SYMBOL INDEX (262 symbols across 34 files)

FILE: examples/MyopicTrap/appendix_exp_cosine_sim.py
  function add_eos (line 212) | def add_eos(input_examples):

FILE: examples/MyopicTrap/commercial_embedding_api.py
  class CommercialEncoder (line 14) | class CommercialEncoder(ABC):
    method encode (line 15) | def encode(
    method _generate_cache_key (line 24) | def _generate_cache_key(self, model_name: str, text: str, **kwargs):
  class OpenAIEncoder (line 29) | class OpenAIEncoder(CommercialEncoder):
    method __init__ (line 30) | def __init__(self):
    method encode (line 37) | def encode(
    method test (line 85) | def test():
  class CohereEncoder (line 94) | class CohereEncoder(CommercialEncoder):
    method __init__ (line 95) | def __init__(self):
    method encode (line 101) | def encode(
    method test (line 146) | def test():
  class VoyageEncoder (line 157) | class VoyageEncoder(CommercialEncoder):
    method __init__ (line 158) | def __init__(self):
    method encode (line 164) | def encode(
    method test (line 210) | def test():
  class JinaEncoder (line 222) | class JinaEncoder(CommercialEncoder):
    method __init__ (line 223) | def __init__(self):
    method encode (line 227) | def encode(
    method test (line 288) | def test():

FILE: examples/MyopicTrap/utils.py
  function compute_colbert_score (line 22) | def compute_colbert_score(q_reps, p_reps, q_mask=None):
  function convert_numpy_to_tensor (line 37) | def convert_numpy_to_tensor(output_1):
  function find_topk_via_faiss (line 59) | def find_topk_via_faiss(source_vecs: np.ndarray, target_vecs: np.ndarray...
  function find_topk_by_bm25 (line 66) | def find_topk_by_bm25(
  function find_topk_by_single_vecs (line 86) | def find_topk_by_single_vecs(
  function find_topk_by_multi_vecs (line 319) | def find_topk_by_multi_vecs(
  function find_topk_by_reranker (line 517) | def find_topk_by_reranker(

FILE: examples/distill_llm_to_bert_reranker/create_distill_data.py
  function get_train_data (line 11) | def get_train_data(train_data_path):

FILE: examples/distill_llm_to_bert_reranker/model_llm_generate.py
  class LLMGenerateDecoder (line 7) | class LLMGenerateDecoder(nn.Module):
    method __init__ (line 8) | def __init__(
    method forward (line 27) | def forward(self, batch, labels = None):
    method compute_score (line 41) | def compute_score(
    method preprocess (line 61) | def preprocess(self, sentences,max_len):
    method from_pretrained (line 81) | def from_pretrained(
  function test_relecance (line 103) | def test_relecance():

FILE: examples/stella_embedding_distill/create_distill_data.py
  function get_train_data (line 14) | def get_train_data(train_data_path):

FILE: examples/synthetic_data_embedding/get_lm_probs_dataset.py
  class LMProb (line 16) | class LMProb:
    method __init__ (line 23) | def __init__(self, config):
    method run (line 29) | def run(self, dataset):
    method calculate_prob (line 54) | def calculate_prob(self, prompt, answer):
  function main (line 59) | def main(

FILE: rag_retrieval/infer/reranker_models/cross_encoder_ranker.py
  class CorssEncoderRanker (line 13) | class CorssEncoderRanker(BaseRanker):
    method __init__ (line 14) | def __init__(self,
    method compute_score (line 43) | def compute_score(self,
    method rerank (line 77) | def rerank(self,
    method __max_length_truncation_rerank (line 99) | def __max_length_truncation_rerank(self,
    method __max_score_slice_rerank (line 119) | def __max_score_slice_rerank(self,
    method __reranker_tokenize_preproc (line 170) | def __reranker_tokenize_preproc(self,

