Hugging Face's logo

Optimum documentation

Models

You are viewing v1.21.0 version. A newer version v1.25.2 is available.
Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

Models

Natural Language Processing

The following classes are available for the following natural language processing tasks.

OVModelForCausalLM

class optimum.intel.OVModelForCausalLM

< >

( model: Model config: PretrainedConfig = None device: str = 'CPU' dynamic_shapes: bool = True ov_config: typing.Union[typing.Dict[str, str], NoneType] = None model_save_dir: typing.Union[str, pathlib.Path, tempfile.TemporaryDirectory, NoneType] = None quantization_config: typing.Union[optimum.intel.openvino.configuration.OVWeightQuantizationConfig, typing.Dict, NoneType] = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

OpenVINO Model with a causal language modeling head on top (linear layer with weights tied to the input embeddings).

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

forward

< >

( input_ids: LongTensor attention_mask: typing.Optional[torch.LongTensor] = None past_key_values: typing.Optional[typing.Tuple[typing.Tuple[torch.FloatTensor]]] = None position_ids: typing.Optional[torch.LongTensor] = None **kwargs )

generate

< >

( inputs: typing.Optional[torch.Tensor] = None generation_config: typing.Optional[transformers.generation.configuration_utils.GenerationConfig] = None logits_processor: typing.Optional[transformers.generation.logits_process.LogitsProcessorList] = None stopping_criteria: typing.Optional[transformers.generation.stopping_criteria.StoppingCriteriaList] = None prefix_allowed_tokens_fn: typing.Union[typing.Callable[[int, torch.Tensor], typing.List[int]], NoneType] = None synced_gpus: typing.Optional[bool] = None assistant_model: typing.Optional[ForwardRef('PreTrainedModel')] = None streamer: typing.Optional[ForwardRef('BaseStreamer')] = None negative_prompt_ids: typing.Optional[torch.Tensor] = None negative_prompt_attention_mask: typing.Optional[torch.Tensor] = None **kwargs )

OVModelForMaskedLM

class optimum.intel.OVModelForMaskedLM

< >

( model = None config = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

OpenVINO Model with a MaskedLMOutput for masked language modeling tasks.

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

forward

< >

( input_ids: typing.Union[torch.Tensor, numpy.ndarray] attention_mask: typing.Union[torch.Tensor, numpy.ndarray] token_type_ids: typing.Union[torch.Tensor, numpy.ndarray, NoneType] = None **kwargs )

Parameters

  • input_ids (torch.Tensor) — Indices of input sequence tokens in the vocabulary. Indices can be obtained using AutoTokenizer. What are input IDs?
  • attention_mask (torch.Tensor), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:
  • token_type_ids (torch.Tensor, optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

The OVModelForMaskedLM forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of masked language modeling using transformers.pipelines:

>>> from transformers import AutoTokenizer, pipeline
>>> from optimum.intel import OVModelForMaskedLM

>>> tokenizer = AutoTokenizer.from_pretrained("roberta-base")
>>> model = OVModelForMaskedLM.from_pretrained("roberta-base", export=True)
>>> mask_token = tokenizer.mask_token
>>> pipe = pipeline("fill-mask", model=model, tokenizer=tokenizer)
>>> outputs = pipe("The goal of life is" + mask_token)

OVModelForSeq2SeqLM

class optimum.intel.OVModelForSeq2SeqLM

< >

( encoder: Model decoder: Model decoder_with_past: Model = None config: PretrainedConfig = None **kwargs )

Parameters

  • encoder (openvino.runtime.Model) — The OpenVINO Runtime model associated to the encoder.
  • decoder (openvino.runtime.Model) — The OpenVINO Runtime model associated to the decoder.
  • decoder_with_past (openvino.runtime.Model) — The OpenVINO Runtime model associated to the decoder with past key values.
  • config (transformers.PretrainedConfig) — PretrainedConfig is an instance of the configuration associated to the model. Initializing with a config file does not load the weights associated with the model, only the configuration.

Sequence-to-sequence model with a language modeling head for OpenVINO inference.

forward

< >

( input_ids: LongTensor = None attention_mask: typing.Optional[torch.FloatTensor] = None decoder_input_ids: typing.Optional[torch.LongTensor] = None decoder_attention_mask: typing.Optional[torch.LongTensor] = None encoder_outputs: typing.Optional[typing.Tuple[typing.Tuple[torch.Tensor]]] = None past_key_values: typing.Optional[typing.Tuple[typing.Tuple[torch.Tensor]]] = None **kwargs )

