image-match/python/py/Lib/site-packages/transformers/models/siglip2/configuration_siglip2.py

#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
#           This file was automatically generated from src/transformers/models/siglip2/modular_siglip2.py.
#               Do NOT edit this file manually as any edits will be overwritten by the generation of
#             the file from the modular. If any change should be done, please apply the change to the
#                          modular_siglip2.py file directly. One of our CI enforces this.
#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
# Copyright 2025 The HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.


from huggingface_hub.dataclasses import strict

from ...configuration_utils import PreTrainedConfig
from ...utils import auto_docstring, logging


logger = logging.get_logger(__name__)


@auto_docstring(checkpoint="google/siglip2-base-patch16-naflex")
@strict
class Siglip2TextConfig(PreTrainedConfig):
    r"""
    Example:

    ```python
    >>> from transformers import Siglip2TextConfig, Siglip2TextModel

    >>> # Initializing a Siglip2TextConfig with google/siglip2-base-patch16-224 style configuration
    >>> configuration = Siglip2TextConfig()

    >>> # Initializing a Siglip2TextModel (with random weights) from the google/siglip2-base-patch16-224 style configuration
    >>> model = Siglip2TextModel(configuration)

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

    model_type = "siglip2_text_model"
    base_config_key = "text_config"

    vocab_size: int = 32000
    hidden_size: int = 768
    intermediate_size: int = 3072
    num_hidden_layers: int = 12
    num_attention_heads: int = 12
    max_position_embeddings: int = 64
    hidden_act: str = "gelu_pytorch_tanh"
    layer_norm_eps: float = 1e-6
    attention_dropout: float | int = 0.0
    # This differs from `CLIPTokenizer`'s default and from openai/siglip2
    # See https://github.com/huggingface/transformers/pull/24773#issuecomment-1632287538
    pad_token_id: int | None = 1
    bos_token_id: int | None = 49406
    eos_token_id: int | list[int] | None = 49407
    projection_size: int | None = None

    def __post_init__(self, **kwargs):
        self.projection_size = self.projection_size if self.projection_size is not None else self.hidden_size
        super().__post_init__(**kwargs)


@auto_docstring(checkpoint="google/siglip2-base-patch16-naflex")
@strict
class Siglip2VisionConfig(PreTrainedConfig):
    r"""
    num_patches (`int`, *optional*, defaults to 256):
        The number of patches in the image with the size of (`patch_size`, `patch_size`).
        The image is resized to fill maximum of this number of patches, and to preserve
        the aspect ratio. In case the resulted number of patches is lower, the image is
        padded in "patch" dimension.

    Example:

    ```python
    >>> from transformers import Siglip2VisionConfig, Siglip2VisionModel

    >>> # Initializing a Siglip2VisionConfig with google/siglip2-base-patch16-naflex style configuration
    >>> configuration = Siglip2VisionConfig()

    >>> # Initializing a Siglip2VisionModel (with random weights) from the google/siglip2-base-patch16-naflex style configuration
    >>> model = Siglip2VisionModel(configuration)

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

    model_type = "siglip2_vision_model"
    base_config_key = "vision_config"

    hidden_size: int = 768
    intermediate_size: int = 3072
    num_hidden_layers: int = 12
    num_attention_heads: int = 12
    num_channels: int = 3
    patch_size: int | list[int] | tuple[int, int] = 16
    hidden_act: str = "gelu_pytorch_tanh"
    layer_norm_eps: float = 1e-6
    attention_dropout: float | int = 0.0

    num_patches: int = 256


@auto_docstring(checkpoint="google/siglip2-base-patch16-naflex")
@strict
class Siglip2Config(PreTrainedConfig):
    r"""
    Example:

    ```python
    >>> from transformers import Siglip2Config, Siglip2Model

    >>> # Initializing a Siglip2Config with google/siglip2-base-patch16-224 style configuration
    >>> configuration = Siglip2Config()

    >>> # Initializing a Siglip2Model (with random weights) from the google/siglip2-base-patch16-224 style configuration
    >>> model = Siglip2Model(configuration)

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

    >>> # We can also initialize a Siglip2Config from a Siglip2TextConfig and a Siglip2VisionConfig
    >>> from transformers import Siglip2TextConfig, Siglip2VisionConfig

    >>> # Initializing a Siglip2Text and Siglip2Vision configuration
    >>> config_text = Siglip2TextConfig()
    >>> config_vision = Siglip2VisionConfig()

    >>> config = Siglip2Config(text_config=config_text, vision_config=config_vision)
    ```"""

    model_type = "siglip2"
    sub_configs = {"text_config": Siglip2TextConfig, "vision_config": Siglip2VisionConfig}

    text_config: dict | PreTrainedConfig | None = None
    vision_config: dict | PreTrainedConfig | None = None
    initializer_factor: float = 1.0

    def __post_init__(self, **kwargs):
        if self.text_config is None:
            self.text_config = Siglip2TextConfig()
            logger.info("`text_config` is `None`. Initializing the `Siglip2TextConfig` with default values.")
        elif isinstance(self.text_config, dict):
            self.text_config = Siglip2TextConfig(**self.text_config)

        if self.vision_config is None:
            self.vision_config = Siglip2VisionConfig()
            logger.info("`vision_config` is `None`. initializing the `Siglip2VisionConfig` with default values.")
        elif isinstance(self.vision_config, dict):
            self.vision_config = Siglip2VisionConfig(**self.vision_config)

        super().__post_init__(**kwargs)


__all__ = ["Siglip2Config", "Siglip2TextConfig", "Siglip2VisionConfig"]