MCPcopy Create free account

hub / github.com/google-research/language / functions

Functions6,154 in github.com/google-research/language

↓ 1 callersFunction_grow_alive
Given sequences and scores, will gather the top k=beam size sequences. Args: curr_seq: current topk sequence that has been grown by one positio
language/xsp/model/beam_search.py:164
↓ 1 callersFunction_grow_finished
Given sequences and scores, will gather the top k=beam size sequences. Args: finished_seq: Current finished sequences. [batch_size, beam_
language/xsp/model/beam_search.py:127
↓ 1 callersFunction_grow_topk
r"""Inner beam seach loop. This function takes the current alive sequences, and grows them to topk sequences where k = 2*beam. We use 2*beam beca
language/xsp/model/beam_search.py:191
↓ 1 callersFunction_has_leading_or_trailing_punctuation
(source)
language/compgen/csl/tasks/exact_match_utils.py:28
↓ 1 callersFunction_has_nested_from_clause
Returns True if the outer sql statement has nested from clause.
language/xsp/data_preprocessing/abstract_sql.py:628
↓ 1 callersMethod_hash_bucket_tensors
Converts ids to hash bucket ids via multiple hashing. Args: ids: The codepoints or other IDs to be hashed. num_hashes: The number of
language/canine/modeling.py:530
↓ 1 callersFunction_improve_answer_span
Returns tokenized answer spans that better match the annotated answer.
language/labs/drkit/hotpotqa/answer_extractor.py:578
↓ 1 callersFunction_improve_answer_span
Returns tokenized answer spans that better match the annotated answer.
language/labs/drkit/hotpotqa/preprocessing/create_tfrecords.py:239
↓ 1 callersFunction_improve_answer_span
Returns tokenized answer spans that better match the annotated answer.
language/labs/drkit/preprocessing/preprocess_qa.py:239
↓ 1 callersFunction_improve_answer_span
Returns tokenized answer spans that better match the annotated answer.
language/bert_extraction/steal_bert_qa/models/run_squad.py:488
↓ 1 callersFunction_improve_answer_span
Returns tokenized answer spans that better match the annotated answer.
language/conpono/evals/run_squad.py:475
↓ 1 callersFunction_induce_rules
(pipeline)
language/compgen/csl/induction/search_main_beam.py:154
↓ 1 callersFunction_induce_rules_for_examples
Iteratively searches for rules to optimize codelength objective.
language/compgen/nqg/model/induction/induction_utils.py:188
↓ 1 callersMethod_init_bm25
Initialize BM25 retriever.
language/quest/bm25/bm25_retriever.py:31
↓ 1 callersMethod_init_strategy
Initialize the distribution strategy (e.g. TPU/GPU/Mirrored).
language/serene/training.py:281
↓ 1 callersMethod_initial_reward
(self)
language/search_agents/muzero/env.py:799
↓ 1 callersFunction_int64_feature
(value)
language/fruit/tf_utils.py:33
↓ 1 callersFunction_int64_list_feature
(values)
language/question_answering/decatt_docreader/preprocessing/create_nq_long_examples.py:43
↓ 1 callersMethod_intermediate_reward
Computes an 'intermediate' reward after each step. r_t = S(d_t | q) - S(d_t-1 | q). Intermediate rewards (which make up the bulk of our rewa
language/search_agents/muzero/env.py:715
↓ 1 callersMethod_invert_binary_conjunct
Inverts a binary conjunct for an IR to its original representation.
language/compir/dataset_parsers/cfq_parser.py:219
↓ 1 callersMethod_invert_unary_conjunct
Inverts an unary conjunct for an IR to its original representation.
language/compir/dataset_parsers/cfq_parser.py:215
↓ 1 callersFunction_is_balanced_paren_candidate
(rule)
language/compgen/nqg/model/induction/induction_utils.py:81
↓ 1 callersFunction_is_column_mention
Returns True if sqlparse_token represents a column.
