SYMBOL INDEX (552 symbols across 71 files)

FILE: entity_candidate/biencoder/biencoder_.py
  function load_biencoder (line 26) | def load_biencoder(params):
  class BiEncoderModule (line 32) | class BiEncoderModule(torch.nn.Module):
    method __init__ (line 33) | def __init__(self, params):
    method forward (line 51) | def forward(
  class BiEncoderRanker (line 73) | class BiEncoderRanker(torch.nn.Module):
    method __init__ (line 74) | def __init__(self, params, shared=None):
    method load_model (line 100) | def load_model(self, fname, cpu=False):
    method build_model (line 108) | def build_model(self):
    method get_optimizer (line 120) | def get_optimizer(self, optim_states=None, saved_optim_type=None):
    method encode_context (line 128) | def encode_context(self, cands):
    method encode_candidate (line 137) | def encode_candidate(self, cands):
    method score_candidate (line 147) | def score_candidate(
    method forward (line 296) | def forward(self, context_input, cand_input_A, cand_input_B=None, labe...
  function to_bert_input (line 335) | def to_bert_input(token_idx, null_idx):

FILE: entity_candidate/biencoder/data_process_.py
  function select_field (line 21) | def select_field(data, key1, key2=None):
  function get_context_representation (line 28) | def get_context_representation(
  function get_candidate_representation (line 87) | def get_candidate_representation(
  function process_mention_data (line 121) | def process_mention_data(
  function process_mention_data_ (line 224) | def process_mention_data_(
  function process_mention_data__ (line 347) | def process_mention_data__(

FILE: entity_candidate/biencoder/eval_biencoder_.py
  function load_entity_dict (line 28) | def load_entity_dict(logger, params, is_zeshel):
  function get_candidate_pool_tensor_zeshel (line 65) | def get_candidate_pool_tensor_zeshel(
  function get_candidate_pool_tensor_helper (line 86) | def get_candidate_pool_tensor_helper(
  function get_candidate_pool_tensor (line 109) | def get_candidate_pool_tensor(
  function encode_candidate (line 139) | def encode_candidate(
  function load_or_generate_candidate_pool (line 188) | def load_or_generate_candidate_pool(
  function make_ids (line 223) | def make_ids(data_path):
  function main (line 292) | def main(params):

FILE: entity_candidate/biencoder/nn_prediction_.py
  function get_topk_predictions (line 18) | def get_topk_predictions(

FILE: entity_candidate/biencoder/train_biencoder_.py
  function evaluate (line 45) | def evaluate(reranker, eval_dataloader, params, device, logger,):
  function evaluate_ (line 89) | def evaluate_(reranker, eval_dataloader, params, device, logger,):
  function get_optimizer (line 133) | def get_optimizer(model, params):
  function get_optimizer (line 141) | def get_optimizer(model, params):
  function get_scheduler (line 150) | def get_scheduler(params, optimizer, len_train_data, logger):
  function main (line 166) | def main(params):

FILE: entity_candidate/biencoder/zeshel_utils.py
  function load_entity_dict_zeshel (line 36) | def load_entity_dict_zeshel(logger, params):
  class Stats (line 70) | class Stats():
    method __init__ (line 71) | def __init__(self, top_k=1000):
    method add (line 80) | def add(self, idx):
    method extend (line 88) | def extend(self, stats):
    method output (line 93) | def output(self):

FILE: entity_candidate/build_faiss_index.py
  function main (line 17) | def main(params):

FILE: entity_candidate/candidate_data_fetcher.py
  function get_model (line 11) | def get_model(parameters):
  class Wikimedia_Data_Fetcher (line 15) | class Wikimedia_Data_Fetcher:
    method __init__ (line 16) | def __init__(self, path_to_data):
    method get_data_for_entity (line 19) | def get_data_for_entity(self, entity_data):

FILE: entity_candidate/candidate_generation.py
  function get_model (line 15) | def get_model(params):
  class Candidate_Generator (line 19) | class Candidate_Generator:
    method __init__ (line 20) | def __init__(self, parameters=None):
    method get_candidates (line 23) | def get_candidates(self, mention_data):
  class BM45_Candidate_Generator (line 28) | class BM45_Candidate_Generator(Candidate_Generator):
    method __init__ (line 31) | def __init__(self, params):
    method _filter_result (line 48) | def _filter_result(self, cand, detailed=True):
    method get_candidates (line 68) | def get_candidates(self, mention_data):
    method process_mentions_for_candidate_generator (line 118) | def process_mentions_for_candidate_generator(sentences, mentions):
    method solr_escape (line 124) | def solr_escape(string):

FILE: entity_candidate/candidate_ranking/bert_reranking.py
  class BertForReranking (line 26) | class BertForReranking(BertPreTrainedModel):
    method __init__ (line 95) | def __init__(self, config):
    method forward (line 103) | def forward(
  class BertReranker (line 160) | class BertReranker:
    method __init__ (line 161) | def __init__(self, parameters):
    method rerank (line 179) | def rerank(self, mentions, sentences):
    method get_scheduler_and_optimizer (line 231) | def get_scheduler_and_optimizer(self, parameters, train_tensor_data, l...
    method get_model (line 282) | def get_model(parameters):
    method get_tokenizer (line 296) | def get_tokenizer(parameters):
    method _get_candidate_representation (line 303) | def _get_candidate_representation(
    method _get_mention_context_end2end (line 337) | def _get_mention_context_end2end(mention, sentences):
    method _select_field (line 356) | def _select_field(samples, field):
    method _get_context_token_representation (line 363) | def _get_context_token_representation(
    method _process_mentions_for_model (line 393) | def _process_mentions_for_model(

