Functions — github.com/cedrickchee/transformers-llama

hub / github.com/cedrickchee/transformers-llama / functions

Functions36,568 in github.com/cedrickchee/transformers-llama

⨍ Functions 36,568◇ Types & classes 9,398↳ Endpoints 539

↓ 17 callersMethod_get_uniform_logits

(self, batch_size: int, length: int)

tests/generation/test_tf_logits_process.py:50

↓ 17 callersMethod_prepare_for_class

(self, inputs_dict, model_class, return_labels=False)

tests/models/bert/test_modeling_bert.py:463

↓ 17 callersMethod_prepare_for_class

(self, inputs_dict, model_class, return_labels=False)

tests/models/vilt/test_modeling_vilt.py:244

↓ 17 callersMethod_shape

(self, tensor, seq_len, bsz)

src/transformers/models/fsmt/modeling_fsmt.py:882

↓ 17 callersMethodcheck_model_type

Check if the model class is in supported by the pipeline. Args: supported_models (`List[str]` or `dict`):

src/transformers/pipelines/base.py:932

↓ 17 callersFunctiongelu

This is the gelu implementation from the original ESM repo. Using F.gelu yields subtly wrong results.

src/transformers/models/esm/modeling_esm.py:63

↓ 17 callersMethodget_config_dict

(self, stage)

tests/deepspeed/test_deepspeed.py:256

↓ 17 callersMethodget_lr

(self)

src/transformers/optimization.py:729

↓ 17 callersFunctionget_scheduler

Unified API to get any scheduler from its name. Args: name (`str` or `SchedulerType`): The name of the scheduler to use.

src/transformers/optimization.py:315

↓ 17 callersFunctionis_tensorboard_available

()

src/transformers/integrations.py:86

↓ 17 callersFunctionpermute_final_dims

(tensor: torch.Tensor, inds: List[int])

src/transformers/models/esm/modeling_esmfold.py:221

↓ 17 callersMethodprepare_config_and_inputs

(self)

tests/models/bert/test_modeling_bert.py:96

↓ 17 callersMethodprepare_config_and_inputs

(self)

tests/models/bert/test_modeling_tf_bert.py:95

↓ 17 callersMethodquestion_encoder

(self)

src/transformers/models/rag/modeling_rag.py:1204

↓ 17 callersFunctionrgb_to_id

Converts RGB color to unique ID.

src/transformers/image_transforms.py:566

↓ 17 callersMethodset_defaults

(cls, prefix, defaults)

src/transformers/utils/hp_naming.py:25

↓ 17 callersMethodset_seed

(self)

tests/models/xlnet/test_modeling_xlnet.py:170

↓ 17 callersMethodstart

start tracking for the caller's stage

src/transformers/trainer_utils.py:453

↓ 17 callersMethodword_to_chars

Get the character span in the original string corresponding to given word in a sequence of the batch. Character spans are returned a

src/transformers/tokenization_utils_base.py:578

↓ 16 callersMethod__init__

(self, config)

src/transformers/models/markuplm/modeling_markuplm.py:523

↓ 16 callersMethod__init__

( self, config: GroupViTVisionConfig, hidden_size: Optional[int] = None, inter

src/transformers/models/groupvit/modeling_groupvit.py:571

↓ 16 callersMethod__init__

(self, config: DeiTConfig, **kwargs)

src/transformers/models/deit/modeling_tf_deit.py:374

↓ 16 callersMethod__init__

(self, config: HubertConfig)

src/transformers/models/hubert/modeling_hubert.py:956

↓ 16 callersMethod__init__

(self, config)

src/transformers/models/esm/modeling_esm.py:480

↓ 16 callersMethod__init__

(self, tensors, mask: Optional[Tensor])

src/transformers/models/table_transformer/modeling_table_transformer.py:1909

↓ 16 callersMethod__init__

The Mask2Former Loss. The loss is computed very similar to DETR. The process happens in two steps: 1) we compute hungarian assignment

src/transformers/models/mask2former/modeling_mask2former.py:503

↓ 16 callersMethod__init__

(self, config, layer_id=0)

