MCPcopy Create free account

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

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

↓ 1 callersFunctionmax_scoring_span
Compute max scoring span, using the sum of start and end scores. Args: start_scores: <float32> [batch_size, seq_len] end_scores: <float32>
language/capwap/utils/reward_utils.py:48
↓ 1 callersFunctionmaximize_divergence
Approx. maximizes compound divergence by iteratively swapping examples.
language/compgen/nqg/tasks/mcd_utils.py:207
↓ 1 callersMethodmaybe_add_child
Adds a new node for a given child symbol if not already in Trie.
language/compgen/csl/cky/cfg_parser.py:45
↓ 1 callersMethodmaybe_add_child
Adds a new node for a given child symbol if not already in Trie.
language/compgen/nqg/common/cky/trie_utils.py:41
↓ 1 callersMethodmaybe_load_memory
Load memory (passages and mention positions) if it's not loaded yet.
language/mentionmemory/encoders/mention_memory_encoder.py:79
↓ 1 callersMethodmeaning
(self)
language/canine/config_utils_test.py:65
↓ 1 callersMethodmemory_layer
(self)
language/labs/exemplar_decoding/models/attention.py:184
↓ 1 callersFunctionmerge_predictions
Merge multiple predcition files into one.
language/compgen/csl/model/inference/eval_model_beam.py:78
↓ 1 callersFunctionmerge_validation_gold_and_candidate
Merge gold and candidate examples for the validation fold. Note: This mutates the members of examples, specifically adds the tfidf_candidate fi
language/serene/fever_tfds.py:50
↓ 1 callersFunctionmetric_fn
(per_example_loss, label_ids, logits)
language/bert_extraction/steal_bert_qa/utils/run_bert_boolq_diff.py:483
↓ 1 callersFunctionmetric_max_over_ground_truths
(metric_fn, prediction, ground_truths)
language/orqa/utils/eval_utils.py:58
↓ 1 callersFunctionmm3hash
Returns a murmur hash for given string.
language/labs/drkit/search_utils.py:35
↓ 1 callersFunctionmodel_fn
Model function for use with tf.learn.Estimator.
language/xsp/model/model_builder.py:57
↓ 1 callersMethodmodel_fn
Estimator model_fn produced by ModelBuilder. Args: features: passed to config_model_prediction labels: passed to config_model_[
language/nql/nql/util.py:184
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/labs/drkit/run_dualencoder_lsf.py:1414
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/labs/drkit/hotpotqa/answer_extractor.py:652
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/bert_extraction/steal_bert_classifier/models/run_classifier_membership.py:592
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/bert_extraction/steal_bert_classifier/models/run_classifier_distillation.py:560
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/bert_extraction/steal_bert_classifier/models/run_classifier.py:814
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/bert_extraction/steal_bert_qa/utils/run_bert_boolq_diff.py:419
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/bert_extraction/steal_bert_qa/models/run_squad.py:602
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/bert_extraction/steal_bert_qa/models/run_squad_membership.py:522
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/bert_extraction/steal_bert_qa/models/run_bert_boolq_distill.py:456
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/bert_extraction/steal_bert_qa/models/run_bert_boolq.py:440
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/cpc/run_cc_cpc.py:558
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/cpc/run_bilin_cpc.py:614
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/cpc/run_cpc.py:280
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/evals/run_squad.py:589
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/evals/run_finetune_coherence.py:237
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/evals/run_hellaswag.py:192
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/evals/run_classifier.py:721
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/evals/run_multichoice.py:192
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/evals/run_record.py:303
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/evals/run_concat_classifier.py:832
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/binary_order/run_binary_coherence.py:238
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/conpono/reconstruct/run_paragraph_reconstruct.py:261
↓ 1 callersFunctionmodel_fn_builder
(bert_config, init_checkpoint, learning_rate, num_train_steps, num_warmup_steps, use_tpu,
language/multivec/models/ranking_model_experiment_inbatch.py:584
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/boolq/run_bert_boolq.py:425
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/question_answering/bert_joint/run_nq.py:896
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/question_answering/b2t2/run_b2t2.py:253
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/question_answering/b2t2/run_dual_encoder.py:186
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for tf.Estimator.
language/relation_learning/models/bert_fewshot_classifier.py:329
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/capwap/synthetic/generate_answers.py:307
↓ 1 callersFunctionmodel_fn_builder
Returns `model_fn` closure for TPUEstimator.
language/capwap/synthetic/generate_questions.py:231
↓ 1 callersFunctionmodel_fn_builder_dual
Returns model_fn for TPUEstimator.
language/multivec/models/ranking_model_experiment_inbatch.py:597
↓ 1 callersFunctionmodel_fn_builder_single
Returns `model_fn` closure for TPUEstimator.
