Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/google-research/language
/ functions
Functions
6,154 in github.com/google-research/language
⨍
Functions
6,154
◇
Types & classes
800
↳
Endpoints
12
↓ 1 callers
Function
write_predictions
Writing out the predictions to a file.
language/multivec/models/metrics.py:165
↓ 1 callers
Function
write_predictions
Writes one-best predictions in Spider-evaluation compatible format.
language/xsp/evaluation/convert_preds_for_spider_eval.py:127
↓ 1 callers
Function
write_result_to_json
write results to json file.
language/multivec/predict/retrieval.py:66
↓ 1 callers
Function
write_result_to_tsv
write results to tsv file.
language/multivec/predict/retrieval.py:42
↓ 1 callers
Function
write_summary
Write summary statistics about the examples in df. For example, this will output how many of each label there is of each type. Args: report_
language/serene/analysis.py:199
↓ 1 callers
Method
write_tf_feature_files
Converts TyDi examples to features and writes them to files.
language/canine/tydiqa/run_tydi_lib.py:299
↓ 1 callers
Function
write_tsv
Write examples to tsv file.
language/compgen/nqg/tasks/tsv_utils.py:44
↓ 1 callers
Function
yield_predictions_from_estimator
Yields predictions from Estimator.predict, with added error correction. This function handles the case of Estimator.predict occasionally restarting
language/realm/retrieval.py:434
Method
PopExample
(self, request, context)
language/realm/preprocessing.py:41
Method
PopExample
Return a tf.Example given an unused (usually empty) input tf.Example.
language/realm/preprocessing_pb2_grpc.py:44
Method
PopExample
(request, target, options=(), channel_credentials=None,
language/realm/preprocessing_pb2_grpc.py:72
Method
__and__
Alias for __mul__ which is more appropriate for hard sets. Args: other: A NeuralQueryExpression Returns: A NeuralQueryExpression
language/nql/nql/__init__.py:460
Method
__call__
(self, args)
language/labs/exemplar_decoding/models/linear.py:100
Method
__call__
Score the query based on the keys and values. Args: query: Tensor of dtype matching `self.values` and shape `[batch_size, query_dep
language/labs/exemplar_decoding/models/attention.py:363
Method
__call__
Score the query based on the keys and values. Args: query: Tensor of dtype matching `self.values` and shape `[batch_size, query_dep
language/labs/exemplar_decoding/models/attention.py:464
Method
__call__
Score the query based on the keys and values. Args: query: Tensor of dtype matching `self.values` and shape `[batch_size, query_dep
language/labs/consistent_zero_shot_nmt/modules/attention_wrappers.py:88
Method
__call__
(self, reuse=None, **kwargs)
language/labs/consistent_zero_shot_nmt/modules/base.py:42
Method
__call__
Run the cell with bottom layer's attention copied to all upper layers.
language/labs/consistent_zero_shot_nmt/utils/model_utils.py:130
Method
__call__
Builds the loss and returns <float32> [] loss tensor.
language/labs/consistent_zero_shot_nmt/models/losses.py:47
Method
__call__
(self, input_data)
language/realm/generate_retrieval_corpus.py:61
Method
__call__
(self, word)
language/gscan/xattn_model/dataset/preprocess.py:88
Method
__call__
Construct the Transformer embeddings. Args: input_ids: <int32>[batch_size, len]. pos_ids: <int32>[batch_size, len]. token_type_
language/gscan/xattn_model/model/layers.py:58
Method
__call__
Apply attention block including attention and an output layer. Args: inputs_q: <float32>[batch_size, len, dim]. inputs_kv: <float32>[
language/gscan/xattn_model/model/layers.py:135
Method
__call__
Apply attention block including cross-attention and an output layer. Args: input1: <float32>[batch_size, len, dim]. input2: <float32>
language/gscan/xattn_model/model/layers.py:194
Method
__call__
Apply the Transformer intermediate layer. Args: hidden: <float32>[batch_size, len, dim]. Returns: output: <float32>[batch_size,
language/gscan/xattn_model/model/layers.py:252
Method
__call__
Apply the Transformer output layer. Args: hidden: <float32>[batch_size, len, dim]. x: <float32>[batch_size, len, dim]. Returns:
language/gscan/xattn_model/model/layers.py:291
Method
__call__
Apply the Transformer bi-output layer. Args: hidden1: <float32>[batch_size, len, dim]. x1: <float32>[batch_size, len, dim]. hid
language/gscan/xattn_model/model/layers.py:337
Method
__call__
Apply the Transformer pooler. Args: x: <float32>[batch_size, len, dim]. mask: <int32>[batch_size, len]. Returns: pooled: <
language/gscan/xattn_model/model/layers.py:382
Method
__call__
Apply the Transformer layer. Args: inputs_q: <float32>[batch_size, len, dim]. inputs_kv: <float32>[batch_size, len, dim]. mask:
language/gscan/xattn_model/model/layers.py:443
Method
__call__
Apply the Transformer connection layer. Args: input1: <float32>[batch_size, len, dim] input2: <float32>[batch_size, len, dim] m
language/gscan/xattn_model/model/layers.py:516
Method
__call__
Apply the transfomer encoder-decoder layer. Args: x: <float32>[batch_size, len, dim]. encoded: <float32>[batch_size, len, dim].
