Transformers documentation

BLIP

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v4.27.2).
Join the Hugging Face community

to get started

# BLIP

## Overview

The BLIP model was proposed in BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation by Junnan Li, Dongxu Li, Caiming Xiong, Steven Hoi.

BLIP is a model that is able to perform various multi-modal tasks including

• Image-Text retrieval (Image-text matching)
• Image Captioning

The abstract from the paper is the following:

Vision-Language Pre-training (VLP) has advanced the performance for many vision-language tasks. However, most existing pre-trained models only excel in either understanding-based tasks or generation-based tasks. Furthermore, performance improvement has been largely achieved by scaling up the dataset with noisy image-text pairs collected from the web, which is a suboptimal source of supervision. In this paper, we propose BLIP, a new VLP framework which transfers flexibly to both vision-language understanding and generation tasks. BLIP effectively utilizes the noisy web data by bootstrapping the captions, where a captioner generates synthetic captions and a filter removes the noisy ones. We achieve state-of-the-art results on a wide range of vision-language tasks, such as image-text retrieval (+2.7% in average recall@1), image captioning (+2.8% in CIDEr), and VQA (+1.6% in VQA score). BLIP also demonstrates strong generalization ability when directly transferred to videolanguage tasks in a zero-shot manner. Code, models, and datasets are released.

This model was contributed by ybelkada. The original code can be found here.

## Resources

• Jupyter notebook on how to fine-tune BLIP for image captioning on a custom dataset

## BlipConfig

### class transformers.BlipConfig

< >

( text_config = None vision_config = None projection_dim = 512 logit_scale_init_value = 2.6592 image_text_hidden_size = 256 **kwargs )

Parameters

• text_config (dict, optional) — Dictionary of configuration options used to initialize BlipTextConfig.
• vision_config (dict, optional) — Dictionary of configuration options used to initialize BlipVisionConfig.
• projection_dim (int, optional, defaults to 512) — Dimentionality of text and vision projection layers.
• logit_scale_init_value (float, optional, defaults to 2.6592) — The inital value of the logit_scale paramter. Default is used as per the original BLIP implementation.
• image_text_hidden_size (int, optional, defaults to 768) — Dimentionality of the hidden state of the image-text fusion layer.
• kwargs (optional) — Dictionary of keyword arguments.

BlipConfig is the configuration class to store the configuration of a BlipModel. It is used to instantiate a BLIP model according to the specified arguments, defining the text model and vision model configs. Instantiating a configuration with the defaults will yield a similar configuration to that of the BLIP-base Salesforce/blip-vqa-base architecture.

Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.

Example:

>>> from transformers import BlipConfig, BlipModel

>>> # Initializing a BlipConfig with Salesforce/blip-vqa-base style configuration
>>> configuration = BlipConfig()

>>> # Initializing a BlipPModel (with random weights) from the Salesforce/blip-vqa-base style configuration
>>> model = BlipModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

>>> # We can also initialize a BlipConfig from a BlipTextConfig and a BlipVisionConfig

>>> # Initializing a BLIPText and BLIPVision configuration
>>> config_text = BlipTextConfig()
>>> config_vision = BlipVisionConfig()

>>> config = BlipConfig.from_text_vision_configs(config_text, config_vision)

#### from_text_vision_configs

< >

( text_config: BlipTextConfig vision_config: BlipVisionConfig **kwargs ) BlipConfig

Returns

BlipConfig

An instance of a configuration object

Instantiate a BlipConfig (or a derived class) from blip text model configuration and blip vision model configuration.

## BlipTextConfig

### class transformers.BlipTextConfig

< >

( vocab_size = 30524 hidden_size = 768 encoder_hidden_size = 768 intermediate_size = 3072 projection_dim = 768 num_hidden_layers = 12 num_attention_heads = 8 max_position_embeddings = 512 hidden_act = 'gelu' layer_norm_eps = 1e-12 hidden_dropout_prob = 0.0 attention_probs_dropout_prob = 0.0 initializer_range = 0.02 bos_token_id = 30522 eos_token_id = 2 pad_token_id = 0 sep_token_id = 102 is_decoder = True use_cache = True **kwargs )