Parameters

  • input_ids (torch.LongTensor) — Indices of input sequence tokens in the vocabulary of shape (batch_size, encoder_sequence_length).
  • attention_mask (torch.LongTensor) — Mask to avoid performing attention on padding token indices, of shape (batch_size, encoder_sequence_length). Mask values selected in [0, 1].
  • decoder_input_ids (torch.LongTensor) — Indices of decoder input sequence tokens in the vocabulary of shape (batch_size, decoder_sequence_length).
  • encoder_outputs (torch.FloatTensor) — The encoder last_hidden_state of shape (batch_size, encoder_sequence_length, hidden_size).
  • past_key_values (tuple(tuple(torch.FloatTensor), *optional*) — Contains the precomputed key and value hidden states of the attention blocks used to speed up decoding. The tuple is of length config.n_layers with each tuple having 2 tensors of shape (batch_size, num_heads, decoder_sequence_length, embed_size_per_head) and 2 additional tensors of shape (batch_size, num_heads, encoder_sequence_length, embed_size_per_head).

The OVModelForSeq2SeqLM forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of text generation:

>>> from transformers import AutoTokenizer
>>> from optimum.intel import OVModelForSeq2SeqLM

>>> tokenizer = AutoTokenizer.from_pretrained("echarlaix/t5-small-openvino")
>>> model = OVModelForSeq2SeqLM.from_pretrained("echarlaix/t5-small-openvino")
>>> text = "He never went out without a book under his arm, and he often came back with two."
>>> inputs = tokenizer(text, return_tensors="pt")
>>> gen_tokens = model.generate(**inputs)
>>> outputs = tokenizer.batch_decode(gen_tokens)

Example using transformers.pipeline:

>>> from transformers import AutoTokenizer, pipeline
>>> from optimum.intel import OVModelForSeq2SeqLM

>>> tokenizer = AutoTokenizer.from_pretrained("echarlaix/t5-small-openvino")
>>> model = OVModelForSeq2SeqLM.from_pretrained("echarlaix/t5-small-openvino")
>>> pipe = pipeline("translation_en_to_fr", model=model, tokenizer=tokenizer)
>>> text = "He never went out without a book under his arm, and he often came back with two."
>>> outputs = pipe(text)

OVModelForQuestionAnswering

class optimum.intel.OVModelForQuestionAnswering

< >

( model = None config = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

OpenVINO Model with a QuestionAnsweringModelOutput for extractive question-answering tasks.

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

forward

< >

( input_ids: typing.Union[torch.Tensor, numpy.ndarray] attention_mask: typing.Union[torch.Tensor, numpy.ndarray] token_type_ids: typing.Union[torch.Tensor, numpy.ndarray, NoneType] = None **kwargs )

Parameters

  • input_ids (torch.Tensor) — Indices of input sequence tokens in the vocabulary. Indices can be obtained using AutoTokenizer. What are input IDs?
  • attention_mask (torch.Tensor), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:
  • token_type_ids (torch.Tensor, optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

The OVModelForQuestionAnswering forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of question answering using transformers.pipeline:

>>> from transformers import AutoTokenizer, pipeline
>>> from optimum.intel import OVModelForQuestionAnswering

>>> tokenizer = AutoTokenizer.from_pretrained("distilbert-base-cased-distilled-squad")
>>> model = OVModelForQuestionAnswering.from_pretrained("distilbert-base-cased-distilled-squad", export=True)
>>> pipe = pipeline("question-answering", model=model, tokenizer=tokenizer)
>>> question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet"
>>> outputs = pipe(question, text)

OVModelForSequenceClassification

class optimum.intel.OVModelForSequenceClassification

< >

( model = None config = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

OpenVINO Model with a SequenceClassifierOutput for sequence classification tasks.

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

forward

< >

( input_ids: typing.Union[torch.Tensor, numpy.ndarray] attention_mask: typing.Union[torch.Tensor, numpy.ndarray] token_type_ids: typing.Union[torch.Tensor, numpy.ndarray, NoneType] = None **kwargs )

Parameters

  • input_ids (torch.Tensor) — Indices of input sequence tokens in the vocabulary. Indices can be obtained using AutoTokenizer. What are input IDs?
  • attention_mask (torch.Tensor), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:
  • token_type_ids (torch.Tensor, optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

The OVModelForSequenceClassification forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of sequence classification using transformers.pipeline:

>>> from transformers import AutoTokenizer, pipeline
>>> from optimum.intel import OVModelForSequenceClassification

>>> tokenizer = AutoTokenizer.from_pretrained("distilbert-base-uncased-finetuned-sst-2-english")
>>> model = OVModelForSequenceClassification.from_pretrained("distilbert-base-uncased-finetuned-sst-2-english", export=True)
>>> pipe = pipeline("text-classification", model=model, tokenizer=tokenizer)
>>> outputs = pipe("Hello, my dog is cute")

OVModelForTokenClassification

class optimum.intel.OVModelForTokenClassification

< >

( model = None config = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

OpenVINO Model with a TokenClassifierOutput for token classification tasks.

