# SigLIP[[siglip]]

## 개요[[overview]]

SigLIP 모델은 Xiaohua Zhai, Basil Mustafa, Alexander Kolesnikov, Lucas Beyer의 [Sigmoid Loss for Language Image Pre-Training](https://huggingface.co/papers/2303.15343) 논문에서 제안되었습니다. SigLIP은 [CLIP](clip)에서 사용된 손실 함수를 간단한 쌍별 시그모이드 손실(pairwise sigmoid loss)로 대체할 것을 제안합니다. 이는 ImageNet에서 제로샷 분류 정확도 측면에서 더 나은 성능을 보입니다.

논문의 초록은 다음과 같습니다:

*우리는 언어-이미지 사전 학습(Language-Image Pre-training, SigLIP)을 위한 간단한 쌍별 시그모이드 손실을 제안합니다. 소프트맥스 정규화를 사용하는 표준 대조 학습과 달리, 시그모이드 손실은 이미지-텍스트 쌍에만 작용하며 정규화를 위해 쌍별 유사성의 전역적 관점을 필요로 하지 않습니다. 시그모이드 손실은 배치 크기를 더욱 확장할 수 있게 하는 동시에 작은 배치 크기에서도 더 나은 성능을 보입니다. Locked-image Tuning과 결합하여, 단 4개의 TPUv4 칩만으로 이틀 만에 84.5%의 ImageNet 제로샷 정확도를 달성하는 SigLiT 모델을 학습했습니다. 손실 함수에서 배치 크기를 분리함으로써 예제 대 쌍의 영향과 Negative 대 Positive 비율을 연구할 수 있게 되었습니다. 마지막으로, 우리는 배치 크기를 100만 개까지 극단적으로 늘려보았고, 배치 크기 증가의 이점이 빠르게 감소하며 32k의 더 합리적인 배치 크기로도 충분하다는 것을 발견했습니다.*

## 사용 팁[[usage-tips]]

- SigLIP의 사용법은 [CLIP](clip)과 유사합니다. 주요 차이점은 학습 손실 함수로, 배치 내 모든 이미지와 텍스트 간의 쌍별 유사성에 대한 전역적 관점이 필요하지 않습니다. 소프트맥스 대신 로짓에 시그모이드 활성화 함수를 적용해야 합니다.
- 학습은 지원되지만 `torch.distributed` 유틸리티를 사용하지 않아 배치 크기의 확장성이 제한될 수 있습니다. 그러나 단일 노드 다중 GPU 설정에서는 DDP와 FDSP가 작동합니다.
- 독립형 [SiglipTokenizer](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipTokenizer) 또는 [SiglipProcessor](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipProcessor)를 사용할 때는 모델이 그렇게 학습되었으므로 `padding="max_length"`를 전달해야 합니다.
- 파이프라인과 동일한 결과를 얻으려면 "This is a photo of {label}."의 프롬프트 템플릿을 사용해야 합니다.

 CLIP과 비교한 SigLIP 평가 결과. 원본 논문에서 발췌.

이 모델은 [nielsr](https://huggingface.co/nielsr)가 기여했습니다.
원본 코드는 [여기](https://github.com/google-research/big_vision/tree/main)에서 찾을 수 있습니다.

## 사용 예시[[usage-example]]

SigLIP을 사용하는 방법에는 두 가지 주요 방법이 있습니다: 모든 복잡성을 추상화하는 파이프라인 API를 사용하거나, 직접 `SiglipModel` 클래스를 사용하는 방법입니다.

