MCPcopy Create free account

hub / github.com/huggingface/sentence-transformers / functions

Functions4,575 in github.com/huggingface/sentence-transformers

↓ 2 callersMethodapply_chat_template
(self, messages, tokenize=False, chat_template=None, **kwargs)
tests/base/test_modality.py:1577
↓ 2 callersFunctionargs_class
(request)
tests/base/test_training_args.py:26
↓ 2 callersFunctionb
(e)
docs/_static/js/custom.js:28
↓ 2 callersFunctionbackend_warn_to_save
Warns the user to save the model if they just exported it. Args: model_name_or_path: The model name or path is_local: Whethe
sentence_transformers/backend/utils.py:185
↓ 2 callersFunctionbuild
(**kwargs)
tests/multi_vector_encoder/evaluation/test_information_retrieval.py:198
↓ 2 callersMethodcalculate_loss
Generalized function to calculate the cross-entropy loss without caching gradients.
sentence_transformers/sentence_transformer/losses/cached_gist_embed.py:252
↓ 2 callersMethodcalculate_loss
Calculate the all-pairs InfoNCE loss, chunked over ``mini_batch_size`` rows of the score matrix. The labels are unused: the positive pair is t
sentence_transformers/sentence_transformer/losses/cached_multiple_negatives_ranking.py:244
↓ 2 callersMethodcalculate_loss
( self, reps: list[list[Tensor]], masks_chunks: list[list[Tensor]], *,
sentence_transformers/multi_vector_encoder/losses/cached_multiple_negatives_ranking.py:215
↓ 2 callersFunctioncat_padded_token_embeddings
Concatenate ``(B_i, T_i, D)`` token-embedding chunks and ``(B_i, T_i)`` mask chunks along ``dim=0``, padding each chunk up to the chunk-wide max t
sentence_transformers/util/tensor.py:300
↓ 2 callersFunctioncheck_package_availability
Checks if a package is available from the correct owner.
sentence_transformers/util/environment.py:92
↓ 2 callersFunctioncheck_teacher_targets
Inspect a distillation loss's teacher target once, on its first forward. Takes the target the loss itself computed rather than recomputing a soft
sentence_transformers/util/misc.py:89
↓ 2 callersFunctionclear_warning_once_cache
transformers caches ``warning_once`` globally, so without this a warning emitted by one test silences it in every later test. Request the fixture
tests/conftest.py:38
↓ 2 callersFunctioncolbert_scores_pairwise
Pairwise ColBERT (MaxSim) scoring for matched ``(query_i, document_i)`` pairs. Takes ``(batch_size, q_tokens, dim)`` query embeddings and ``(batc
sentence_transformers/multi_vector_encoder/scoring/colbert.py:105
↓ 2 callersMethodcompute_loss_from_embeddings
Compute the Triplet loss from embeddings. Args: embeddings: List of embeddings labels: Unused, the triplet s
sentence_transformers/sentence_transformer/losses/triplet.py:88
↓ 2 callersMethodcompute_loss_from_embeddings
(self, embeddings: list[Tensor], labels: Tensor)
sentence_transformers/sentence_transformer/losses/batch_hard_soft_margin_triplet.py:97
↓ 2 callersMethodcompute_loss_from_embeddings
(self, embeddings: torch.Tensor)
sentence_transformers/sparse_encoder/losses/flops.py:47
↓ 2 callersMethodcompute_metrics
(self, model: SentenceTransformer)
sentence_transformers/sentence_transformer/evaluation/binary_classification.py:224
↓ 2 callersMethodcompute_metrics
Compute MRR, NDCG, and AP using all known positives for normalization.
sentence_transformers/sparse_encoder/evaluation/reciprocal_rank_fusion.py:315
↓ 2 callersFunctionconvert
(embeddings: np.ndarray, input_format: str)
tests/util/test_retrieval.py:51
↓ 2 callersMethodconvert
(tensor: Tensor)
sentence_transformers/base/model.py:1727
↓ 2 callersFunctioncreate_markov_matrix
(weights_matrix)
examples/sentence_transformer/applications/text-summarization/lex_rank.py:74
↓ 2 callersMethoddelete
(text, del_ratio=0.6)
sentence_transformers/sentence_transformer/datasets/denoising_auto_encoder.py:49
↓ 2 callersMethodembed_inputs
( self, model: SentenceTransformer, sentences: SingleInput | Sequence[SingleInput] | n
sentence_transformers/sentence_transformer/evaluation/reranking.py:362
↓ 2 callersMethodembed_inputs
( self, model: SentenceTransformer, sentences: str | list[str] | np.ndarray, *
sentence_transformers/sentence_transformer/evaluation/mse.py:143
↓ 2 callersMethodembed_inputs
( self, model: SentenceTransformer, sentences: str | list[str] | np.ndarray, *
sentence_transformers/sentence_transformer/evaluation/mse_from_dataframe.py:121
↓ 2 callersMethodembed_minibatch
Embed a mini-batch of inputs.
