modeling/siglip/configuration_siglip.py

# Copyright 2024 The HuggingFace Inc. team.
# SPDX-License-Identifier: Apache-2.0

"""Siglip model configuration"""

import os
from typing import Union

from transformers.configuration_utils import PretrainedConfig
from transformers.utils import logging


logger = logging.get_logger(__name__)


class SiglipTextConfig(PretrainedConfig):
    r"""
    This is the configuration class to store the configuration of a [`SiglipTextModel`]. It is used to instantiate a
    Siglip 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 Siglip
    [google/siglip-base-patch16-224](https://huggingface.co/google/siglip-base-patch16-224) architecture.

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

    Args:
        vocab_size (`int`, *optional*, defaults to 32000):
            Vocabulary size of the Siglip text model. Defines the number of different tokens that can be represented by
            the `inputs_ids` passed when calling [`SiglipModel`].
        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.
        max_position_embeddings (`int`, *optional*, defaults to 64):
            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_pytorch_tanh"`):
            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-06):
            The epsilon used by the layer normalization layers.
        attention_dropout (`float`, *optional*, defaults to 0.0):
            The dropout ratio for the attention probabilities.
        pad_token_id (`int`, *optional*, defaults to 1):
            The id of the padding token in the vocabulary.
        bos_token_id (`int`, *optional*, defaults to 49406):
            The id of the beginning-of-sequence token in the vocabulary.
        eos_token_id (`int`, *optional*, defaults to 49407):
            The id of the end-of-sequence token in the vocabulary.

    Example:

    ```python
    >>> from transformers import SiglipTextConfig, SiglipTextModel

    >>> # Initializing a SiglipTextConfig with google/siglip-base-patch16-224 style configuration
    >>> configuration = SiglipTextConfig()

    >>> # Initializing a SiglipTextModel (with random weights) from the google/siglip-base-patch16-224 style configuration
    >>> model = SiglipTextModel(configuration)

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

    model_type = "siglip_text_model"

    def __init__(
        self,
        vocab_size=32000,
        hidden_size=768,
        intermediate_size=3072,
        num_hidden_layers=12,
        num_attention_heads=12,
        max_position_embeddings=64,
        hidden_act="gelu_pytorch_tanh",
        layer_norm_eps=1e-6,
        attention_dropout=0.0,
        # This differs from `CLIPTokenizer`'s default and from openai/siglip
        # See https://github.com/huggingface/transformers/pull/24773#issuecomment-1632287538
        pad_token_id=1,
        bos_token_id=49406,
        eos_token_id=49407,
        **kwargs,
    ):
        super().__init__(pad_token_id=pad_token_id, bos_token_id=bos_token_id, eos_token_id=eos_token_id, **kwargs)

        self.vocab_size = vocab_size
        self.hidden_size = hidden_size
        self.intermediate_size = intermediate_size
        self.num_hidden_layers = num_hidden_layers
        self.num_attention_heads = num_attention_heads
        self.max_position_embeddings = max_position_embeddings
        self.layer_norm_eps = layer_norm_eps
        self.hidden_act = hidden_act
        self.attention_dropout = attention_dropout

    @classmethod
    def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os.PathLike], **kwargs) -> "PretrainedConfig":
        cls._set_token_in_kwargs(kwargs)

        config_dict, kwargs = cls.get_config_dict(pretrained_model_name_or_path, **kwargs)

        # get the text config dict if we are loading from SiglipConfig
        if config_dict.get("model_type") == "siglip":
            config_dict = config_dict["text_config"]

        if "model_type" in config_dict and hasattr(cls, "model_type") and config_dict["model_type"] != cls.model_type:
            logger.warning(
                f"You are using a model of type {config_dict['model_type']} to instantiate a model of type "
                f"{cls.model_type}. This is not supported for all configurations of models and can yield errors."
            )

        return cls.from_dict(config_dict, **kwargs)


class SiglipVisionConfig(PretrainedConfig):
    r"""
    This is the configuration class to store the configuration of a [`SiglipVisionModel`]. It is used to instantiate a
    Siglip 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 Siglip
    [google/siglip-base-patch16-224](https://huggingface.co/google/siglip-base-patch16-224) architecture.

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

    Args:
        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.
        num_channels (`int`, *optional*, defaults to 3):
            Number of channels in the input images.
        image_size (`int`, *optional*, defaults to 224):
            The size (resolution) of each image.
        patch_size (`int`, *optional*, defaults to 16):
            The size (resolution) of each patch.
        hidden_act (`str` or `function`, *optional*, defaults to `"gelu_pytorch_tanh"`):
            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-06):
            The epsilon used by the layer normalization layers.
        attention_dropout (`float`, *optional*, defaults to 0.0):
            The dropout ratio for the attention probabilities.