### 파이프라인 API[[pipeline-API]]

파이프라인을 사용하면 몇 줄의 코드로 모델을 사용할 수 있습니다:

```python
>>> from transformers import pipeline
>>> from PIL import Image
>>> import requests

>>> # 파이프라인 로드
>>> image_classifier = pipeline(task="zero-shot-image-classification", model="google/siglip-base-patch16-224")

>>> # 이미지 로드
>>> url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> # 추론
>>> candidate_labels = ["2 cats", "a plane", "a remote"]
>>> outputs = image_classifier(image, candidate_labels=candidate_labels)
>>> outputs = [{"score": round(output["score"], 4), "label": output["label"] } for output in outputs]
>>> print(outputs)
[{'score': 0.1979, 'label': '2 cats'}, {'score': 0.0, 'label': 'a remote'}, {'score': 0.0, 'label': 'a plane'}]
```

### 직접 모델 사용하기[[using-the-model-yourself]]

전처리와 후처리를 직접 수행하려면 다음과 같이 하면 됩니다:

```python
>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, AutoModel
>>> import torch

>>> model = AutoModel.from_pretrained("google/siglip-base-patch16-224")
>>> processor = AutoProcessor.from_pretrained("google/siglip-base-patch16-224")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> candidate_labels = ["2 cats", "2 dogs"]
# 파이프라인 프롬프트 템플릿을 따라 동일한 결과를 얻습니다
>>> texts = [f'This is a photo of {label}.' for label in candidate_labels]
# 중요: 모델이 이렇게 학습되었으므로 `padding=max_length`를 전달합니다
>>> inputs = processor(text=texts, images=image, padding="max_length", return_tensors="pt")

>>> with torch.no_grad():
...     outputs = model(**inputs)

>>> logits_per_image = outputs.logits_per_image
>>> probs = torch.sigmoid(logits_per_image) # 시그모이드 활성화 함수를 적용한 확률입니다
>>> print(f"{probs[0][0]:.1%} that image 0 is '{candidate_labels[0]}'")
19.8% that image 0 is '2 cats'
```

## 리소스[[resources]]

SigLIP을 시작하는 데 도움이 되는 공식 Hugging Face 및 커뮤니티(🌎로 표시) 리소스 목록입니다.

- [제로샷 이미지 분류 작업 가이드](../tasks/zero_shot_image_classification)
- SigLIP에 대한 데모 노트북은 [여기](https://github.com/NielsRogge/Transformers-Tutorials/tree/master/SigLIP)에서 찾을 수 있습니다. 🌎

여기에 포함될 리소스를 제출하는 데 관심이 있으시면 Pull Request를 열어주시면 검토하겠습니다! 리소스는 이상적으로 기존 리소스를 복제하는 대신 새로운 것을 보여주어야 합니다.

## SigLIP과 Flash Attention 2 결합하기[[combining-siglip-with-flash-attention-2]]

먼저 Flash Attention 2의 최신 버전을 설치해야 합니다.

```bash
pip install -U flash-attn --no-build-isolation
```

또한 Flash-Attention 2와 호환되는 하드웨어가 있는지 확인하세요. flash-attn 저장소의 공식 문서에서 자세히 알아보세요. 또한 모델을 반정밀도(예: `torch.float16`)로 로드해야 합니다.

Flash Attention 2를 사용하여 모델을 로드하고 실행하려면 아래 코드를 참조하세요:

```python
>>> import torch
>>> import requests
>>> from PIL import Image
>>> from transformers import SiglipProcessor, SiglipModel
>>> device = "cuda" # 모델을 로드할 장치

>>> model = SiglipModel.from_pretrained(
...     "google/siglip-so400m-patch14-384",
...     attn_implementation="flash_attention_2",
...     dtype=torch.float16,
...     device_map=device,
... )
>>> processor = SiglipProcessor.from_pretrained("google/siglip-so400m-patch14-384")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> candidate_labels = ["2 cats", "2 dogs"]
# 파이프라인 프롬프트 템플릿을 따라 동일한 결과를 얻습니다
>>> texts = [f'This is a photo of {label}.' for label in candidate_labels]
# 중요: 모델이 이렇게 학습되었으므로 `padding=max_length`를 전달합니다
>>> inputs = processor(text=texts, images=image, padding="max_length", return_tensors="pt").to(device)

>>> with torch.no_grad():
...     with torch.autocast(device):
...         outputs = model(**inputs)

>>> logits_per_image = outputs.logits_per_image
>>> probs = torch.sigmoid(logits_per_image) # 시그모이드 활성화 함수를 적용한 확률입니다
>>> print(f"{probs[0][0]:.1%} that image 0 is '{candidate_labels[0]}'")
19.8% that image 0 is '2 cats'
```