Parameters

• vocab_size (int, optional, defaults to 30522) — Vocabulary size of the Blip text model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling BlipModel.
• hidden_size (int, optional, defaults to 768) — Dimensionality of the encoder layers and the pooler layer.
• encoder_hidden_size (int, optional, defaults to 768) — Dimensionality of the encoder layers from the vision model.
• intermediate_size (int, optional, defaults to 3072) — Dimensionality of the “intermediate” (i.e., feed-forward) layer in the Transformer encoder.
• num_hidden_layers (int, optional, defaults to 12) — Number of hidden layers in the Transformer encoder.
• num_attention_heads (int, optional, defaults to 8) — Number of attention heads for each attention layer in the Transformer encoder.
• max_position_embeddings (int, optional, defaults to 77) — The maximum sequence length that this model might ever be used with. Typically set this to something large just in case (e.g., 512 or 1024 or 2048).
• hidden_act (str or function, optional, defaults to "gelu") — The non-linear activation function (function or string) in the encoder and pooler. If string, "gelu", "relu", "selu" and "gelu_new" "gelu" are supported.
• layer_norm_eps (float, optional, defaults to 1e-12) — The epsilon used by the layer normalization layers.
• hidden_dropout_prob (float, optional, defaults to 0.0) — The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
• attention_dropout (float, optional, defaults to 0.0) — The dropout ratio for the attention probabilities.
• initializer_range (float, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
• bos_token_id (int, optional, defaults to 30522) — The id of the beginning-of-sequence token.
• eos_token_id (int, optional, defaults to 2) — The id of the end-of-sequence token.
• pad_token_id (int, optional, defaults to 0) — The id of the padding token.
• sep_token_id (int, optional, defaults to 102) — The id of the separator token.
• is_decoder (bool, optional, defaults to False) — Whether the model is used as a decoder.
• use_cache (bool, optional, defaults to True) — Whether or not the model should return the last key/values attentions (not used by all models).

This is the configuration class to store the configuration of a BlipTextModel. It is used to instantiate a BLIP text model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the BlipText used by the base architectures.

Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.

Example:

>>> from transformers import BlipTextConfig, BlipTextModel

>>> # Initializing a BlipTextConfig with Salesforce/blip-vqa-base style configuration
>>> configuration = BlipTextConfig()

>>> # Initializing a BlipTextModel (with random weights) from the Salesforce/blip-vqa-base style configuration
>>> model = BlipTextModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

## BlipVisionConfig

### class transformers.BlipVisionConfig

< >

( hidden_size = 768 intermediate_size = 3072 projection_dim = 512 num_hidden_layers = 12 num_attention_heads = 12 num_channels = 3 image_size = 384 patch_size = 16 hidden_act = 'gelu' layer_norm_eps = 1e-05 attention_dropout = 0.0 initializer_range = 1e-10 **kwargs )

Parameters

• hidden_size (int, optional, defaults to 768) — Dimensionality of the encoder layers and the pooler layer.
• intermediate_size (int, optional, defaults to 3072) — Dimensionality of the “intermediate” (i.e., feed-forward) layer in the Transformer encoder.
• num_hidden_layers (int, optional, defaults to 12) — Number of hidden layers in the Transformer encoder.
• num_attention_heads (int, optional, defaults to 12) — Number of attention heads for each attention layer in the Transformer encoder.
• image_size (int, optional, defaults to 224) — The size (resolution) of each image.
• patch_size (int, optional, defaults to 32) — The size (resolution) of each patch.
• hidden_act (str or function, optional, defaults to "gelu") — The non-linear activation function (function or string) in the encoder and pooler. If string, "gelu", "relu", "selu" and "gelu_new" "gelu" are supported.
• layer_norm_eps (float, optional, defaults to 1e-5) — The epsilon used by the layer normalization layers.
• attention_dropout (float, optional, defaults to 0.0) — The dropout ratio for the attention probabilities.
• initializer_range (float, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices.

This is the configuration class to store the configuration of a BlipVisionModel. It is used to instantiate a BLIP vision model according to the specified arguments, defining the model architecture. Instantiating a configuration defaults will yield a similar configuration to that of the Blip-base Salesforce/blip-vqa-base architecture.

Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.