FILE: entity_candidate/candidate_ranking/evaluate.py
  function evaluate_model_on_dataset (line 20) | def evaluate_model_on_dataset(
  function evaluate (line 104) | def evaluate(parameters, logger=None):

FILE: entity_candidate/candidate_ranking/train.py
  function main (line 41) | def main(parameters):

FILE: entity_candidate/candidate_ranking/utils.py
  function new_read_dataset (line 24) | def new_read_dataset(dataset_name, preprocessed_json_data_parent_folder,...
  function filter_samples (line 101) | def filter_samples(samples, top_k, gold_key="gold_pos"):
  function _truncate_seq_pair (line 113) | def _truncate_seq_pair(tokens_a, tokens_b, max_length):
  function eval_precision_bm45_dataloader (line 125) | def eval_precision_bm45_dataloader(dataloader, ks=[1, 5, 10], number_of_...
  function accuracy (line 148) | def accuracy(out, labels):
  function remove_module_from_state_dict (line 153) | def remove_module_from_state_dict(state_dict):
  function save_model (line 161) | def save_model(model, tokenizer, output_dir):
  function get_logger (line 174) | def get_logger(output_dir=None):
  function write_to_file (line 201) | def write_to_file(path, string, mode="w"):
  function get_reranker (line 206) | def get_reranker(parameters):
  function get_biencoder (line 210) | def get_biencoder(parameters):

FILE: entity_candidate/candidate_retrieval/candidate_generators.py
  function mention_data_summary (line 12) | def mention_data_summary(mention):
  class Simple_Candidate_Generator (line 16) | class Simple_Candidate_Generator:
    method __init__ (line 17) | def __init__(self, params):
    method _filter_result (line 37) | def _filter_result(self, cand):
    method get_candidates (line 59) | def get_candidates(
  class Pregenerated_Candidates_Data_Fetcher (line 119) | class Pregenerated_Candidates_Data_Fetcher:
    method __init__ (line 120) | def __init__(self, parameters):
    method get_candidates_data (line 132) | def get_candidates_data(self, candidates_wikidata_ids):
    method filter_result (line 145) | def filter_result(cand, detailed=True):
    method get_candidate_data_for_wikidata_id (line 165) | def get_candidate_data_for_wikidata_id(self, wikidata_id):

FILE: entity_candidate/candidate_retrieval/data_ingestion.py
  function remove_all_docs (line 51) | def remove_all_docs():
  function load_data (line 55) | def load_data():
  function get_data_for_key (line 59) | def get_data_for_key(data, title):

FILE: entity_candidate/candidate_retrieval/dataset.py
  function read_csv_file (line 18) | def read_csv_file(path, added_params):
  function read_conll_file (line 73) | def read_conll_file(data, path):
  function load_person_names (line 145) | def load_person_names(path):
  function find_coref (line 153) | def find_coref(ment, mentlist, person_names):
  function with_coref (line 174) | def with_coref(dataset, person_names):
  function eval (line 193) | def eval(testset, system_pred, nel=False):
  function get_candidate_generator (line 225) | def get_candidate_generator(added_params):
  class CoNLLDataset (line 235) | class CoNLLDataset:
    method __init__ (line 240) | def __init__(self, path, person_path, conll_path, added_params):
  class FetchCandidateEntities (line 300) | class FetchCandidateEntities(object):
    method __init__ (line 305) | def __init__(self, p_e_m_data_path="data/basic_data/p_e_m_data/"):
    method process (line 322) | def process(self, span):

FILE: entity_candidate/candidate_retrieval/evaluator.py
  class Evaluator (line 12) | class Evaluator:
    method __init__ (line 13) | def __init__(self, data):
    method candidate_generation (line 16) | def candidate_generation(
    method candidate_generation_recall_at (line 92) | def candidate_generation_recall_at(self, ax=None, max_rank=None):

FILE: entity_candidate/candidate_retrieval/perform_and_evaluate_candidate_retrieval_multithreaded.py
  function run_thread (line 26) | def run_thread(arguments):
  function split (line 61) | def split(a, n):
  function main (line 66) | def main(args):
  function get_parameters (line 157) | def get_parameters(args):

FILE: entity_candidate/candidate_retrieval/utils.py
  function solr_escape (line 18) | def solr_escape(string):
  function get_wikidata_id_from_link_name (line 36) | def get_wikidata_id_from_link_name(link):
  function get_datasets (line 52) | def get_datasets(get_test_dataset=False, get_pregenereted_candidates_wik...
  function get_sent_context (line 119) | def get_sent_context(mention, key, solr_escaped=True):
  function get_list_of_mentions (line 156) | def get_list_of_mentions(dev_datasets):
  function write_candidate_generation_results_for_a_run_to_file (line 272) | def write_candidate_generation_results_for_a_run_to_file(run, results_du...
  function write_candidate_generation_execution_time_to_file (line 285) | def write_candidate_generation_execution_time_to_file(
  function write_candidate_generation_results_to_file (line 294) | def write_candidate_generation_results_to_file(

FILE: entity_candidate/common/optimizer.py
  function get_bert_optimizer (line 37) | def get_bert_optimizer(models, type_optimization, learning_rate, fp16=Fa...
  function ellipse (line 82) | def ellipse(lst, max_display=5, sep='|'):

FILE: entity_candidate/common/params.py
  class BlinkParser (line 21) | class BlinkParser(argparse.ArgumentParser):
    method __init__ (line 35) | def __init__(
    method add_blink_args (line 60) | def add_blink_args(self, args=None):
    method add_model_args (line 93) | def add_model_args(self, args=None):
    method add_training_args (line 168) | def add_training_args(self, args=None):
    method add_eval_args (line 237) | def add_eval_args(self, args=None):