FILE: rag_retrieval/infer/reranker_models/llm_rankers.py
  function sigmoid (line 11) | def sigmoid(x):
  class LLMRanker (line 14) | class LLMRanker(BaseRanker):
    method __init__ (line 16) | def __init__(self,
    method compute_score (line 54) | def compute_score(self,
    method rerank (line 91) | def rerank(self,
    method __max_length_truncation_rerank (line 114) | def __max_length_truncation_rerank(self,
    method __max_score_slice_rerank (line 144) | def __max_score_slice_rerank(self,
    method __reranker_tokenize_preproc (line 200) | def __reranker_tokenize_preproc(self,
    method get_inputs (line 289) | def get_inputs(self,

FILE: rag_retrieval/infer/reranker_models/ranker.py
  class BaseRanker (line 4) | class BaseRanker(ABC):
    method __init__ (line 6) | def __init__(self,
    method rerank (line 13) | def rerank(
    method compute_score (line 24) | def compute_score(

FILE: rag_retrieval/infer/reranker_models/result.py
  class Result (line 5) | class Result(BaseModel):
    method check_score_or_rank_exists (line 12) | def check_score_or_rank_exists(cls, v, values):
  class RankedResults (line 18) | class RankedResults(BaseModel):
    method results_count (line 23) | def results_count(self) -> int:
    method top_k (line 27) | def top_k(self, k: int) -> List[Result]:
    method get_score_by_docid (line 41) | def get_score_by_docid(self, doc_id: Union[int, str]) -> Optional[float]:

FILE: rag_retrieval/infer/reranker_models/utils.py
  function vprint (line 4) | def vprint(txt: str, verbose: int) -> None:
  function get_dtype (line 13) | def get_dtype(
  function get_device (line 35) | def get_device(

FILE: rag_retrieval/reranker.py
  function _get_model_type (line 28) | def _get_model_type(
  function Reranker (line 60) | def Reranker(

FILE: rag_retrieval/train/colbert/data.py
  class ColBERTDTripletataset (line 14) | class ColBERTDTripletataset(Dataset):
    method __init__ (line 15) | def __init__(self,
    method read_train_data (line 29) | def read_train_data(self,train_data_path):
    method __len__ (line 51) | def __len__(self):
    method __getitem__ (line 54) | def __getitem__(self,idx):
    method collate_fn (line 57) | def collate_fn(self,batch):
  function test_ColBERTDTripletataset (line 93) | def test_ColBERTDTripletataset():

FILE: rag_retrieval/train/colbert/model.py
  class ColBERT (line 12) | class ColBERT(nn.Module):
    method __init__ (line 13) | def __init__(self,
    method get_embedding (line 36) | def get_embedding(self,input_ids,attention_mask):
    method punctuation_padding_mask (line 52) | def punctuation_padding_mask(self,input_ids):
    method score (line 58) | def score(self,query_embedding,doc_embedding,query_attention_mask):
    method forward (line 69) | def forward(
    method compute_score (line 105) | def compute_score(self,
    method preprocess (line 128) | def preprocess(self,
    method save_pretrained (line 154) | def save_pretrained(self,
    method from_pretrained (line 168) | def from_pretrained(
  function test_relecance (line 192) | def test_relecance():

FILE: rag_retrieval/train/colbert/train_colbert.py
  function create_adamw_optimizer (line 15) | def create_adamw_optimizer(
  function parse_args (line 36) | def parse_args():
  function main (line 68) | def main():

FILE: rag_retrieval/train/colbert/trainer.py
  class Trainer (line 23) | class Trainer:
    method __init__ (line 24) | def __init__(
    method train (line 59) | def train(self):
    method log_metrics (line 114) | def log_metrics(self, metrics: dict[str, float], step: int):
    method add_prefix (line 119) | def add_prefix(values: dict[str, Any], prefix: str):
    method get_checkpoint_dir (line 122) | def get_checkpoint_dir(self, current_epoch):
  function evaluate (line 146) | def evaluate(
  class DummyProgressBar (line 162) | class DummyProgressBar:
    method update (line 163) | def update(self, n: int = 1) -> None:
    method close (line 166) | def close(self) -> None:
    method set_description (line 169) | def set_description(self, description: str) -> None:
  class DistributedTqdmProgressBar (line 173) | class DistributedTqdmProgressBar:
    method __init__ (line 174) | def __init__(self, epochs: int, num_steps_per_epoch: int | None, **kwa...
    method on_epoch_start (line 181) | def on_epoch_start(self):
    method update (line 187) | def update(self, n: int = 1) -> None:
    method close (line 190) | def close(self) -> None:
    method on_epoch_end (line 193) | def on_epoch_end(self) -> None:
    method show_metrics (line 197) | def show_metrics(self, metrics: dict[str, float]) -> None:
  class LossTracker (line 204) | class LossTracker:
    method __init__ (line 205) | def __init__(
    method update (line 214) | def update(self, loss_tensor: torch.Tensor):
    method reset (line 219) | def reset(self):
    method on_epoch_end (line 223) | def on_epoch_end(self, reset: bool = True):
    method loss (line 229) | def loss(self) -> float:

FILE: rag_retrieval/train/embedding/data.py
  class EmbeddingDataset (line 13) | class EmbeddingDataset(Dataset):
    method __init__ (line 14) | def __init__(
    method read_train_data (line 44) | def read_train_data(self, train_data_path):
    method __len__ (line 74) | def __len__(self):
    method __getitem__ (line 77) | def __getitem__(self, idx):
    method triplet_collate_fn (line 80) | def triplet_collate_fn(self, batch):
    method pair_collate_fn (line 103) | def pair_collate_fn(self, batch):
    method pair_score_collate_fn (line 121) | def pair_score_collate_fn(self, batch):
  class EmbeddingDistillDataset (line 142) | class EmbeddingDistillDataset(Dataset):
    method __init__ (line 143) | def __init__(
    method read_train_data (line 165) | def read_train_data(self, train_data_path):
    method __len__ (line 178) | def __len__(self):
    method __getitem__ (line 181) | def __getitem__(self, idx):
    method collate_fn (line 186) | def collate_fn(self, batch):
  function test_EmbeddingDataset (line 205) | def test_EmbeddingDataset():
  function test_EmbeddingDistillDataset (line 222) | def test_EmbeddingDistillDataset():

FILE: rag_retrieval/train/embedding/model.py
  class Embedding (line 14) | class Embedding(nn.Module):
    method __init__ (line 15) | def __init__(
    method get_embedding (line 33) | def get_embedding(self, input_ids, attention_mask):
    method forward (line 41) | def forward(
    method pair_inbatch_softmax_loss (line 205) | def pair_inbatch_softmax_loss(
    method triplet_inbatch_softmax_loss (line 231) | def triplet_inbatch_softmax_loss(
    method pair_kl_loss (line 270) | def pair_kl_loss(
    method encode (line 296) | def encode(
    method preprocess (line 329) | def preprocess(
    method _text_length (line 342) | def _text_length(self, text):
    method save_pretrained (line 358) | def save_pretrained(
    method from_pretrained (line 366) | def from_pretrained(
  function test_model_embedding (line 408) | def test_model_embedding():

FILE: rag_retrieval/train/embedding/model_distill.py
  class DistillEmbedding (line 13) | class DistillEmbedding(nn.Module):
    method __init__ (line 14) | def __init__(
    method get_embedding (line 26) | def get_embedding(self, input_ids, attention_mask):
    method forward (line 33) | def forward(
    method cosine_embedding_loss (line 71) | def cosine_embedding_loss(
    method pair_inbatch_similarity_loss (line 84) | def pair_inbatch_similarity_loss(
    method pair_inbatch_triplet_loss (line 97) | def pair_inbatch_triplet_loss(
    method get_score_diff (line 108) | def get_score_diff(
    method encode (line 119) | def encode(
    method preprocess (line 152) | def preprocess(
    method _text_length (line 165) | def _text_length(self, text):
    method save_pretrained (line 181) | def save_pretrained(
    method from_pretrained (line 190) | def from_pretrained(
  function test_model_embedding (line 215) | def test_model_embedding():

FILE: rag_retrieval/train/embedding/train_embedding.py
  function create_adamw_optimizer (line 16) | def create_adamw_optimizer(
  function parse_args (line 37) | def parse_args():
  function main (line 94) | def main():

