languagebind 0.1.0__tar.gz

languagebind-0.1.0/PKG-INFO

@@ -0,0 +1,71 @@
+Metadata-Version: 2.4
+Name: languagebind
+Version: 0.1.0
+Summary: LanguageBind: multimodal (video/audio/image/depth/thermal) embedding model aligned to text via language-based semantic alignment. Packaged for pip-installability with compatibility patches for modern transformers/torchvision/torchaudio.
+Project-URL: Homepage, https://github.com/PKU-YuanGroup/LanguageBind
+Project-URL: Source, https://github.com/embeddings-benchmark/languagebind
+License: MIT
+Requires-Python: >=3.9
+Requires-Dist: einops>=0.8.0
+Requires-Dist: peft>=0.11.0
+Requires-Dist: torch>=2.0.0
+Requires-Dist: transformers<5.0.0,>=4.40.0
+Provides-Extra: all
+Requires-Dist: decord>=0.6.0; extra == 'all'
+Requires-Dist: opencv-python-headless>=4.5.0; extra == 'all'
+Requires-Dist: pytorchvideo>=0.1.5; extra == 'all'
+Requires-Dist: soundfile>=0.12.0; extra == 'all'
+Requires-Dist: torchaudio>=0.13.0; extra == 'all'
+Requires-Dist: torchvision>=0.15.0; extra == 'all'
+Provides-Extra: audio
+Requires-Dist: soundfile>=0.12.0; extra == 'audio'
+Requires-Dist: torchaudio>=0.13.0; extra == 'audio'
+Provides-Extra: video
+Requires-Dist: decord>=0.6.0; extra == 'video'
+Requires-Dist: opencv-python-headless>=4.5.0; extra == 'video'
+Requires-Dist: pytorchvideo>=0.1.5; extra == 'video'
+Requires-Dist: torchvision>=0.15.0; extra == 'video'
+Description-Content-Type: text/markdown
+
+# languagebind
+
+[LanguageBind](https://github.com/PKU-YuanGroup/LanguageBind) (ICLR 2024) packaged as a pip-installable library with compatibility patches for modern `transformers`, `torchvision`, and `torchaudio`.
+
+The original LanguageBind repo has no `pyproject.toml`, so it cannot be installed via pip. This package provides that, plus inline patches for five breaking changes introduced in newer dependency versions.
+
+## Installation
+
+```bash
+pip install languagebind
+```
+
+For video support:
+```bash
+pip install "languagebind[video]"
+```
+
+For audio support:
+```bash
+pip install "languagebind[audio]"
+```
+
+## Usage
+
+```python
+from languagebind import (
+    LanguageBindVideo, LanguageBindVideoProcessor, LanguageBindVideoTokenizer,
+    LanguageBindAudio, LanguageBindAudioProcessor, LanguageBindAudioTokenizer,
+    LanguageBindImage, LanguageBindImageProcessor, LanguageBindImageTokenizer,
+)
+```
+
+## Compatibility patches
+
+- `_expand_mask` / `clip_loss`: removed from `transformers` 4.40+, re-implemented in `languagebind._compat`
+- `torchaudio.set_audio_backend()`: deprecated, guarded with `try/except`
+- `torchvision.transforms._transforms_video`: private API fallback to public `torchvision.transforms` equivalents
+- `CLIPTokenizer.__init__` positional args: changed to keyword args for `transformers` 4.40+ compatibility
+
+## License
+
+MIT — same as the original [LanguageBind](https://github.com/PKU-YuanGroup/LanguageBind).

languagebind-0.1.0/README.md

@@ -0,0 +1,42 @@
+# languagebind
+
+[LanguageBind](https://github.com/PKU-YuanGroup/LanguageBind) (ICLR 2024) packaged as a pip-installable library with compatibility patches for modern `transformers`, `torchvision`, and `torchaudio`.
+
+The original LanguageBind repo has no `pyproject.toml`, so it cannot be installed via pip. This package provides that, plus inline patches for five breaking changes introduced in newer dependency versions.
+
+## Installation
+
+```bash
+pip install languagebind
+```
+
+For video support:
+```bash
+pip install "languagebind[video]"
+```
+
+For audio support:
+```bash
+pip install "languagebind[audio]"
+```
+
+## Usage
+
+```python
+from languagebind import (
+    LanguageBindVideo, LanguageBindVideoProcessor, LanguageBindVideoTokenizer,
+    LanguageBindAudio, LanguageBindAudioProcessor, LanguageBindAudioTokenizer,
+    LanguageBindImage, LanguageBindImageProcessor, LanguageBindImageTokenizer,
+)
+```
+
+## Compatibility patches
+
+- `_expand_mask` / `clip_loss`: removed from `transformers` 4.40+, re-implemented in `languagebind._compat`
+- `torchaudio.set_audio_backend()`: deprecated, guarded with `try/except`
+- `torchvision.transforms._transforms_video`: private API fallback to public `torchvision.transforms` equivalents
+- `CLIPTokenizer.__init__` positional args: changed to keyword args for `transformers` 4.40+ compatibility
+
+## License
+
+MIT — same as the original [LanguageBind](https://github.com/PKU-YuanGroup/LanguageBind).