FILE: entity_candidate/common/ranker_base.py
  function get_model_obj (line 10) | def get_model_obj(model):
  class BertEncoder (line 14) | class BertEncoder(nn.Module):
    method __init__ (line 15) | def __init__(
    method forward (line 28) | def forward(self, token_ids, segment_ids, attention_mask):

FILE: entity_candidate/crossencoder/crossencoder.py
  function load_crossencoder (line 36) | def load_crossencoder(params):
  class CrossEncoderModule (line 42) | class CrossEncoderModule(torch.nn.Module):
    method __init__ (line 43) | def __init__(self, params, tokenizer):
    method forward (line 59) | def forward(
  class CrossEncoderRanker (line 66) | class CrossEncoderRanker(torch.nn.Module):
    method __init__ (line 67) | def __init__(self, params, shared=None):
    method load_model (line 104) | def load_model(self, fname, cpu=False):
    method save (line 111) | def save(self, output_dir):
    method build_model (line 115) | def build_model(self):
    method save_model (line 118) | def save_model(self, output_dir):
    method get_optimizer (line 127) | def get_optimizer(self, optim_states=None, saved_optim_type=None):
    method score_candidate (line 135) | def score_candidate(self, text_vecs, context_len):
    method forward (line 146) | def forward(self, input_idx, label_input, context_len):
  function to_bert_input (line 152) | def to_bert_input(token_idx, null_idx, segment_pos):

FILE: entity_candidate/crossencoder/data_process.py
  function prepare_crossencoder_mentions (line 17) | def prepare_crossencoder_mentions(
  function prepare_crossencoder_candidates (line 46) | def prepare_crossencoder_candidates(
  function filter_crossencoder_tensor_input (line 89) | def filter_crossencoder_tensor_input(
  function prepare_crossencoder_data (line 115) | def prepare_crossencoder_data(

FILE: entity_candidate/crossencoder/train_cross.py
  function modify (line 43) | def modify(context_input, candidate_input, max_seq_length):
  function evaluate (line 63) | def evaluate(reranker, eval_dataloader, device, logger, context_length, ...
  function get_optimizer (line 134) | def get_optimizer(model, params):
  function get_scheduler (line 143) | def get_scheduler(params, optimizer, len_train_data, logger):
  function main (line 159) | def main(params):

FILE: entity_candidate/indexer/faiss_indexer.py
  class DenseIndexer (line 22) | class DenseIndexer(object):
    method __init__ (line 23) | def __init__(self, buffer_size: int = 50000):
    method index_data (line 28) | def index_data(self, data: np.array):
    method search_knn (line 31) | def search_knn(self, query_vectors: np.array, top_docs: int):
    method serialize (line 34) | def serialize(self, index_file: str):
    method deserialize_from (line 38) | def deserialize_from(self, index_file: str):
  class DenseFlatIndexer (line 47) | class DenseFlatIndexer(DenseIndexer):
    method __init__ (line 48) | def __init__(self, vector_sz: int = 1, buffer_size: int = 50000):
    method index_data (line 52) | def index_data(self, data: np.array):
    method search_knn (line 65) | def search_knn(self, query_vectors, top_k):
  class DenseHNSWFlatIndexer (line 71) | class DenseHNSWFlatIndexer(DenseIndexer):
    method __init__ (line 76) | def __init__(
    method index_data (line 94) | def index_data(self, data: np.array):
    method search_knn (line 131) | def search_knn(self, query_vectors, top_k):
    method deserialize_from (line 138) | def deserialize_from(self, file: str):

FILE: entity_candidate/main_dense.py
  function _print_colorful_text (line 42) | def _print_colorful_text(input_sentence, samples):
  function _print_colorful_prediction (line 65) | def _print_colorful_prediction(
  function _annotate (line 75) | def _annotate(ner_model, input_sentences):
  function _load_candidates (line 99) | def _load_candidates(
  function __map_test_entities (line 154) | def __map_test_entities(test_entities_path, title2id, logger):
  function __load_test (line 173) | def __load_test(test_filename, kb2id, wikipedia_id2local_id, logger):
  function _get_test_samples (line 210) | def _get_test_samples(
  function _process_biencoder_dataloader (line 220) | def _process_biencoder_dataloader(samples, tokenizer, biencoder_params):
  function _run_biencoder (line 237) | def _run_biencoder(biencoder, dataloader, candidate_encoding, top_k=100,...
  function _process_crossencoder_dataloader (line 263) | def _process_crossencoder_dataloader(context_input, label_input, crossen...
  function _run_crossencoder (line 272) | def _run_crossencoder(crossencoder, dataloader, logger, context_len, dev...
  function load_models (line 289) | def load_models(args, logger=None):
  function run (line 341) | def run(

FILE: entity_candidate/main_solr.py
  function main (line 19) | def main(parameters):

FILE: entity_candidate/ner.py
  function get_model (line 11) | def get_model(parameters=None):
  class NER_model (line 15) | class NER_model:
    method __init__ (line 16) | def __init__(self, parameters=None):
    method predict (line 19) | def predict(self, sents):
  class Flair (line 29) | class Flair(NER_model):
    method __init__ (line 30) | def __init__(self, parameters=None):
    method predict (line 33) | def predict(self, sentences):

FILE: entity_candidate/reranker.py
  function get_model (line 10) | def get_model(params):