FILE: rag_retrieval/train/embedding/trainer.py
  class Trainer (line 20) | class Trainer:
    method __init__ (line 21) | def __init__(
    method train (line 59) | def train(self):
    method log_metrics (line 154) | def log_metrics(self, metrics: dict[str, float], step: int):
    method add_prefix (line 159) | def add_prefix(values: dict[str, Any], prefix: str):
    method get_checkpoint_dir (line 162) | def get_checkpoint_dir(self, current_epoch, is_step=False):
  function evaluate (line 186) | def evaluate(
  class DummyProgressBar (line 214) | class DummyProgressBar:
    method update (line 215) | def update(self, n: int = 1) -> None:
    method close (line 218) | def close(self) -> None:
    method set_description (line 221) | def set_description(self, description: str) -> None:
  class DistributedTqdmProgressBar (line 225) | class DistributedTqdmProgressBar:
    method __init__ (line 226) | def __init__(self, accelerator, epochs: int, num_steps_per_epoch: int ...
    method on_epoch_start (line 233) | def on_epoch_start(self):
    method update (line 239) | def update(self, n: int = 1) -> None:
    method close (line 242) | def close(self) -> None:
    method on_epoch_end (line 245) | def on_epoch_end(self) -> None:
    method show_metrics (line 249) | def show_metrics(self, metrics: dict[str, float]) -> None:
  class LossTracker (line 256) | class LossTracker:
    method __init__ (line 257) | def __init__(
    method update (line 266) | def update(self, loss_tensor: torch.Tensor):
    method reset (line 271) | def reset(self):
    method on_epoch_end (line 275) | def on_epoch_end(self, reset: bool = True):
    method loss (line 281) | def loss(self) -> float:

FILE: rag_retrieval/train/reranker/data.py
  class PointwiseRankerDataset (line 10) | class PointwiseRankerDataset(Dataset):
    method __init__ (line 11) | def __init__(self, data_path=None, label_key="label", target_model=Non...
    method read_data (line 26) | def read_data(self, data_path):
    method __len__ (line 51) | def __len__(self):
    method __getitem__ (line 54) | def __getitem__(self, idx):
    method collate_fn (line 57) | def collate_fn(self, batch):
  class GroupedRankerDataset (line 73) | class GroupedRankerDataset(Dataset):
    method __init__ (line 74) | def __init__(self, data_path=None, label_key=None, target_model=None, ...
    method read_data (line 89) | def read_data(self, data_path):
    method __len__ (line 155) | def __len__(self):
    method __getitem__ (line 158) | def __getitem__(self, idx):
    method collate_fn (line 161) | def collate_fn(self, batch):
  function test_PointwiseRankerDataset (line 179) | def test_PointwiseRankerDataset():
  function test_GroupedRankerDataset (line 208) | def test_GroupedRankerDataset():

FILE: rag_retrieval/train/reranker/model_bert.py
  class CrossEncoder (line 9) | class CrossEncoder(nn.Module):
    method __init__ (line 10) | def __init__(
    method forward (line 27) | def forward(self, batch, labels=None):
    method compute_score (line 50) | def compute_score(
    method preprocess (line 71) | def preprocess(self, sentences_pairs, max_len):
    method from_pretrained (line 90) | def from_pretrained(
    method save_pretrained (line 112) | def save_pretrained(self, save_dir, safe_serialization=False):
  function test_CrossEncoder (line 127) | def test_CrossEncoder():

FILE: rag_retrieval/train/reranker/model_llm.py
  class LLMDecoder (line 8) | class LLMDecoder(nn.Module):
    method __init__ (line 9) | def __init__(
    method forward (line 36) | def forward(self, batch, labels=None):
    method compute_score (line 58) | def compute_score(
    method preprocess (line 80) | def preprocess(self, sentences_pairs, max_len):
    method from_pretrained (line 119) | def from_pretrained(
    method save_pretrained (line 160) | def save_pretrained(self, save_dir, safe_serialization=False):
  function test_LLMDecoder (line 175) | def test_LLMDecoder():

