# BiT

## Big Transfer (BiT)

### Overview

The BiT model was proposed in [Big Transfer (BiT): General Visual Representation Learning](https://arxiv.org/abs/1912.11370) by Alexander Kolesnikov, Lucas Beyer, Xiaohua Zhai, Joan Puigcerver, Jessica Yung, Sylvain Gelly, Neil Houlsby. BiT is a simple recipe for scaling up pre-training of [ResNet](https://huggingface.co/docs/transformers/model_doc/resnet)-like architectures (specifically, ResNetv2). The method results in significant improvements for transfer learning.

The abstract from the paper is the following:

*Transfer of pre-trained representations improves sample efficiency and simplifies hyperparameter tuning when training deep neural networks for vision. We revisit the paradigm of pre-training on large supervised datasets and fine-tuning the model on a target task. We scale up pre-training, and propose a simple recipe that we call Big Transfer (BiT). By combining a few carefully selected components, and transferring using a simple heuristic, we achieve strong performance on over 20 datasets. BiT performs well across a surprisingly wide range of data regimes — from 1 example per class to 1M total examples. BiT achieves 87.5% top-1 accuracy on ILSVRC-2012, 99.4% on CIFAR-10, and 76.3% on the 19 task Visual Task Adaptation Benchmark (VTAB). On small datasets, BiT attains 76.8% on ILSVRC-2012 with 10 examples per class, and 97.0% on CIFAR-10 with 10 examples per class. We conduct detailed analysis of the main components that lead to high transfer performance.*

Tips:

* BiT models are equivalent to ResNetv2 in terms of architecture, except that: 1) all batch normalization layers are replaced by [group normalization](https://arxiv.org/abs/1803.08494), 2) [weight standardization](https://arxiv.org/abs/1903.10520) is used for convolutional layers. The authors show that the combination of both is useful for training with large batch sizes, and has a significant impact on transfer learning.

This model was contributed by [nielsr](https://huggingface.co/nielsr). The original code can be found [here](https://github.com/google-research/big_transfer).

### Resources

A list of official BOINC AI and community (indicated by 🌎) resources to help you get started with BiT.

Image Classification

* [BitForImageClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/bit#transformers.BitForImageClassification) is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/image-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/image_classification.ipynb).
* See also: [Image classification task guide](https://huggingface.co/docs/transformers/tasks/image_classification)

If you’re interested in submitting a resource to be included here, please feel free to open a Pull Request and we’ll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.

### BitConfig

#### class transformers.BitConfig

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/bit/configuration_bit.py#L29)

( num\_channels = 3embedding\_size = 64hidden\_sizes = \[256, 512, 1024, 2048]depths = \[3, 4, 6, 3]layer\_type = 'preactivation'hidden\_act = 'relu'global\_padding = Nonenum\_groups = 32drop\_path\_rate = 0.0embedding\_dynamic\_padding = Falseoutput\_stride = 32width\_factor = 1out\_features = Noneout\_indices = None\*\*kwargs )

Parameters

* **num\_channels** (`int`, *optional*, defaults to 3) — The number of input channels.
* **embedding\_size** (`int`, *optional*, defaults to 64) — Dimensionality (hidden size) for the embedding layer.
* **hidden\_sizes** (`List[int]`, *optional*, defaults to `[256, 512, 1024, 2048]`) — Dimensionality (hidden size) at each stage.
* **depths** (`List[int]`, *optional*, defaults to `[3, 4, 6, 3]`) — Depth (number of layers) for each stage.
* **layer\_type** (`str`, *optional*, defaults to `"preactivation"`) — The layer to use, it can be either `"preactivation"` or `"bottleneck"`.
* **hidden\_act** (`str`, *optional*, defaults to `"relu"`) — The non-linear activation function in each block. If string, `"gelu"`, `"relu"`, `"selu"` and `"gelu_new"` are supported.
* **global\_padding** (`str`, *optional*) — Padding strategy to use for the convolutional layers. Can be either `"valid"`, `"same"`, or `None`.
* **num\_groups** (`int`, *optional*, defaults to `32`) — Number of groups used for the `BitGroupNormActivation` layers.
* **drop\_path\_rate** (`float`, *optional*, defaults to 0.0) — The drop path rate for the stochastic depth.
* **embedding\_dynamic\_padding** (`bool`, *optional*, defaults to `False`) — Whether or not to make use of dynamic padding for the embedding layer.
* **output\_stride** (`int`, *optional*, defaults to 32) — The output stride of the model.
* **width\_factor** (`int`, *optional*, defaults to 1) — The width factor for the model.
* **out\_features** (`List[str]`, *optional*) — If used as backbone, list of features to output. Can be any of `"stem"`, `"stage1"`, `"stage2"`, etc. (depending on how many stages the model has). If unset and `out_indices` is set, will default to the corresponding stages. If unset and `out_indices` is unset, will default to the last stage.
* **out\_indices** (`List[int]`, *optional*) — If used as backbone, list of indices of features to output. Can be any of 0, 1, 2, etc. (depending on how many stages the model has). If unset and `out_features` is set, will default to the corresponding stages. If unset and `out_features` is unset, will default to the last stage.