FILE: entity_candidate/utils.py
  function read_sentences_from_file (line 15) | def read_sentences_from_file(path_to_file, one_sentence_per_line=True):
  function get_candidate_summary (line 34) | def get_candidate_summary(candidate):
  function present_sentence_mentions (line 42) | def present_sentence_mentions(sentence, mentions, output_file):
  function sentence_mentions_pairs (line 84) | def sentence_mentions_pairs(sentences, mentions):
  function present_annotated_sentences (line 103) | def present_annotated_sentences(sentences, mentions, output_file=None):
  function write_dicts_as_json_per_line (line 110) | def write_dicts_as_json_per_line(list_of_dicts, txt_file_path):
  function get_mentions_txt_file_path (line 120) | def get_mentions_txt_file_path(output_folder_path):
  function get_sentences_txt_file_path (line 128) | def get_sentences_txt_file_path(output_folder_path):
  function get_end2end_pickle_output_file_path (line 136) | def get_end2end_pickle_output_file_path(output_folder_path):
  function write_end2end_pickle_output (line 144) | def write_end2end_pickle_output(sentences, mentions, output_file_id):
  function get_end2end_pretty_output_file_path (line 150) | def get_end2end_pretty_output_file_path(output_folder_path):

FILE: entity_candidate/utils/tokenization.py
  function convert_to_unicode (line 27) | def convert_to_unicode(text):
  function printable_text (line 47) | def printable_text(text):
  function load_vocab (line 70) | def load_vocab(vocab_file):
  function convert_by_vocab (line 85) | def convert_by_vocab(vocab, items):
  function convert_tokens_to_ids (line 95) | def convert_tokens_to_ids(vocab, tokens):
  function convert_ids_to_tokens (line 99) | def convert_ids_to_tokens(inv_vocab, ids):
  function whitespace_tokenize (line 103) | def whitespace_tokenize(text):
  class FullTokenizer (line 112) | class FullTokenizer(object):
    method __init__ (line 115) | def __init__(self, vocab_file, do_lower_case=True):
    method tokenize (line 121) | def tokenize(self, text):
    method convert_tokens_to_ids (line 129) | def convert_tokens_to_ids(self, tokens):
    method convert_ids_to_tokens (line 132) | def convert_ids_to_tokens(self, ids):
  class BasicTokenizer (line 136) | class BasicTokenizer(object):
    method __init__ (line 139) | def __init__(self, do_lower_case=True):
    method tokenize (line 147) | def tokenize(self, text):
    method _run_strip_accents (line 171) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 182) | def _run_split_on_punc(self, text):
    method _tokenize_chinese_chars (line 202) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 215) | def _is_chinese_char(self, cp):
    method _clean_text (line 237) | def _clean_text(self, text):
  class WordpieceTokenizer (line 251) | class WordpieceTokenizer(object):
    method __init__ (line 254) | def __init__(self, vocab, unk_token="[UNK]", max_input_chars_per_word=...
    method tokenize (line 259) | def tokenize(self, text):
  function _is_whitespace (line 313) | def _is_whitespace(char):
  function _is_control (line 325) | def _is_control(char):
  function _is_punctuation (line 337) | def _is_punctuation(char):

FILE: entity_candidate/utils/utils.py
  function timer (line 9) | def timer(func):
  function set_seed (line 27) | def set_seed(seed=123):
  function set_logger (line 39) | def set_logger(log_path):

FILE: entity_sort/blink/biencoder/biencoder.py
  function load_biencoder (line 26) | def load_biencoder(params):
  class BiEncoderModule (line 32) | class BiEncoderModule(torch.nn.Module):
    method __init__ (line 33) | def __init__(self, params):
    method forward (line 51) | def forward(
  class BiEncoderRanker (line 73) | class BiEncoderRanker(torch.nn.Module):
    method __init__ (line 74) | def __init__(self, params, shared=None):
    method load_model (line 99) | def load_model(self, fname, cpu=False):
    method build_model (line 106) | def build_model(self):
    method get_optimizer (line 118) | def get_optimizer(self, optim_states=None, saved_optim_type=None):
    method encode_context (line 126) | def encode_context(self, cands):
    method encode_candidate (line 135) | def encode_candidate(self, cands):
    method score_candidate (line 148) | def score_candidate(
    method forward (line 188) | def forward(self, context_input, cand_input, label_input=None):
  function to_bert_input (line 203) | def to_bert_input(token_idx, null_idx):

FILE: entity_sort/blink/biencoder/data_process.py
  function select_field (line 18) | def select_field(data, key1, key2=None):
  function get_context_representation (line 25) | def get_context_representation(
  function get_candidate_representation (line 84) | def get_candidate_representation(
  function process_mention_data (line 119) | def process_mention_data(

FILE: entity_sort/blink/build_faiss_index.py
  function main (line 17) | def main(params):

FILE: entity_sort/blink/candidate_ranking/bert_reranking.py
  class BertForReranking (line 26) | class BertForReranking(BertPreTrainedModel):
    method __init__ (line 95) | def __init__(self, config):
    method forward (line 103) | def forward(
  class BertReranker (line 160) | class BertReranker:
    method __init__ (line 161) | def __init__(self, parameters):
    method rerank (line 179) | def rerank(self, mentions, sentences):
    method get_scheduler_and_optimizer (line 231) | def get_scheduler_and_optimizer(self, parameters, train_tensor_data, l...
    method get_model (line 282) | def get_model(parameters):
    method get_tokenizer (line 296) | def get_tokenizer(parameters):
    method _get_candidate_representation (line 303) | def _get_candidate_representation(
    method _get_mention_context_end2end (line 337) | def _get_mention_context_end2end(mention, sentences):
    method _select_field (line 356) | def _select_field(samples, field):
    method _get_context_token_representation (line 363) | def _get_context_token_representation(
    method _process_mentions_for_model (line 393) | def _process_mentions_for_model(