Example:

>>> from transformers import BlipVisionConfig, BlipVisionModel

>>> # Initializing a BlipVisionConfig with Salesforce/blip-vqa-base style configuration
>>> configuration = BlipVisionConfig()

>>> # Initializing a BlipVisionModel (with random weights) from the Salesforce/blip-vqa-base style configuration
>>> model = BlipVisionModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

## BlipProcessor

### class transformers.BlipProcessor

< >

( image_processor tokenizer )

Parameters

• image_processor (BlipImageProcessor) — An instance of BlipImageProcessor. The image processor is a required input.
• tokenizer (BertTokenizerFast) — An instance of [‘BertTokenizerFast]. The tokenizer is a required input.

Constructs a BLIP processor which wraps a BERT tokenizer and BLIP image processor into a single processor.

BlipProcessor offers all the functionalities of BlipImageProcessor and BertTokenizerFast. See the docstring of __call__() and decode() for more information.

#### batch_decode

< >

( *args **kwargs )

This method forwards all its arguments to BertTokenizerFast’s batch_decode(). Please refer to the docstring of this method for more information.

#### decode

< >

( *args **kwargs )

This method forwards all its arguments to BertTokenizerFast’s decode(). Please refer to the docstring of this method for more information.

## BlipImageProcessor

### class transformers.BlipImageProcessor

< >

( do_resize: bool = True size: typing.Dict[str, int] = None resample: Resampling = <Resampling.BICUBIC: 3> do_rescale: bool = True rescale_factor: typing.Union[int, float] = 0.00392156862745098 do_normalize: bool = True image_mean: typing.Union[float, typing.List[float], NoneType] = None image_std: typing.Union[float, typing.List[float], NoneType] = None do_convert_rgb: bool = True **kwargs )

Parameters

• do_resize (bool, optional, defaults to True) — Whether to resize the image’s (height, width) dimensions to the specified size. Can be overridden by the do_resize parameter in the preprocess method.
• size (dict, optional, defaults to {"height" -- 384, "width": 384}): Size of the output image after resizing. Can be overridden by the size parameter in the preprocess method.
• resample (PILImageResampling, optional, defaults to PILImageResampling.BICUBIC) — Resampling filter to use if resizing the image. Only has an effect if do_resize is set to True. Can be overridden by the resample parameter in the preprocess method.
• do_rescale (bool, optional, defaults to True) — Wwhether to rescale the image by the specified scale rescale_factor. Can be overridden by the do_rescale parameter in the preprocess method.
• rescale_factor (int or float, optional, defaults to 1/255) — Scale factor to use if rescaling the image. Only has an effect if do_rescale is set to True. Can be overridden by the rescale_factor parameter in the preprocess method.
• do_normalize (bool, optional, defaults to True) — Whether to normalize the image. Can be overridden by the do_normalize parameter in the preprocess method. Can be overridden by the do_normalize parameter in the preprocess method.
• image_mean (float or List[float], optional, defaults to IMAGENET_STANDARD_MEAN) — Mean to use if normalizing the image. This is a float or list of floats the length of the number of channels in the image. Can be overridden by the image_mean parameter in the preprocess method. Can be overridden by the image_mean parameter in the preprocess method.
• image_std (float or List[float], optional, defaults to IMAGENET_STANDARD_STD) — Standard deviation to use if normalizing the image. This is a float or list of floats the length of the number of channels in the image. Can be overridden by the image_std parameter in the preprocess method. Can be overridden by the image_std parameter in the preprocess method.
• do_convert_rgb (bool, optional, defaults to True) — Whether to convert the image to RGB.

Constructs a BLIP image processor.

#### preprocess

< >

( images: typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), typing.List[ForwardRef('PIL.Image.Image')], typing.List[numpy.ndarray], typing.List[ForwardRef('torch.Tensor')]] do_resize: typing.Optional[bool] = None size: typing.Union[typing.Dict[str, int], NoneType] = None resample: Resampling = None do_rescale: typing.Optional[bool] = None rescale_factor: typing.Optional[float] = None do_normalize: typing.Optional[bool] = None image_mean: typing.Union[float, typing.List[float], NoneType] = None image_std: typing.Union[float, typing.List[float], NoneType] = None return_tensors: typing.Union[str, transformers.utils.generic.TensorType, NoneType] = None do_convert_rgb: bool = None data_format: ChannelDimension = <ChannelDimension.FIRST: 'channels_first'> **kwargs )