## Scaled Dot Product Attention(SDPA) 사용하기[using-scaled-dot-product-attention(SDPA)]]

PyTorch는 `torch.nn.functional`의 일부로 스케일된 점곱 어텐션(SDPA) 연산자를 포함합니다. 이 함수는 
입력과 사용 중인 하드웨어에 따라 적용할 수 있는 여러 구현을 포함합니다. 자세한 내용은 
[공식 문서](https://pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html) 
또는 [GPU 추론](https://huggingface.co/docs/transformers/main/en/perf_infer_gpu_one#pytorch-scaled-dot-product-attention) 
페이지를 참조하세요.

`from_pretrained()`에서 `attn_implementation="sdpa"`를 설정하여 SDPA를 명시적으로 요청할 수 있습니다. `torch>=2.1.1`이 설치되어 있는지 확인하세요.

```python
>>> from transformers import SiglipModel

>>> model = SiglipModel.from_pretrained(
...     "google/siglip-so400m-patch14-384",
...     attn_implementation="sdpa",
...     dtype=torch.float16,
...     device_map=device,
... )
```

최상의 속도 향상을 위해 모델을 반정밀도(예: `torch.float16` 또는 `torch.bfloat16`)로 로드하는 것이 좋습니다.

## 예상 속도 향상[[expected-speedups]]

아래는 `google/siglip-so400m-patch14-384` 체크포인트를 `float16` 정밀도로 사용하는 transformers의 네이티브 구현과 Flash Attention 2 / SDPA 버전의 모델을 다양한 배치 크기로 비교한 추론 시간의 예상 속도 향상 다이어그램입니다.

## SiglipConfig[[transformers.SiglipConfig]]

#### transformers.SiglipConfig[[transformers.SiglipConfig]]

[Source](https://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/configuration_siglip.py#L197)

[SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig) is the configuration class to store the configuration of a [SiglipModel](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.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](/docs/transformers/v5.1.0/ko/main_classes/configuration#transformers.PreTrainedConfig) and can be used to control the model outputs. Read the
documentation from [PreTrainedConfig](/docs/transformers/v5.1.0/ko/main_classes/configuration#transformers.PreTrainedConfig) for more information.

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(text_config=config_text, vision_config=config_vision)
```

**Parameters:**

text_config (`dict`, *optional*) : Dictionary of configuration options used to initialize [SiglipTextConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipTextConfig).

vision_config (`dict`, *optional*) : Dictionary of configuration options used to initialize [SiglipVisionConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipVisionConfig).

kwargs (*optional*) : Dictionary of keyword arguments.

## SiglipTextConfig[[transformers.SiglipTextConfig]]

#### transformers.SiglipTextConfig[[transformers.SiglipTextConfig]]

[Source](https://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/configuration_siglip.py#L23)

This is the configuration class to store the configuration of a [SiglipTextModel](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.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](/docs/transformers/v5.1.0/ko/main_classes/configuration#transformers.PreTrainedConfig) and can be used to control the model outputs. Read the
documentation from [PreTrainedConfig](/docs/transformers/v5.1.0/ko/main_classes/configuration#transformers.PreTrainedConfig) for more information.

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
```

**Parameters:**

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](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.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.

projection_size (`int`, *optional*, defaults to `hidden_size`) : The size of the projection head.

## SiglipVisionConfig[[transformers.SiglipVisionConfig]]

#### transformers.SiglipVisionConfig[[transformers.SiglipVisionConfig]]

[Source](https://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/configuration_siglip.py#L118)

This is the configuration class to store the configuration of a [SiglipVisionModel](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.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](/docs/transformers/v5.1.0/ko/main_classes/configuration#transformers.PreTrainedConfig) and can be used to control the model outputs. Read the
documentation from [PreTrainedConfig](/docs/transformers/v5.1.0/ko/main_classes/configuration#transformers.PreTrainedConfig) for more information.