language/xsp/data_preprocessing/abstract_sql.py:178
↓ 1 callersFunction_is_named_tuple
Check if we're dealing with a NamedTuple.
language/canine/tensor_contracts.py:1332
↓ 1 callersFunction_is_operator
(item)
language/xsp/data_preprocessing/sql_parsing.py:61
↓ 1 callersFunction_is_ragged_tensor
Returns true if the value is a RaggedTensor *or* a normal tensor.
language/canine/tensor_contracts.py:966
↓ 1 callersFunction_is_scalar
Returns true for tensors that are a scalar.
language/canine/tensor_contracts.py:1004
↓ 1 callersFunction_is_value_literal
Returns true if sqlparse_token represents a literal to copy from query.
language/xsp/data_preprocessing/abstract_sql.py:269
↓ 1 callersFunction_is_whitespace
(sqlparse_token)
language/compgen/nqg/tasks/spider/sql_tokenizer.py:20
↓ 1 callersFunction_is_wildcard
(item)
language/xsp/data_preprocessing/sql_parsing.py:69
↓ 1 callersFunction_iterate_sql
(sql_parse)
language/xsp/data_preprocessing/sql_parsing.py:232
↓ 1 callersFunction_layer_softmax
(entities)
language/labs/drkit/model_fns.py:1497
↓ 1 callersFunction_lcs
Computes the length of the LCS between two seqs. The implementation below uses a DP programming algorithm and runs in O(nm) time where n = len(x)
language/labs/exemplar_decoding/utils/rouge_utils.py:50
↓ 1 callersFunction_lcs
Computes the length of the LCS between two seqs. The implementation below uses a DP programming algorithm and runs in O(nm) time where n = len(x)
language/table_text_eval/table_text_eval.py:263
↓ 1 callersFunction_lcs
Computes the length of the LCS between two seqs. The implementation below uses a DP programming algorithm and runs in O(nm) time where n = len(x)
language/totto/totto_parent_eval.py:243
↓ 1 callersFunction_len_lcs
Returns the length of the Longest Common Subsequence between two seqs. Source: http://www.algorithmist.com/index.php/Longest_Common_Subsequence
language/labs/exemplar_decoding/utils/rouge_utils.py:33
↓ 1 callersFunction_len_lcs
Returns the length of the Longest Common Subsequence between two seqs. Source: http://www.algorithmist.com/index.php/Longest_Common_Subsequence
language/table_text_eval/table_text_eval.py:246
↓ 1 callersFunction_len_lcs
Returns the length of the Longest Common Subsequence between two seqs. Source: http://www.algorithmist.com/index.php/Longest_Common_Subsequence
language/totto/totto_parent_eval.py:226
↓ 1 callersFunction_link_entities
Return list of matching entities in text.
language/labs/drkit/metaqa/preprocessing/process_wiki.py:50
↓ 1 callersMethod_load_fever_embeddings
Load fever claim embeddings and ids.
language/serene/wiki_index.py:274
↓ 1 callersFunction_load_json
(filepath)
language/xsp/evaluation/convert_preds_for_spider_eval.py:65
↓ 1 callersFunction_load_json_from_file
(filename)
language/xsp/data_preprocessing/convert_to_examples.py:72
↓ 1 callersFunction_load_predictions
Loads predictions from file.
language/compir/utils/io_utils.py:37
↓ 1 callersFunction_load_predictions_dicts
Returns list of predictions dicts.
language/xsp/evaluation/restore_from_asql.py:35
↓ 1 callersFunction_load_predictions_dicts
Returns list of predictions dicts.
language/xsp/evaluation/convert_preds_for_spider_eval.py:70
↓ 1 callersFunction_load_records
(filenames)
language/question_answering/decatt_docreader/datasets/nq_long_dataset.py:74
↓ 1 callersFunction_load_records
(filenames)
language/question_answering/decatt_docreader/datasets/nq_short_pipeline_dataset.py:72
↓ 1 callersMethod_load_wiki_embeddings
Build an index from embeddings.
