AltCLIP
概要
AltCLIPモデルは、「AltCLIP: Altering the Language Encoder in CLIP for Extended Language Capabilities」という論文でZhongzhi Chen、Guang Liu、Bo-Wen Zhang、Fulong Ye、Qinghong Yang、Ledell Wuによって提案されました。AltCLIP(CLIPの言語エンコーダーの代替)は、様々な画像-テキストペアおよびテキスト-テキストペアでトレーニングされたニューラルネットワークです。CLIPのテキストエンコーダーを事前学習済みの多言語テキストエンコーダーXLM-Rに置き換えることで、ほぼ全てのタスクでCLIPに非常に近い性能を得られ、オリジナルのCLIPの能力を多言語理解などに拡張しました。
論文の要旨は以下の通りです:
この研究では、強力なバイリンガルマルチモーダル表現モデルを訓練するための概念的に単純で効果的な方法を提案します。OpenAIによってリリースされたマルチモーダル表現モデルCLIPから開始し、そのテキストエンコーダを事前学習済みの多言語テキストエンコーダXLM-Rに交換し、教師学習と対照学習からなる2段階のトレーニングスキーマを用いて言語と画像の表現を整合させました。幅広いタスクの評価を通じて、我々の方法を検証します。ImageNet-CN、Flicker30k-CN、COCO-CNを含む多くのタスクで新たな最先端の性能を達成しました。さらに、ほぼすべてのタスクでCLIPに非常に近い性能を得ており、これはCLIPのテキストエンコーダを変更するだけで、多言語理解などの拡張を実現できることを示唆しています。
このモデルはjongjyhにより提供されました。
使用上のヒントと使用例
AltCLIPの使用方法はCLIPに非常に似ています。CLIPとの違いはテキストエンコーダーにあります。私たちはカジュアルアテンションではなく双方向アテンションを使用し、XLM-Rの[CLS]トークンをテキスト埋め込みを表すものとして取ることに留意してください。
AltCLIPはマルチモーダルな視覚言語モデルです。これは画像とテキストの類似度や、ゼロショット画像分類に使用できます。AltCLIPはViTのようなTransformerを使用して視覚的特徴を、双方向言語モデルを使用してテキスト特徴を取得します。テキストと視覚の両方の特徴は、同一の次元を持つ潜在空間に射影されます。射影された画像とテキスト特徴間のドット積が類似度スコアとして使用されます。
Transformerエンコーダーに画像を与えるには、各画像を固定サイズの重複しないパッチの系列に分割し、それらを線形に埋め込みます。画像全体を表現するための[CLS]トークンが追加されます。著者は絶対位置埋め込みも追加し、結果として得られるベクトルの系列を標準的なTransformerエンコーダーに供給します。CLIPImageProcessorを使用して、モデルのために画像のサイズ変更(または拡大縮小)と正規化を行うことができます。
AltCLIPProcessorは、テキストのエンコードと画像の前処理を両方行うために、CLIPImageProcessorとXLMRobertaTokenizer
を単一のインスタンスにラップします。以下の例は、AltCLIPProcessorとAltCLIPModelを使用して画像-テキスト類似スコアを取得する方法を示しています。
>>> from PIL import Image
>>> import requests
>>> from transformers import AltCLIPModel, AltCLIPProcessor
>>> model = AltCLIPModel.from_pretrained("BAAI/AltCLIP")
>>> processor = AltCLIPProcessor.from_pretrained("BAAI/AltCLIP")
>>> 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
このモデルはCLIPModel
をベースにしており、オリジナルのCLIPと同じように使用してください。
AltCLIPConfig
class transformers.AltCLIPConfig
< source >( text_config = None vision_config = None projection_dim = 768 logit_scale_init_value = 2.6592 **kwargs )
Parameters
- text_config (
dict
, optional) — Dictionary of configuration options used to initialize AltCLIPTextConfig. - vision_config (
dict
, optional) — Dictionary of configuration options used to initialize AltCLIPVisionConfig. - projection_dim (
int
, optional, defaults to 768) — 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 CLIP implementation. - kwargs (optional) — Dictionary of keyword arguments.