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
```

**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.

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.

## SiglipTokenizer[[transformers.SiglipTokenizer]]

#### transformers.SiglipTokenizer[[transformers.SiglipTokenizer]]

[Source](https://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/tokenization_siglip.py#L44)

Construct a Siglip tokenizer. Based on [SentencePiece](https://github.com/google/sentencepiece).

This tokenizer inherits from [PreTrainedTokenizer](/docs/transformers/v5.1.0/ko/main_classes/tokenizer#transformers.PythonBackend) which contains most of the main methods. Users should refer to
this superclass for more information regarding those methods.

build_inputs_with_special_tokenstransformers.SiglipTokenizer.build_inputs_with_special_tokenshttps://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/tokenization_siglip.py#L203[{"name": "token_ids_0", "val": ": list"}, {"name": "token_ids_1", "val": ": list[int] | None = None"}]- **token_ids_0** (`list[int]`) --
  List of IDs to which the special tokens will be added.
- **token_ids_1** (`list[int]`, *optional*) --
  Optional second list of IDs for sequence pairs.0`list[int]`List of [input IDs](../glossary#input-ids) with the appropriate special tokens.

Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
adding special tokens. A sequence has the following format:

- single sequence: `X `
- pair of sequences: `A  B `

**Parameters:**

vocab_file (`str`) : [SentencePiece](https://github.com/google/sentencepiece) file (generally has a *.spm* extension) that contains the vocabulary necessary to instantiate a tokenizer.

eos_token (`str`, *optional*, defaults to `""`) : The end of sequence token.

unk_token (`str`, *optional*, defaults to `""`) : The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this token instead.

pad_token (`str`, *optional*, defaults to `""`) : The token used for padding, for example when batching sequences of different lengths.

additional_special_tokens (`list[str]`, *optional*) : Additional special tokens used by the tokenizer.

sp_model_kwargs (`dict`, *optional*) : Will be passed to the `SentencePieceProcessor.__init__()` method. The [Python wrapper for SentencePiece](https://github.com/google/sentencepiece/tree/master/python) can be used, among other things, to set:  - `enable_sampling`: Enable subword regularization. - `nbest_size`: Sampling parameters for unigram. Invalid for BPE-Dropout.  - `nbest_size = {0,1}`: No sampling is performed. - `nbest_size > 1`: samples from the nbest_size results. - `nbest_size >> from PIL import Image
>>> import httpx
>>> from io import BytesIO
>>> from transformers import AutoProcessor, AutoModel
>>> import torch

>>> model = AutoModel.from_pretrained("google/siglip-base-patch16-224")
>>> processor = AutoProcessor.from_pretrained("google/siglip-base-patch16-224")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> with httpx.stream("GET", url) as response:
...     image = Image.open(BytesIO(response.read()))

>>> texts = ["a photo of 2 cats", "a photo of 2 dogs"]
>>> # important: we pass `padding=max_length` since the model was trained with this
>>> inputs = processor(text=texts, images=image, padding="max_length", return_tensors="pt")

>>> with torch.no_grad():
...     outputs = model(**inputs)

>>> logits_per_image = outputs.logits_per_image
>>> probs = torch.sigmoid(logits_per_image) # these are the probabilities
>>> print(f"{probs[0][0]:.1%} that image 0 is '{texts[0]}'")
31.9% that image 0 is 'a photo of 2 cats'
```

**Parameters:**

config ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v5.1.0/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

**Returns:**

``transformers.models.siglip.modeling_siglip.SiglipOutput` or `tuple(torch.FloatTensor)``

A `transformers.models.siglip.modeling_siglip.SiglipOutput` 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 ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) and inputs.