Parameters

• images (ImageInput) — Image to preprocess.
• do_resize (bool, optional, defaults to self.do_resize) — Whether to resize the image.
• size (Dict[str, int], optional, defaults to self.size) — Controls the size of the image after resize. The shortest edge of the image is resized to size["shortest_edge"] whilst preserving the aspect ratio. If the longest edge of this resized image is > int(size["shortest_edge"] * (1333 / 800)), then the image is resized again to make the longest edge equal to int(size["shortest_edge"] * (1333 / 800)).
• resample (PILImageResampling, optional, defaults to self.resample) — Resampling filter to use if resizing the image. Only has an effect if do_resize is set to True.
• do_rescale (bool, optional, defaults to self.do_rescale) — Whether to rescale the image values between [0 - 1].
• rescale_factor (float, optional, defaults to self.rescale_factor) — Rescale factor to rescale the image by if do_rescale is set to True.
• do_normalize (bool, optional, defaults to self.do_normalize) — Whether to normalize the image.
• image_mean (float or List[float], optional, defaults to self.image_mean) — Image mean to normalize the image by if do_normalize is set to True.
• image_std (float or List[float], optional, defaults to self.image_std) — Image standard deviation to normalize the image by if do_normalize is set to True.
• do_convert_rgb (bool, optional, defaults to self.do_convert_rgb) — Whether to convert the image to RGB.
• return_tensors (str or TensorType, optional) — The type of tensors to return. Can be one of:
• Unset: Return a list of np.ndarray.
• TensorType.TENSORFLOW or 'tf': Return a batch of type tf.Tensor.
• TensorType.PYTORCH or 'pt': Return a batch of type torch.Tensor.
• TensorType.NUMPY or 'np': Return a batch of type np.ndarray.
• TensorType.JAX or 'jax': Return a batch of type jax.numpy.ndarray.
• data_format (ChannelDimension or str, optional, defaults to ChannelDimension.FIRST) — The channel dimension format for the output image. Can be one of:
• ChannelDimension.FIRST: image in (num_channels, height, width) format.
• ChannelDimension.LAST: image in (height, width, num_channels) format.

Preprocess an image or batch of images.

## BlipModel

### class transformers.BlipModel

< >

( config: BlipConfig )

Parameters

• config (BlipConfig) — 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 from_pretrained() method to load the model weights.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

#### forward

< >

( input_ids: typing.Optional[torch.LongTensor] = None pixel_values: typing.Optional[torch.FloatTensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.LongTensor] = None return_loss: typing.Optional[bool] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) transformers.models.blip.modeling_blip.BlipOutput or tuple(torch.FloatTensor)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it.

Indices can be obtained using AutoProcessor. See BlipProcessor.__call__() for details.

What are input IDs?

• attention_mask (torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.max_position_embeddings - 1].

What are position IDs?

• pixel_values (torch.FloatTensor of shape (batch_size, num_channels, height, width)) — Pixel values. Padding will be ignored by default should you provide it. Pixel values can be obtained using BlipImageProcessor. See BlipImageProcessor.call() for details.
• return_loss (bool, optional) — Whether or not to return the contrastive loss.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.

Returns

transformers.models.blip.modeling_blip.BlipOutput or tuple(torch.FloatTensor)

A transformers.models.blip.modeling_blip.BlipOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (<class 'transformers.models.blip.configuration_blip.BlipConfig'>) and inputs.

• loss (torch.FloatTensor of shape (1,), optional, returned when return_loss is True) — Contrastive loss for image-text similarity.
• logits_per_image:(torch.FloatTensor of shape (image_batch_size, text_batch_size)) — The scaled dot product scores between image_embeds and text_embeds. This represents the image-text similarity scores.
• logits_per_text:(torch.FloatTensor of shape (text_batch_size, image_batch_size)) — The scaled dot product scores between text_embeds and image_embeds. This represents the text-image similarity scores.
• text_embeds(torch.FloatTensor of shape (batch_size, output_dim) — The text embeddings obtained by applying the projection layer to the pooled output of BlipTextModel.
• image_embeds(torch.FloatTensor of shape (batch_size, output_dim) — The image embeddings obtained by applying the projection layer to the pooled output of BlipVisionModel.
• text_model_output(BaseModelOutputWithPooling): The output of the BlipTextModel.
• vision_model_output(BaseModelOutputWithPooling): The output of the BlipVisionModel.