language/serene/wiki_index.py:289
↓ 1 callersFunction_log_matmul_exp
Numerically stable equivalent of log-matmul-exp computation. Is equivalent to: tf.math.log( tf.matmul(tf.math.exp(tensor_a),
language/compgen/csl/model/weighted_model.py:20
↓ 1 callersFunction_log_prob_from_logits
(logits)
language/xsp/model/beam_search.py:56
↓ 1 callersFunction_make_bert_vocab_productions
Generates productions for the `unconstrained` vocabulary setting. To ensure that non-initial word-pieces never begin a word, we use a grammar cor
language/search_agents/muzero/grammar_lib.py:73
↓ 1 callersFunction_make_comparative
Heuristic function to build comparative adjectives.
language/wino_dict/utils.py:236
↓ 1 callersFunction_make_sparse_tensor_dict
()
language/nql/nql/io_test.py:28
↓ 1 callersFunction_make_symbol_table_dict
()
language/nql/nql/io_test.py:39
↓ 1 callersFunction_map_years_to_objects
Map each year between min, max to the corresponding object in facts. Args: facts: a list of facts with the same subject and relation. qid_n
language/templama/templama.py:296
↓ 1 callersMethod_maybe_actually_init
Lazily create example converter.
language/capwap/preprocessing/text_synthetic_qa.py:75
↓ 1 callersMethod_maybe_actually_init
Lazily create example converter.
language/capwap/preprocessing/text_synthetic_qa.py:106
↓ 1 callersFunction_maybe_get_candidate_pair
Returns candidate rule pair if proposed sources and targets are valid.
language/compgen/csl/induction/unification_utils.py:88
↓ 1 callersFunction_maybe_get_candidate_pair
Returns candidate rule pair if proposed sources and targets are valid.
language/compgen/nqg/model/induction/split_utils.py:61
↓ 1 callersFunction_maybe_list_replace
Replace first occurrence of sublist in lst with replacement.
language/compgen/nqg/tasks/geoquery/entity_utils.py:18
↓ 1 callersFunction_maybe_make_rule
Canoncalize NT indexes and return QCFGRule.
language/compgen/csl/induction/unification_utils.py:77
↓ 1 callersFunction_maybe_mask
Mask the sequence with m.
language/labs/exemplar_decoding/models/attention.py:78
↓ 1 callersFunction_maybe_mask_score
(score, memory_sequence_length, score_mask,
language/labs/exemplar_decoding/models/attention.py:94
↓ 1 callersFunction_maybe_replace_entity
Replace entity identifiers so that they can be generated using copy.
language/compgen/nqg/tasks/geoquery/entity_utils.py:36
↓ 1 callersFunction_maybe_unpack
(x)
language/fruit/scripts/evaluate_t5x_jsonl.py:45
↓ 1 callersFunction_maybe_unpack
(x)
language/fruit/scripts/evaluate_direct_jsonls.py:67
↓ 1 callersFunction_merge_nested_dicts_rec
Procedure to merge two nested dictionaries.
language/mentionmemory/utils/checkpoint_utils.py:61
↓ 1 callersMethod_min_remaining_steps_from_stack
(self, stack)
language/search_agents/muzero/state_tree.py:270
↓ 1 callersMethod_min_steps_to_terminal
(self, node)
language/search_agents/muzero/state_tree.py:128
↓ 1 callersFunction_morph_by_rules
Creates the morphology dictionary for a word based on the rules. Note that this dictionary is from a part of speech to that word's conjugation fo
language/wino_dict/create_new_words.py:253
↓ 1 callersFunction_most_frequent_answer
(year2obj)
language/templama/templama.py:388
↓ 1 callersFunction_most_recent_answer
(yr2obj)
language/templama/templama.py:394
↓ 1 callersFunction_nested_statement_end_idx
If acion_idx is start of nested statement, return end_idx, else None.