languagebind-0.1.0/languagebind/__init__.py

@@ -0,0 +1,91 @@
+import torch
+from torch import nn
+from transformers import AutoConfig
+
+from .image.configuration_image import LanguageBindImageConfig
+from .image.modeling_image import LanguageBindImage
+from .image.tokenization_image import LanguageBindImageTokenizer
+from .image.processing_image import LanguageBindImageProcessor
+
+from .video.configuration_video import LanguageBindVideoConfig
+from .video.modeling_video import LanguageBindVideo
+from .video.tokenization_video import LanguageBindVideoTokenizer
+from .video.processing_video import LanguageBindVideoProcessor
+
+from .depth.configuration_depth import LanguageBindDepthConfig
+from .depth.modeling_depth import LanguageBindDepth
+from .depth.tokenization_depth import LanguageBindDepthTokenizer
+from .depth.processing_depth import LanguageBindDepthProcessor
+
+from .audio.configuration_audio import LanguageBindAudioConfig
+from .audio.modeling_audio import LanguageBindAudio
+from .audio.tokenization_audio import LanguageBindAudioTokenizer
+from .audio.processing_audio import LanguageBindAudioProcessor
+
+from .thermal.configuration_thermal import LanguageBindThermalConfig
+from .thermal.modeling_thermal import LanguageBindThermal
+from .thermal.tokenization_thermal import LanguageBindThermalTokenizer
+from .thermal.processing_thermal import LanguageBindThermalProcessor
+
+
+
+config_dict = {
+    'thermal': LanguageBindThermalConfig,
+    'image': LanguageBindImageConfig,
+    'video': LanguageBindVideoConfig,
+    'depth': LanguageBindDepthConfig,
+    'audio': LanguageBindAudioConfig
+}
+model_dict = {
+    'thermal': LanguageBindThermal,
+    'image': LanguageBindImage,
+    'video': LanguageBindVideo,
+    'depth': LanguageBindDepth,
+    'audio': LanguageBindAudio
+}
+transform_dict = {
+    'video': LanguageBindVideoProcessor,
+    'audio': LanguageBindAudioProcessor,
+    'depth': LanguageBindDepthProcessor,
+    'thermal': LanguageBindThermalProcessor,
+    'image': LanguageBindImageProcessor,
+}
+
+class LanguageBind(nn.Module):
+    def __init__(self, clip_type, use_temp=True, cache_dir='./cache_dir'):
+        super(LanguageBind, self).__init__()
+        self.use_temp = use_temp
+        self.modality_encoder = {}
+        self.modality_proj = {}
+        self.modality_scale = {}
+        self.modality_config = {}
+        for k, v in clip_type.items():
+            pretrained_ckpt = f'LanguageBind/{v}'
+            model = model_dict[k].from_pretrained(pretrained_ckpt, cache_dir=cache_dir)
+            self.modality_encoder[k] = model.vision_model
+            self.modality_proj[k] = model.visual_projection
+            self.modality_scale[k] = model.logit_scale
+            self.modality_config[k] = model.config
+        self.modality_encoder['language'] = model.text_model
+        self.modality_proj['language'] = model.text_projection
+
+        self.modality_encoder = nn.ModuleDict(self.modality_encoder)
+        self.modality_proj = nn.ModuleDict(self.modality_proj)
+
+    def forward(self, inputs):
+        outputs = {}
+        for key, value in inputs.items():
+            value = self.modality_encoder[key](**value)[1]
+            value = self.modality_proj[key](value)
+            value = value / value.norm(p=2, dim=-1, keepdim=True)
+            if self.use_temp:
+                if key != 'language':
+                    value = value * self.modality_scale[key].exp()
+            outputs[key] = value
+        return outputs
+
+def to_device(x, device):
+    out_dict = {k: v.to(device) for k, v in x.items()}
+    return out_dict
+
+