language/multivec/models/ranking_model_experiment_inbatch.py:684
↓ 1 callersFunctionmodel_function
A generic model, satisfying the tf.Estimator API. Args: features: Dictionary of input tensors. Key "seqs": For the pattern task:
language/labs/memory/run_models.py:116
↓ 1 callersMethodmodel_mixture
Jointly train four tasks. m_xxx stands for feature and label to train the set membership task. p_xxx stands for a pair of set to train inters
language/emql/model.py:295
↓ 1 callersFunctionmodel_recall_by_position
Compute the evidence recall for each claim and list of ranked evidence. Args: verifiable_dev: The subset of validation set that is verifiable.
language/serene/callbacks.py:135
↓ 1 callersMethodne
(self)
language/nql/nql/nql_test.py:430
↓ 1 callersMethodnext_inputs
Compute next inputs and state.
language/labs/consistent_zero_shot_nmt/modules/helpers.py:195
↓ 1 callersMethodno_transformation
Applies no transformation by returning the raw data.
language/compir/dataset_parsers/dataset_parser.py:80
↓ 1 callersFunctionnode_fn
Represent nodes as target strings.
language/compgen/csl/qcfg/qcfg_parser.py:141
↓ 1 callersFunctionnode_fn
Represent node as applied rules.
language/compgen/csl/qcfg/qcfg_target_parser.py:132
↓ 1 callersFunctionnode_fn
Represent nodes as target strings.
language/compgen/nqg/model/qcfg/qcfg_parser.py:121
↓ 1 callersFunctionnonneg_crossentropy
A cross entropy operator that is appropriate for NQL outputs. Query expressions often evaluate to sparse vectors. This evaluates cross entropy s
language/nql/nql/__init__.py:1778
↓ 1 callersFunctionnonneg_softmax
A softmax operator that is appropriate for NQL outputs. NeuralQueryExpressions often evaluate to sparse vectors of small, nonnegative values. Sof
language/nql/nql/__init__.py:1755
↓ 1 callersFunctionnormalize_funql
Recursively parse FunQL string to re-formatted string.
language/compgen/nqg/tasks/geoquery/funql_normalization.py:100
↓ 1 callersFunctionobject_graph_key_mapping
Return name to key mappings from the checkpoint. Args: checkpoint_path: string, path to object-based checkpoint Returns: Dictionary mapp
language/search_agents/muzero/transformer_encoder.py:332
↓ 1 callersFunctionone_hop
One hop of propagation from input to output entities.
language/labs/drkit/model_fns.py:752
↓ 1 callersMethodoov_metric
Metric for checking the OOV rate. Args: word_ids: int32 Tensor of arbitrary shape with word indices. Returns: metric: A metric c
language/common/inputs/embedding_utils.py:254
↓ 1 callersMethodorig_tmp_file_pairs
Creates a list of tuples of (original_path, tmp_path). Returns: A list of tuples where first path is original and second is tmp location
language/serene/wiki_index.py:93
↓ 1 callersFunctionoriginal_query_state_part
Computes the original query part of the BERT state.
language/search_agents/muzero/bert_state_lib.py:117
↓ 1 callersFunctionoutput_order_with_beam
Compute the predicted output for each paragraph.
language/conpono/evals/coherence_eval.py:116
↓ 1 callersFunctionpad_or_truncate_pair
Pad or truncate pair.
language/multivec/predict/encode_blocks.py:93
↓ 1 callersMethodpadded_shapes
(self)
language/serene/preprocessing.py:542
↓ 1 callersFunctionparent
Metric for comparing predictions to references given tables. Args: predictions: An iterator over tokenized predictions. Each prediction i
language/table_text_eval/table_text_eval.py:304
↓ 1 callersFunctionparent
Metric for comparing predictions to references given tables. Args: predictions: An iterator over tokenized predictions. Each prediction is a
language/totto/totto_parent_eval.py:284
↓ 1 callersFunctionparse
Run bottom up parser. Args: tokens: List of strings for input (terminals or nonterminals). rules: List of QCFGRule instances. node_fn:
language/compgen/csl/qcfg/qcfg_parser.py:34
↓ 1 callersFunctionparse
Run bottom up parser on QCFG target using target CFG. Args: tokens: List of strings for input. rules: List of TargetCfgRule instances.
language/compgen/csl/qcfg/qcfg_target_parser.py:29
↓ 1 callersFunctionparse
Run bottom up parser. Args: tokens: List of strings for input. rules: List of QCFGRule instances. node_fn: Function with input argument
language/compgen/nqg/model/qcfg/qcfg_parser.py:41
↓ 1 callersFunctionparse_activation
( name)
language/serene/model.py:38
↓ 1 callersFunctionparse_args
(argv=None)
language/asqa/scoring.py:381
↓ 1 callersFunctionparse_args
Parse input arguments.