This is the configuration class to store the configuration of a [BitModel](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/bit#transformers.BitModel). It is used to instantiate an BiT model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the BiT [google/bit-50](https://huggingface.co/google/bit-50) architecture.

Configuration objects inherit from [PretrainedConfig](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/configuration#transformers.PretrainedConfig) and can be used to control the model outputs. Read the documentation from [PretrainedConfig](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/configuration#transformers.PretrainedConfig) for more information.

Example:

Copied

```
>>> from transformers import BitConfig, BitModel

>>> # Initializing a BiT bit-50 style configuration
>>> configuration = BitConfig()

>>> # Initializing a model (with random weights) from the bit-50 style configuration
>>> model = BitModel(configuration)

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

### BitImageProcessor

#### class transformers.BitImageProcessor

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/bit/image_processing_bit.py#L50)

( do\_resize: bool = Truesize: typing.Dict\[str, int] = Noneresample: Resampling = \<Resampling.BICUBIC: 3>do\_center\_crop: bool = Truecrop\_size: typing.Dict\[str, int] = Nonedo\_rescale: bool = Truerescale\_factor: typing.Union\[int, float] = 0.00392156862745098do\_normalize: bool = Trueimage\_mean: typing.Union\[float, typing.List\[float], NoneType] = Noneimage\_std: typing.Union\[float, typing.List\[float], NoneType] = Nonedo\_convert\_rgb: bool = True\*\*kwargs )

Parameters

* **do\_resize** (`bool`, *optional*, defaults to `True`) — Whether to resize the image’s (height, width) dimensions to the specified `size`. Can be overridden by `do_resize` in the `preprocess` method.
* **size** (`Dict[str, int]` *optional*, defaults to `{"shortest_edge" -- 224}`): Size of the image after resizing. The shortest edge of the image is resized to size\[“shortest\_edge”], with the longest edge resized to keep the input aspect ratio. Can be overridden by `size` in the `preprocess` method.
* **resample** (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BICUBIC`) — Resampling filter to use if resizing the image. Can be overridden by `resample` in the `preprocess` method.
* **do\_center\_crop** (`bool`, *optional*, defaults to `True`) — Whether to center crop the image to the specified `crop_size`. Can be overridden by `do_center_crop` in the `preprocess` method.
* **crop\_size** (`Dict[str, int]` *optional*, defaults to 224) — Size of the output image after applying `center_crop`. Can be overridden by `crop_size` in the `preprocess` method.
* **do\_rescale** (`bool`, *optional*, defaults to `True`) — Whether to rescale the image by the specified scale `rescale_factor`. Can be overridden by `do_rescale` in the `preprocess` method.
* **rescale\_factor** (`int` or `float`, *optional*, defaults to `1/255`) — Scale factor to use if rescaling the image. Can be overridden by `rescale_factor` in the `preprocess` method. do\_normalize — Whether to normalize the image. Can be overridden by `do_normalize` in the `preprocess` method.
* **image\_mean** (`float` or `List[float]`, *optional*, defaults to `OPENAI_CLIP_MEAN`) — Mean to use if normalizing the image. This is a float or list of floats the length of the number of channels in the image. Can be overridden by the `image_mean` parameter in the `preprocess` method.
* **image\_std** (`float` or `List[float]`, *optional*, defaults to `OPENAI_CLIP_MEAN`) — Standard deviation to use if normalizing the image. This is a float or list of floats the length of the number of channels in the image. Can be overridden by the `image_std` parameter in the `preprocess` method. Can be overridden by the `image_std` parameter in the `preprocess` method.
* **do\_convert\_rgb** (`bool`, *optional*, defaults to `True`) — Whether to convert the image to RGB.

Constructs a BiT image processor.

**preprocess**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/bit/image_processing_bit.py#L165)

( images: typing.Union\[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), typing.List\[ForwardRef('PIL.Image.Image')], typing.List\[numpy.ndarray], typing.List\[ForwardRef('torch.Tensor')]]do\_resize: bool = Nonesize: typing.Dict\[str, int] = Noneresample: Resampling = Nonedo\_center\_crop: bool = Nonecrop\_size: int = Nonedo\_rescale: bool = Nonerescale\_factor: float = Nonedo\_normalize: bool = Noneimage\_mean: typing.Union\[float, typing.List\[float], NoneType] = Noneimage\_std: typing.Union\[float, typing.List\[float], NoneType] = Nonedo\_convert\_rgb: bool = Nonereturn\_tensors: typing.Union\[str, transformers.utils.generic.TensorType, NoneType] = Nonedata\_format: typing.Optional\[transformers.image\_utils.ChannelDimension] = \<ChannelDimension.FIRST: 'channels\_first'>input\_data\_format: typing.Union\[str, transformers.image\_utils.ChannelDimension, NoneType] = None\*\*kwargs )

Parameters

* **images** (`ImageInput`) — Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If passing in images with pixel values between 0 and 1, set `do_rescale=False`.