src/transformers/models/longformer/modeling_longformer.py:1231

↓ 16 callersMethod__init__

(self, config, **kwargs)

src/transformers/models/mpnet/modeling_tf_mpnet.py:333

↓ 16 callersMethod__init__

(self, config)

src/transformers/models/ibert/modeling_ibert.py:497

↓ 16 callersMethod__init__

(self, matcher, num_classes, eos_coef, losses)

src/transformers/models/yolos/modeling_yolos.py:940

↓ 16 callersMethod__init__

(self, config: FunnelConfig, block_index: int)

src/transformers/models/funnel/modeling_funnel.py:615

↓ 16 callersMethod__init__

(self, config)

src/transformers/models/layoutlm/modeling_layoutlm.py:362

↓ 16 callersMethodapply_gradients

(self, features, labels, nb_instances_in_global_batch)

src/transformers/trainer_tf.py:651

↓ 16 callersMethodconvert

(k, v)

examples/research_projects/seq2seq-distillation/_test_seq2seq_examples_multi_gpu.py:130

↓ 16 callersFunctionconvert_pytorch_state_dict_to_flax

(pt_state_dict, flax_model)

src/transformers/modeling_flax_pytorch_utils.py:126

↓ 16 callersMethodfill_with_past_key_values_

(self, inputs_or_outputs: Mapping[str, Mapping[int, str]], direction: str)

src/transformers/models/blenderbot/configuration_blenderbot.py:382

↓ 16 callersFunctionfind_labels

Find the labels used by a given model. Args: model_class (`type`): The class of the model.

src/transformers/utils/generic.py:385

↓ 16 callersMethodget_env

Return a copy of the `os.environ` object that sets up `PYTHONPATH` correctly, depending on the test suite it's invoked from. This is

src/transformers/testing_utils.py:1173

↓ 16 callersFunctionget_gpu_count

Return the number of available gpus (regardless of whether torch, tf or jax is used)

src/transformers/testing_utils.py:751

↓ 16 callersFunctionget_logger

Return a logger with the specified name. This function is not supposed to be directly accessed unless you are writing a custom transformers

src/transformers/utils/logging.py:114

↓ 16 callersFunctionget_user_field

A utility function that asks a question to the user to get an answer, potentially looping until it gets a valid answer. Args: qu

src/transformers/commands/add_new_model_like.py:1403

↓ 16 callersFunctionmeshgrid

Wrapper around torch.meshgrid to avoid warning messages about the introduced `indexing` argument. Reference: https://pytorch.org/docs/1.13/g

src/transformers/pytorch_utils.py:277

↓ 16 callersMethodpad

Pad an image to make the height and width divisible by `size`. Args: image (`np.ndarray`): Image to pad.

src/transformers/models/swin2sr/image_processing_swin2sr.py:82

↓ 16 callersMethodprepare_config_and_inputs

(self)

tests/models/reformer/test_modeling_reformer.py:141

↓ 16 callersMethodprepare_config_and_inputs

(self)

tests/models/prophetnet/test_modeling_prophetnet.py:111

↓ 16 callersMethodprepare_config_and_inputs_for_common

(self)

tests/models/speecht5/test_modeling_speecht5.py:137

↓ 16 callersMethodregister_for_auto_class

Register this class with a given auto class. This should only be used for custom models as the ones in the library are already mapped

src/transformers/modeling_utils.py:3094

↓ 15 callersMethod__init__

(self, dim: int, keepdim: bool = False, minimum_scale: float = 1e-5)

src/transformers/models/time_series_transformer/modeling_time_series_transformer.py:273