FILE: rag_retrieval/train/reranker/ranking_loss.py
  function pointwise_mse (line 6) | def pointwise_mse(logits, labels):
  function pointwise_bce (line 12) | def pointwise_bce(logits, labels):
  function pairwise_ranknet (line 16) | def pairwise_ranknet(logits, labels, group_size):
  function listwise_ce (line 51) | def listwise_ce(logits, labels, group_size):

FILE: rag_retrieval/train/reranker/train_reranker.py
  function create_adamw_optimizer (line 16) | def create_adamw_optimizer(
  function parse_args (line 37) | def parse_args():
  function main (line 104) | def main():

FILE: rag_retrieval/train/reranker/trainer.py
  class Trainer (line 21) | class Trainer:
    method __init__ (line 22) | def __init__(
    method train (line 59) | def train(self):
    method log_metrics (line 143) | def log_metrics(self, metrics: dict[str, float], step: int):
    method add_prefix (line 148) | def add_prefix(values: dict[str, Any], prefix: str):
    method get_checkpoint_dir (line 151) | def get_checkpoint_dir(self, current_epoch):
  function evaluate (line 183) | def evaluate(
  class DummyProgressBar (line 198) | class DummyProgressBar:
    method update (line 199) | def update(self, n: int = 1) -> None:
    method close (line 202) | def close(self) -> None:
    method set_description (line 205) | def set_description(self, description: str) -> None:
  class DistributedTqdmProgressBar (line 209) | class DistributedTqdmProgressBar:
    method __init__ (line 210) | def __init__(
    method on_epoch_start (line 219) | def on_epoch_start(self):
    method update (line 225) | def update(self, n: int = 1) -> None:
    method close (line 228) | def close(self) -> None:
    method on_epoch_end (line 231) | def on_epoch_end(self) -> None:
    method show_metrics (line 235) | def show_metrics(self, metrics: dict[str, float]) -> None:
  class LossTracker (line 242) | class LossTracker:
    method __init__ (line 243) | def __init__(
    method update (line 252) | def update(self, loss_tensor: torch.Tensor):
    method reset (line 257) | def reset(self):
    method on_epoch_end (line 261) | def on_epoch_end(self, reset: bool = True):
    method loss (line 267) | def loss(self) -> float:

FILE: rag_retrieval/train/reranker/utils.py
  function map_label_to_continuous (line 4) | def map_label_to_continuous(label, min_label, max_label):
  function visualize_label_distribution (line 22) | def visualize_label_distribution(label_distribution):
  function shuffle_text (line 65) | def shuffle_text(text, shuffle_ratio=0.15):
  function create_adamw_optimizer_with_special_lr_groups (line 85) | def create_adamw_optimizer_with_special_lr_groups(

FILE: tests/test_cross_encoder_reranker_bce.py
  function test_rag_retrieval_cross_encode (line 10) | def test_rag_retrieval_cross_encode(query,docs):
  function test_bce_cross_encoder (line 29) | def test_bce_cross_encoder(query,docs):

FILE: tests/test_cross_encoder_reranker_bge.py
  function test_rag_retrieval_cross_encode (line 10) | def test_rag_retrieval_cross_encode(query,docs):
  function test_bge_cross_encode (line 29) | def test_bge_cross_encode(query,docs):

FILE: tests/test_cross_encoder_reranker_bge_m3.py
  function test_rag_retrieval_cross_encode (line 10) | def test_rag_retrieval_cross_encode(query,docs):
  function test_bge_cross_encode (line 29) | def test_bge_cross_encode(query,docs):

FILE: tests/test_llm_reranker_bge_cpm.py
  function test_rag_retrieval_cpm (line 10) | def test_rag_retrieval_cpm(query,docs):
  function test_bge_cpm (line 28) | def test_bge_cpm(query,docs):

FILE: tests/test_llm_reranker_bge_gemma.py
  function test_rag_retrieval_gemma (line 10) | def test_rag_retrieval_gemma(query,docs):
  function test_bge_gemma (line 28) | def test_bge_gemma(query,docs):