The BlipModel 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.

Examples:

>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, BlipModel

>>> model = BlipModel.from_pretrained("Salesforce/blip-image-captioning-base")
>>> processor = AutoProcessor.from_pretrained("Salesforce/blip-image-captioning-base")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> inputs = processor(
...     text=["a photo of a cat", "a photo of a dog"], images=image, return_tensors="pt", padding=True
... )

>>> outputs = model(**inputs)
>>> logits_per_image = outputs.logits_per_image  # this is the image-text similarity score
>>> probs = logits_per_image.softmax(dim=1)  # we can take the softmax to get the label probabilities

#### get_text_features

< >

( input_ids: typing.Optional[torch.Tensor] = None attention_mask: typing.Optional[torch.Tensor] = None position_ids: typing.Optional[torch.Tensor] = None return_dict: typing.Optional[bool] = None ) text_features (torch.FloatTensor of shape (batch_size, output_dim)

Parameters

• input_ids (torch.LongTensor of shape (batch_size, sequence_length)) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide it.

Indices can be obtained using AutoProcessor. See BlipProcessor.__call__() for details.

What are input IDs?

• attention_mask (torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked.

• position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.max_position_embeddings - 1].

What are position IDs?

• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.

Returns

text_features (torch.FloatTensor of shape (batch_size, output_dim)

The text embeddings obtained by applying the projection layer to the pooled output of BlipTextModel.

The BlipModel 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.

Examples:

>>> from transformers import AutoProcessor, BlipModel

>>> model = BlipModel.from_pretrained("Salesforce/blip-image-captioning-base")
>>> processor = AutoProcessor.from_pretrained("Salesforce/blip-image-captioning-base")

>>> inputs = processor(text=["a photo of a cat", "a photo of a dog"], padding=True, return_tensors="pt")
>>> text_features = model.get_text_features(**inputs)

#### get_image_features

< >

( pixel_values: typing.Optional[torch.FloatTensor] = None return_dict: typing.Optional[bool] = None ) image_features (torch.FloatTensor of shape (batch_size, output_dim)

Parameters

• pixel_values (torch.FloatTensor of shape (batch_size, num_channels, height, width)) — Pixel values. Padding will be ignored by default should you provide it. Pixel values can be obtained using BlipImageProcessor. See BlipImageProcessor.call() for details.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.

Returns

image_features (torch.FloatTensor of shape (batch_size, output_dim)

The image embeddings obtained by applying the projection layer to the pooled output of BlipVisionModel.

The BlipModel 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.

Examples:

>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, BlipModel

>>> model = BlipModel.from_pretrained("Salesforce/blip-image-captioning-base")
>>> processor = AutoProcessor.from_pretrained("Salesforce/blip-image-captioning-base")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> inputs = processor(images=image, return_tensors="pt")

>>> image_features = model.get_image_features(**inputs)

## BlipTextModel

### class transformers.BlipTextModel

< >

( config add_pooling_layer = True )

The model can behave as an encoder (with only self-attention) as well as a decoder, in which case a layer of cross-attention is added between the self-attention layers, following the architecture described in Attention is all you need by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser and Illia Polosukhin. argument and is_decoder set to True; an encoder_hidden_states is then expected as an input to the forward pass.

#### forward

< >

( input_ids = None attention_mask = None position_ids = None head_mask = None inputs_embeds = None encoder_embeds = None encoder_hidden_states = None encoder_attention_mask = None past_key_values = None use_cache = None output_attentions = None output_hidden_states = None return_dict = None is_decoder = False )

encoder_hidden_states (torch.FloatTensor, optional): Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention if the model is configured as a decoder. encoder_attention_mask (torch.FloatTensor, optional): Mask to avoid performing attention on the padding token indices of the encoder input. This mask is used in the cross-attention if the model is configured as a decoder. Mask values selected in [0, 1]:

• 1 for tokens that are not masked,
• 0 for tokens that are masked. past_key_values (tuple(tuple(torch.FloatTensor)), optional): Contains precomputed key and value hidden states of the attention blocks. Can be used to speed up decoding. If past_key_values are used, the user can optionally input only the last decoder_input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, 1) instead of all decoder_input_ids of shape (batch_size, sequence_length). use_cache (bool, optional): If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).