    Example:

    ```python
    >>> from transformers import SiglipVisionConfig, SiglipVisionModel

    >>> # Initializing a SiglipVisionConfig with google/siglip-base-patch16-224 style configuration
    >>> configuration = SiglipVisionConfig()

    >>> # Initializing a SiglipVisionModel (with random weights) from the google/siglip-base-patch16-224 style configuration
    >>> model = SiglipVisionModel(configuration)

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

    model_type = "siglip_vision_model"

    def __init__(
        self,
        hidden_size=768,
        intermediate_size=3072,
        num_hidden_layers=12,
        num_attention_heads=12,
        num_channels=3,
        image_size=224,
        patch_size=16,
        hidden_act="gelu_pytorch_tanh",
        layer_norm_eps=1e-6,
        attention_dropout=0.0,
        **kwargs,
    ):
        super().__init__(**kwargs)

        self.hidden_size = hidden_size
        self.intermediate_size = intermediate_size
        self.num_hidden_layers = num_hidden_layers
        self.num_attention_heads = num_attention_heads
        self.num_channels = num_channels
        self.patch_size = patch_size
        self.image_size = image_size
        self.attention_dropout = attention_dropout
        self.layer_norm_eps = layer_norm_eps
        self.hidden_act = hidden_act

    @classmethod
    def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os.PathLike], **kwargs) -> "PretrainedConfig":
        cls._set_token_in_kwargs(kwargs)

        config_dict, kwargs = cls.get_config_dict(pretrained_model_name_or_path, **kwargs)

        # get the vision config dict if we are loading from SiglipConfig
        if config_dict.get("model_type") == "siglip":
            config_dict = config_dict["vision_config"]

        if "model_type" in config_dict and hasattr(cls, "model_type") and config_dict["model_type"] != cls.model_type:
            logger.warning(
                f"You are using a model of type {config_dict['model_type']} to instantiate a model of type "
                f"{cls.model_type}. This is not supported for all configurations of models and can yield errors."
            )

        return cls.from_dict(config_dict, **kwargs)


class SiglipConfig(PretrainedConfig):
    r"""
    [`SiglipConfig`] is the configuration class to store the configuration of a [`SiglipModel`]. It is used to
    instantiate a Siglip 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 Siglip
    [google/siglip-base-patch16-224](https://huggingface.co/google/siglip-base-patch16-224) architecture.

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

    Args:
        text_config (`dict`, *optional*):
            Dictionary of configuration options used to initialize [`SiglipTextConfig`].
        vision_config (`dict`, *optional*):
            Dictionary of configuration options used to initialize [`SiglipVisionConfig`].
        kwargs (*optional*):
            Dictionary of keyword arguments.

    Example:

    ```python
    >>> from transformers import SiglipConfig, SiglipModel

    >>> # Initializing a SiglipConfig with google/siglip-base-patch16-224 style configuration
    >>> configuration = SiglipConfig()

    >>> # Initializing a SiglipModel (with random weights) from the google/siglip-base-patch16-224 style configuration
    >>> model = SiglipModel(configuration)

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

    >>> # We can also initialize a SiglipConfig from a SiglipTextConfig and a SiglipVisionConfig
    >>> from transformers import SiglipTextConfig, SiglipVisionConfig

    >>> # Initializing a SiglipText and SiglipVision configuration
    >>> config_text = SiglipTextConfig()
    >>> config_vision = SiglipVisionConfig()

    >>> config = SiglipConfig.from_text_vision_configs(config_text, config_vision)
    ```"""

    model_type = "siglip"

    def __init__(self, text_config=None, vision_config=None, **kwargs):
        super().__init__(**kwargs)

        if text_config is None:
            text_config = {}
            logger.info("`text_config` is `None`. Initializing the `SiglipTextConfig` with default values.")

        if vision_config is None:
            vision_config = {}
            logger.info("`vision_config` is `None`. initializing the `SiglipVisionConfig` with default values.")

        self.text_config = SiglipTextConfig(**text_config)
        self.vision_config = SiglipVisionConfig(**vision_config)

        self.initializer_factor = 1.0

    @classmethod
    def from_text_vision_configs(cls, text_config: SiglipTextConfig, vision_config: SiglipVisionConfig, **kwargs):
        r"""
        Instantiate a [`SiglipConfig`] (or a derived class) from siglip text model configuration and siglip vision
        model configuration.

        Returns:
            [`SiglipConfig`]: An instance of a configuration object
        """

        return cls(text_config=text_config.to_dict(), vision_config=vision_config.to_dict(), **kwargs)