FILE: entity_sort/blink/candidate_ranking/evaluate.py
  function evaluate_model_on_dataset (line 20) | def evaluate_model_on_dataset(
  function evaluate (line 104) | def evaluate(parameters, logger=None):

FILE: entity_sort/blink/candidate_ranking/train.py
  function main (line 41) | def main(parameters):

FILE: entity_sort/blink/candidate_ranking/utils.py
  function new_read_dataset (line 24) | def new_read_dataset(dataset_name, preprocessed_json_data_parent_folder,...
  function filter_samples (line 101) | def filter_samples(samples, top_k, gold_key="gold_pos"):
  function _truncate_seq_pair (line 113) | def _truncate_seq_pair(tokens_a, tokens_b, max_length):
  function eval_precision_bm45_dataloader (line 125) | def eval_precision_bm45_dataloader(dataloader, ks=[1, 5, 10], number_of_...
  function accuracy (line 148) | def accuracy(out, labels):
  function remove_module_from_state_dict (line 153) | def remove_module_from_state_dict(state_dict):
  function save_model (line 161) | def save_model(model, tokenizer, output_dir):
  function get_logger (line 174) | def get_logger(output_dir=None):
  function write_to_file (line 201) | def write_to_file(path, string, mode="w"):
  function get_reranker (line 206) | def get_reranker(parameters):
  function get_biencoder (line 210) | def get_biencoder(parameters):

FILE: entity_sort/blink/common/optimizer.py
  function get_bert_optimizer (line 37) | def get_bert_optimizer(models, type_optimization, learning_rate, fp16=Fa...
  function ellipse (line 82) | def ellipse(lst, max_display=5, sep='|'):

FILE: entity_sort/blink/common/params.py
  class BlinkParser (line 21) | class BlinkParser(argparse.ArgumentParser):
    method __init__ (line 35) | def __init__(
    method add_blink_args (line 60) | def add_blink_args(self, args=None):
    method add_model_args (line 93) | def add_model_args(self, args=None):
    method add_training_args (line 168) | def add_training_args(self, args=None):
    method add_eval_args (line 237) | def add_eval_args(self, args=None):

FILE: entity_sort/blink/common/ranker_base.py
  function get_model_obj (line 10) | def get_model_obj(model):
  class BertEncoder (line 14) | class BertEncoder(nn.Module):
    method __init__ (line 15) | def __init__(
    method forward (line 28) | def forward(self, token_ids, segment_ids, attention_mask):

FILE: entity_sort/blink/indexer/faiss_indexer.py
  class DenseIndexer (line 22) | class DenseIndexer(object):
    method __init__ (line 23) | def __init__(self, buffer_size: int = 50000):
    method index_data (line 28) | def index_data(self, data: np.array):
    method search_knn (line 31) | def search_knn(self, query_vectors: np.array, top_docs: int):
    method serialize (line 34) | def serialize(self, index_file: str):
    method deserialize_from (line 38) | def deserialize_from(self, index_file: str):
  class DenseFlatIndexer (line 47) | class DenseFlatIndexer(DenseIndexer):
    method __init__ (line 48) | def __init__(self, vector_sz: int = 1, buffer_size: int = 50000):
    method index_data (line 52) | def index_data(self, data: np.array):
    method search_knn (line 65) | def search_knn(self, query_vectors, top_k):
  class DenseHNSWFlatIndexer (line 71) | class DenseHNSWFlatIndexer(DenseIndexer):
    method __init__ (line 76) | def __init__(
    method index_data (line 94) | def index_data(self, data: np.array):
    method search_knn (line 131) | def search_knn(self, query_vectors, top_k):
    method deserialize_from (line 138) | def deserialize_from(self, file: str):

FILE: entity_sort/blink/utils/tokenization.py
  function convert_to_unicode (line 27) | def convert_to_unicode(text):
  function printable_text (line 47) | def printable_text(text):
  function load_vocab (line 70) | def load_vocab(vocab_file):
  function convert_by_vocab (line 85) | def convert_by_vocab(vocab, items):
  function convert_tokens_to_ids (line 95) | def convert_tokens_to_ids(vocab, tokens):
  function convert_ids_to_tokens (line 99) | def convert_ids_to_tokens(inv_vocab, ids):
  function whitespace_tokenize (line 103) | def whitespace_tokenize(text):
  class FullTokenizer (line 112) | class FullTokenizer(object):
    method __init__ (line 115) | def __init__(self, vocab_file, do_lower_case=True):
    method tokenize (line 121) | def tokenize(self, text):
    method convert_tokens_to_ids (line 129) | def convert_tokens_to_ids(self, tokens):
    method convert_ids_to_tokens (line 132) | def convert_ids_to_tokens(self, ids):
  class BasicTokenizer (line 136) | class BasicTokenizer(object):
    method __init__ (line 139) | def __init__(self, do_lower_case=True):
    method tokenize (line 147) | def tokenize(self, text):
    method _run_strip_accents (line 171) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 182) | def _run_split_on_punc(self, text):
    method _tokenize_chinese_chars (line 202) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 215) | def _is_chinese_char(self, cp):
    method _clean_text (line 237) | def _clean_text(self, text):
  class WordpieceTokenizer (line 251) | class WordpieceTokenizer(object):
    method __init__ (line 254) | def __init__(self, vocab, unk_token="[UNK]", max_input_chars_per_word=...
    method tokenize (line 259) | def tokenize(self, text):
  function _is_whitespace (line 313) | def _is_whitespace(char):
  function _is_control (line 325) | def _is_control(char):
  function _is_punctuation (line 337) | def _is_punctuation(char):