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

forward

< >

( input_ids: typing.Union[torch.Tensor, numpy.ndarray] attention_mask: typing.Union[torch.Tensor, numpy.ndarray] token_type_ids: typing.Union[torch.Tensor, numpy.ndarray, NoneType] = None **kwargs )

Parameters

  • input_ids (torch.Tensor) — Indices of input sequence tokens in the vocabulary. Indices can be obtained using AutoTokenizer. What are input IDs?
  • attention_mask (torch.Tensor), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:
  • token_type_ids (torch.Tensor, optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

The OVModelForTokenClassification forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of token classification using transformers.pipelines:

>>> from transformers import AutoTokenizer, pipeline
>>> from optimum.intel import OVModelForTokenClassification

>>> tokenizer = AutoTokenizer.from_pretrained("dslim/bert-base-NER")
>>> model = OVModelForTokenClassification.from_pretrained("dslim/bert-base-NER", export=True)
>>> pipe = pipeline("token-classification", model=model, tokenizer=tokenizer)
>>> outputs = pipe("My Name is Peter and I live in New York.")

Audio

The following classes are available for the following audio tasks.

OVModelForAudioClassification

class optimum.intel.OVModelForAudioClassification

< >

( model = None config = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

OpenVINO Model with a SequenceClassifierOutput for audio classification tasks.

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

forward

< >

( input_values: typing.Union[torch.Tensor, numpy.ndarray] attention_mask: typing.Union[torch.Tensor, numpy.ndarray, NoneType] = None **kwargs )

Parameters

  • input_ids (torch.Tensor) — Indices of input sequence tokens in the vocabulary. Indices can be obtained using AutoTokenizer. What are input IDs?
  • attention_mask (torch.Tensor), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:
  • token_type_ids (torch.Tensor, optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

The OVModelForAudioClassification forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of audio classification using transformers.pipelines:

>>> from datasets import load_dataset
>>> from transformers import AutoFeatureExtractor, pipeline
>>> from optimum.intel import OVModelForAudioClassification

>>> preprocessor = AutoFeatureExtractor.from_pretrained("superb/hubert-base-superb-er")
>>> model = OVModelForAudioClassification.from_pretrained("superb/hubert-base-superb-er", export=True)
>>> pipe = pipeline("audio-classification", model=model, feature_extractor=preprocessor)
>>> dataset = load_dataset("superb", "ks", split="test")
>>> audio_file = dataset[3]["audio"]["array"]
>>> outputs = pipe(audio_file)

OVModelForAudioFrameClassification

class optimum.intel.OVModelForAudioFrameClassification

< >

( model: Model config: PretrainedConfig = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

OpenVINO Model for with a frame classification head on top for tasks like Speaker Diarization.

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

Audio Frame Classification model for OpenVINO.

forward

< >

( input_values: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None **kwargs )

Parameters

  • input_values (torch.Tensor of shape (batch_size, sequence_length)) — Float values of input raw speech waveform.. Input values can be obtained from audio file loaded into an array using AutoFeatureExtractor.

The OVModelForAudioFrameClassification forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of audio frame classification:

>>> from transformers import AutoFeatureExtractor
>>> from optimum.intel import OVModelForAudioFrameClassification
>>> from datasets import load_dataset
>>> import torch

>>> dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation")
>>> dataset = dataset.sort("id")
>>> sampling_rate = dataset.features["audio"].sampling_rate

>>> feature_extractor = AutoFeatureExtractor.from_pretrained("anton-l/wav2vec2-base-superb-sd")
>>> model =  OVModelForAudioFrameClassification.from_pretrained("anton-l/wav2vec2-base-superb-sd", export=True)

>>> inputs = feature_extractor(dataset[0]["audio"]["array"], return_tensors="pt", sampling_rate=sampling_rate)
>>>    logits = model(**inputs).logits

>>> probabilities = torch.sigmoid(torch.as_tensor(logits)[0])
>>> labels = (probabilities > 0.5).long()
>>> labels[0].tolist()

OVModelForCTC

class optimum.intel.OVModelForCTC

< >

( model: Model config: PretrainedConfig = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

Onnx Model with a language modeling head on top for Connectionist Temporal Classification (CTC).

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

CTC model for OpenVINO.

forward

< >

( input_values: typing.Optional[torch.Tensor] = None attention_mask: typing.Union[torch.Tensor, numpy.ndarray, NoneType] = None **kwargs )

Parameters

  • input_values (torch.Tensor of shape (batch_size, sequence_length)) — Float values of input raw speech waveform.. Input values can be obtained from audio file loaded into an array using AutoFeatureExtractor.