- **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `return_loss` is `True`) -- Contrastive loss for image-text similarity.
- **logits_per_image** (`torch.FloatTensor` of shape `(image_batch_size, text_batch_size)`) -- The scaled dot product scores between `image_embeds` and `text_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 between `text_embeds` and `image_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 [SiglipTextModel](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipTextModel).
- **image_embeds** (`torch.FloatTensor` of shape `(batch_size, output_dim`) -- The image embeddings obtained by applying the projection layer to the pooled output of [SiglipVisionModel](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipVisionModel).
- **text_model_output** (`.text_model_output`, defaults to `None`) -- The output of the [SiglipTextModel](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipTextModel).
- **vision_model_output** (`.vision_model_output`, defaults to `None`) -- The output of the [SiglipVisionModel](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipVisionModel).
#### get_text_features[[transformers.SiglipModel.get_text_features]]

[Source](https://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/modeling_siglip.py#L756)

Examples:

```python
>>> from transformers import AutoTokenizer, AutoModel
>>> import torch

>>> model = AutoModel.from_pretrained("google/siglip-base-patch16-224")
>>> tokenizer = AutoTokenizer.from_pretrained("google/siglip-base-patch16-224")

>>> # important: make sure to set padding="max_length" as that's how the model was trained
>>> inputs = tokenizer(["a photo of a cat", "a photo of a dog"], padding="max_length", return_tensors="pt")
>>> with torch.no_grad():
...     text_features = model.get_text_features(**inputs)
```

**Parameters:**

input_ids (`torch.Tensor` of shape `(batch_size, sequence_length)`) : Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.  Indices can be obtained using [AutoTokenizer](/docs/transformers/v5.1.0/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.1.0/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v5.1.0/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details.  [What are input IDs?](../glossary#input-ids)

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**.  [What are attention masks?](../glossary#attention-mask)

position_ids (`torch.Tensor` 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.n_positions - 1]`.  [What are position IDs?](../glossary#position-ids)

**Returns:**

`[transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)``

A [transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#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 ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) 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 when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.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 when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.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.
#### get_image_features[[transformers.SiglipModel.get_image_features]]

[Source](https://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/modeling_siglip.py#L787)

Examples:

```python
>>> import torch
>>> from transformers import AutoProcessor, AutoModel
>>> from transformers.image_utils import load_image

>>> model = AutoModel.from_pretrained("google/siglip-base-patch16-224")
>>> processor = AutoProcessor.from_pretrained("google/siglip-base-patch16-224")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = load_image(url)

>>> inputs = processor(images=image, return_tensors="pt")

>>> with torch.no_grad():
...     image_features = model.get_image_features(**inputs)
```

**Parameters:**

pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`) : The tensors corresponding to the input images. Pixel values can be obtained using [SiglipImageProcessorFast](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipImageProcessorFast). See `SiglipImageProcessorFast.__call__()` for details ([SiglipProcessor](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipProcessor) uses [SiglipImageProcessorFast](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipImageProcessorFast) for processing images).

interpolate_pos_encoding (`bool`, *optional*, defaults to `False`) : Whether to interpolate the pre-trained position encodings.

**Returns:**

`[transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)``

A [transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#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 ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) 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 when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.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 when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.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.

## SiglipTextModel[[transformers.SiglipTextModel]]

#### transformers.SiglipTextModel[[transformers.SiglipTextModel]]

[Source](https://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/modeling_siglip.py#L539)

The text model from SigLIP without any head or projection on top.

This model inherits from [PreTrainedModel](/docs/transformers/v5.1.0/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the
library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
etc.)

This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
and behavior.

forwardtransformers.SiglipTextModel.forwardhttps://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/modeling_siglip.py#L555[{"name": "input_ids", "val": ": torch.Tensor | None = None"}, {"name": "attention_mask", "val": ": torch.Tensor | None = None"}, {"name": "position_ids", "val": ": torch.Tensor | None = None"}, {"name": "**kwargs", "val": ": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}]- **input_ids** (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*) --
  Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

  Indices can be obtained using [AutoTokenizer](/docs/transformers/v5.1.0/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.1.0/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and
  [PreTrainedTokenizer.__call__()](/docs/transformers/v5.1.0/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details.

  [What are input IDs?](../glossary#input-ids)
- **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**.

  [What are attention masks?](../glossary#attention-mask)
- **position_ids** (`torch.Tensor` 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.n_positions - 1]`.