language/xsp/data_preprocessing/abstract_sql_converters.py:200
↓ 1 callersFunction_next_token
()
language/xsp/data_preprocessing/sql_parsing.py:347
↓ 1 callersFunction_ngrams
Yields all ngrams of given order in sequence.
language/table_text_eval/table_text_eval.py:290
↓ 1 callersFunction_ngrams
Yields all ngrams of given order in sequence.
language/totto/totto_parent_eval.py:270
↓ 1 callersFunction_noop_context_manager
A `ContextManager` that does nothing.
language/canine/tensor_contracts.py:1274
↓ 1 callersFunction_num_context_mapper
Wrapper for mapper that computes number of contexts.
language/question_answering/decatt_docreader/models/nq_long_model.py:215
↓ 1 callersFunction_numpy_dict_to_sparse_matrix
Convert a dictionary of numpy arrays into a scipy sparse matrix. Args: numpy_dict: A dictionary representing the data. Returns: A scipy
language/nql/nql/io.py:169
↓ 1 callersMethod_oov_bucket
(self, token)
language/serene/preprocessing.py:295
↓ 1 callersFunction_open
(path)
language/question_answering/bert_joint/run_nq.py:819
↓ 1 callersMethod_pad_or_truncate
(self, token_ids)
language/serene/preprocessing.py:672
↓ 1 callersMethod_pad_or_truncate_pair
Pad or truncate pair.
language/serene/preprocessing.py:684
↓ 1 callersFunction_pad_values
(values, padded_length)
language/compgen/csl/model/data/example_converter.py:33
↓ 1 callersFunction_pad_values
(values, padded_length)
language/compgen/nqg/model/parser/data/example_converter.py:26
↓ 1 callersFunction_page_name_to_title
(page_name)
language/serene/wikipedia_processing.py:50
↓ 1 callersFunction_parse_mr
Parse MR into a tuple.
language/table_text_eval/preprocess_webnlg.py:62
↓ 1 callersFunction_parse_sentence
Parses a single `sentence` string from the FEVER WikipediaDump.
language/serene/wikipedia_processing.py:54
↓ 1 callersFunction_parse_sentences
Parses the `lines` values from the FEVER Wikipedia dump.
language/serene/wikipedia_processing.py:73
↓ 1 callersFunction_parse_to_str_inner
(node, indent)
language/compgen/nqg/tasks/spider/sql_parser.py:72
↓ 1 callersFunction_parse_where
Parse the part relative to the WHERE clause of the SQL query.
language/xsp/data_preprocessing/sql_parsing.py:666
↓ 1 callersFunction_placeholder_to_feature
(placeholder)
language/common/utils/nest_utils.py:91
↓ 1 callersMethod_pool
Grab the [CLS] molecule for use in classification tasks.
language/canine/modeling.py:472
↓ 1 callersFunction_populate_spans_for_token
Add SqlSpan tuples to sql_spans given grouped_tokens and schema.
language/xsp/data_preprocessing/abstract_sql.py:277
↓ 1 callersFunction_prepare_memory
Convert to tensor and possibly mask `memory`. Args: memory: `Tensor`, shaped `[batch_size, max_time, ...]`. memory_sequence_length: `int32`
language/labs/exemplar_decoding/models/attention.py:39
↓ 1 callersMethod_preprocess
Preprocesses features for multilingual translation.
language/labs/consistent_zero_shot_nmt/models/agreement.py:80
↓ 1 callersMethod_preprocess
Preprocesses features for multilingual translation.
language/labs/consistent_zero_shot_nmt/models/basic.py:106
↓ 1 callersFunction_prettyprint_ngrams
(output: List[str])
language/wino_dict/create_new_words.py:109
↓ 1 callersFunction_print_debug_info
(results: List[tf.train.Example], creator_fn: tf_io.CreateTFExampleFn)
language/canine/tydiqa/tf_io_test.py:34
↓ 1 callersFunction_print_tree
Print tree for debugging.
language/compgen/csl/model/data/example_converter.py:25
← previousnext →1,501–1,600 of 6,154, ranked by callers