sentence_transformers/base/losses/gradcache.py:439
↓ 2 callersMethodembed_minibatch_iter
Do a forward pass on every mini-batch of the input features and yield the embeddings.
sentence_transformers/base/losses/gradcache.py:458
↓ 2 callersMethodencode_pre_act
Args: x: input data (shape: [batch, input_dim]) Returns: autoencoder latents before activation (shape: [batc
sentence_transformers/sparse_encoder/modules/sparse_auto_encoder.py:94
↓ 2 callersFunctioneuclidean_sim
Computes the euclidean similarity (i.e., negative distance) between two tensors. Handles sparse tensors without converting to dense when poss
sentence_transformers/util/similarity.py:181
↓ 2 callersFunctionevaluate_stsb_test
( model: SparseEncoder, expected_score: float, test_dataset: Dataset, num_test_samples: int = -1 )
tests/sparse_encoder/test_train_stsb.py:20
↓ 2 callersFunctionexport_dynamic_quantized_onnx_model
Export a quantized ONNX model from a SentenceTransformer, SparseEncoder, CrossEncoder, or MultiVectorEncoder model. This function applie
sentence_transformers/backend/quantize.py:24
↓ 2 callersFunctionexport_static_quantized_openvino_model
Export a quantized OpenVINO model from a SentenceTransformer, SparseEncoder, CrossEncoder, or MultiVectorEncoder model. This function ap
sentence_transformers/backend/quantize.py:109
↓ 2 callersMethodextra_repr
(self)
sentence_transformers/base/modules/router.py:513
↓ 2 callersFunctionf
(e,t,o)
docs/_static/js/custom.js:28
↓ 2 callersFunctionformat_log
(value: float | int | str)
sentence_transformers/base/model_card.py:314
↓ 2 callersMethodforward
Route ``features`` through the resolved sub-module pipeline. Resolves the route from ``task`` and ``modality`` (falling back to values stored
sentence_transformers/base/modules/router.py:460
↓ 2 callersMethodfrom_distillation
r""" Creates a StaticEmbedding instance from a distillation process using the `model2vec` package. Args: model_name (str)
sentence_transformers/sentence_transformer/modules/static_embedding.py:191
↓ 2 callersFunctionfrom_pretrained_decorator
(method)
tests/sentence_transformer/test_backends.py:183
↓ 2 callersFunctionfrom_pretrained_decorator
(method)
tests/cross_encoder/test_backends.py:191
↓ 2 callersFunctionfrom_pretrained_decorator
(method)
tests/sparse_encoder/test_backends.py:186
↓ 2 callersMethodget_anchor_negative_triplet_mask
Return a 2D mask where mask[a, n] is True iff a and n have distinct labels. Args: labels: tf.int32 `Tensor` with shape [batch_size
sentence_transformers/sentence_transformer/losses/batch_hard_triplet.py:260
↓ 2 callersMethodget_config_dict
(self)
sentence_transformers/multi_vector_encoder/losses/token_pooling.py:253
↓ 2 callersMethodget_config_dict
(self)
sentence_transformers/multi_vector_encoder/evaluation/distillation.py:274
↓ 2 callersMethodget_config_dict
(self)
sentence_transformers/cross_encoder/losses/cached_multiple_negatives_ranking.py:284
↓ 2 callersMethodget_config_dict
Returns a dictionary of the configuration parameters of the module. These parameters are used to save the module's configuration whe
sentence_transformers/base/modules/module.py:113
↓ 2 callersMethodget_device_name
Choose a default device for model loading and multi-process pools.
