SYMBOL INDEX (64 symbols across 18 files) FILE: examples/chat_gradio.py function user (line 23) | def user(user_message, history): function bot (line 25) | def bot(history,temperature,top_p,slider_context_times): FILE: examples/chat_gradio_no_merge.py function user (line 24) | def user(user_message, history): function bot (line 26) | def bot(history,temperature,top_p,slider_context_times): FILE: examples/llama2_for_langchain.py class Llama2 (line 7) | class Llama2(LLM): method __init__ (line 14) | def __init__(self, model_name_or_path, bit4=False): method _llm_type (line 31) | def _llm_type(self) -> str: method _call (line 34) | def _call(self, prompt: str, stop: Optional[List[str]] = None) -> str: FILE: inference-speed/GPU/TensorRT-LLM_example/atom_inference.py class AtomTRTApi (line 17) | class AtomTRTApi: method __init__ (line 18) | def __init__(self,engine_dir,tokenizer_dir,max_input_length=4096): method ask (line 49) | def ask(self,input_text,temperature=0.4,top_p=0.95,max_new_tokens=1024... method ask_streaming (line 117) | def ask_streaming(self,input_text,temperature=0.8,top_p=0.95,max_new_t... FILE: inference-speed/GPU/TensorRT-LLM_example/utils.py function get_engine_version (line 54) | def get_engine_version(engine_dir: str) -> Union[None, str]: function read_model_name (line 65) | def read_model_name(engine_dir: str): function throttle_generator (line 77) | def throttle_generator(generator, stream_interval): function load_tokenizer (line 86) | def load_tokenizer(tokenizer_dir: Optional[str] = None, FILE: inference-speed/GPU/lmdeploy_example/test_api_server.py function input_prompt (line 5) | def input_prompt(chat_history, system_prompt: str): function main (line 20) | def main(tritonserver_addr: str, FILE: inference-speed/GPU/vllm_example/api_server.py function generate (line 20) | async def generate(request: Request) -> Response: FILE: inference-speed/GPU/vllm_example/client_test.py function get_prompt_llama_chinese (line 14) | def get_prompt_llama_chinese( function get_prompt_llama2_meta (line 30) | def get_prompt_llama2_meta(chat_history, system_prompt=""): function get_prompt_llama3_meta (line 63) | def get_prompt_llama3_meta(chat_history, system_prompt=""): function test_api_server (line 80) | def test_api_server(chat_history=[], system_prompt=""): FILE: scripts/api/accelerate_client.py function test_api_server (line 7) | def test_api_server(input_text): FILE: scripts/api/accelerate_server.py function get_prompt_llama2chinese (line 32) | def get_prompt_llama2chinese( function get_prompt (line 48) | def get_prompt(chat_history, system_prompt=""): function get_prompt_llama3 (line 81) | def get_prompt_llama3(chat_history, system_prompt=""): function create_item (line 99) | async def create_item(request: Request): function get_world_size (line 147) | def get_world_size() -> int: function print_rank0 (line 153) | def print_rank0(*msg): FILE: scripts/convert2hf/convert_llama_weights_to_hf.py function compute_intermediate_size (line 75) | def compute_intermediate_size(n, ffn_dim_multiplier=1, multiple_of=256): function read_json (line 79) | def read_json(path): function write_json (line 84) | def write_json(text, path): function write_model (line 89) | def write_model(model_path, input_base_path, model_size, safe_serializat... function write_tokenizer (line 268) | def write_tokenizer(tokenizer_path, input_tokenizer_path): function main (line 276) | def main(): FILE: train/merge_peft_model/merge_muilt_peft_adapter.py class ScriptArguments (line 11) | class ScriptArguments: FILE: train/merge_peft_model/merge_peft_adapter.py class ScriptArguments (line 11) | class ScriptArguments: FILE: train/pretrain/accuracy.py class Accuracy (line 81) | class Accuracy(evaluate.Metric): method _info (line 82) | def _info(self): method _compute (line 101) | def _compute(self, predictions, references, normalize=True, sample_wei... FILE: train/pretrain/pretrain_clm.py class ModelArguments (line 76) | class ModelArguments: method __post_init__ (line 140) | def __post_init__(self): class DataTrainingArguments (line 148) | class DataTrainingArguments: method __post_init__ (line 210) | def __post_init__(self): function main (line 224) | def main(): function _mp_fn (line 608) | def _mp_fn(index): FILE: train/sft/accuracy.py class Accuracy (line 81) | class Accuracy(evaluate.Metric): method _info (line 82) | def _info(self): method _compute (line 101) | def _compute(self, predictions, references, normalize=True, sample_wei... FILE: train/sft/finetune_clm.py class ModelArguments (line 86) | class ModelArguments: method __post_init__ (line 151) | def __post_init__(self): class DataTrainingArguments (line 160) | class DataTrainingArguments: method __post_init__ (line 224) | def __post_init__(self): function main (line 238) | def main(): function _mp_fn (line 608) | def _mp_fn(index): FILE: train/sft/finetune_clm_lora.py class ModelArguments (line 86) | class ModelArguments: method __post_init__ (line 161) | def __post_init__(self): class DataTrainingArguments (line 171) | class DataTrainingArguments: method __post_init__ (line 235) | def __post_init__(self): class SavePeftModelCallback (line 249) | class SavePeftModelCallback(TrainerCallback): method on_save (line 250) | def on_save( function main (line 269) | def main(): function _mp_fn (line 688) | def _mp_fn(index):