# SHViT

**Category:** Efficient/Mobile · **Code:**
[`timm/models/shvit.py`](https://github.com/huggingface/pytorch-image-models/blob/main/timm/models/shvit.py) ·
**Architectures:** 4 · **Pretrained weights:** 4 (4 with ImageNet-1k results)

SHViT (Single-Head Vision Transformer) uses a memory efficient macro design and single-head attention applied to part of
the channels.

## Papers

* [SHViT: Single-Head Vision Transformer with Memory Efficient Macro Design](https://arxiv.org/abs/2401.16456)

## Usage

See the [Quickstart](../quickstart) for loading pretrained weights, image preprocessing and fine-tuning, and [Feature
Extraction](../feature_extraction) for using the models as backbones.

```py
>>> import timm
>>> model = timm.create_model('shvit_s4.in1k', pretrained=True)
>>> data_config = timm.data.resolve_model_data_config(model)
```

## Model API support

| Feature | `SHViT` |
|---|---|
| [`features_only=True` feature maps](../model_api#intermediate-features) | ✓ (strides 16, 32, 64) |
| [`forward_intermediates()`](../model_api#intermediate-features) | ✓ |
| [`prune_intermediate_layers()`](../model_api#intermediate-features) | ✓ |
| [Any input size, no changes needed](../model_api#input-size) | ✓ |
| [`set_input_size()`](../model_api#input-size) | – |
| [`dynamic_img_size=True`](../model_api#input-size) | – |
| [`set_grad_checkpointing()`](../model_api#training-hooks) | ✓ |
| [Layer-wise LR decay (`group_matcher()`)](../model_api#training-hooks) | ✓ |
| [`reset_classifier()`](../model_api#classifier) | ✓ |

✓ supported, ✓ (square) works for square sizes only, ✗ not supported, – not needed (the model already accepts any input
size). Checked by running each feature on a randomly initialized model (`SHViT` with `shvit_s1` at 224), other
architectures of the same class are expected to behave the same. See the [Model API](../model_api) page for details.

## Architectures

| Architecture | Params (M) | GMACs | Input size | Pretrained weights |
|---|---:|---:|---:|---:|
| `shvit_s1` | 6.31 | 0.24 | 224 | 1 |
| `shvit_s2` | 11.45 | 0.37 | 224 | 1 |
| `shvit_s3` | 14.21 | 0.60 | 224 | 1 |
| `shvit_s4` | 16.55 | 0.99 | 256 | 1 |

GMACs and parameter counts are for the default input size and classifier of each architecture, from the [benchmark
results](https://github.com/huggingface/pytorch-image-models/tree/main/results).

## Pretrained weights

Top-1 is ImageNet-1k validation accuracy from
[`results-imagenet.csv`](https://github.com/huggingface/pytorch-image-models/blob/main/results/results-imagenet.csv) at
the listed eval size, - if the weight has no ImageNet-1k result (e.g. pretraining or image-text weights). Classes is the
size of the pretrained classifier, none for weights without one.

| Weight | Input size | Top-1 (%) | Eval size | Classes | License | Hub |
|---|---:|---:|---:|---:|---|---|
| `shvit_s1.in1k` | 224 | 72.780 | 224 | 1000 | mit | [timm](https://huggingface.co/timm/shvit_s1.in1k) |
| `shvit_s2.in1k` | 224 | 75.184 | 224 | 1000 | mit | [timm](https://huggingface.co/timm/shvit_s2.in1k) |
| `shvit_s3.in1k` | 224 | 77.356 | 224 | 1000 | mit | [timm](https://huggingface.co/timm/shvit_s3.in1k) |
| `shvit_s4.in1k` | 256 | 79.362 | 256 | 1000 | mit | [timm](https://huggingface.co/timm/shvit_s4.in1k) |

## Citation

```bibtex
@inproceedings{yun2024shvit,
  author={Yun, Seokju and Ro, Youngmin},
  title={SHViT: Single-Head Vision Transformer with Memory Efficient Macro Design},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  pages={5756--5767},
  year={2024}
}
```