## BlipVisionModel

### class transformers.BlipVisionModel

< >

( config: BlipVisionConfig )

#### forward

< >

( pixel_values: typing.Optional[torch.FloatTensor] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) transformers.modeling_outputs.BaseModelOutputWithPooling or tuple(torch.FloatTensor)

Parameters

• pixel_values (torch.FloatTensor of shape (batch_size, num_channels, height, width)) — Pixel values. Padding will be ignored by default should you provide it. Pixel values can be obtained using BlipImageProcessor. See BlipImageProcessor.call() for details.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.

Returns

transformers.modeling_outputs.BaseModelOutputWithPooling or tuple(torch.FloatTensor)

A transformers.modeling_outputs.BaseModelOutputWithPooling or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (<class 'transformers.models.blip.configuration_blip.BlipVisionConfig'>) and inputs.

• last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

• pooler_output (torch.FloatTensor of shape (batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processing through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns the classification token after processing through a linear layer and a tanh activation function. The linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

• hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

• attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The BlipVisionModel 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.

## BlipForConditionalGeneration

### class transformers.BlipForConditionalGeneration

< >

( config: BlipConfig )

Parameters

• config (BlipConfig) — 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 from_pretrained() method to load the model weights.

BLIP Model for image captioning. The model consists of a vision encoder and a text decoder. One can optionally pass input_ids to the model, which serve as a text prompt, to make the text decoder continue the prompt. Otherwise, the decoder starts generating text from the [BOS] (beginning-of-sequence) token. will start generating the caption from the text input. If no text input is provided, the decoder will start with the [BOS] token only.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

#### forward

< >

( pixel_values: FloatTensor input_ids: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.LongTensor] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None labels: typing.Optional[torch.LongTensor] = None return_dict: typing.Optional[bool] = None ) transformers.models.blip.modeling_blip.BlipForConditionalGenerationModelOutput or tuple(torch.FloatTensor)

Parameters

• pixel_values (torch.FloatTensor of shape (batch_size, num_channels, height, width)) — Pixel values. Padding will be ignored by default should you provide it. Pixel values can be obtained using BlipImageProcessor. See BlipImageProcessor.call() for details.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.

Returns

transformers.models.blip.modeling_blip.BlipForConditionalGenerationModelOutput or tuple(torch.FloatTensor)

A transformers.models.blip.modeling_blip.BlipForConditionalGenerationModelOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (<class 'transformers.models.blip.configuration_blip.BlipVisionConfig'>) and inputs.

• loss (torch.FloatTensor, optional, returned when labels is provided, torch.FloatTensor of shape (1,)) — Languge modeling loss from the text decoder.

• decoder_logits (torch.FloatTensor of shape (batch_size, sequence_length, config.vocab_size), optional) — Prediction scores of the language modeling head of the text decoder model.

• image_embeds (torch.FloatTensor of shape (batch_size, output_dim), optional) — The image embeddings obtained after applying the Vision Transformer model to the input image.

• last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Sequence of hidden-states at the output of the last layer of the model.

• hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

• attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The BlipForConditionalGeneration 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.

Examples:

>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, BlipForConditionalGeneration

>>> processor = AutoProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
>>> model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)
>>> text = "A picture of"

>>> inputs = processor(images=image, text=text, return_tensors="pt")

>>> outputs = model(**inputs)

## BlipForImageTextRetrieval

### class transformers.BlipForImageTextRetrieval

< >

( config: BlipConfig )

Parameters

• config (BlipConfig) — 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 from_pretrained() method to load the model weights.