The OVModelForCTC forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of CTC:

>>> from transformers import AutoFeatureExtractor
>>> from optimum.intel import OVModelForCTC
>>> from datasets import load_dataset

>>> dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation")
>>> dataset = dataset.sort("id")
>>> sampling_rate = dataset.features["audio"].sampling_rate

>>> processor = AutoFeatureExtractor.from_pretrained("facebook/hubert-large-ls960-ft")
>>> model = OVModelForCTC.from_pretrained("facebook/hubert-large-ls960-ft", export=True)

>>> # audio file is decoded on the fly
>>> inputs = processor(dataset[0]["audio"]["array"], sampling_rate=sampling_rate, return_tensors="np")
>>> logits = model(**inputs).logits
>>> predicted_ids = np.argmax(logits, axis=-1)

>>> transcription = processor.batch_decode(predicted_ids)

OVModelForAudioXVector

class optimum.intel.OVModelForAudioXVector

< >

( model: Model config: PretrainedConfig = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

Onnx Model with an XVector feature extraction head on top for tasks like Speaker Verification.

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

Audio XVector model for OpenVINO.

forward

< >

( input_values: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None **kwargs )

Parameters

  • input_values (torch.Tensor of shape (batch_size, sequence_length)) — Float values of input raw speech waveform.. Input values can be obtained from audio file loaded into an array using AutoFeatureExtractor.

The OVModelForAudioXVector forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of Audio XVector:

>>> from transformers import AutoFeatureExtractor
>>> from optimum.intel import OVModelForAudioXVector
>>> from datasets import load_dataset
>>> import torch

>>> dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation")
>>> dataset = dataset.sort("id")
>>> sampling_rate = dataset.features["audio"].sampling_rate

>>> feature_extractor = AutoFeatureExtractor.from_pretrained("anton-l/wav2vec2-base-superb-sv")
>>> model = OVModelForAudioXVector.from_pretrained("anton-l/wav2vec2-base-superb-sv", export=True)

>>> # audio file is decoded on the fly
>>> inputs = feature_extractor(
...     [d["array"] for d in dataset[:2]["audio"]], sampling_rate=sampling_rate, return_tensors="pt", padding=True
... )
>>>     embeddings = model(**inputs).embeddings

>>> embeddings = torch.nn.functional.normalize(embeddings, dim=-1).cpu()

>>> cosine_sim = torch.nn.CosineSimilarity(dim=-1)
>>> similarity = cosine_sim(embeddings[0], embeddings[1])
>>> threshold = 0.7
>>> if similarity < threshold:
...     print("Speakers are not the same!")
>>> round(similarity.item(), 2)

OVModelForSpeechSeq2Seq

class optimum.intel.OVModelForSpeechSeq2Seq

< >

( encoder: Model decoder: Model decoder_with_past: Model = None config: PretrainedConfig = None **kwargs )

Parameters

  • encoder (openvino.runtime.Model) — The OpenVINO Runtime model associated to the encoder.
  • decoder (openvino.runtime.Model) — The OpenVINO Runtime model associated to the decoder.
  • decoder_with_past (openvino.runtime.Model) — The OpenVINO Runtime model associated to the decoder with past key values.
  • config (transformers.PretrainedConfig) — PretrainedConfig is an instance of the configuration associated to the model. Initializing with a config file does not load the weights associated with the model, only the configuration.

Speech Sequence-to-sequence model with a language modeling head for OpenVINO inference. This class officially supports whisper, speech_to_text.

forward

< >

( input_features: typing.Optional[torch.FloatTensor] = None attention_mask: typing.Optional[torch.LongTensor] = None decoder_input_ids: typing.Optional[torch.LongTensor] = None decoder_attention_mask: typing.Optional[torch.BoolTensor] = None encoder_outputs: typing.Optional[typing.Tuple[typing.Tuple[torch.Tensor]]] = None past_key_values: typing.Optional[typing.Tuple[typing.Tuple[torch.Tensor]]] = None **kwargs )

Parameters

  • input_features (torch.FloatTensor) — Mel features extracted from the raw speech waveform. (batch_size, feature_size, encoder_sequence_length).
  • decoder_input_ids (torch.LongTensor) — Indices of decoder input sequence tokens in the vocabulary of shape (batch_size, decoder_sequence_length).
  • encoder_outputs (torch.FloatTensor) — The encoder last_hidden_state of shape (batch_size, encoder_sequence_length, hidden_size).
  • past_key_values (tuple(tuple(torch.FloatTensor), *optional*, defaults to None) — Contains the precomputed key and value hidden states of the attention blocks used to speed up decoding. The tuple is of length config.n_layers with each tuple having 2 tensors of shape (batch_size, num_heads, decoder_sequence_length, embed_size_per_head) and 2 additional tensors of shape (batch_size, num_heads, encoder_sequence_length, embed_size_per_head).