This is the configuration class to store the configuration of a AltCLIPModel. It is used to instantiate an AltCLIP 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 AltCLIP BAAI/AltCLIP 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 AltCLIPConfig, AltCLIPModel
>>> # Initializing a AltCLIPConfig with BAAI/AltCLIP style configuration
>>> configuration = AltCLIPConfig()
>>> # Initializing a AltCLIPModel (with random weights) from the BAAI/AltCLIP style configuration
>>> model = AltCLIPModel(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config
>>> # We can also initialize a AltCLIPConfig from a AltCLIPTextConfig and a AltCLIPVisionConfig
>>> # Initializing a AltCLIPText and AltCLIPVision configuration
>>> config_text = AltCLIPTextConfig()
>>> config_vision = AltCLIPVisionConfig()
>>> config = AltCLIPConfig.from_text_vision_configs(config_text, config_vision)
from_text_vision_configs
< source >( text_config: AltCLIPTextConfig vision_config: AltCLIPVisionConfig **kwargs ) → AltCLIPConfig
Instantiate a AltCLIPConfig (or a derived class) from altclip text model configuration and altclip vision model configuration.
AltCLIPTextConfig
class transformers.AltCLIPTextConfig
< source >( vocab_size = 250002 hidden_size = 1024 num_hidden_layers = 24 num_attention_heads = 16 intermediate_size = 4096 hidden_act = 'gelu' hidden_dropout_prob = 0.1 attention_probs_dropout_prob = 0.1 max_position_embeddings = 514 type_vocab_size = 1 initializer_range = 0.02 initializer_factor = 0.02 layer_norm_eps = 1e-05 pad_token_id = 1 bos_token_id = 0 eos_token_id = 2 position_embedding_type = 'absolute' use_cache = True project_dim = 768 **kwargs )
Parameters
- vocab_size (
int
, optional, defaults to 250002) — Vocabulary size of the AltCLIP model. Defines the number of different tokens that can be represented by theinputs_ids
passed when calling AltCLIPTextModel. - hidden_size (
int
, optional, defaults to 1024) — Dimensionality of the encoder layers and the pooler layer. - num_hidden_layers (
int
, optional, defaults to 24) — Number of hidden layers in the Transformer encoder. - num_attention_heads (
int
, optional, defaults to 16) — Number of attention heads for each attention layer in the Transformer encoder. - intermediate_size (
int
, optional, defaults to 4096) — Dimensionality of the “intermediate” (often named feed-forward) layer in the Transformer encoder. - hidden_act (
str
orCallable
, optional, defaults to"gelu"
) — The non-linear activation function (function or string) in the encoder and pooler. If string,"gelu"
,"relu"
,"silu"
and"gelu_new"
are supported. - hidden_dropout_prob (
float
, optional, defaults to 0.1) — The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. - attention_probs_dropout_prob (
float
, optional, defaults to 0.1) — The dropout ratio for the attention probabilities. - max_position_embeddings (
int
, optional, defaults to 514) — 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). - type_vocab_size (
int
, optional, defaults to 1) — The vocabulary size of thetoken_type_ids
passed when calling AltCLIPTextModel - initializer_range (
float
, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices. - initializer_factor (
float
, optional, defaults to 0.02) — A factor for initializing all weight matrices (should be kept to 1, used internally for initialization testing). - layer_norm_eps (
float
, optional, defaults to 1e-05) — The epsilon used by the layer normalization layers. - pad_token_id (
int
, optional, defaults to 1) — The id of the padding token. - bos_token_id (
int
, optional, defaults to 0) — The id of the beginning-of-sequence token. - eos_token_id (
Union[int, List[int]]
, optional, defaults to 2) — The id of the end-of-sequence token. Optionally, use a list to set multiple end-of-sequence tokens. - position_embedding_type (
str
, optional, defaults to"absolute"
) — Type of position embedding. Choose one of"absolute"
,"relative_key"
,"relative_key_query"
. For positional embeddings use"absolute"
. For more information on"relative_key"
, please refer to Self-Attention with Relative Position Representations (Shaw et al.). For more information on"relative_key_query"
, please refer to Method 4 in Improve Transformer Models with Better Relative Position Embeddings (Huang et al.). - use_cache (
bool
, optional, defaults toTrue
) — Whether or not the model should return the last key/values attentions (not used by all models). Only relevant ifconfig.is_decoder=True
. - project_dim (
int
, optional, defaults to 768) — The dimentions of the teacher model before the mapping layer.
