults):
    method shortname_for_word (line 31) | def shortname_for_word(info, word):
    method shortname_for_key (line 70) | def shortname_for_key(info, param_name):
    method add_new_param_name (line 89) | def add_new_param_name(info, param_name):
    method build_naming_info (line 95) | def build_naming_info(cls):
    method shortname (line 114) | def shortname(cls, params):
    method parse_repr (line 138) | def parse_repr(cls, repr):

FILE: transformers/utils/hub.py
  function is_offline_mode (line 70) | def is_offline_mode():
  function is_remote_url (line 130) | def is_remote_url(url_or_filename):
  function get_cached_models (line 135) | def get_cached_models(cache_dir: Union[str, Path] = None) -> List[Tuple]:
  function define_sagemaker_information (line 170) | def define_sagemaker_information():
  function http_user_agent (line 196) | def http_user_agent(user_agent: Union[Dict, str, None] = None) -> str:
  function extract_commit_hash (line 219) | def extract_commit_hash(resolved_file: Optional[str], commit_hash: Optio...
  function try_to_load_from_cache (line 233) | def try_to_load_from_cache(
  function cached_file (line 300) | def cached_file(
  function get_file_from_repo (line 480) | def get_file_from_repo(
  function download_url (line 561) | def download_url(url, proxies=None):
  function has_file (line 587) | def has_file(
  class PushToHubMixin (line 628) | class PushToHubMixin:
    method _create_repo (line 633) | def _create_repo(
    method _get_files_timestamps (line 668) | def _get_files_timestamps(self, working_dir: Union[str, os.PathLike]):
    method _upload_modified_files (line 674) | def _upload_modified_files(
    method push_to_hub (line 733) | def push_to_hub(
  function get_full_repo_name (line 826) | def get_full_repo_name(model_id: str, organization: Optional[str] = None...
  function send_example_telemetry (line 834) | def send_example_telemetry(example_name, *example_args, framework="pytor...
  function convert_file_size_to_int (line 872) | def convert_file_size_to_int(size: Union[int, str]):
  function get_checkpoint_shard_files (line 905) | def get_checkpoint_shard_files(
  function get_all_cached_files (line 993) | def get_all_cached_files(cache_dir=None):
  function extract_info_from_url (line 1019) | def extract_info_from_url(url):
  function clean_files_for (line 1031) | def clean_files_for(file):
  function move_to_new_cache (line 1040) | def move_to_new_cache(file, repo, filename, revision, etag, commit_hash):
  function move_cache (line 1066) | def move_cache(cache_dir=None, new_cache_dir=None, token=None):