The OVModelForSpeechSeq2Seq forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of text generation:

>>> from transformers import AutoProcessor
>>> from optimum.intel import OVModelForSpeechSeq2Seq
>>> from datasets import load_dataset

>>> processor = AutoProcessor.from_pretrained("openai/whisper-tiny")
>>> model = OVModelForSpeechSeq2Seq.from_pretrained("openai/whisper-tiny")

>>> ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
>>> inputs = processor.feature_extractor(ds[0]["audio"]["array"], return_tensors="pt")

>>> gen_tokens = model.generate(inputs=inputs.input_features)
>>> outputs = processor.tokenizer.batch_decode(gen_tokens)

Example using transformers.pipeline:

>>> from transformers import AutoProcessor, pipeline
>>> from optimum.intel import OVModelForSpeechSeq2Seq
>>> from datasets import load_dataset

>>> processor = AutoProcessor.from_pretrained("openai/whisper-tiny")
>>> model = OVModelForSpeechSeq2Seq.from_pretrained("openai/whisper-tiny")
>>> speech_recognition = pipeline("automatic-speech-recognition", model=model, tokenizer=processor.tokenizer, feature_extractor=processor.feature_extractor)

>>> ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
>>> pred = speech_recognition(ds[0]["audio"]["array"])

Computer Vision

The following classes are available for the following computer vision tasks.

OVModelForImageClassification

class optimum.intel.OVModelForImageClassification

< >

( model = None config = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

OpenVINO Model with a ImageClassifierOutput for image classification tasks.

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

forward

< >

( pixel_values: typing.Union[torch.Tensor, numpy.ndarray] **kwargs )

Parameters

  • pixel_values (torch.Tensor) — Pixel values corresponding to the images in the current batch. Pixel values can be obtained from encoded images using AutoFeatureExtractor.

The OVModelForImageClassification forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of image classification using transformers.pipelines:

>>> from transformers import AutoFeatureExtractor, pipeline
>>> from optimum.intel import OVModelForImageClassification

>>> preprocessor = AutoFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
>>> model = OVModelForImageClassification.from_pretrained("google/vit-base-patch16-224", export=True)
>>> model.reshape(batch_size=1, sequence_length=3, height=224, width=224)
>>> pipe = pipeline("image-classification", model=model, feature_extractor=preprocessor)
>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> outputs = pipe(url)
This class can also be used with [timm](https://github.com/huggingface/pytorch-image-models)

models hosted on HuggingFaceHub. Example:

>>> from transformers import pipeline
>>> from optimum.intel.openvino.modeling_timm import TimmImageProcessor
>>> from optimum.intel import OVModelForImageClassification

>>> model_id = "timm/vit_tiny_patch16_224.augreg_in21k"
>>> preprocessor = TimmImageProcessor.from_pretrained(model_id)
>>> model = OVModelForImageClassification.from_pretrained(model_id, export=True)
>>> pipe = pipeline("image-classification", model=model, feature_extractor=preprocessor)
>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> outputs = pipe(url)

Multimodal

The following classes are available for the following multimodal tasks.

OVModelForVision2Seq

class optimum.intel.OVModelForVision2Seq

< >

( encoder: Model decoder: Model decoder_with_past: Model = None config: PretrainedConfig = None **kwargs )

Parameters

  • encoder (openvino.runtime.Model) — The OpenVINO Runtime model associated to the encoder.
  • decoder (openvino.runtime.Model) — The OpenVINO Runtime model associated to the decoder.
  • decoder_with_past (openvino.runtime.Model) — The OpenVINO Runtime model associated to the decoder with past key values.
  • config (transformers.PretrainedConfig) — PretrainedConfig is an instance of the configuration associated to the model. Initializing with a config file does not load the weights associated with the model, only the configuration.

VisionEncoderDecoder Sequence-to-sequence model with a language modeling head for OpenVINO inference.

forward

< >

( pixel_values: typing.Optional[torch.FloatTensor] = None attention_mask: typing.Optional[torch.LongTensor] = None decoder_input_ids: typing.Optional[torch.LongTensor] = None decoder_attention_mask: typing.Optional[torch.BoolTensor] = None encoder_outputs: typing.Optional[typing.Tuple[typing.Tuple[torch.Tensor]]] = None past_key_values: typing.Optional[typing.Tuple[typing.Tuple[torch.Tensor]]] = None **kwargs )

Parameters

  • pixel_values (torch.FloatTensor) — Features extracted from an Image. This tensor should be of shape (batch_size, num_channels, height, width).
  • decoder_input_ids (torch.LongTensor) — Indices of decoder input sequence tokens in the vocabulary of shape (batch_size, decoder_sequence_length).
  • encoder_outputs (torch.FloatTensor) — The encoder last_hidden_state of shape (batch_size, encoder_sequence_length, hidden_size).
  • past_key_values (tuple(tuple(torch.FloatTensor), *optional*, defaults to None) — Contains the precomputed key and value hidden states of the attention blocks used to speed up decoding. The tuple is of length config.n_layers with each tuple having 2 tensors of shape (batch_size, num_heads, decoder_sequence_length, embed_size_per_head) and 2 additional tensors of shape (batch_size, num_heads, encoder_sequence_length, embed_size_per_head).