This is the configuration class to store the configuration of a AltCLIPTextModel. It is used to instantiate a AltCLIP 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 AltCLIP BAAI/AltCLIP architecture.
Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.
Examples:
>>> from transformers import AltCLIPTextModel, AltCLIPTextConfig
>>> # Initializing a AltCLIPTextConfig with BAAI/AltCLIP style configuration
>>> configuration = AltCLIPTextConfig()
>>> # Initializing a AltCLIPTextModel (with random weights) from the BAAI/AltCLIP style configuration
>>> model = AltCLIPTextModel(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config
AltCLIPVisionConfig
class transformers.AltCLIPVisionConfig
< source >( hidden_size = 768 intermediate_size = 3072 projection_dim = 512 num_hidden_layers = 12 num_attention_heads = 12 num_channels = 3 image_size = 224 patch_size = 32 hidden_act = 'quick_gelu' layer_norm_eps = 1e-05 attention_dropout = 0.0 initializer_range = 0.02 initializer_factor = 1.0 **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. - projection_dim (
int
, optional, defaults to 512) — Dimentionality of text and vision projection layers. - 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. - num_channels (
int
, optional, defaults to 3) — The number of input channels. - 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
orfunction
, optional, defaults to"quick_gelu"
) — The non-linear activation function (function or string) in the encoder and pooler. If string,"gelu"
,"relu"
,"selu"
and"gelu_new"
`"quick_gelu"
are supported. - layer_norm_eps (
float
, optional, defaults to 1e-05) — 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. - initializer_factor (
float
, optional, defaults to 1.0) — A factor for initializing all weight matrices (should be kept to 1, used internally for initialization testing).
This is the configuration class to store the configuration of a AltCLIPModel. It is used to instantiate an AltCLIP 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 AltCLIP BAAI/AltCLIP 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 AltCLIPVisionConfig, AltCLIPVisionModel
>>> # Initializing a AltCLIPVisionConfig with BAAI/AltCLIP style configuration
>>> configuration = AltCLIPVisionConfig()
>>> # Initializing a AltCLIPVisionModel (with random weights) from the BAAI/AltCLIP style configuration
>>> model = AltCLIPVisionModel(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config
AltCLIPProcessor
class transformers.AltCLIPProcessor
< source >( image_processor = None tokenizer = None **kwargs )
Parameters
- image_processor (CLIPImageProcessor, optional) — The image processor is a required input.
- tokenizer (
XLMRobertaTokenizerFast
, optional) — The tokenizer is a required input.
Constructs a AltCLIP processor which wraps a CLIP image processor and a XLM-Roberta tokenizer into a single processor.
AltCLIPProcessor offers all the functionalities of CLIPImageProcessor and XLMRobertaTokenizerFast
. See
the __call__()
and decode() for more information.
This method forwards all its arguments to XLMRobertaTokenizerFast’s batch_decode(). Please refer to the docstring of this method for more information.
This method forwards all its arguments to XLMRobertaTokenizerFast’s decode(). Please refer to the docstring of this method for more information.
AltCLIPModel
forward
< source >( input_ids: Optional = None pixel_values: Optional = None attention_mask: Optional = None position_ids: Optional = None token_type_ids: Optional = None return_loss: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = None ) → transformers.models.altclip.modeling_altclip.AltCLIPOutput
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 AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- 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]
. - 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 AutoImageProcessor. See CLIPImageProcessor.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. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_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.altclip.modeling_altclip.AltCLIPOutput
or tuple(torch.FloatTensor)
A transformers.models.altclip.modeling_altclip.AltCLIPOutput
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.altclip.configuration_altclip.AltCLIPConfig'>
) and inputs.