FILE: entity_sort/blink/utils/utils.py
  function timer (line 9) | def timer(func):
  function set_seed (line 27) | def set_seed(seed=123):
  function set_logger (line 39) | def set_logger(log_path):

FILE: entity_sort/el_config.py
  class Args (line 4) | class Args:
    method parse (line 6) | def parse():
    method initialize (line 11) | def initialize(parser):
    method get_parser (line 71) | def get_parser(self):

FILE: entity_sort/el_dataset.py
  class ELDataset (line 10) | class ELDataset(Dataset):
    method __init__ (line 11) | def __init__(self, features):  # 特征向量
    method __len__ (line 15) | def __len__(self):
    method __getitem__ (line 18) | def __getitem__(self, index):

FILE: entity_sort/el_main.py
  function _load_candidates (line 48) | def _load_candidates(faiss_index=None, index_path=None):
  function load_models (line 64) | def load_models(args):
  function _process_biencoder_dataloader (line 83) | def _process_biencoder_dataloader(samples, tokenizer, biencoder_params):
  function _run_biencoder (line 96) | def _run_biencoder(biencoder, dataloader, top_k=10, indexer=None):
  function _run_biencoder_test (line 113) | def _run_biencoder_test(biencoder, dataloader, entity_to_ids, subject_id...
  class Trainer (line 183) | class Trainer:
    method __init__ (line 184) | def __init__(self, args, train_loader, dev_loader, test_loader):
    method configure_optimizers (line 201) | def configure_optimizers(self):
    method load_ckp (line 226) | def load_ckp(self, model, checkpoint_path):
    method save_ckp (line 233) | def save_ckp(self, state, checkpoint_path):
    method train (line 245) | def train(self):
    method dev (line 344) | def dev(self):
    method test (line 368) | def test(self, checkpoint_path):
    method convert_example_to_feature (line 397) | def convert_example_to_feature(self,
    method predict (line 505) | def predict(self,
    method get_metrics (line 665) | def get_metrics(self, outputs, targets):
    method get_classification_report (line 672) | def get_classification_report(self, outputs, targets):

FILE: entity_sort/el_models.py
  class BertForEntityLinking (line 12) | class BertForEntityLinking(nn.Module):
    method _forward_unimplemented (line 13) | def _forward_unimplemented(self, *input: Any) -> None:
    method __init__ (line 16) | def __init__(self, args):
    method find_entity_span (line 45) | def find_entity_span(self, entity_mask, flag=1):
    method forward (line 52) | def forward(self, token_ids, attention_masks, token_type_ids, seq_labe...
  class Args (line 98) | class Args:

FILE: entity_sort/el_preprocess.py
  class InputExample (line 20) | class InputExample:
    method __init__ (line 21) | def __init__(self, set_type, text, seq_label, entity_label):
  class BaseFeature (line 28) | class BaseFeature:
    method __init__ (line 29) | def __init__(self, token_ids, attention_masks, token_type_ids):
  class BertFeature (line 35) | class BertFeature(BaseFeature):
    method __init__ (line 36) | def __init__(self, token_ids, attention_masks, token_type_ids, seq_lab...
  class ELProcessor (line 46) | class ELProcessor:
    method __init__ (line 47) | def __init__(self):
    method read_json (line 53) | def read_json(self, path):
    method get_result (line 58) | def get_result(self, lines, set_type):
    method get_text_pair (line 90) | def get_text_pair(self, word, kb_id, text):
    method get_info (line 124) | def get_info(self, subject_id, maxlength):
  function convert_bert_example (line 166) | def convert_bert_example(ex_idx, example: InputExample, tokenizer: BertT...
  function convert_examples_to_features (line 269) | def convert_examples_to_features(examples, max_seq_len, bert_dir):
  function split_train_test (line 302) | def split_train_test(examples, train_rate):
  function get_out (line 317) | def get_out(processor, txt_path, args, mode):

FILE: entity_sort/el_processor.py
  class ELProcessor (line 10) | class ELProcessor:
    method __init__ (line 11) | def __init__(self):
    method read_json (line 17) | def read_json(self, path):  # 读取train.json
    method get_result (line 22) | def get_result(self, lines):
    method get_text_pair (line 45) | def get_text_pair(self, word, kb_id, text):
    method get_info (line 72) | def get_info(self, subject_id):

FILE: entity_sort/el_service.py
  function dict_to_obj (line 15) | def dict_to_obj(dictObj):
  class EntityLinking (line 25) | class EntityLinking:
    method __init__ (line 26) | def __init__(self):
    method predict (line 45) | def predict(self, text):
    method parse_result (line 58) | def parse_result(self, result):
  function get_result (line 75) | def get_result():

