SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Improve vision models docs (#19103)

Browse Source 
* Add tips

* Add BEiT figure

* Fix URL

* Move tip to start

* Add tip to TF model as well

Co-authored-by: Niels Rogge <nielsrogge@Nielss-MacBook-Pro.local>
This commit is contained in:
NielsRogge
2022-09-19 19:22:34 +02:00
committed by GitHub
parent 0d1ba2dd0b
commit e7206ceab9
9 changed files with 75 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
docs/source/en/model_doc/vit.mdx
View File

@@ -12,13 +12,6 @@ specific language governing permissions and limitations under the License.
# Vision Transformer (ViT)
<Tip>
This is a recently introduced model so the API hasn't been tested extensively. There may be some bugs or slight
breaking changes to fix it in the future. If you see something strange, file a [Github Issue](https://github.com/huggingface/transformers/issues/new?assignees=&labels=&template=bug-report.md&title).
</Tip>
## Overview
The Vision Transformer (ViT) model was proposed in [An Image is Worth 16x16 Words: Transformers for Image Recognition
@@ -63,6 +56,11 @@ Tips:
language modeling). With this approach, the smaller ViT-B/16 model achieves 79.9% accuracy on ImageNet, a significant
improvement of 2% to training from scratch, but still 4% behind supervised pre-training.
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/vit_architecture.jpg"
alt="drawing" width="600"/>
<small> ViT architecture. Taken from the <a href="https://arxiv.org/abs/2010.11929">original paper.</a> </small>
Following the original Vision Transformer, some follow-up works have been made:
- [DeiT](deit) (Data-efficient Image Transformers) by Facebook AI. DeiT models are distilled vision transformers.