- loss (
torch.FloatTensor
of shape(1,)
, optional, returned whenreturn_loss
isTrue
) — Contrastive loss for image-text similarity. - logits_per_image:(
torch.FloatTensor
of shape(image_batch_size, text_batch_size)
) — The scaled dot product scores betweenimage_embeds
andtext_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 betweentext_embeds
andimage_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 AltCLIPTextModel. - image_embeds(
torch.FloatTensor
of shape(batch_size, output_dim
) — The image embeddings obtained by applying the projection layer to the pooled output of AltCLIPVisionModel. - text_model_output(
BaseModelOutputWithPooling
): The output of the AltCLIPTextModel. - vision_model_output(
BaseModelOutputWithPooling
): The output of the AltCLIPVisionModel.
The AltCLIPModel 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, AltCLIPModel
>>> model = AltCLIPModel.from_pretrained("BAAI/AltCLIP")
>>> processor = AutoProcessor.from_pretrained("BAAI/AltCLIP")
>>> 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
< source >( input_ids: Optional = None attention_mask: Optional = None position_ids: Optional = None token_type_ids = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = 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 AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- 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]
. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_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 AltCLIPTextModel.
The AltCLIPModel 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, AltCLIPModel
>>> model = AltCLIPModel.from_pretrained("BAAI/AltCLIP")
>>> processor = AutoProcessor.from_pretrained("BAAI/AltCLIP")
>>> 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
< source >( pixel_values: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = 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 AutoImageProcessor. See CLIPImageProcessor.call() for details. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_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 AltCLIPVisionModel.
The AltCLIPModel 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, AltCLIPModel
>>> model = AltCLIPModel.from_pretrained("BAAI/AltCLIP")
>>> processor = AutoProcessor.from_pretrained("BAAI/AltCLIP")
>>> 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)
AltCLIPTextModel
forward
< source >( input_ids: Optional = None attention_mask: Optional = None token_type_ids: Optional = None position_ids: Optional = None head_mask: Optional = None inputs_embeds: Optional = None encoder_hidden_states: Optional = None encoder_attention_mask: Optional = None output_attentions: Optional = None return_dict: Optional = None output_hidden_states: Optional = None ) → transformers.modeling_outputs.BaseModelOutputWithPoolingAndProjection
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 AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.
- 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]
. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_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.BaseModelOutputWithPoolingAndProjection
or tuple(torch.FloatTensor)
A transformers.modeling_outputs.BaseModelOutputWithPoolingAndProjection
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.altclip.configuration_altclip.AltCLIPTextConfig'>
) 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 whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) — Tuple oftorch.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 whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple oftorch.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.
-
projection_state (
tuple(torch.FloatTensor)
, returned whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple oftorch.FloatTensor
of shape(batch_size,config.project_dim)
.Text embeddings before the projection layer, used to mimic the last hidden state of the teacher encoder.
The AltCLIPTextModel 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, AltCLIPTextModel
>>> model = AltCLIPTextModel.from_pretrained("BAAI/AltCLIP")
>>> processor = AutoProcessor.from_pretrained("BAAI/AltCLIP")
>>> texts = ["it's a cat", "it's a dog"]
>>> inputs = processor(text=texts, padding=True, return_tensors="pt")
>>> outputs = model(**inputs)
>>> last_hidden_state = outputs.last_hidden_state
>>> pooled_output = outputs.pooler_output # pooled CLS states
AltCLIPVisionModel
forward
< source >( pixel_values: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = 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 AutoImageProcessor. See CLIPImageProcessor.call() for details. - output_attentions (
bool
, optional) — Whether or not to return the attentions tensors of all attention layers. Seeattentions
under returned tensors for more detail. - output_hidden_states (
bool
, optional) — Whether or not to return the hidden states of all layers. Seehidden_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.altclip.configuration_altclip.AltCLIPVisionConfig'>
) 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 whenoutput_hidden_states=True
is passed or whenconfig.output_hidden_states=True
) — Tuple oftorch.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 whenoutput_attentions=True
is passed or whenconfig.output_attentions=True
) — Tuple oftorch.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 AltCLIPVisionModel 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, AltCLIPVisionModel
>>> model = AltCLIPVisionModel.from_pretrained("BAAI/AltCLIP")
>>> processor = AutoProcessor.from_pretrained("BAAI/AltCLIP")
>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)
>>> inputs = processor(images=image, return_tensors="pt")
>>> outputs = model(**inputs)
>>> last_hidden_state = outputs.last_hidden_state
>>> pooled_output = outputs.pooler_output # pooled CLS states