  [What are position IDs?](../glossary#position-ids)0[transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)`A [transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#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 ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) 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 when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.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 when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.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 [SiglipTextModel](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipTextModel) 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:

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

>>> model = SiglipTextModel.from_pretrained("google/siglip-base-patch16-224")
>>> tokenizer = AutoTokenizer.from_pretrained("google/siglip-base-patch16-224")

>>> # important: make sure to set padding="max_length" as that's how the model was trained
>>> inputs = tokenizer(["a photo of a cat", "a photo of a dog"], padding="max_length", return_tensors="pt")

>>> outputs = model(**inputs)
>>> last_hidden_state = outputs.last_hidden_state
>>> pooled_output = outputs.pooler_output  # pooled (EOS token) states
```

**Parameters:**

config ([SiglipTextConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipTextConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v5.1.0/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

**Returns:**

`[transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)``

A [transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#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 ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) 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 when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.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 when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.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.

## SiglipVisionModel[[transformers.SiglipVisionModel]]

#### transformers.SiglipVisionModel[[transformers.SiglipVisionModel]]

[Source](https://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/modeling_siglip.py#L660)

The vision model from SigLIP without any head or projection on top.

This model inherits from [PreTrainedModel](/docs/transformers/v5.1.0/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the
library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
etc.)

This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
and behavior.

forwardtransformers.SiglipVisionModel.forwardhttps://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/modeling_siglip.py#L676[{"name": "pixel_values", "val": ""}, {"name": "interpolate_pos_encoding", "val": ": bool = False"}, {"name": "**kwargs", "val": ": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}]- **pixel_values** (`` of shape `(batch_size, num_channels, image_size, image_size)`) --
  The tensors corresponding to the input images. Pixel values can be obtained using
  [SiglipImageProcessorFast](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipImageProcessorFast). See `SiglipImageProcessorFast.__call__()` for details ([SiglipProcessor](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipProcessor) uses
  [SiglipImageProcessorFast](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipImageProcessorFast) for processing images).
- **interpolate_pos_encoding** (`bool`, *optional*, defaults to `False`) --
  Whether to interpolate the pre-trained position encodings.0[transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)`A [transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#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 ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) 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 when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.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 when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.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 [SiglipVisionModel](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipVisionModel) 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:

```python
>>> from PIL import Image
>>> import httpx
>>> from io import BytesIO
>>> from transformers import AutoProcessor, SiglipVisionModel

>>> model = SiglipVisionModel.from_pretrained("google/siglip-base-patch16-224")
>>> processor = AutoProcessor.from_pretrained("google/siglip-base-patch16-224")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> with httpx.stream("GET", url) as response:
...     image = Image.open(BytesIO(response.read()))

>>> inputs = processor(images=image, return_tensors="pt")

