Hugging Face's logo

Transformers documentation

🤗 Transformers 모델을 ONNX로 내보내기

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v5.14.0).
Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

🤗 Transformers 모델을 ONNX로 내보내기

🤗 트랜스포머는 transformers.onnx 패키지를 제공하며, 이 패키지는 설정 객체를 활용하여 모델 체크포인트를 ONNX 그래프로 변환할 수 있게 합니다.

🤗 Transformers에 대한 자세한 내용은 이 가이드를 참조하세요.

ONNX 설정

내보내려는(export) 모델 아키텍처의 유형에 따라 상속받아야 할 세 가지 추상 클래스를 제공합니다:

OnnxConfig

class transformers.onnx.OnnxConfig

< >

( config: PreTrainedConfig task: str = 'default' patching_specs: typing.Optional[list[transformers.onnx.config.PatchingSpec]] = None )

Base class for ONNX exportable model describing metadata on how to export the model through the ONNX format.

flatten_output_collection_property

< >

( name: str field: Iterable ) → (dict[str, Any])

Parameters

  • name — The name of the nested structure
  • field — The structure to, potentially, be flattened

Returns

(dict[str, Any])

Outputs with flattened structure and key mapping this new structure.

Flatten any potential nested structure expanding the name of the field with the index of the element within the structure.

from_model_config

< >

( config: PreTrainedConfig task: str = 'default' )

Parameters

  • config — The model’s configuration to use when exporting to ONNX

Instantiate a OnnxConfig for a specific model

generate_dummy_inputs

< >

( preprocessor: typing.Union[ForwardRef('PreTrainedTokenizerBase'), ForwardRef('FeatureExtractionMixin'), ForwardRef('ImageProcessingMixin')] batch_size: int = -1 seq_length: int = -1 num_choices: int = -1 is_pair: bool = False num_channels: int = 3 image_width: int = 40 image_height: int = 40 sampling_rate: int = 22050 time_duration: float = 5.0 frequency: int = 220 tokenizer: typing.Optional[ForwardRef('PreTrainedTokenizerBase')] = None )

Parameters

  • preprocessor — (PreTrainedTokenizerBase, FeatureExtractionMixin, or ImageProcessingMixin): The preprocessor associated with this model configuration.
  • batch_size (int, optional, defaults to -1) — The batch size to export the model for (-1 means dynamic axis).
  • num_choices (int, optional, defaults to -1) — The number of candidate answers provided for multiple choice task (-1 means dynamic axis).
  • seq_length (int, optional, defaults to -1) — The sequence length to export the model for (-1 means dynamic axis).
  • is_pair (bool, optional, defaults to False) — Indicate if the input is a pair (sentence 1, sentence 2)
  • num_channels (int, optional, defaults to 3) — The number of channels of the generated images.
  • image_width (int, optional, defaults to 40) — The width of the generated images.
  • image_height (int, optional, defaults to 40) — The height of the generated images.
  • sampling_rate (int, optional defaults to 22050) — The sampling rate for audio data generation.
  • time_duration (float, optional defaults to 5.0) — Total seconds of sampling for audio data generation.
  • frequency (int, optional defaults to 220) — The desired natural frequency of generated audio.

Generate inputs to provide to the ONNX exporter

generate_dummy_inputs_onnxruntime

< >

( reference_model_inputs: Mapping ) → Mapping[str, Tensor]

Parameters

  • reference_model_inputs ([Mapping[str, Tensor]) — Reference inputs for the model.

Returns

Mapping[str, Tensor]

The mapping holding the kwargs to provide to the model’s forward function

Generate inputs for ONNX Runtime using the reference model inputs. Override this to run inference with seq2seq models which have the encoder and decoder exported as separate ONNX files.

use_external_data_format

< >

( num_parameters: int )

Parameters

  • num_parameters — Number of parameter on the model

Flag indicating if the model requires using external data format

OnnxConfigWithPast

class transformers.onnx.OnnxConfigWithPast

< >

( config: PreTrainedConfig task: str = 'default' patching_specs: typing.Optional[list[transformers.onnx.config.PatchingSpec]] = None use_past: bool = False )

fill_with_past_key_values_

< >

( inputs_or_outputs: Mapping direction: str inverted_values_shape: bool = False )

Parameters

  • inputs_or_outputs — The mapping to fill.
  • direction — either “inputs” or “outputs”, it specifies whether input_or_outputs is the input mapping or the output mapping, this is important for axes naming.
  • inverted_values_shape — If True, store values on dynamic axis 1, else on axis 2.

Fill the input_or_outputs mapping with past_key_values dynamic axes considering.

with_past

< >

( config: PreTrainedConfig task: str = 'default' )

Parameters

  • config — The underlying model’s config to use when exporting to ONNX

Instantiate a OnnxConfig with use_past attribute set to True

OnnxSeq2SeqConfigWithPast

class transformers.onnx.OnnxSeq2SeqConfigWithPast

< >

( config: PreTrainedConfig task: str = 'default' patching_specs: typing.Optional[list[transformers.onnx.config.PatchingSpec]] = None use_past: bool = False )

ONNX 특징

각 ONNX 설정은 다양한 유형의 토폴로지나 작업에 대해 모델을 내보낼 수 있게(exporting) 해주는 features 세트와 연관되어 있습니다.

Update on GitHub

🤗 Transformers→