MCPcopy Create free account

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

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

Functionconvert_nql_to_tf
(value, dtype=None, name=None, as_ref=False)
language/nql/nql/__init__.py:1931
Methodconvert_relid
(q2b_relid)
language/emql/data_loader.py:1071
Functionconvert_shard
Multithreading helper to serialize an individual shard to disk. Args: shard: Tuple of shard id (int) and examples (ImageMetadata list). sha
language/capwap/utils/io_utils.py:196
Functionconvert_to_rc
Convert dataset of ImageMetadata items to RC-formatted JSON file. Args: dataset: List of ImageMetadata. filename: Path to write to.
language/capwap/utils/io_utils.py:262
Functionconvert_to_tfrecords
Convert a dataset to sharded TFRecords. Args: dataset: List of ImageMetadata. num_shards: Number of randomized shards to write dataset to.
language/capwap/utils/io_utils.py:232
Functioncopy_file_to_tmp
(path_pair)
language/serene/wiki_index.py:57
Functioncopy_files_to_dir
Copies files matching `source_filepattern` into `dest_dir`.
language/common/utils/file_utils.py:33
Functioncoreference_resolution_loss
Compute coreference resolution loss over global batch (across all devices). Args: mention_encodings: [n_mentions, hidden_dim] mention encodings
language/mentionmemory/modules/mention_losses.py:180
Functioncosine_distance
(x, y, normalize=True, reduce_axis=-1)
language/labs/consistent_zero_shot_nmt/models/losses.py:199
Functioncosine_similarity
Computes batched cosine similarity between two 2D arrays.
language/mentionmemory/utils/jax_utils.py:136
Functioncount_tf_records_parallel
Counts number of records in TFRecord files in parallel. Args: file_paths: a list of paths, where each path points to a GZIP-ed TFRecord f
language/realm/retrieval.py:145
Functioncount_tf_records_parallel_helper
Just a helper function for count_tf_records_parallel.
language/realm/retrieval.py:139
Functioncounts_to_idfs
Compute IDFs for the vocab given a V x N count matrix. Args: count_matrix: Sparse matrix from vocab to documents. cutoff: Fraction of lowes
language/labs/drkit/search_utils.py:65
Functioncounts_to_tfidf
Convert counts to Tf-Idf scores given diagonal idf matrix.
language/labs/drkit/search_utils.py:87
Methodcreate
(self)
language/serene/scrape_db.py:61
Functioncreate_agent
Creates an agent.
language/search_agents/muzero/agent_lib.py:92
Functioncreate_block_info
Create block info.
language/orqa/preprocessing/preprocess_wiki_extractor.py:61
Functioncreate_block_info
Create block info.
language/multivec/utils/data_processor.py:81
Functioncreate_bytes_feature
(value)
language/conpono/cpc/preproc/raw_books_preproc_pipeline.py:90
Methodcreate_bytes_feature
(value)
language/labs/drkit/input_fns.py:110
Methodcreate_bytes_feature
(value)
language/labs/drkit/run_dualencoder_lsf.py:2037
Methodcreate_bytes_feature
(value)
language/labs/drkit/hotpotqa/preprocessing/create_tfrecords.py:212
Methodcreate_bytes_feature
(value)
language/labs/drkit/preprocessing/preprocess_qa.py:212
Functioncreate_datasets
Create datasets for training and evaluation. Args: config: ml_collections.ConfigDict to use. data_rng: PRNGKey for seeding operations in th
language/gscan/xattn_model/dataset/input_pipeline.py:60
Functioncreate_dict_mask
(input_dict, mask_keys)
language/mentionmemory/utils/optim_utils.py:57
Methodcreate_encoder_model
(self, is_training, input_ids, input_mask, segment_ids)
language/canine/tydiqa/tydi_modeling.py:206
Functioncreate_environment
Creates an NQEnvironment. Args: task: Used to set the random seed of the environment. In SeedRL context, it makes sense to pass in the
language/search_agents/muzero/env.py:1011
Functioncreate_eval_input_fn
Creates an input function that can be used with tf.learn estimators.
language/xsp/model/input_pipeline.py:336
Functioncreate_experiment
Create Experiment.
language/labs/consistent_zero_shot_nmt/utils/t2t_tweaks.py:114
Functioncreate_float_feature
(values)
language/conpono/evals/model_builder.py:293
Functioncreate_from_dict
Create a new SymbolTable from a dictionary for deserialization. Args: symbol_table_dict: A dict created from to_dict. Returns: A newly c
language/nql/nql/symbol.py:221
Functioncreate_gnmt_rnn_cell
Returns an instance of an GNMT-style RNN cell. Args: unit_type: A string that specifies the type of the recurrent unit. Must be one of {"
language/labs/consistent_zero_shot_nmt/utils/model_utils.py:257
Functioncreate_hotpot_answer_model
Creates a classification model.