The OVModelForVision2Seq forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of text generation:

>>> from transformers import AutoProcessor, AutoTokenizer
>>> from optimum.intel import OVModelForVision2Seq
>>> from PIL import Image
>>> import requests


>>> processor = AutoProcessor.from_pretrained("microsoft/trocr-small-handwritten")
>>> tokenizer = AutoTokenizer.from_pretrained("microsoft/trocr-small-handwritten")
>>> model = OVModelForVision2Seq.from_pretrained("microsoft/trocr-small-handwritten", export=True)

>>> url = "https://fki.tic.heia-fr.ch/static/img/a01-122-02-00.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)
>>> inputs = processor(image, return_tensors="pt")

>>> gen_tokens = model.generate(**inputs)
>>> outputs = tokenizer.batch_decode(gen_tokens, skip_special_tokens=True)

Example using transformers.pipeline:

>>> from transformers import AutoProcessor, AutoTokenizer, pipeline
>>> from optimum.intel import OVModelForVision2Seq
>>> from PIL import Image
>>> import requests


>>> processor = AutoProcessor.from_pretrained("microsoft/trocr-small-handwritten")
>>> tokenizer = AutoTokenizer.from_pretrained("microsoft/trocr-small-handwritten")
>>> model = OVModelForVision2Seq.from_pretrained("microsoft/trocr-small-handwritten", export=True)

>>> url = "https://fki.tic.heia-fr.ch/static/img/a01-122-02-00.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> image_to_text = pipeline("image-to-text", model=model, tokenizer=tokenizer, feature_extractor=processor, image_processor=processor)
>>> pred = image_to_text(image)

OVModelForPix2Struct

class optimum.intel.OVModelForPix2Struct

< >

( encoder: Model decoder: Model decoder_with_past: Model = None config: PretrainedConfig = None **kwargs )

Parameters

  • encoder (openvino.runtime.Model) — The OpenVINO Runtime model associated to the encoder.
  • decoder (openvino.runtime.Model) — The OpenVINO Runtime model associated to the decoder.
  • decoder_with_past (openvino.runtime.Model) — The OpenVINO Runtime model associated to the decoder with past key values.
  • config (transformers.PretrainedConfig) — PretrainedConfig is an instance of the configuration associated to the model. Initializing with a config file does not load the weights associated with the model, only the configuration.

Pix2Struct model with a language modeling head for OpenVINO inference.

forward

< >

( flattened_patches: typing.Optional[torch.FloatTensor] = None attention_mask: typing.Optional[torch.LongTensor] = None decoder_input_ids: typing.Optional[torch.LongTensor] = None decoder_attention_mask: typing.Optional[torch.BoolTensor] = None encoder_outputs: typing.Optional[typing.Tuple[typing.Tuple[torch.Tensor]]] = None past_key_values: typing.Optional[typing.Tuple[typing.Tuple[torch.Tensor]]] = None **kwargs )

Parameters

  • flattened_patches (torch.FloatTensor of shape (batch_size, seq_length, hidden_size)) — Flattened pixel patches. the hidden_size is obtained by the following formula: hidden_size = num_channels patch_size patch_size The process of flattening the pixel patches is done by Pix2StructProcessor.
  • attention_mask (torch.FloatTensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices.
  • decoder_input_ids (torch.LongTensor of shape (batch_size, target_sequence_length), optional) — Indices of decoder input sequence tokens in the vocabulary. Pix2StructText uses the pad_token_id as the starting token for decoder_input_ids generation. If past_key_values is used, optionally only the last decoder_input_ids have to be input (see past_key_values).
  • decoder_attention_mask (torch.BoolTensor of shape (batch_size, target_sequence_length), optional) — Default behavior: generate a tensor that ignores pad tokens in decoder_input_ids. Causal mask will also be used by default.
  • encoder_outputs (tuple(tuple(torch.FloatTensor), optional) — Tuple consists of (last_hidden_state, optional: hidden_states, optional: attentions) last_hidden_state of shape (batch_size, sequence_length, hidden_size) is a sequence of hidden states at the output of the last layer of the encoder. Used in the cross-attention of the decoder.
  • past_key_values (tuple(tuple(torch.FloatTensor), *optional*, defaults to None) — Contains the precomputed key and value hidden states of the attention blocks used to speed up decoding. The tuple is of length config.n_layers with each tuple having 2 tensors of shape (batch_size, num_heads, decoder_sequence_length, embed_size_per_head) and 2 additional tensors of shape (batch_size, num_heads, encoder_sequence_length, embed_size_per_head).