↓ 15 callersMethod__init__

(self, config)

src/transformers/models/blip_2/modeling_blip_2.py:210

↓ 15 callersMethod__init__

(self, config: BigBirdPegasusConfig)

src/transformers/models/bigbird_pegasus/modeling_bigbird_pegasus.py:2342

↓ 15 callersMethod__init__

(self, config)

src/transformers/models/mpnet/modeling_mpnet.py:283

↓ 15 callersMethod__init__

(self, config, hidden_size, num_attention_heads, drop_path, sequence_reduction_ratio, mlp_ratio)

src/transformers/models/glpn/modeling_glpn.py:294

↓ 15 callersMethod__init__

( self, config: CvtConfig, num_heads: int, embed_dim: int, kernel_size

src/transformers/models/cvt/modeling_tf_cvt.py:456

↓ 15 callersMethod__init__

(self, config: DetrConfig)

src/transformers/models/maskformer/modeling_maskformer.py:676

↓ 15 callersMethod__init__

(self, config, dim, input_resolution, num_heads, shift_size=0)

src/transformers/models/swin/modeling_swin.py:586

↓ 15 callersMethod__init__

(self, config, has_relative_attention_bias=False)

src/transformers/models/longt5/modeling_longt5.py:1151

↓ 15 callersMethod_check_zero_mean_unit_variance

(self, input_vector)

tests/models/speecht5/test_feature_extraction_speecht5.py:157

↓ 15 callersMethod_check_zero_mean_unit_variance

(self, input_vector)

tests/models/speech_to_text/test_feature_extraction_speech_to_text.py:112

↓ 15 callersMethod_check_zero_mean_unit_variance

(self, input_vector)

tests/models/mctct/test_feature_extraction_mctct.py:112

↓ 15 callersMethod_check_zero_mean_unit_variance

(self, input_vector)

tests/models/wav2vec2/test_feature_extraction_wav2vec2.py:104

↓ 15 callersMethod_prepare_for_class

(self, inputs_dict, model_class)

tests/test_modeling_flax_common.py:154

↓ 15 callersMethodchar_to_token

Get the index of the token in the encoded output comprising a character in the original string for a sequence of the batch.

src/transformers/tokenization_utils_base.py:537

↓ 15 callersMethodchar_to_word

Get the word in the original string corresponding to a character in the original string of a sequence of the batch. Can be c

src/transformers/tokenization_utils_base.py:623

↓ 15 callersMethodconvert_tokens_to_string

Converts a sequence of tokens (string) in a single string.

src/transformers/models/t5/tokenization_t5.py:317

↓ 15 callersMethoddecode

This method forwards all its arguments to PreTrainedTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of thi

src/transformers/models/layoutxlm/processing_layoutxlm.py:155

↓ 15 callersMethodfill_match

A utility method that massages the config file and can optionally verify that the values match. 1. Replace "auto" values with `Train

src/transformers/deepspeed.py:93

↓ 15 callersMethodfor_model

(cls, model_type: str, *args, **kwargs)

src/transformers/models/auto/configuration_auto.py:794

↓ 15 callersMethodfrom_pretrained

(cls, pretrained_model_name_or_path: Optional[Union[str, os.PathLike]], *model_args, **kwargs)

src/transformers/models/markuplm/modeling_markuplm.py:734

↓ 15 callersMethodfrom_question_encoder_generator_configs

r""" Instantiate a [`EncoderDecoderConfig`] (or a derived class) from a pre-trained encoder model configuration and decoder model conf

src/transformers/models/rag/configuration_rag.py:171

↓ 15 callersMethodfrom_text_vision_configs

r""" Instantiate a [`CLIPConfig`] (or a derived class) from clip text model configuration and clip vision model configuration.

src/transformers/models/clip/configuration_clip.py:326

↓ 15 callersMethodget_rust_tokenizer

(self, **kwargs)

tests/models/clip/test_processor_clip.py:71

↓ 15 callersMethodmean

Returns the mean of the distribution.

src/transformers/models/time_series_transformer/modeling_time_series_transformer.py:61

↓ 15 callersMethodpost_process_semantic_segmentation

Converts the output of [`DPTForSemanticSegmentation`] into semantic segmentation maps. Only supports PyTorch. Args: outp

src/transformers/models/dpt/image_processing_dpt.py:342

↓ 15 callersMethodprepare_config_and_inputs

(self)

tests/models/unispeech_sat/test_modeling_unispeech_sat.py:124

↓ 15 callersMethodprepare_config_and_inputs_for_common

(self)

tests/models/blip/test_modeling_blip.py:382

↓ 15 callersFunctionsqueeze

Framework-agnostic version of `numpy.squeeze` that will work on torch/TensorFlow/Jax tensors as well as NumPy arrays.