* **do\_resize** (`bool`, *optional*, defaults to `self.do_resize`) — Whether to resize the image.
* **size** (`Dict[str, int]`, *optional*, defaults to `self.size`) — Size of the image after resizing. Shortest edge of the image is resized to size\[“shortest\_edge”], with the longest edge resized to keep the input aspect ratio.
* **resample** (`int`, *optional*, defaults to `self.resample`) — Resampling filter to use if resizing the image. This can be one of the enum `PILImageResampling`. Only has an effect if `do_resize` is set to `True`.
* **do\_center\_crop** (`bool`, *optional*, defaults to `self.do_center_crop`) — Whether to center crop the image.
* **crop\_size** (`Dict[str, int]`, *optional*, defaults to `self.crop_size`) — Size of the center crop. Only has an effect if `do_center_crop` is set to `True`.
* **do\_rescale** (`bool`, *optional*, defaults to `self.do_rescale`) — Whether to rescale the image.
* **rescale\_factor** (`float`, *optional*, defaults to `self.rescale_factor`) — Rescale factor to rescale the image by if `do_rescale` is set to `True`.
* **do\_normalize** (`bool`, *optional*, defaults to `self.do_normalize`) — Whether to normalize the image.
* **image\_mean** (`float` or `List[float]`, *optional*, defaults to `self.image_mean`) — Image mean to use for normalization. Only has an effect if `do_normalize` is set to `True`.
* **image\_std** (`float` or `List[float]`, *optional*, defaults to `self.image_std`) — Image standard deviation to use for normalization. Only has an effect if `do_normalize` is set to `True`.
* **do\_convert\_rgb** (`bool`, *optional*, defaults to `self.do_convert_rgb`) — Whether to convert the image to RGB.
* **return\_tensors** (`str` or `TensorType`, *optional*) — The type of tensors to return. Can be one of:
  * Unset: Return a list of `np.ndarray`.
  * `TensorType.TENSORFLOW` or `'tf'`: Return a batch of type `tf.Tensor`.
  * `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`.
  * `TensorType.NUMPY` or `'np'`: Return a batch of type `np.ndarray`.
  * `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
* **data\_format** (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`) — The channel dimension format for the output image. Can be one of:
  * `"channels_first"` or `ChannelDimension.FIRST`: image in (num\_channels, height, width) format.
  * `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num\_channels) format.
  * Unset: Use the channel dimension format of the input image.
* **input\_data\_format** (`ChannelDimension` or `str`, *optional*) — The channel dimension format for the input image. If unset, the channel dimension format is inferred from the input image. Can be one of:
  * `"channels_first"` or `ChannelDimension.FIRST`: image in (num\_channels, height, width) format.
  * `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num\_channels) format.
  * `"none"` or `ChannelDimension.NONE`: image in (height, width) format.

Preprocess an image or batch of images.

### BitModel

#### class transformers.BitModel

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/bit/modeling_bit.py#L706)

( config )

Parameters

* **config** ([BitConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/bit#transformers.BitConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

The bare BiT model outputting raw features without any specific head on top. This model is 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.

**forward**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/bit/modeling_bit.py#L724)

( pixel\_values: Tensoroutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = None ) → `transformers.modeling_outputs.BaseModelOutputWithPoolingAndNoAttention` or `tuple(torch.FloatTensor)`

Parameters

* **pixel\_values** (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`) — Pixel values. Pixel values can be obtained using [AutoImageProcessor](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoImageProcessor). See [BitImageProcessor.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/deit#transformers.DeiTFeatureExtractor.__call__) for details.
* **output\_hidden\_states** (`bool`, *optional*) — Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.

Returns

`transformers.modeling_outputs.BaseModelOutputWithPoolingAndNoAttention` or `tuple(torch.FloatTensor)`

A `transformers.modeling_outputs.BaseModelOutputWithPoolingAndNoAttention` 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 ([BitConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/bit#transformers.BitConfig)) and inputs.

* **last\_hidden\_state** (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`) — 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 after a pooling operation on the spatial dimensions.
* **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, num_channels, height, width)`.

  Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

The [BitModel](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/bit#transformers.BitModel) 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.

Example:

Copied