The OVModelForPix2Struct forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of pix2struct:

>>> from transformers import AutoProcessor
>>> from optimum.intel import OVModelForPix2Struct
>>> from PIL import Image
>>> import requests

>>> processor = AutoProcessor.from_pretrained("google/pix2struct-ai2d-base")
>>> model = OVModelForPix2Struct.from_pretrained("google/pix2struct-ai2d-base", export=True)

>>> url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)
>>> question = "What does the label 15 represent? (1) lava (2) core (3) tunnel (4) ash cloud"
>>> inputs = processor(images=image, text=question, return_tensors="pt")

>>> gen_tokens = model.generate(**inputs)
>>> outputs = processor.batch_decode(gen_tokens, skip_special_tokens=True)

Custom Tasks

OVModelForCustomTasks

class optimum.intel.OVModelForCustomTasks

< >

( model: Model config: PretrainedConfig = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

OpenVINO Model for custom tasks. It can be used to leverage the inference acceleration for any single-file OpenVINO model, that may use custom inputs and outputs.

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

forward

< >

( **kwargs )

The OVModelForCustomTasks forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of custom tasks (e.g. a sentence transformers with a pooler head):

>>> from transformers import AutoTokenizer
>>> from optimum.intel import OVModelForCustomTasks

>>> tokenizer = AutoTokenizer.from_pretrained("IlyasMoutawwakil/sbert-all-MiniLM-L6-v2-with-pooler")
>>> model = OVModelForCustomTasks.from_pretrained("IlyasMoutawwakil/sbert-all-MiniLM-L6-v2-with-pooler")

>>> inputs = tokenizer("I love burritos!", return_tensors="np")

>>> outputs = model(**inputs)
>>> last_hidden_state = outputs.last_hidden_state
>>> pooler_output = outputs.pooler_output

OVModelForFeatureExtraction

class optimum.intel.OVModelForFeatureExtraction

< >

( model = None config = None **kwargs )

Parameters

  • model (openvino.runtime.Model) — is the main class used to run OpenVINO Runtime inference.
  • config (transformers.PretrainedConfig) — PretrainedConfig is the Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the ~intel.openvino.modeling.OVBaseModel.from_pretrained method to load the model weights.
  • device (str, defaults to "CPU") — The device type for which the model will be optimized for. The resulting compiled model will contains nodes specific to this device.
  • dynamic_shapes (bool, defaults to True) — All the model’s dimension will be set to dynamic when set to True. Should be set to False for the model to not be dynamically reshaped by default.
  • ov_config (Optional[Dict], defaults to None) — The dictionnary containing the informations related to the model compilation.
  • compile (bool, defaults to True) — Disable the model compilation during the loading step when set to False. Can be useful to avoid unnecessary compilation, in the case where the model needs to be statically reshaped, the device modified or if FP16 conversion is enabled.

OpenVINO Model with a BaseModelOutput for feature extraction tasks.

This model inherits from optimum.intel.openvino.modeling.OVBaseModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving)

forward

< >

( input_ids: typing.Union[torch.Tensor, numpy.ndarray] attention_mask: typing.Union[torch.Tensor, numpy.ndarray] token_type_ids: typing.Union[torch.Tensor, numpy.ndarray, NoneType] = None **kwargs )

Parameters

  • input_ids (torch.Tensor) — Indices of input sequence tokens in the vocabulary. Indices can be obtained using AutoTokenizer. What are input IDs?
  • attention_mask (torch.Tensor), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:
  • token_type_ids (torch.Tensor, optional) — Segment token indices to indicate first and second portions of the inputs. Indices are selected in [0, 1]:

The OVModelForFeatureExtraction forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example of feature extraction using transformers.pipelines:

>>> from transformers import AutoTokenizer, pipeline
>>> from optimum.intel import OVModelForFeatureExtraction

>>> tokenizer = AutoTokenizer.from_pretrained("sentence-transformers/all-MiniLM-L6-v2")
>>> model = OVModelForFeatureExtraction.from_pretrained("sentence-transformers/all-MiniLM-L6-v2", export=True)
>>> pipe = pipeline("feature-extraction", model=model, tokenizer=tokenizer)
>>> outputs = pipe("My Name is Peter and I live in New York.")