>>> outputs = model(**inputs)
>>> last_hidden_state = outputs.last_hidden_state
>>> pooled_output = outputs.pooler_output  # pooled features
```

**Parameters:**

config ([SiglipVisionConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipVisionConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v5.1.0/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

**Returns:**

`[transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#transformers.modeling_outputs.BaseModelOutputWithPooling) or `tuple(torch.FloatTensor)``

A [transformers.modeling_outputs.BaseModelOutputWithPooling](/docs/transformers/v5.1.0/ko/main_classes/output#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 ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) 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 when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.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 when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.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.

## SiglipForImageClassification[[transformers.SiglipForImageClassification]]

#### transformers.SiglipForImageClassification[[transformers.SiglipForImageClassification]]

[Source](https://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/modeling_siglip.py#L919)

SigLIP vision encoder with an image classification head on top (a linear layer on top of the pooled final hidden states of
the patch tokens) e.g. for ImageNet.

This model inherits from [PreTrainedModel](/docs/transformers/v5.1.0/ko/main_classes/model#transformers.PreTrainedModel). Check the superclass documentation for the generic methods the
library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
etc.)

This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
and behavior.

forwardtransformers.SiglipForImageClassification.forwardhttps://github.com/huggingface/transformers/blob/v5.1.0/src/transformers/models/siglip/modeling_siglip.py#L947[{"name": "pixel_values", "val": ": torch.Tensor | None = None"}, {"name": "labels", "val": ": torch.Tensor | None = None"}, {"name": "interpolate_pos_encoding", "val": ": bool = False"}, {"name": "**kwargs", "val": ": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}]- **pixel_values** (`torch.Tensor` of shape `(batch_size, num_channels, image_size, image_size)`, *optional*) --
  The tensors corresponding to the input images. Pixel values can be obtained using
  [SiglipImageProcessorFast](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipImageProcessorFast). See `SiglipImageProcessorFast.__call__()` for details ([SiglipProcessor](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipProcessor) uses
  [SiglipImageProcessorFast](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipImageProcessorFast) for processing images).
- **labels** (`torch.LongTensor` of shape `(batch_size,)`, *optional*) --
  Labels for computing the image classification/regression loss. Indices should be in `[0, ...,
  config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
  `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
- **interpolate_pos_encoding** (`bool`, *optional*, defaults to `False`) --
  Whether to interpolate the pre-trained position encodings.0[transformers.modeling_outputs.ImageClassifierOutput](/docs/transformers/v5.1.0/ko/main_classes/output#transformers.modeling_outputs.ImageClassifierOutput) or `tuple(torch.FloatTensor)`A [transformers.modeling_outputs.ImageClassifierOutput](/docs/transformers/v5.1.0/ko/main_classes/output#transformers.modeling_outputs.ImageClassifierOutput) 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 ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) and inputs.

- **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) -- Classification (or regression if config.num_labels==1) loss.
- **logits** (`torch.FloatTensor` of shape `(batch_size, config.num_labels)`) -- Classification (or regression if config.num_labels==1) scores (before SoftMax).
- **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
  one for the output of each stage) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states
  (also called feature maps) of the model at the output of each stage.
- **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, patch_size,
  sequence_length)`.

  Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
  heads.
The [SiglipForImageClassification](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipForImageClassification) 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:

```python
>>> from transformers import AutoImageProcessor, SiglipForImageClassification
>>> import torch
>>> from PIL import Image
>>> import httpx
>>> from io import BytesIO

>>> torch.manual_seed(3)
>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> with httpx.stream("GET", url) as response:
...     image = Image.open(BytesIO(response.read()))

>>> # note: we are loading a `SiglipModel` from the hub here,
>>> # so the head will be randomly initialized, hence the predictions will be random if seed is not set above.
>>> image_processor = AutoImageProcessor.from_pretrained("google/siglip-base-patch16-224")
>>> model = SiglipForImageClassification.from_pretrained("google/siglip-base-patch16-224")

>>> inputs = image_processor(images=image, return_tensors="pt")
>>> outputs = model(**inputs)
>>> logits = outputs.logits
>>> # model predicts one of the two classes
>>> predicted_class_idx = logits.argmax(-1).item()
>>> print("Predicted class:", model.config.id2label[predicted_class_idx])
Predicted class: LABEL_1
```

**Parameters:**

config ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) : Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the [from_pretrained()](/docs/transformers/v5.1.0/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

**Returns:**

`[transformers.modeling_outputs.ImageClassifierOutput](/docs/transformers/v5.1.0/ko/main_classes/output#transformers.modeling_outputs.ImageClassifierOutput) or `tuple(torch.FloatTensor)``

A [transformers.modeling_outputs.ImageClassifierOutput](/docs/transformers/v5.1.0/ko/main_classes/output#transformers.modeling_outputs.ImageClassifierOutput) 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 ([SiglipConfig](/docs/transformers/v5.1.0/ko/model_doc/siglip#transformers.SiglipConfig)) and inputs.

- **loss** (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided) -- Classification (or regression if config.num_labels==1) loss.
- **logits** (`torch.FloatTensor` of shape `(batch_size, config.num_labels)`) -- Classification (or regression if config.num_labels==1) scores (before SoftMax).
- **hidden_states** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`) -- Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
  one for the output of each stage) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states
  (also called feature maps) of the model at the output of each stage.
- **attentions** (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`) -- Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, patch_size,
  sequence_length)`.

  Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
  heads.