BLIP Model with a vision and text projector, and a classification head on top. The model is used in the context of image-text retrieval. Given an image and a text, the model returns the probability of the text being relevant to the image.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

#### forward

< >

( input_ids: LongTensor pixel_values: FloatTensor use_itm_head: typing.Optional[bool] = True attention_mask: typing.Optional[torch.LongTensor] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None return_dict: typing.Optional[bool] = None ) transformers.models.blip.modeling_blip.BlipTextVisionModelOutput or tuple(torch.FloatTensor)

Parameters

• pixel_values (torch.FloatTensor of shape (batch_size, num_channels, height, width)) — Pixel values. Padding will be ignored by default should you provide it. Pixel values can be obtained using BlipImageProcessor. See BlipImageProcessor.call() for details.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.

Returns

transformers.models.blip.modeling_blip.BlipTextVisionModelOutput or tuple(torch.FloatTensor)

A transformers.models.blip.modeling_blip.BlipTextVisionModelOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (<class 'transformers.models.blip.configuration_blip.BlipVisionConfig'>) and inputs.

• loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Languge modeling loss from the text decoder.

• image_embeds (torch.FloatTensor of shape (batch_size, output_dim) optional returned when model is initialized with with_projection=True) — The image embeddings obtained by applying the projection layer to the pooler_output.

• last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

• hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

• attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The BlipForImageTextRetrieval 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.

Examples:

>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, BlipForImageTextRetrieval

>>> model = BlipForImageTextRetrieval.from_pretrained("Salesforce/blip-itm-base-coco")
>>> processor = AutoProcessor.from_pretrained("Salesforce/blip-itm-base-coco")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)
>>> text = "an image of a cat"

>>> inputs = processor(images=image, text=text, return_tensors="pt")
>>> outputs = model(**inputs)

< >

( config: BlipConfig )

Parameters

• config (BlipConfig) — 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 from_pretrained() method to load the model weights.

BLIP Model for visual question answering. The model consists of a vision encoder, a text encoder as well as a text decoder. The vision encoder will encode the input image, the text encoder will encode the input question together with the encoding of the image, and the text decoder will output the answer to the question.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

#### forward

< >

( input_ids: LongTensor pixel_values: FloatTensor decoder_input_ids: typing.Optional[torch.LongTensor] = None decoder_attention_mask: typing.Optional[torch.LongTensor] = None attention_mask: typing.Optional[torch.LongTensor] = None output_attentions: typing.Optional[bool] = None output_hidden_states: typing.Optional[bool] = None labels: typing.Optional[torch.LongTensor] = None return_dict: typing.Optional[bool] = None ) transformers.models.blip.modeling_blip.BlipTextVisionModelOutput or tuple(torch.FloatTensor)

Parameters

• pixel_values (torch.FloatTensor of shape (batch_size, num_channels, height, width)) — Pixel values. Padding will be ignored by default should you provide it. Pixel values can be obtained using BlipImageProcessor. See BlipImageProcessor.call() for details.
• output_attentions (bool, optional) — Whether or not to return the attentions tensors of all attention layers. See attentions under returned tensors for more detail.
• output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
• return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.

Returns

transformers.models.blip.modeling_blip.BlipTextVisionModelOutput or tuple(torch.FloatTensor)

A transformers.models.blip.modeling_blip.BlipTextVisionModelOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (<class 'transformers.models.blip.configuration_blip.BlipVisionConfig'>) and inputs.

• loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Languge modeling loss from the text decoder.

• image_embeds (torch.FloatTensor of shape (batch_size, output_dim) optional returned when model is initialized with with_projection=True) — The image embeddings obtained by applying the projection layer to the pooler_output.

• last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

• hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

• attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

The BlipForQuestionAnswering 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.

Examples:

>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, BlipForQuestionAnswering

>>> processor = AutoProcessor.from_pretrained("Salesforce/blip-vqa-base")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> # training
>>> text = "How many cats are in the picture?"
>>> label = "2"
>>> inputs = processor(images=image, text=text, return_tensors="pt")
>>> labels = processor(text=label, return_tensors="pt").input_ids

>>> inputs["labels"] = labels
>>> outputs = model(**inputs)
>>> loss = outputs.loss
>>> loss.backward()

>>> # inference
>>> text = "How many cats are in the picture?"
>>> inputs = processor(images=image, text=text, return_tensors="pt")
>>> outputs = model.generate(**inputs)
>>> print(processor.decode(outputs[0], skip_special_tokens=True))
2`