language/gscan/xattn_model/model/layers.py:598
Method
__call__
Apply the two-layer MLP. Args: x: <float32>[batch_size, ..., dim]. Returns: output: <float32>[batch_size, ..., output_size].
language/gscan/xattn_model/model/layers.py:664
Method
__call__
(self, x)
language/gscan/xattn_model/model/models.py:68
Method
__call__
(self, batch)
language/gscan/xattn_model/model/models.py:132
Method
__call__
(self, batch)
language/gscan/xattn_model/model/models.py:168
Method
__call__
(self, x, encoded, pos_ids=None, token_type_ids=No
language/gscan/xattn_model/model/models.py:214
Method
__call__
(self, batch)
language/gscan/xattn_model/model/models.py:302
Method
__call__
Perform attention update over entity embedding table. Args: encoded_input: [batch_size, n_tokens, hidden_size]. mention_batch_positio
language/mentionmemory/modules/entity_attention_layer.py:54
Method
__call__
Incorporates retrieval into input. Args: encoded_input: [batch_size, n_tokens, hidden_size] input representation. retrieval_values: [
language/mentionmemory/modules/retrieval_update_layers.py:38
Method
__call__
( self, encoded_input: Array, retrieval_values: Array, retrieval_scores: Array,
language/mentionmemory/modules/retrieval_update_layers.py:86
Method
__call__
( self, encoded_input: Array, retrieval_values: Array, retrieval_scores: Array,
language/mentionmemory/modules/retrieval_update_layers.py:118
Method
__call__
( self, encoded_input: Array, retrieval_values: Array, retrieval_scores: Array,
language/mentionmemory/modules/retrieval_update_layers.py:201
Method
__call__
Perform masked language modeling scoring. Args: encoded_input: [bsz, n_tokens, hidden_size]. mlm_target_positions: [bsz, max_mlm_targ
language/mentionmemory/modules/mlm_layer.py:53
Method
__call__
Perform attention update over memory table. Args: queries: [n_mentions, hidden_size] query vectors. memory_keys: [rows, values per ro
language/mentionmemory/modules/memory_retrieval_layer.py:65
Method
__call__
Perform attention update over memory table. Args: encoded_input: [batch_size, n_tokens, hidden_size] input representation. mention_ba
language/mentionmemory/modules/memory_attention_layer.py:97
Method
__call__
Perform attention update over memory table. Args: encoding: [batch_size, n_tokens, hidden_size] input representation. mention_batch_p
language/mentionmemory/modules/batch_memory_attention_layer.py:82
Method
__call__
Perform approximate top-k similarity search over vector table. Args: queries: [n_queries, vector_dim]. table: [rows, values per row,
language/mentionmemory/modules/topk_similarity_layer.py:51
Method
__call__
Self attention layer forward. Args: encoding: [bsz, seq_len, model_dim] model state. attention_mask: [bsz, seq_len]. determinis
language/mentionmemory/modules/attention.py:56
Method
__call__
Applies MLP block update. Args: x: [..., input_dim]. deterministic: don't apply dropout if true. Returns: Updated array x
language/mentionmemory/modules/mlp.py:54
Method
__call__
Transformer layer forward. Args: encoding: [bsz, seq_len, model_dim] model state. attention_mask: [bsz, seq_len]. deterministic
language/mentionmemory/modules/transformer.py:71
Method
__call__
(self, encoding: Array, *args, **kwargs)
language/mentionmemory/modules/transformer.py:114
Method
__call__
Transformer layer forward. Args: encoding: [bsz, seq_len, model_dim] model state. attention_mask: [bsz, seq_len]. deterministic
language/mentionmemory/modules/transformer.py:164
Method
__call__
Perform approximate top-k similarity search over vector table. Args: queries: [n_queries, vector_dim]. table: [n_clusters, rows, valu
language/mentionmemory/modules/sparse_topk_similarity_layer.py:53
Method
__call__
. Args: encoding: [batch_size, n_tokens, hidden_size]. mention_batch_positions: [n_mentions]. mention_start_positions: [n_menti
language/mentionmemory/modules/memory_extraction_layer.py:56
Method
__call__
Embeds the inputs along the last dimension. Args: embedding_input: [..., embedding_dim] integer array of embedding indices. Returns:
language/mentionmemory/modules/embedding.py:47
Method
__call__
(self, inputs: Dict[str, Array])
language/mentionmemory/modules/embedding.py:67
Method
__call__
Produces memory texts and mention positions given memory IDs.