language/asqa/convert_to_roberta_format.py:43
↓ 1 callersFunctionparse_args
Argument parser.
language/asqa/human_annotation/analysis.py:132
↓ 1 callersFunctionparse_args
Argument parser.
language/asqa/human_annotation/preparation.py:70
↓ 1 callersFunctionparse_bounding_box
Decodes an image, resizes and scales RGB values to be in [0, 1].
language/question_answering/b2t2/run_b2t2.py:506
↓ 1 callersFunctionparse_by_regexp
(regexp_string: str, response: str)
language/search_agents/t5/t5_agent_lib.py:186
↓ 1 callersFunctionparse_fold
Parse the examples in the model predictions. Args: fold_name: Name of fold that examples in rows are from model_predictions: Map from examp
language/serene/analysis.py:116
↓ 1 callersFunctionparse_sparse_situation
Parse sparse situation to dense grid. Code adapted from https://github.com/LauraRuis/multimodal_seq2seq_gSCAN/ blob/master/read_gscan/read_gsca
language/gscan/xattn_model/dataset/preprocess.py:234
↓ 1 callersFunctionparse_symbols
Run bottom up parser. Let T be an arbitrary type for chart entries, specified by the return type of populate_fn. Examples for T are simple types
language/compgen/csl/cky/cfg_parser.py:126
↓ 1 callersFunctionparse_to_nested_list
Parse target to nested lists based on parens.
language/compgen/csl/tasks/smcalflow/tools/generate_target_cfg.py:150
↓ 1 callersFunctionparse_to_str
(node)
language/compgen/nqg/tasks/spider/sql_parser.py:80
↓ 1 callersFunctionpartition_preds_by_scrape_type
Partition predictions by which scrape_type they come from. The validation fold contains four sets of evidence: drqa, lucene, ukp_pred, and ukp_wi
language/serene/callbacks.py:164
↓ 1 callersFunctionpassage_texts
(pid_list, passages)
language/multivec/preprocessing/create_training_data.py:259
↓ 1 callersFunctionperturbed_chunks
Perturbed chunks.
language/orqa/datasets/ict_dataset.py:88
↓ 1 callersFunctionpick_answer_from_nbest
Picks from nbest uniformly at random, or return s.
language/capwap/synthetic/generate_answers.py:567
↓ 1 callersFunctionpipeline
Maps all dev-set query ids to their episodes.
language/search_agents/muzero/run_inference_beam.py:270
↓ 1 callersFunctionpipeline
(root)
language/search_agents/t5/run_inference_beam.py:264
↓ 1 callersMethodplace_target_reference_object
Place and return target's refenrece object and its position.
language/gscan/data/dataset.py:486
↓ 1 callersFunctionpopulate_example_from_sql_spans
Creates a sequence of output / decoder actions from sql_spans. Args: sql_spans: List of SqlSpan tuples. example: The NLToSQLExample object
language/xsp/data_preprocessing/abstract_sql_converters.py:166
↓ 1 callersMethodpostprocess
Get final prediction from logits.
language/labs/drkit/hotpotqa/demo.py:545
↓ 1 callersMethodpostprocess
Apply postpostprocess_fn to a chart cell.
language/compgen/nqg/common/cky/trie_utils.py:106
↓ 1 callersFunctionpostprocess_candidate_chains
Postprocess shortest paths and keep the one that leads to answers. Args: kb: dictionary to store the KB {subject: {relation: objects}} topi
language/labs/drkit/metaqa/preprocessing/metaqa_preprocess.py:130
↓ 1 callersFunctionpostprocess_candidate_chains
Postprocess shortest paths and keep the one that leads to answers. Args: kb: dictionary to store the KB {subject: {relation: objects}} topi
language/emql/preprocess/metaqa_preprocess.py:130
↓ 1 callersFunctionpostprocess_candidates
Perform additional processing on the candidates retrieved for a query. Args: candidates (list[Document]): a list of retrieved documents. qu
language/realm/example_generator.py:295
↓ 1 callersFunctionpostprocess_fn
(nodes)
language/compgen/nqg/model/parser/data/parsing_utils.py:110
↓ 1 callersMethodpostprocess_full
Inverts predicted reversible IRs to programs and postprocesses them.
language/compir/dataset_parsers/dataset_parser.py:62
↓ 1 callersMethodpostprocess_program
Postprocesses a single predicted program.
language/compir/dataset_parsers/dataset_parser.py:51
↓ 1 callersMethodpredict
Runs a prediction on a set of features. Calling multiple times does *not* regenerate the graph which makes predict much faster. Args:
language/labs/drkit/bert_utils_v2.py:98
← previousnext →2,401–2,500 of 6,154, ranked by callers