easydel.modules.clip.__init__

class easydel.modules.clip.__init__.CLIPConfig(text_config=None, vision_config=None, projection_dim=512, logit_scale_init_value=2.6592, **kwargs)

Bases: EasyDeLBaseConfig

[CLIPConfig] is the configuration class to store the configuration of a [CLIPModel]. It is used to instantiate a CLIP 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 CLIP [openai/clip-vit-base-patch32](https://huggingface.co/openai/clip-vit-base-patch32) architecture.

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

Parameters

• text_config (dict, optional) – Dictionary of configuration options used to initialize [CLIPTextConfig].

• vision_config (dict, optional) – Dictionary of configuration options used to initialize [CLIPVisionConfig].

• projection_dim (int, optional, defaults to 512) – Dimensionality of text and vision projection layers.

• logit_scale_init_value (float, optional, defaults to 2.6592) – The initial value of the logit_scale parameter. Default is used as per the original CLIP implementation.

• kwargs (optional) – Dictionary of keyword arguments.

Example:

```python >>> from transformers import CLIPConfig, CLIPModel

>>> # Initializing a CLIPConfig with openai/clip-vit-base-patch32 style configuration
>>> configuration = CLIPConfig()

>>> # Initializing a CLIPModel (with random weights) from the openai/clip-vit-base-patch32 style configuration
>>> model = CLIPModel(configuration)

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

>>> # We can also initialize a CLIPConfig from a CLIPTextConfig and a CLIPVisionConfig
>>> from transformers import CLIPTextConfig, CLIPVisionConfig

>>> # Initializing a CLIPText and CLIPVision configuration
>>> config_text = CLIPTextConfig()
>>> config_vision = CLIPVisionConfig()

>>> config = CLIPConfig.from_text_vision_configs(config_text, config_vision)
```

classmethod from_text_vision_configs(text_config: CLIPTextConfig, vision_config: CLIPVisionConfig, **kwargs)

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

Parameters

• text_config (CLIPTextConfig) – The text model configuration.

• vision_config (CLIPVisionConfig) – The vision model configuration.

• **kwargs – Additional keyword arguments.

Returns

An instance of a configuration object

Return type

[CLIPConfig]

get_partition_rules(*arg, **kwargs)

Generic partition rules for CLIP text and vision models.

Parameters

• self – The configuration object (unused but part of method signature).

• *arg – Additional positional arguments (unused).

• **kwargs – Additional keyword arguments (unused).

Returns

A tuple of partition rules for model parameters.

Return type

Tuple

model_type: str = 'clip'

sub_configs: dict[str, 'PretrainedConfig'] = {'text_config': <class 'easydel.modules.clip.clip_configuration.CLIPTextConfig'>, 'vision_config': <class 'easydel.modules.clip.clip_configuration.CLIPVisionConfig'>}

class easydel.modules.clip.__init__.CLIPForImageClassification(*args: Any, **kwargs: Any)

Bases: EasyDeLBaseModule

CLIP vision model with an image classification head on top (a linear layer on the pooled final hidden state).

config

Configuration object.

Type

CLIPVisionConfig

dtype

Data type for computation.

Type

jnp.dtype

param_dtype

Data type for parameters.

Type

jnp.dtype

precision

JAX precision level.

Type

jax.lax.PrecisionLike

rngs

Random number generators.

Type

nn.Rngs

class easydel.modules.clip.__init__.CLIPModel(*args: Any, **kwargs: Any)

Bases: EasyDeLBaseModule

compute_loss(*, labels=None, loss_config=None, loss_kwargs=None, **batch) → Tuple[Any, CLIPOutput]

Computes the loss for the model given a batch of inputs and labels.

This method performs a forward pass using the provided batch arguments, then calculates the loss using the determined loss_function. It handles potential label inference (e.g., using input_ids as labels for Causal LM) and default loss configurations.

Parameters

• labels (tp.Optional[chex.Array], optional) – The target labels. If None and the task is Causal LM, input_ids from the batch might be used. Defaults to None.

• loss_config (tp.Optional[LossConfig], optional) – Specific configuration for the loss calculation. If None, defaults might be inferred (e.g., for sequence classification). Defaults to None.

• loss_kwargs (tp.Optional[tp.Dict], optional) – Additional keyword arguments to pass directly to the loss function. Defaults to None.

• **batch – Keyword arguments representing the input batch (e.g., input_ids, attention_mask).

Returns

A tuple containing:

• The model’s output ( Pytree typically including logits, hidden states etc.)

• A LossMetrics object containing the calculated loss and potentially other metrics.

Return type

tp.Tuple[tp.Any, LossMetrics]

Raises

• AssertionError – If labels are required for the loss function but are not provided or inferred.