language/mentionmemory/encoders/mention_memory_encoder.py:84
Method
__call__
(self, batch: Dict[str, Array], deterministic: bool)
language/mentionmemory/tasks/mention_classifier_task.py:54
Method
__call__
( self, batch: Dict[str, Array], deterministic: bool)
language/mentionmemory/tasks/mention_memory_task.py:55
Method
__call__
( self, batch: Dict[str, Array], deterministic: bool)
language/mentionmemory/tasks/eae_task.py:51
Method
__call__
( self, batch: Dict[str, Array], deterministic: bool)
language/mentionmemory/tasks/embedding_based_entity_qa_task.py:39
Method
__call__
(self, batch, deterministic)
language/mentionmemory/tasks/readtwice_task.py:53
Method
__call__
( self, batch: Dict[str, Array], deterministic: bool)
language/mentionmemory/tasks/mauto_task.py:56
Method
__call__
(self, batch, deterministic)
language/mentionmemory/tasks/example_task.py:39
Method
__call__
(self, batch: Dict[str, Array], deterministic: bool)
language/mentionmemory/tasks/relation_classifier_task.py:83
Method
__call__
(self, batch: Dict[str, Array], deterministic: bool)
language/mentionmemory/tasks/text_classifier.py:65
Method
__call__
( self, batch: Dict[str, Array], deterministic: bool)
language/mentionmemory/tasks/memory_generation_task.py:52
Method
__call__
( self, batch: Dict[str, Array], deterministic: bool)
language/mentionmemory/tasks/mention_based_entity_qa_task.py:76
Method
__call__
( self, article_pair, )
language/fruit/rendering_utils.py:668
Method
__class__
(self)
language/gscan/data/grammar.py:41
Method
__class__
(self)
language/gscan/data/grammar.py:57
Method
__class__
(self)
language/gscan/data/grammar.py:302
Method
__contains__
(self, claim_id)
language/serene/scrape_db.py:95
Method
__contains__
(self, token)
language/capwap/utils/text_utils.py:84
Method
__enter__
(self)
language/realm/parallel.py:156
Method
__enter__
(self)
language/realm/profile.py:50
Method
__exit__
(self, exc_type, exc_val, exc_tb)
language/realm/parallel.py:159
Method
__exit__
(self, exc_type, exc_val, exc_tb)
language/realm/profile.py:54
Method
__getattr__
Allow x.rel_name() to be used instead of x.follow('rel_name'). Args: rel_name: a string naming a declared KG relation Returns: F
language/nql/nql/__init__.py:314
Method
__getitem__
Get the proto for the wikipedia page by url. Args: claim_id: The fever claim_id to get evidence for Returns: Returns proto of pa
language/serene/scrape_db.py:76
Method
__getitem__
(self, key)
language/search_agents/muzero/utils.py:270
Method
__init__
(self, vocab_file, max_size, dataset)
language/labs/exemplar_decoding/utils/data.py:351
Method
__init__
(self, units, mem_input, hps, use_beam=False,
language/labs/exemplar_decoding/models/linear.py:157
Method
__init__
Create a cell with output projection. Args: cell: an RNNCell, a projection to output_size is added to it. num_layers: number of MLP l
language/labs/exemplar_decoding/models/output_wrapper.py:31
Method
__init__
Construct base AttentionMechanism class. Args: query_layer: Callable. Instance of `tf.layers.Layer`. The layer's depth must match
language/labs/exemplar_decoding/models/attention.py:110
Method
__init__
Construct the AttentionMechanism mechanism. Args: num_units: The depth of the attention mechanism. mem_input: mem_input. hps: h
language/labs/exemplar_decoding/models/attention.py:403
Method
__init__
Initialize the HyperLSTM cell. Args: num_units: int, The number of units in the LSTM cell. mem_input: mem_input. use_peepholes:
language/labs/exemplar_decoding/models/hyperlstm.py:36
Method
__init__
Constructor for BertModel which adds an option to reuse variables. Args: config: `BertConfig` instance. is_training: bool. true for t
language/labs/drkit/run_dualencoder_qa.py:213
Method
__init__
(self, doc_layers_to_use, doc_aggregation_fn, qry_layers_to_use, qry_aggregation_fn, projection
language/labs/drkit/run_dualencoder_qa.py:324
Method
__init__
(self, qas_id, question_text, doc_tokens, orig_ans
language/labs/drkit/run_dualencoder_qa.py:1155
Method
__init__
(self, unique_id, qas_id, example_index, doc_span_
language/labs/drkit/run_dualencoder_qa.py:1192
Method
__init__
(self, qas_id, question_text, subject_entity, rela
language/labs/drkit/input_fns.py:30
Method
__init__
(self, qas_id, qry_tokens, qry_input_ids, qry_inpu
language/labs/drkit/input_fns.py:66
Method
__init__
(self, filename, is_training, has_bridge)
language/labs/drkit/input_fns.py:94
Method
__init__
Initialize dataset.
language/labs/drkit/input_fns.py:267
Method
__init__
Initialize dataset.
language/labs/drkit/input_fns.py:346
Method
__init__
Initialize dataset.
language/labs/drkit/input_fns.py:437
Method
__init__
Initialize dataset.
language/labs/drkit/input_fns.py:531
Method
__init__
Initialize dataset.
language/labs/drkit/input_fns.py:636
Method
__init__
(self, estimator, input_fn)
language/labs/drkit/bert_utils.py:87
← previous
next →
2,801–2,900 of 6,154, ranked by callers