languagebind-0.1.0/languagebind/_compat.py

@@ -0,0 +1,24 @@
+"""Compatibility shims for symbols removed from newer versions of transformers/torchaudio/torchvision."""
+from __future__ import annotations
+
+import torch
+import torch.nn.functional as F
+
+try:
+    from transformers.models.clip.modeling_clip import _expand_mask
+except ImportError:
+    def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: int | None = None) -> torch.Tensor:
+        bsz, src_len = mask.size()
+        tgt_len = tgt_len if tgt_len is not None else src_len
+        expanded_mask = mask[:, None, None, :].expand(bsz, 1, tgt_len, src_len).to(dtype)
+        inverted_mask = 1.0 - expanded_mask
+        return inverted_mask.masked_fill(inverted_mask.to(torch.bool), torch.finfo(dtype).min)
+
+try:
+    from transformers.models.clip.modeling_clip import clip_loss
+except ImportError:
+    def _contrastive_loss(logits: torch.Tensor) -> torch.Tensor:
+        return F.cross_entropy(logits, torch.arange(len(logits), device=logits.device))
+
+    def clip_loss(similarity: torch.Tensor) -> torch.Tensor:
+        return (_contrastive_loss(similarity) + _contrastive_loss(similarity.t())) / 2.0

languagebind-0.1.0/languagebind/audio/configuration_audio.py

@@ -0,0 +1,420 @@
+import copy
+import os
+from typing import Union
+
+from transformers import PretrainedConfig
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+
+
+
+
+
+
+class CLIPTextConfig(PretrainedConfig):
+    r"""
+    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 [`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 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.
+        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-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.
+        initializer_factor (`float`, *optional*, defaults to 1):
+            A factor for initializing all weight matrices (should be kept to 1, used internally for initialization
+            testing).
+
+    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
+    ```"""
+    model_type = "clip_text_model"
+
+    def __init__(
+        self,
+        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-5,
+        attention_dropout=0.0,
+        initializer_range=0.02,
+        initializer_factor=1.0,
+        # This differs from `CLIPTokenizer`'s default and from openai/clip
+        # 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.projection_dim = projection_dim
+        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.initializer_range = initializer_range
+        self.initializer_factor = initializer_factor
+        self.attention_dropout = attention_dropout
+        self.add_time_attn = False  ######################################
+
+    @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 CLIPConfig
+        if config_dict.get("model_type") == "clip":
+            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 CLIPVisionConfig(PretrainedConfig):
+    r"""
+    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 [`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.
+        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-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.
+        initializer_factor (`float`, *optional*, defaults to 1):
+            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
+    ```"""
+
+    model_type = "clip_vision_model"
+
+    def __init__(
+        self,
+        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-5,
+        attention_dropout=0.0,
+        initializer_range=0.02,
+        initializer_factor=1.0,
+
+        add_time_attn=False, ################################
+        num_frames=1, ################################
+        force_patch_dropout=0.0, ################################
+        lora_r=2, ################################
+        lora_alpha=16, ################################
+        lora_dropout=0.0, ################################
+        num_mel_bins=0.0, ################################
+        target_length=0.0, ################################
+        video_decode_backend='decord', #########################
+        audio_sample_rate=16000,
+        audio_mean=0.5,
+        audio_std=0.5,
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.projection_dim = projection_dim
+        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.initializer_range = initializer_range
+        self.initializer_factor = initializer_factor
+        self.attention_dropout = attention_dropout
+        self.layer_norm_eps = layer_norm_eps
+        self.hidden_act = hidden_act
+
+        self.add_time_attn = add_time_attn  ################
+        self.num_frames = num_frames  ################
+        self.force_patch_dropout = force_patch_dropout  ################
+        self.lora_r = lora_r  ################
+        self.lora_alpha = lora_alpha  ################
+        self.lora_dropout = lora_dropout  ################
+        self.num_mel_bins = num_mel_bins  ################
+        self.target_length = target_length  ################
+        self.video_decode_backend = video_decode_backend  ################
+
+        self.audio_sample_rate = audio_sample_rate
+        self.audio_mean = audio_mean
+        self.audio_std = audio_std
+
+    @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 CLIPConfig
+        if config_dict.get("model_type") == "clip":
+            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 LanguageBindAudioConfig(PretrainedConfig):
+    r"""
+    [`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 [`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 [`CLIPTextConfig`].
+        vision_config (`dict`, *optional*):
+            Dictionary of configuration options used to initialize [`CLIPVisionConfig`].
+        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 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)
+    ```"""
+