FILE: entity_sort/my_jieba/__init__.py
  function setLogLevel (line 54) | def setLogLevel(log_level):
  class Tokenizer (line 58) | class Tokenizer(object):
    method __init__ (line 60) | def __init__(self, dictionary=DEFAULT_DICT):
    method __repr__ (line 73) | def __repr__(self):
    method gen_pfdict (line 77) | def gen_pfdict(f):
    method initialize (line 99) | def initialize(self, dictionary=None):
    method check_initialized (line 174) | def check_initialized(self):
    method calc (line 178) | def calc(self, sentence, DAG, route):
    method get_DAG (line 186) | def get_DAG(self, sentence):
    method __cut_all (line 204) | def __cut_all(self, sentence):
    method __cut_DAG_NO_HMM (line 232) | def __cut_DAG_NO_HMM(self, sentence):
    method __cut_DAG (line 255) | def __cut_DAG(self, sentence):
    method cut (line 295) | def cut(self, sentence, cut_all=False, HMM=True, use_paddle=False):
    method cut_for_search (line 344) | def cut_for_search(self, sentence, HMM=True):
    method lcut (line 362) | def lcut(self, *args, **kwargs):
    method lcut_for_search (line 365) | def lcut_for_search(self, *args, **kwargs):
    method _lcut_no_hmm (line 371) | def _lcut_no_hmm(self, sentence):
    method _lcut_all (line 374) | def _lcut_all(self, sentence):
    method _lcut_for_search_no_hmm (line 377) | def _lcut_for_search_no_hmm(self, sentence):
    method get_dict_file (line 380) | def get_dict_file(self):
    method load_userdict (line 386) | def load_userdict(self, f):
    method add_word (line 426) | def add_word(self, word, freq=None, tag=None):
    method del_word (line 449) | def del_word(self, word):
    method suggest_freq (line 455) | def suggest_freq(self, segment, tune=False):
    method tokenize (line 486) | def tokenize(self, unicode_sentence, mode="default", HMM=True):
    method set_dictionary (line 519) | def set_dictionary(self, dictionary_path):
  function _lcut_all (line 552) | def _lcut_all(s):
  function _lcut (line 556) | def _lcut(s):
  function _lcut_no_hmm (line 560) | def _lcut_no_hmm(s):
  function _lcut_all (line 564) | def _lcut_all(s):
  function _lcut_for_search (line 568) | def _lcut_for_search(s):
  function _lcut_for_search_no_hmm (line 572) | def _lcut_for_search_no_hmm(s):
  function _pcut (line 576) | def _pcut(sentence, cut_all=False, HMM=True):
  function _pcut_for_search (line 589) | def _pcut_for_search(sentence, HMM=True):
  function enable_parallel (line 600) | def enable_parallel(processnum=None):
  function disable_parallel (line 623) | def disable_parallel():

FILE: entity_sort/my_jieba/__main__.py
  function cutfunc (line 34) | def cutfunc(sentence, _, HMM=True):

FILE: entity_sort/my_jieba/_compat.py
  function setLogLevel (line 11) | def setLogLevel(log_level):
  function enable_paddle (line 27) | def enable_paddle():
  function strdecode (line 76) | def strdecode(sentence):
  function resolve_filename (line 85) | def resolve_filename(f):

FILE: entity_sort/my_jieba/analyse/__init__.py
  function set_stop_words (line 16) | def set_stop_words(stop_words_path):

FILE: entity_sort/my_jieba/analyse/analyzer.py
  class ChineseTokenizer (line 19) | class ChineseTokenizer(Tokenizer):
    method __call__ (line 21) | def __call__(self, text, **kargs):
  function ChineseAnalyzer (line 34) | def ChineseAnalyzer(stoplist=STOP_WORDS, minsize=1, stemfn=stem, cachesi...

FILE: entity_sort/my_jieba/analyse/textrank.py
  class UndirectWeightedGraph (line 13) | class UndirectWeightedGraph:
    method __init__ (line 16) | def __init__(self):
    method addEdge (line 19) | def addEdge(self, start, end, weight):
    method rank (line 24) | def rank(self):
  class TextRank (line 57) | class TextRank(KeywordExtractor):
    method __init__ (line 59) | def __init__(self):
    method pairfilter (line 65) | def pairfilter(self, wp):
    method textrank (line 69) | def textrank(self, sentence, topK=20, withWeight=False, allowPOS=('ns'...

FILE: entity_sort/my_jieba/analyse/tfidf.py
  class KeywordExtractor (line 15) | class KeywordExtractor(object):
    method set_stop_words (line 23) | def set_stop_words(self, stop_words_path):
    method extract_tags (line 31) | def extract_tags(self, *args, **kwargs):
  class IDFLoader (line 35) | class IDFLoader(object):
    method __init__ (line 37) | def __init__(self, idf_path=None):
    method set_new_path (line 44) | def set_new_path(self, new_idf_path):
    method get_idf (line 55) | def get_idf(self):
  class TFIDF (line 59) | class TFIDF(KeywordExtractor):
    method __init__ (line 61) | def __init__(self, idf_path=None):
    method set_idf_path (line 68) | def set_idf_path(self, idf_path):
    method extract_tags (line 75) | def extract_tags(self, sentence, topK=20, withWeight=False, allowPOS=(...

FILE: entity_sort/my_jieba/finalseg/__init__.py
  function load_model (line 23) | def load_model():
  function viterbi (line 37) | def viterbi(obs, states, start_p, trans_p, emit_p):
  function __cut (line 59) | def __cut(sentence):
  function add_force_split (line 81) | def add_force_split(word):
  function cut (line 85) | def cut(sentence):

FILE: entity_sort/my_jieba/lac_small/creator.py
  function create_model (line 28) | def create_model(vocab_size, num_labels, mode='train'):

FILE: entity_sort/my_jieba/lac_small/nets.py
  function lex_net (line 25) | def lex_net(word, vocab_size, num_labels, for_infer=True, target=None):

FILE: entity_sort/my_jieba/lac_small/predict.py
  function get_sent (line 51) | def get_sent(str1):
  function get_result (line 68) | def get_result(str1):

FILE: entity_sort/my_jieba/lac_small/reader_small.py
  function load_kv_dict (line 24) | def load_kv_dict(dict_path,
  class Dataset (line 50) | class Dataset(object):
    method __init__ (line 52) | def __init__(self):
    method vocab_size (line 66) | def vocab_size(self):
    method num_labels (line 71) | def num_labels(self):
    method word_to_ids (line 75) | def word_to_ids(self, words):
    method label_to_ids (line 85) | def label_to_ids(self, labels):
    method get_vars (line 95) | def get_vars(self,str1):