sentence_transformers/base/model.py:1843
↓ 2 callersMethodget_embedding_dimension
(self)
sentence_transformers/sentence_transformer/modules/pooling.py:340
↓ 2 callersMethodget_frames_at
(self, indices: list[int])
tests/base/test_model_card.py:1101
↓ 2 callersMethodget_in_batch_negatives
( self, anchors: list[str], candidates: list[list[str]] )
sentence_transformers/cross_encoder/losses/multiple_negatives_ranking.py:128
↓ 2 callersMethodget_labels
()
sentence_transformers/sentence_transformer/readers/nli_data.py:50
↓ 2 callersFunctionget_sample_values
(index: int)
tests/base/samplers/test_no_duplicates_batch_sampler.py:84
↓ 2 callersFunctionget_world_size
The number of processes in the distributed group, or ``1`` when not distributed.
sentence_transformers/util/distributed.py:31
↓ 2 callersFunctionh
(e,t,o)
docs/_static/js/custom.js:28
↓ 2 callersFunctionhas_overlap
(sample_values: set[str], batch_values: set[str])
tests/base/samplers/test_no_duplicates_batch_sampler.py:87
↓ 2 callersMethodinfer_modalities_edge_cases
Return a ``(modality_config, module_output_name)`` for model types that cannot be handled by the general :meth:`infer_modalities` inference pa
sentence_transformers/base/modules/transformer.py:2517
↓ 2 callersMethodinfer_modalities_from_processor
Determine which modalities the processor supports by inspecting its attributes or type.
sentence_transformers/base/modules/transformer.py:2556
↓ 2 callersFunctionis_accelerate_available
Returns True if the Huggingface accelerate library is available.
sentence_transformers/util/environment.py:104
↓ 2 callersMethodis_singular_input
Check if the input is a single example rather than a batch. Redeclared with :class:`~typing_extensions.TypeIs` so type checkers narrow the bra
sentence_transformers/sentence_transformer/model.py:571
↓ 2 callersMethodis_text_pair
Whether a single input is a text pair, i.e. cross-encoder ``(query, document)`` input.
sentence_transformers/base/modality.py:612
↓ 2 callersFunctionkNN
(x, y, k, use_ann_search=False, ann_num_clusters=32768, ann_num_cluster_probe=3)
examples/sentence_transformer/applications/parallel-sentence-mining/bitext_mining_utils.py:31
↓ 2 callersMethodload
Load this module from a model checkpoint. The checkpoint can be either a local directory or a model id on Hugging Face. Args:
sentence_transformers/base/modules/module.py:127
↓ 2 callersFunctionloss_and_grads
(**kwargs)
tests/sentence_transformer/losses/test_mega_batch_margin.py:255
↓ 2 callersFunctionloss_and_grads
(inner: torch.nn.Module)
tests/sentence_transformer/losses/test_cmnrl.py:495
↓ 2 callersFunctionmake_model
Build a tiny model that adds a few weight tensors to its input. The weights are held as initializers of the graph, as values of Constant node att
tests/backend/test_utils.py:27
↓ 2 callersFunctionmanhattan_sim
Computes the manhattan similarity (i.e., negative distance) between two tensors. Handles sparse tensors without converting to dense when poss
sentence_transformers/util/similarity.py:135
↓ 2 callersFunctionmean_maxsim
Computes the MeanMaxSim score between two collections of multi-vector embeddings: :func:`maxsim` divided by each query's real token count, wi
sentence_transformers/util/similarity.py:660
↓ 2 callersFunctionmean_maxsim_pairwise
Computes the pairwise MeanMaxSim score for each query-document pair: :func:`maxsim_pairwise` divided by each query's real token count, with i
sentence_transformers/util/similarity.py:692
↓ 2 callersFunctionmodel
()
tests/test_merging.py:25
↓ 2 callersMethodold_fit
Deprecated training method from before Sentence Transformers v4.0, it is recommended to use :class:`~sentence_transformers.cross_enco
sentence_transformers/cross_encoder/fit_mixin.py:416
↓ 2 callersFunctionp
(e)
docs/_static/js/custom.js:28
↓ 2 callersMethodpairwise
Score aligned query and document pairs, as in ``util.maxsim_pairwise``.
sentence_transformers/multi_vector_encoder/scoring/colbert.py:378
↓ 2 callersMethodpredict_minibatch_iter
Do forward pass on all the minibatches of the input features and yield corresponding embeddings.