• AssertionError – If sequence classification loss is used without num_labels in the config.

get_image_features(pixel_values: Union[Array, ndarray, bool, number])

get_text_features(input_ids: Union[Array, ndarray, bool, number], attention_mask: Optional[Union[Array, ndarray, bool, number]] = None, position_ids: Optional[Union[Array, ndarray, bool, number]] = None)

class easydel.modules.clip.__init__.CLIPTextConfig(vocab_size=49408, hidden_size=512, intermediate_size=2048, projection_dim=512, num_hidden_layers=12, num_attention_heads=8, max_position_embeddings=77, hidden_act='quick_gelu', layer_norm_eps=1e-05, attention_dropout=0.0, initializer_range=0.02, initializer_factor=1.0, pad_token_id=1, bos_token_id=49406, eos_token_id=49407, **kwargs)

Bases: EasyDeLBaseConfig

This is the configuration class to store the configuration of a [CLIPTextModel]. It is used to instantiate a CLIP text encoder according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the text encoder of the CLIP [openai/clip-vit-base-patch32](https://huggingface.co/openai/clip-vit-base-patch32) architecture.

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

Parameters

• vocab_size (int, optional, defaults to 49408) – Vocabulary size of the CLIP text model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling [CLIPModel].

• hidden_size (int, optional, defaults to 512) – Dimensionality of the encoder layers and the pooler layer.

• intermediate_size (int, optional, defaults to 2048) – Dimensionality of the “intermediate” (i.e., feed-forward) layer in the Transformer encoder.

• projection_dim (int, optional, defaults to 512) – Dimensionality 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 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 “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).

• pad_token_id (int, optional, defaults to 1) – Padding token id.

• bos_token_id (int, optional, defaults to 49406) – Beginning of stream token id.

• eos_token_id (int, optional, defaults to 49407) – End of stream token id.

Example:

```python >>> from transformers import CLIPTextConfig, CLIPTextModel

>>> # Initializing a CLIPTextConfig with openai/clip-vit-base-patch32 style configuration
>>> configuration = CLIPTextConfig()

>>> # Initializing a CLIPTextModel (with random weights) from the openai/clip-vit-base-patch32 style configuration
>>> model = CLIPTextModel(configuration)

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

base_config_key: str = 'text_config'

get_partition_rules(*arg, **kwargs)

Generic partition rules for CLIP text and vision models.

Parameters

• self – The configuration object (unused but part of method signature).

• *arg – Additional positional arguments (unused).

• **kwargs – Additional keyword arguments (unused).

Returns

A tuple of partition rules for model parameters.

Return type

Tuple

model_type: str = 'clip_text_model'

class easydel.modules.clip.__init__.CLIPTextModel(*args: Any, **kwargs: Any)

Bases: EasyDeLBaseModule

Bare CLIP text model (transformer) outputting raw hidden-states without any specific head on top.

config

Configuration object.

Type

CLIPTextConfig

dtype

Data type for computation.

Type

jnp.dtype

param_dtype

Data type for parameters.

Type

jnp.dtype

precision

JAX precision level.

Type

jax.lax.PrecisionLike

rngs

Random number generators.

Type

nn.Rngs

class easydel.modules.clip.__init__.CLIPTextModelWithProjection(*args: Any, **kwargs: Any)

Bases: EasyDeLBaseModule

CLIP text model with a projection layer on top.

config

Configuration object.

Type

CLIPTextConfig

dtype

Data type for computation.

Type

jnp.dtype

param_dtype

Data type for parameters.

Type

jnp.dtype

precision

JAX precision level.

Type

jax.lax.PrecisionLike

rngs

Random number generators.

Type

nn.Rngs

class easydel.modules.clip.__init__.CLIPVisionConfig(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)

Bases: EasyDeLBaseConfig

This is the configuration class to store the configuration of a [CLIPVisionModel]. It is used to instantiate a CLIP vision encoder according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the vision encoder of the CLIP [openai/clip-vit-base-patch32](https://huggingface.co/openai/clip-vit-base-patch32) architecture.

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

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) – Dimensionality 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 or function, 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).

Example:

```python >>> from transformers import CLIPVisionConfig, CLIPVisionModel

>>> # Initializing a CLIPVisionConfig with openai/clip-vit-base-patch32 style configuration
>>> configuration = CLIPVisionConfig()

>>> # Initializing a CLIPVisionModel (with random weights) from the openai/clip-vit-base-patch32 style configuration
>>> model = CLIPVisionModel(configuration)

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

base_config_key: str = 'vision_config'

get_partition_rules(*arg, **kwargs)

Generic partition rules for CLIP text and vision models.

Parameters

• self – The configuration object (unused but part of method signature).

• *arg – Additional positional arguments (unused).

• **kwargs – Additional keyword arguments (unused).

Returns

A tuple of partition rules for model parameters.

Return type

Tuple

model_type: str = 'clip_vision_model'

class easydel.modules.clip.__init__.CLIPVisionModel(*args: Any, **kwargs: Any)

Bases: EasyDeLBaseModule

Bare CLIP vision model (transformer) outputting raw hidden-states without any specific head on top.

config

Configuration object.

Type

CLIPVisionConfig

dtype

Data type for computation.

Type

jnp.dtype

param_dtype

Data type for parameters.

Type

jnp.dtype

precision

JAX precision level.

Type

jax.lax.PrecisionLike

rngs

Random number generators.

Type

nn.Rngs