FILE: transformers/utils/import_utils.py
  function _is_package_available (line 41) | def _is_package_available(pkg_name: str, return_version: bool = False) -...
  function is_kenlm_available (line 218) | def is_kenlm_available():
  function is_torch_available (line 222) | def is_torch_available():
  function get_torch_version (line 226) | def get_torch_version():
  function is_torchvision_available (line 230) | def is_torchvision_available():
  function is_pyctcdecode_available (line 234) | def is_pyctcdecode_available():
  function is_librosa_available (line 238) | def is_librosa_available():
  function is_torch_cuda_available (line 242) | def is_torch_cuda_available():
  function is_torch_bf16_gpu_available (line 251) | def is_torch_bf16_gpu_available():
  function is_torch_bf16_cpu_available (line 283) | def is_torch_bf16_cpu_available():
  function is_torch_bf16_available (line 301) | def is_torch_bf16_available():
  function is_torch_tf32_available (line 312) | def is_torch_tf32_available():
  function is_torch_fx_available (line 330) | def is_torch_fx_available():
  function is_peft_available (line 334) | def is_peft_available():
  function is_bs4_available (line 338) | def is_bs4_available():
  function is_tf_available (line 342) | def is_tf_available():
  function is_coloredlogs_available (line 346) | def is_coloredlogs_available():
  function is_tf2onnx_available (line 350) | def is_tf2onnx_available():
  function is_onnx_available (line 354) | def is_onnx_available():
  function is_openai_available (line 358) | def is_openai_available():
  function is_flax_available (line 362) | def is_flax_available():
  function is_ftfy_available (line 366) | def is_ftfy_available():
  function is_torch_tpu_available (line 371) | def is_torch_tpu_available(check_device=True):
  function is_torch_neuroncore_available (line 390) | def is_torch_neuroncore_available(check_device=True):
  function is_torchdynamo_available (line 396) | def is_torchdynamo_available():
  function is_torch_compile_available (line 407) | def is_torch_compile_available():
  function is_torch_tensorrt_fx_available (line 418) | def is_torch_tensorrt_fx_available():
  function is_datasets_available (line 424) | def is_datasets_available():
  function is_detectron2_available (line 428) | def is_detectron2_available():
  function is_rjieba_available (line 432) | def is_rjieba_available():
  function is_psutil_available (line 436) | def is_psutil_available():
  function is_py3nvml_available (line 440) | def is_py3nvml_available():
  function is_sacremoses_available (line 444) | def is_sacremoses_available():
  function is_apex_available (line 448) | def is_apex_available():
  function is_ninja_available (line 452) | def is_ninja_available():
  function is_ipex_available (line 465) | def is_ipex_available():
  function is_bitsandbytes_available (line 483) | def is_bitsandbytes_available():
  function is_torchdistx_available (line 487) | def is_torchdistx_available():
  function is_faiss_available (line 491) | def is_faiss_available():
  function is_scipy_available (line 495) | def is_scipy_available():
  function is_sklearn_available (line 499) | def is_sklearn_available():
  function is_sentencepiece_available (line 503) | def is_sentencepiece_available():
  function is_protobuf_available (line 507) | def is_protobuf_available():
  function is_accelerate_available (line 513) | def is_accelerate_available(min_version: str = None):
  function is_optimum_available (line 519) | def is_optimum_available():
  function is_optimum_neuron_available (line 523) | def is_optimum_neuron_available():
  function is_safetensors_available (line 527) | def is_safetensors_available():
  function is_tokenizers_available (line 533) | def is_tokenizers_available():
  function is_vision_available (line 537) | def is_vision_available():
  function is_pytesseract_available (line 548) | def is_pytesseract_available():
  function is_spacy_available (line 552) | def is_spacy_available():
  function is_tensorflow_text_available (line 556) | def is_tensorflow_text_available():
  function is_keras_nlp_available (line 560) | def is_keras_nlp_available():
  function is_in_notebook (line 564) | def is_in_notebook():
  function is_pytorch_quantization_available (line 582) | def is_pytorch_quantization_available():
  function is_tensorflow_probability_available (line 586) | def is_tensorflow_probability_available():
  function is_pandas_available (line 590) | def is_pandas_available():
  function is_sagemaker_dp_enabled (line 594) | def is_sagemaker_dp_enabled():
  function is_sagemaker_mp_enabled (line 608) | def is_sagemaker_mp_enabled():
  function is_training_run_on_sagemaker (line 632) | def is_training_run_on_sagemaker():
  function is_soundfile_availble (line 636) | def is_soundfile_availble():
  function is_timm_available (line 640) | def is_timm_available():
  function is_natten_available (line 644) | def is_natten_available():
  function is_torchaudio_available (line 648) | def is_torchaudio_available():
  function is_speech_available (line 652) | def is_speech_available():
  function is_phonemizer_available (line 657) | def is_phonemizer_available():
  function torch_only_method (line 661) | def torch_only_method(fn):
  function is_ccl_available (line 674) | def is_ccl_available():
  function is_decord_available (line 678) | def is_decord_available():
  function is_sudachi_available (line 682) | def is_sudachi_available():
  function is_jumanpp_available (line 686) | def is_jumanpp_available():
  function is_cython_available (line 690) | def is_cython_available():
  function is_jieba_available (line 694) | def is_jieba_available():
  function requires_backends (line 997) | def requires_backends(obj, backends):
  class DummyObject (line 1017) | class DummyObject(type):
    method __getattribute__ (line 1023) | def __getattribute__(cls, key):
  function is_torch_fx_proxy (line 1029) | def is_torch_fx_proxy(x):
  class _LazyModule (line 1037) | class _LazyModule(ModuleType):
    method __init__ (line 1044) | def __init__(self, name, module_file, import_structure, module_spec=No...
    method __dir__ (line 1061) | def __dir__(self):
    method __getattr__ (line 1070) | def __getattr__(self, name: str) -> Any:
    method _get_module (line 1084) | def _get_module(self, module_name: str):
    method __reduce__ (line 1093) | def __reduce__(self):
  class OptionalDependencyNotAvailable (line 1097) | class OptionalDependencyNotAvailable(BaseException):
  function direct_transformers_import (line 1101) | def direct_transformers_import(path: str, file="__init__.py") -> ModuleT...