src/transformers/utils/generic.py:469

↓ 15 callersMethodtokenizer

(self)

tests/models/mbart/test_modeling_tf_mbart.py:250

↓ 14 callersMethod__init__

(self, config: AlbertConfig, **kwargs)

src/transformers/models/albert/modeling_tf_albert.py:319

↓ 14 callersMethod__init__

(self, config, dim, num_heads, drop_path_rate=0.0)

src/transformers/models/nat/modeling_nat.py:438

↓ 14 callersMethod__init__

(self, config)

src/transformers/models/lilt/modeling_lilt.py:428

↓ 14 callersMethod__init__

(self, config, dim, num_heads, dilation, drop_path_rate=0.0)

src/transformers/models/dinat/modeling_dinat.py:451

↓ 14 callersMethod__init__

(self, config, dim, input_resolution, num_heads, shift_size=0, pretrained_window_size=0)

src/transformers/models/swinv2/modeling_swinv2.py:632

↓ 14 callersMethod__init__

(self, config: SegformerConfig, **kwargs)

src/transformers/models/segformer/modeling_tf_segformer.py:714

↓ 14 callersMethod__init__

(self, config: LayoutLMv3Config, **kwargs)

src/transformers/models/layoutlmv3/modeling_tf_layoutlmv3.py:471

↓ 14 callersMethod__init__

( self, num_heads, embed_dim, kernel_size, padding_q, padding_

src/transformers/models/cvt/modeling_cvt.py:386

↓ 14 callersMethod__init__

(self, config: SEWConfig)

src/transformers/models/sew/modeling_sew.py:844

↓ 14 callersMethod__init__

(self, cfg, input_shape: Dict[str, ShapeSpec])

examples/research_projects/visual_bert/modeling_frcnn.py:1501

↓ 14 callersMethod__init__

(self, cfg, input_shape: Dict[str, ShapeSpec])

examples/research_projects/lxmert/modeling_frcnn.py:1501

↓ 14 callersMethodaggregate

(self, pre_entities: List[dict], aggregation_strategy: AggregationStrategy)

src/transformers/pipelines/token_classification.py:340

↓ 14 callersMethoddtype

Returns the dtype of the underlying rotation. Returns: The dtype of the underlying rotation

src/transformers/models/esm/openfold_utils/rigid_utils.py:412

↓ 14 callersMethodfrom_vision_text_pretrained

Params: vision_model_name_or_path (`str`, *optional*, defaults to `None`): Information necessary to initiate the

src/transformers/models/vision_text_dual_encoder/modeling_vision_text_dual_encoder.py:415

↓ 14 callersFunctionget_results

(output_dir)

examples/pytorch/test_pytorch_examples.py:85

↓ 14 callersMethodget_special_tokens_mask

Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding special tokens using the

src/transformers/models/t5/tokenization_t5.py:187

↓ 14 callersMethodget_tokenizer

(self, **kwargs)

tests/models/speecht5/test_processor_speecht5.py:67

↓ 14 callersMethodget_tokenizer

(self, **kwargs)

tests/models/speech_to_text/test_processor_speech_to_text.py:63

↓ 14 callersMethodget_tokenizer

(self, **kwargs_init)

tests/models/mctct/test_processor_mctct.py:63

↓ 14 callersMethodget_trainer

(self, a=0, b=0, train_len=64, eval_len=64, callbacks=None, disable_tqdm=False, **kwargs)

tests/trainer/test_trainer_callback.py:90

↓ 14 callersMethodget_words_and_boxes_batch

(self)

tests/models/layoutlmv2/test_tokenization_layoutlmv2.py:77

↓ 14 callersMethodget_words_and_boxes_batch

(self)

tests/models/layoutxlm/test_tokenization_layoutxlm.py:71

↓ 14 callersMethodhf_compute_loss

(self, labels: tf.Tensor, logits: tf.Tensor)

src/transformers/models/mobilebert/modeling_tf_mobilebert.py:98

← previous next →401–500 of 36,568, ranked by callers