language/labs/drkit/model_fns.py:1554
Functioncreate_hotpotqa_model
Creates a classification model.
language/labs/drkit/model_fns.py:1400
Methodcreate_int_feature
(values)
language/labs/drkit/input_fns.py:105
Methodcreate_int_feature
(values)
language/labs/drkit/run_dualencoder_lsf.py:2032
Methodcreate_int_feature
(values)
language/labs/drkit/hotpotqa/answer_extractor.py:1193
Methodcreate_int_feature
(values)
language/labs/drkit/hotpotqa/preprocessing/create_tfrecords.py:207
Methodcreate_int_feature
(values)
language/labs/drkit/preprocessing/preprocess_qa.py:207
Methodcreate_int_feature
(values)
language/bert_extraction/steal_bert_qa/models/run_squad.py:1084
Methodcreate_int_feature
(values)
language/bert_extraction/steal_bert_qa/models/run_squad_membership.py:699
Methodcreate_int_feature
(values)
language/conpono/evals/run_squad.py:1071
Methodcreate_int_feature
(values)
language/question_answering/bert_joint/run_nq.py:759
Methodcreate_int_feature
(values)
language/question_answering/bert_joint/run_nq.py:1076
Methodcreate_int_feature
(values: Sequence[int])
language/canine/tydiqa/tf_io.py:144
Methodcreate_int_feature
(values)
language/canine/tydiqa/tf_io.py:221
Functioncreate_learning_rate_scheduler
Creates learning rate scheduler with options for warmup and linear decay. Args: learning_rate: base learning rate to be modified. warmup: i
language/mentionmemory/utils/optim_utils.py:23
Functioncreate_metrics_ops
Creates metrics ops for evaluation.
language/xsp/model/metrics.py:40
Functioncreate_mips_searcher
Create searcher for returning top-k closest elements.
language/labs/drkit/search_utils.py:258
Functioncreate_model
Creates a classification model. Args: model: the BERT model from modeling.py labels: ground truth paragraph order label_types: which k
language/conpono/cpc/bilin_model_builder.py:26
Functioncreate_model
Creates a classification model. Args: model: the BERT model from modeling.py labels: ground truth paragraph order label_types: which k
language/conpono/cpc/model_builder.py:26
Functioncreate_model
Creates a classification model. Args: model: the BERT model from modeling.py labels: ground truth paragraph order num_choices: number o
language/conpono/evals/model_builder.py:24
Functioncreate_model
Creates a classification model. Args: model: the BERT model from modeling.py labels: ground truth paragraph order decoder_inputs: the i
language/conpono/reconstruct/model_builder.py:35
Functioncreate_model_bilin
Creates a classification model. Args: model: the BERT model from modeling.py labels: ground truth paragraph order num_choices: number o
language/conpono/evals/model_builder.py:57
Methodcreate_one_hot_sketch
Wrapper to create the initial sketch for an one-hot vector.
language/emql/model.py:927
Functioncreate_onehop_model
Creates a classification model.
language/labs/drkit/model_fns.py:925
Functioncreate_optimizer
(unused_final_iteration)
language/search_agents/muzero/muzero_main.py:38
Functioncreate_optimizer
Creates an optimizer training op.
language/canine/bert_optimization.py:24
Functioncreate_serving_input_fn
Sets up input function that accepts serialized tf.SequenceExample.
language/xsp/model/input_pipeline.py:298
Functioncreate_train_state
Create and initialize the model. Args: model: Flax nn module. config: Configuration for model. rng: JAX PRNG Key. inputs: The init
language/gscan/xattn_model/train_utils.py:57
Functioncreate_training_input_fn
Creates an input function that can be used with tf.learn estimators.
language/xsp/model/input_pipeline.py:320
Functioncreate_twohop_model
Creates a classification model.
language/labs/drkit/model_fns.py:1132
Functioncreate_twohopcascade_model
Creates a classification model.
language/labs/drkit/model_fns.py:1002
Functioncreate_wikimovie_model
Creates a classification model.
language/labs/drkit/model_fns.py:1222
Functioncreate_word
Samples a new (word, lemma, root) tuple within a log likelihood bucket.