FILE: transformers/utils/logging.py
  function _get_default_logging_level (line 55) | def _get_default_logging_level():
  function _get_library_name (line 72) | def _get_library_name() -> str:
  function _get_library_root_logger (line 76) | def _get_library_root_logger() -> logging.Logger:
  function _configure_library_root_logger (line 80) | def _configure_library_root_logger() -> None:
  function _reset_library_root_logger (line 97) | def _reset_library_root_logger() -> None:
  function get_log_levels_dict (line 110) | def get_log_levels_dict():
  function get_logger (line 114) | def get_logger(name: Optional[str] = None) -> logging.Logger:
  function get_verbosity (line 128) | def get_verbosity() -> int:
  function set_verbosity (line 151) | def set_verbosity(verbosity: int) -> None:
  function set_verbosity_info (line 170) | def set_verbosity_info():
  function set_verbosity_warning (line 175) | def set_verbosity_warning():
  function set_verbosity_debug (line 180) | def set_verbosity_debug():
  function set_verbosity_error (line 185) | def set_verbosity_error():
  function disable_default_handler (line 190) | def disable_default_handler() -> None:
  function enable_default_handler (line 199) | def enable_default_handler() -> None:
  function add_handler (line 208) | def add_handler(handler: logging.Handler) -> None:
  function remove_handler (line 217) | def remove_handler(handler: logging.Handler) -> None:
  function disable_propagation (line 226) | def disable_propagation() -> None:
  function enable_propagation (line 235) | def enable_propagation() -> None:
  function enable_explicit_format (line 245) | def enable_explicit_format() -> None:
  function reset_format (line 260) | def reset_format() -> None:
  function warning_advice (line 272) | def warning_advice(self, *args, **kwargs):
  function warning_once (line 287) | def warning_once(self, *args, **kwargs):
  class EmptyTqdm (line 301) | class EmptyTqdm:
    method __init__ (line 304) | def __init__(self, *args, **kwargs):  # pylint: disable=unused-argument
    method __iter__ (line 307) | def __iter__(self):
    method __getattr__ (line 310) | def __getattr__(self, _):
    method __enter__ (line 318) | def __enter__(self):
    method __exit__ (line 321) | def __exit__(self, type_, value, traceback):
  class _tqdm_cls (line 325) | class _tqdm_cls:
    method __call__ (line 326) | def __call__(self, *args, **kwargs):
    method set_lock (line 332) | def set_lock(self, *args, **kwargs):
    method get_lock (line 337) | def get_lock(self):
  function is_progress_bar_enabled (line 345) | def is_progress_bar_enabled() -> bool:
  function enable_progress_bar (line 351) | def enable_progress_bar():
  function disable_progress_bar (line 358) | def disable_progress_bar():

FILE: transformers/utils/model_parallel_utils.py
  function assert_device_map (line 19) | def assert_device_map(device_map, num_blocks):
  function get_device_map (line 50) | def get_device_map(n_layers, devices):

FILE: transformers/utils/notebook.py
  function format_time (line 26) | def format_time(t):
  function html_progress_bar (line 33) | def html_progress_bar(value, total, prefix, label, width=300):
  function text_to_html_table (line 44) | def text_to_html_table(items):
  class NotebookProgressBar (line 61) | class NotebookProgressBar:
    method __init__ (line 101) | def __init__(
    method update (line 118) | def update(self, value: int, force_update: bool = False, comment: str ...
    method update_bar (line 169) | def update_bar(self, value, comment=None):
    method display (line 184) | def display(self):
    method close (line 195) | def close(self):
  class NotebookTrainingTracker (line 201) | class NotebookTrainingTracker(NotebookProgressBar):
    method __init__ (line 211) | def __init__(self, num_steps, column_names=None):
    method display (line 216) | def display(self):
    method write_line (line 227) | def write_line(self, values):
    method add_child (line 246) | def add_child(self, total, prefix=None, width=300):
    method remove_child (line 259) | def remove_child(self):
  class NotebookProgressCallback (line 267) | class NotebookProgressCallback(TrainerCallback):
    method __init__ (line 273) | def __init__(self):
    method on_train_begin (line 278) | def on_train_begin(self, args, state, control, **kwargs):
    method on_step_end (line 287) | def on_step_end(self, args, state, control, **kwargs):
    method on_prediction_step (line 296) | def on_prediction_step(self, args, state, control, eval_dataloader=Non...
    method on_predict (line 308) | def on_predict(self, args, state, control, **kwargs):
    method on_log (line 313) | def on_log(self, args, state, control, logs=None, **kwargs):
    method on_evaluate (line 321) | def on_evaluate(self, args, state, control, metrics=None, **kwargs):
    method on_train_end (line 356) | def on_train_end(self, args, state, control, **kwargs):

FILE: transformers/utils/quantization_config.py
  class BitsAndBytesConfig (line 37) | class BitsAndBytesConfig:
    method __init__ (line 87) | def __init__(
    method post_init (line 120) | def post_init(self):
    method is_quantizable (line 151) | def is_quantizable(self):
    method quantization_method (line 157) | def quantization_method(self):
    method from_dict (line 172) | def from_dict(cls, config_dict, return_unused_kwargs, **kwargs):
    method to_json_file (line 204) | def to_json_file(self, json_file_path: Union[str, os.PathLike]):
    method to_dict (line 221) | def to_dict(self) -> Dict[str, Any]:

FILE: transformers/utils/versions.py
  function _compare_versions (line 43) | def _compare_versions(op, got_ver, want_ver, requirement, pkg, hint):
  function require_version (line 55) | def require_version(requirement: str, hint: Optional[str] = None) -> None:
  function require_version_core (line 120) | def require_version_core(requirement):