```
>>> from transformers import AutoImageProcessor, BitModel
>>> import torch
>>> from datasets import load_dataset

>>> dataset = load_dataset("boincai/cats-image")
>>> image = dataset["test"]["image"][0]

>>> image_processor = AutoImageProcessor.from_pretrained("google/bit-50")
>>> model = BitModel.from_pretrained("google/bit-50")

>>> inputs = image_processor(image, return_tensors="pt")

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

>>> last_hidden_states = outputs.last_hidden_state
>>> list(last_hidden_states.shape)
[1, 2048, 7, 7]
```

### BitForImageClassification

#### class transformers.BitForImageClassification

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/bit/modeling_bit.py#L769)

( config )

Parameters

* **config** ([BitConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/bit#transformers.BitConfig)) — 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()](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights.

BiT Model with an image classification head on top (a linear layer on top of the pooled features), e.g. for ImageNet.

This model is 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.

**forward**

[\<source>](https://github.com/huggingface/transformers/blob/v4.34.1/src/transformers/models/bit/modeling_bit.py#L782)

( pixel\_values: typing.Optional\[torch.FloatTensor] = Nonelabels: typing.Optional\[torch.LongTensor] = Noneoutput\_hidden\_states: typing.Optional\[bool] = Nonereturn\_dict: typing.Optional\[bool] = None ) → [transformers.modeling\_outputs.ImageClassifierOutputWithNoAttention](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.ImageClassifierOutputWithNoAttention) or `tuple(torch.FloatTensor)`

Parameters

* **pixel\_values** (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`) — Pixel values. Pixel values can be obtained using [AutoImageProcessor](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/auto#transformers.AutoImageProcessor). See [BitImageProcessor.**call**()](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/deit#transformers.DeiTFeatureExtractor.__call__) for details.
* **output\_hidden\_states** (`bool`, *optional*) — Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail.
* **return\_dict** (`bool`, *optional*) — Whether or not to return a [ModelOutput](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple.
* **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 classification loss is computed (Cross-Entropy).

Returns

[transformers.modeling\_outputs.ImageClassifierOutputWithNoAttention](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.ImageClassifierOutputWithNoAttention) or `tuple(torch.FloatTensor)`

A [transformers.modeling\_outputs.ImageClassifierOutputWithNoAttention](https://huggingface.co/docs/transformers/v4.34.1/en/main_classes/output#transformers.modeling_outputs.ImageClassifierOutputWithNoAttention) 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 ([BitConfig](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/bit#transformers.BitConfig)) 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, num_channels, height, width)`. Hidden-states (also called feature maps) of the model at the output of each stage.

The [BitForImageClassification](https://huggingface.co/docs/transformers/v4.34.1/en/model_doc/bit#transformers.BitForImageClassification) 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.

Example:

Copied

```
>>> from transformers import AutoImageProcessor, BitForImageClassification
>>> import torch
>>> from datasets import load_dataset

>>> dataset = load_dataset("boincai/cats-image")
>>> image = dataset["test"]["image"][0]

>>> image_processor = AutoImageProcessor.from_pretrained("google/bit-50")
>>> model = BitForImageClassification.from_pretrained("google/bit-50")

>>> inputs = image_processor(image, return_tensors="pt")

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

>>> # model predicts one of the 1000 ImageNet classes
>>> predicted_label = logits.argmax(-1).item()
>>> print(model.config.id2label[predicted_label])
tiger cat
```


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://boinc-ai.gitbook.io/transformers/api/models/vision-models/bit.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.