Quantization

OVQuantizer

class optimum.intel.OVQuantizer

< >

( model: PreTrainedModel task: typing.Optional[str] = None seed: int = 42 **kwargs )

Handle the NNCF quantization process.

get_calibration_dataset

< >

( dataset_name: str num_samples: int = 100 dataset_config_name: typing.Optional[str] = None dataset_split: str = 'train' preprocess_function: typing.Optional[typing.Callable] = None preprocess_batch: bool = True use_auth_token: typing.Union[bool, str, NoneType] = None token: typing.Union[bool, str, NoneType] = None cache_dir: str = '/root/.cache/huggingface/hub' trust_remote_code: bool = False )

Parameters

  • dataset_name (str) — The dataset repository name on the Hugging Face Hub or path to a local directory containing data files in generic formats and optionally a dataset script, if it requires some code to read the data files.
  • num_samples (int, defaults to 100) — The maximum number of samples composing the calibration dataset.
  • dataset_config_name (str, optional) — The name of the dataset configuration.
  • dataset_split (str, defaults to "train") — Which split of the dataset to use to perform the calibration step.
  • preprocess_function (Callable, optional) — Processing function to apply to each example after loading dataset.
  • preprocess_batch (bool, defaults to True) — Whether the preprocess_function should be batched.
  • use_auth_token (Optional[Union[bool, str]], defaults to None) — Deprecated. Please use token instead.
  • token (Optional[Union[bool, str]], defaults to None) — The token to use as HTTP bearer authorization for remote files. If True, will use the token generated when running huggingface-cli login (stored in ~/.huggingface).
  • cache_dir (str, optional) — Caching directory for a calibration dataset.
  • trust_remote_code (bool, defaults to False) — Whether or not to allow for custom models defined on the Hub in their own modeling files. This option should only be set to True for repositories you trust and in which you have read the code, as it will execute code present on the Hub on your local machine.

Create the calibration datasets.Dataset to use for the post-training static quantization calibration step.

quantize

< >

( calibration_dataset: typing.Union[ForwardRef('Dataset'), nncf.data.dataset.Dataset, typing.Iterable, NoneType] = None save_directory: typing.Union[pathlib.Path, str, NoneType] = None ov_config: OVConfig = None file_name: typing.Optional[str] = None batch_size: int = 1 data_collator: typing.Optional[DataCollator] = None remove_unused_columns: bool = True weights_only: bool = None **kwargs )

Parameters

  • calibration_dataset (datasets.Dataset or nncf.Dataset or Iterable, optional) — A collection of data samples to use for quantization calibration. Is optional for weight-only quantization and is required for full quantization.
  • save_directory (Union[str, Path], optional) — The directory where the quantized model should be saved.
  • ov_config (OVConfig, optional) — The configuration containing the parameters related to quantization. If not provided, 8-bit symmetric weight-only quantization will be applied.
  • file_name (str, optional) — The model file name to use when saving the model. Overwrites the default file name "model.onnx".
  • batch_size (int, defaults to 1) — The number of calibration samples to load per batch.
  • data_collator (DataCollator, optional) — The function to use to form a batch from a list of elements of the calibration dataset.
  • remove_unused_columns (bool, defaults to True) — Whether to remove the columns unused by the model forward method.
  • weights_only (bool, optional) — Being deprecated. Compress weights to integer precision (8-bit by default) while keeping activations floating-point. Fits best for LLM footprint reduction and performance acceleration.

Quantize a model given the optimization specifications defined in quantization_config.

Examples:

>>> from optimum.intel import OVQuantizer, OVModelForCausalLM
>>> from transformers import AutoModelForCausalLM
>>> model = AutoModelForCausalLM.from_pretrained("databricks/dolly-v2-3b")
>>> quantizer = OVQuantizer.from_pretrained(model, task="text-generation")
>>> ov_config = OVConfig(quantization_config=OVWeightQuantizationConfig())
>>> quantizer.quantize(ov_config=ov_config, save_directory="./quantized_model")
>>> optimized_model = OVModelForCausalLM.from_pretrained("./quantized_model")
>>> from optimum.intel import OVQuantizer, OVModelForSequenceClassification
>>> from transformers import AutoModelForSequenceClassification
>>> model = OVModelForSequenceClassification.from_pretrained("distilbert-base-uncased-finetuned-sst-2-english", export=True)
>>> # or
>>> model = AutoModelForSequenceClassification.from_pretrained("distilbert-base-uncased-finetuned-sst-2-english")
>>> quantizer = OVQuantizer.from_pretrained(model, task="text-classification")
>>> ov_config = OVConfig(quantization_config=OVQuantizationConfig())
>>> quantizer.quantize(calibration_dataset=dataset, ov_config=ov_config, save_directory="./quantized_model")
>>> optimized_model = OVModelForSequenceClassification.from_pretrained("./quantized_model")
< > Update on GitHub

←Supported Models 🤗 Optimum Neuron→