language/wino_dict/utils.py:247
Methodcreate_words
(seed, word, lemma, tag)
language/wino_dict/utils_test.py:190
Functioncross_shard_concat
Concatenates all replicas of `input_tensor` across all shardss. Args: input_tensor: <int32|float32> [local_batch_size, dim1, dim2, ...] Retu
language/common/utils/tpu_utils.py:83
Functiondataset_from_preprocessors
Get dataset of serialized examples from a preprocessing job. Args: preprocessing_servers: BNS addresses for all preprocessing tasks. rpcs_p
language/realm/preprocessing.py:130
Functiondataset_path
Return the dataset path based on its configuration. For example, {data_dir}/type=drqa,n_similar=5,n_background=10,include_nei=true Args: dat
language/serene/fever_tfds.py:707
Functiondate_in_interval
Check if date is within start and end. Args: date: Tuple of (year, month, day). start: Start date (year, month, day). end: End date (ye
language/templama/templama.py:95
Methoddecode
(self, ids)
language/serene/preprocessing.py:260
Functiondecode_features_and_labels
Creates a dataset tensor out of a dataset given placeholders.
language/xsp/model/input_utils.py:129
Methoddecode_func
A closure that builds decoder outputs.
language/labs/consistent_zero_shot_nmt/models/agreement.py:325
Methoddecode_hooks
(self)
language/labs/consistent_zero_shot_nmt/data_generators/translate_multilingual.py:141
Functiondecode_op
Decode set representations to member ids. Args: params: a dictionary of hyper-parameters. set_embs: dense embeddings of set. set_sketch
language/emql/module.py:109
Functiondecode_record
(key, serialized SequenceExample) to dict of <feature name, Tensor>.
language/xsp/model/input_utils.py:142
Functiondecode_steps_from_labels
Returns DecodeSteps given labels dict.
language/xsp/model/decode_utils.py:164
Functiondecode_steps_from_predictions
Returns DecodeSteps given predictions dict.
language/xsp/model/decode_utils.py:179
Functiondecomposable_attention
See https://arxiv.org/abs/1606.01933. Args: emb1: A Tensor with shape [batch_size, max_len1, emb_size] representing the first input seq
language/question_answering/decatt_docreader/layers/decomposable_attention.py:29
Functiondecorator
Invoked by Python when `@contract` is used as a function decorator.
language/canine/tensor_contracts.py:1198
Methoddefault
(self, obj)
language/labs/drkit/evaluate.py:51
Methoddefault
(self, obj)
language/capwap/utils/io_utils.py:283
Methoddefault_training_callback
Called after each training step is executed. Arguments: fd: feed dictionary used for the training step. loss: model's loss on t
language/nql/nql/util.py:76
Functiondefinition_first_strategy
Strategy to add the word definition to the beginning of the prompt.
language/wino_dict/utils.py:677
Functiondefinition_last_strategy
Strategy to add the word definition to the end of the prompt.
language/wino_dict/utils.py:684
Functiondefinition_synonym_first_strategy
Strategy to add the word definition to the beginning of the prompt.
language/wino_dict/utils.py:691
Functiondefinition_synonym_last_strategy
Strategy to add the word definition to the end of the prompt.
language/wino_dict/utils.py:698
Functiondelimiter_f1
Measures p/r/f1 of predicted delimiters.
language/fruit/metrics.py:101
Methoddelimiter_generator
()
language/fruit/rendering_utils.py:155
Methoddeserialize_dictionaries
Restore all symbol tables from a file. Existing tables having the same type_name are overwritten. Args: input_file: Filename string or
language/nql/nql/__init__.py:1508
Functiondeserialize_example
(tensor)
language/realm/example_generator.py:131
Functiondetokenize
(tokens, vocab_mode="downstream_vocab")
language/bert_extraction/steal_bert_classifier/data_generation/preprocess_util.py:92
Methoddict_dot_product
(self, dict1, dict2)
language/multivec/preprocessing/create_training_data.py:745
Functiondict_to_beam_counts
(metrics_dict, namespace)
language/compgen/csl/common/beam_utils.py:20
Functiondict_to_json_file
Saves a dictionary to a json file.
language/compgen/csl/common/json_utils.py:29
Functiondisable_all_checks
Disables all checks (both static and dynamic) by TensorContracts.
language/canine/tensor_contracts.py:81
Functiondisable_dynamic_asserts
Convenience method for `enable_dynamic_asserts(False)`.
language/canine/tensor_contracts.py:56
Functiondisable_predicates
Convenience method for `enable_predicates(False)`.
language/canine/tensor_contracts.py:76
Methoddrop
(self)
language/serene/scrape_db.py:64
Methoddrop
(self)
language/serene/wiki_db.py:63
← previousnext →3,701–3,800 of 6,154, ranked by callers