+    model_type = "LanguageBindAudio"
+    is_composition = True
+
+    def __init__(
+        self, text_config=None, vision_config=None, projection_dim=512, logit_scale_init_value=2.6592, **kwargs
+    ):
+        # If `_config_dict` exist, we use them for the backward compatibility.
+        # We pop out these 2 attributes before calling `super().__init__` to avoid them being saved (which causes a lot
+        # of confusion!).
+        text_config_dict = kwargs.pop("text_config_dict", None)
+        vision_config_dict = kwargs.pop("vision_config_dict", None)
+
+        super().__init__(**kwargs)
+
+        # Instead of simply assigning `[text|vision]_config_dict` to `[text|vision]_config`, we use the values in
+        # `[text|vision]_config_dict` to update the values in `[text|vision]_config`. The values should be same in most
+        # cases, but we don't want to break anything regarding `_config_dict` that existed before commit `8827e1b2`.
+        if text_config_dict is not None:
+            if text_config is None:
+                text_config = {}
+
+            # This is the complete result when using `text_config_dict`.
+            _text_config_dict = CLIPTextConfig(**text_config_dict).to_dict()
+
+            # Give a warning if the values exist in both `_text_config_dict` and `text_config` but being different.
+            for key, value in _text_config_dict.items():
+                if key in text_config and value != text_config[key] and key not in ["transformers_version"]:
+                    # If specified in `text_config_dict`
+                    if key in text_config_dict:
+                        message = (
+                            f"`{key}` is found in both `text_config_dict` and `text_config` but with different values. "
+                            f'The value `text_config_dict["{key}"]` will be used instead.'
+                        )
+                    # If inferred from default argument values (just to be super careful)
+                    else:
+                        message = (
+                            f"`text_config_dict` is provided which will be used to initialize `CLIPTextConfig`. The "
+                            f'value `text_config["{key}"]` will be overriden.'
+                        )
+                    logger.warning(message)
+
+            # Update all values in `text_config` with the ones in `_text_config_dict`.
+            text_config.update(_text_config_dict)
+
+        if vision_config_dict is not None:
+            if vision_config is None:
+                vision_config = {}
+
+            # This is the complete result when using `vision_config_dict`.
+            _vision_config_dict = CLIPVisionConfig(**vision_config_dict).to_dict()
+            # convert keys to string instead of integer
+            if "id2label" in _vision_config_dict:
+                _vision_config_dict["id2label"] = {
+                    str(key): value for key, value in _vision_config_dict["id2label"].items()
+                }
+
+            # Give a warning if the values exist in both `_vision_config_dict` and `vision_config` but being different.
+            for key, value in _vision_config_dict.items():
+                if key in vision_config and value != vision_config[key] and key not in ["transformers_version"]:
+                    # If specified in `vision_config_dict`
+                    if key in vision_config_dict:
+                        message = (
+                            f"`{key}` is found in both `vision_config_dict` and `vision_config` but with different "
+                            f'values. The value `vision_config_dict["{key}"]` will be used instead.'
+                        )
+                    # If inferred from default argument values (just to be super careful)
+                    else:
+                        message = (
+                            f"`vision_config_dict` is provided which will be used to initialize `CLIPVisionConfig`. "
+                            f'The value `vision_config["{key}"]` will be overriden.'
+                        )
+                    logger.warning(message)
+
+            # Update all values in `vision_config` with the ones in `_vision_config_dict`.
+            vision_config.update(_vision_config_dict)
+
+        if text_config is None:
+            text_config = {}
+            logger.info("`text_config` is `None`. Initializing the `CLIPTextConfig` with default values.")
+
+        if vision_config is None:
+            vision_config = {}
+            logger.info("`vision_config` is `None`. initializing the `CLIPVisionConfig` with default values.")
+
+        self.text_config = CLIPTextConfig(**text_config)
+        self.vision_config = CLIPVisionConfig(**vision_config)
+
+        self.projection_dim = projection_dim
+        self.logit_scale_init_value = logit_scale_init_value
+        self.initializer_factor = 1.0
+
+    @classmethod
+    def from_text_vision_configs(cls, text_config: CLIPTextConfig, vision_config: CLIPVisionConfig, **kwargs):
+        r"""
+        Instantiate a [`CLIPConfig`] (or a derived class) from clip text model configuration and clip vision model
+        configuration.
+
+        Returns:
+            [`CLIPConfig`]: An instance of a configuration object
+        """
+
+        return cls(text_config=text_config.to_dict(), vision_config=vision_config.to_dict(), **kwargs)
+
+    def to_dict(self):
+        """
+        Serializes this instance to a Python dictionary. Override the default [`~PretrainedConfig.to_dict`].
+
+        Returns:
+            `Dict[str, any]`: Dictionary of all the attributes that make up this configuration instance,
+        """
+        output = copy.deepcopy(self.__dict__)
+        output["text_config"] = self.text_config.to_dict()
+        output["vision_config"] = self.vision_config.to_dict()
+        output["model_type"] = self.__class__.model_type
+        return output