sentence_transformers/cross_encoder/losses/cached_multiple_negatives_ranking.py:168
↓ 2 callersMethodpreprocess_dataset
Preprocess the dataset by optionally lazily adding a dataset name column, required for multi-dataset training with multiple losses or
sentence_transformers/base/trainer.py:1098
↓ 2 callersMethodprocessor
Property to get the processor that is used by this model
sentence_transformers/base/model.py:2132
↓ 2 callersFunctionpush_to_hub
Run save_or_push_to_hub_model with push_to_hub=True and return the files it hands to the Hub. The folder is uploaded whole and then discarded, so
tests/backend/test_utils.py:123
↓ 2 callersFunctionraise_unsupported_modality_error
Raise a clear ``ValueError`` explaining why ``modality`` is not supported. Shared by :meth:`BaseModel.preprocess` and :meth:`Transformer.preproce
sentence_transformers/base/modality.py:866
↓ 2 callersMethodregister_model
(self, model: BaseModel)
sentence_transformers/base/model_card.py:1326
↓ 2 callersMethodrun
(content, max_length, truncation=True)
tests/base/modules/test_transformer.py:1359
↓ 2 callersMethodrun_usage_snippet
(self)
sentence_transformers/base/model_card.py:1661
↓ 2 callersMethodsave
(self, output_path: str, *args, safe_serialization: bool = True, **kwargs)
sentence_transformers/cross_encoder/modules/logit_score.py:56
↓ 2 callersMethodsave
(self, output_path: str, safe_serialization: bool = True, **kwargs)
sentence_transformers/base/modules/router.py:553
↓ 2 callersFunctionsave_with_model_config
Save a model, then update its config_sentence_transformers.json, e.g. to add requirements.
tests/base/test_model.py:1972
↓ 2 callersMethodscale
(self)
sentence_transformers/sentence_transformer/losses/cached_multiple_negatives_ranking.py:217
↓ 2 callersFunctionscore_candidates
(x, y, candidate_inds, fwd_mean, bwd_mean, margin)
examples/sentence_transformer/applications/parallel-sentence-mining/bitext_mining_utils.py:22
↓ 2 callersMethodset_adapter
Sets a specific adapter by forcing the model to use that adapter and disable the other adapters. Args: *args:
sentence_transformers/base/peft_mixin.py:85
↓ 2 callersMethodset_model_id
(self, model_id: str)
sentence_transformers/base/model_card.py:1357
↓ 2 callersMethodset_vocab
(self, vocab: Iterable[str])
sentence_transformers/sentence_transformer/modules/tokenizer/whitespace.py:28
↓ 2 callersFunctionstatic_embedding
(tokenizer: Tokenizer, embedding_weights)
tests/conftest.py:291
↓ 2 callersMethodstore_metrics_in_model_card_data
( self, model: SparseEncoder, metrics: dict[str, Any], epoch: int = 0,
sentence_transformers/sparse_encoder/evaluation/sparse_mse.py:157
↓ 2 callersFunctionto_scipy_coo
Converts a sparse PyTorch tensor to a SciPy COO sparse matrix. Args: x (torch.Tensor): A 2-dimensional sparse PyTorch tensor in COO
sentence_transformers/util/tensor.py:363
↓ 2 callersMethodtokenize
(self, text: str, **kwargs)
sentence_transformers/sentence_transformer/modules/tokenizer/word.py:408
↓ 2 callersMethodtokenize
.. deprecated:: `tokenize` is deprecated. Use `preprocess` instead.
sentence_transformers/base/model.py:718
↓ 2 callersMethodtokenize
(self, texts, **kwargs)
tests/base/modules/test_input_module.py:16
↓ 2 callersFunctionu
(e,t,o)
docs/_static/js/custom.js:28
↓ 2 callersMethodvalidate_datasets
Validate (i.e. check if the dataset IDs exist on the Hub) and process a list of dataset dictionaries. Args: dataset_list
sentence_transformers/base/model_card.py:437
↓ 2 callersFunctionx
(e,t,o)
docs/_static/js/custom.js:28
↓ 1 callersFunctionC
(e)
docs/_static/js/custom.js:28
↓ 1 callersMethodLN
(self, x: torch.Tensor, eps: float = 1e-5)
sentence_transformers/sparse_encoder/modules/sparse_auto_encoder.py:106
↓ 1 callersMethod__init__
(self, fn)
sentence_transformers/sentence_transformer/losses/adaptive_layer.py:69
↓ 1 callersMethod__init__
(self, model: SentenceTransformer)
sentence_transformers/sentence_transformer/losses/contrastive_tension.py:129
← previousnext →501–600 of 4,575, ranked by callers