FILE: entity_sort/my_jieba/lac_small/utils.py
  function str2bool (line 25) | def str2bool(v):
  function parse_result (line 33) | def parse_result(words, crf_decode, dataset):
  function parse_padding_result (line 76) | def parse_padding_result(words, crf_decode, seq_lens, dataset):
  function init_checkpoint (line 122) | def init_checkpoint(exe, init_checkpoint_path, main_program):

FILE: entity_sort/my_jieba/posseg/__init__.py
  function load_model (line 27) | def load_model():
  class pair (line 45) | class pair(object):
    method __init__ (line 47) | def __init__(self, word, flag):
    method __unicode__ (line 51) | def __unicode__(self):
    method __repr__ (line 54) | def __repr__(self):
    method __str__ (line 57) | def __str__(self):
    method __iter__ (line 63) | def __iter__(self):
    method __lt__ (line 66) | def __lt__(self, other):
    method __eq__ (line 69) | def __eq__(self, other):
    method __hash__ (line 72) | def __hash__(self):
    method encode (line 75) | def encode(self, arg):
  class POSTokenizer (line 79) | class POSTokenizer(object):
    method __init__ (line 81) | def __init__(self, tokenizer=None):
    method __repr__ (line 85) | def __repr__(self):
    method __getattr__ (line 88) | def __getattr__(self, name):
    method initialize (line 94) | def initialize(self, dictionary=None):
    method load_word_tag (line 98) | def load_word_tag(self, f):
    method makesure_userdict_loaded (line 114) | def makesure_userdict_loaded(self):
    method __cut (line 119) | def __cut(self, sentence):
    method __cut_detail (line 137) | def __cut_detail(self, sentence):
    method __cut_DAG_NO_HMM (line 154) | def __cut_DAG_NO_HMM(self, sentence):
    method __cut_DAG (line 177) | def __cut_DAG(self, sentence):
    method __cut_internal (line 217) | def __cut_internal(self, sentence, HMM=True):
    method _lcut_internal (line 244) | def _lcut_internal(self, sentence):
    method _lcut_internal_no_hmm (line 247) | def _lcut_internal_no_hmm(self, sentence):
    method cut (line 250) | def cut(self, sentence, HMM=True):
    method lcut (line 254) | def lcut(self, *args, **kwargs):
  function _lcut_internal (line 267) | def _lcut_internal(s):
  function _lcut_internal_no_hmm (line 271) | def _lcut_internal_no_hmm(s):
  function cut (line 275) | def cut(sentence, HMM=True, use_paddle=False):
  function lcut (line 309) | def lcut(sentence, HMM=True, use_paddle=False):

FILE: entity_sort/my_jieba/posseg/viterbi.py
  function get_top_states (line 10) | def get_top_states(t_state_v, K=4):
  function viterbi (line 14) | def viterbi(obs, states, start_p, trans_p, emit_p):

FILE: entity_sort/service_main.py
  class Trainer (line 7) | class Trainer:
    method __init__ (line 8) | def __init__(self, args):
    method load_ckp (line 16) | def load_ckp(self, model, checkpoint_path):
    method save_ckp (line 23) | def save_ckp(self, state, checkpoint_path):
    method convert_example_to_feature (line 27) | def convert_example_to_feature(self0,
    method predict (line 93) | def predict(self,

FILE: entity_sort/utils/tokenization.py
  function convert_to_unicode (line 27) | def convert_to_unicode(text):
  function printable_text (line 47) | def printable_text(text):
  function load_vocab (line 70) | def load_vocab(vocab_file):
  function convert_by_vocab (line 85) | def convert_by_vocab(vocab, items):
  function convert_tokens_to_ids (line 95) | def convert_tokens_to_ids(vocab, tokens):
  function convert_ids_to_tokens (line 99) | def convert_ids_to_tokens(inv_vocab, ids):
  function whitespace_tokenize (line 103) | def whitespace_tokenize(text):
  class FullTokenizer (line 112) | class FullTokenizer(object):
    method __init__ (line 115) | def __init__(self, vocab_file, do_lower_case=True):
    method tokenize (line 121) | def tokenize(self, text):
    method convert_tokens_to_ids (line 129) | def convert_tokens_to_ids(self, tokens):
    method convert_ids_to_tokens (line 132) | def convert_ids_to_tokens(self, ids):
  class BasicTokenizer (line 136) | class BasicTokenizer(object):
    method __init__ (line 139) | def __init__(self, do_lower_case=True):
    method tokenize (line 147) | def tokenize(self, text):
    method _run_strip_accents (line 171) | def _run_strip_accents(self, text):
    method _run_split_on_punc (line 182) | def _run_split_on_punc(self, text):
    method _tokenize_chinese_chars (line 202) | def _tokenize_chinese_chars(self, text):
    method _is_chinese_char (line 215) | def _is_chinese_char(self, cp):
    method _clean_text (line 237) | def _clean_text(self, text):
  class WordpieceTokenizer (line 251) | class WordpieceTokenizer(object):
    method __init__ (line 254) | def __init__(self, vocab, unk_token="[UNK]", max_input_chars_per_word=...
    method tokenize (line 259) | def tokenize(self, text):
  function _is_whitespace (line 313) | def _is_whitespace(char):
  function _is_control (line 325) | def _is_control(char):
  function _is_punctuation (line 337) | def _is_punctuation(char):

FILE: entity_sort/utils/utils.py
  function timer (line 9) | def timer(func):
  function set_seed (line 27) | def set_seed(seed=123):
  function set_logger (line 39) | def set_logger(log_path):