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

Doc new front (#14590)

Browse Source 
* Convert PretrainedConfig doc to Markdown

* Use syntax

* Add necessary doc files (#14496)

* Doc fixes (#14499)

* Fixes for the new front

* Convert DETR file for table

* Title is needed

* Simplify a bit

* Even simpler

* Remove imports

* Fix typo in toctree (#14516)

* Fix checkpoints badge

* Update versions.yml format (#14517)

* Doc new front github actions (#14512)

* Doc new front github actions

* Fix docstring

* Fix feature extraction utils import (#14515)

* Address Julien's comments

* Push to doc-builder

* Ready for merge

* Remove old build and deploy

* Doc misc fixes (#14583)

* Rm versions.yml from doc

* Fix converting.rst

* Rm pretrained_models from toctree

* Fix index links (#14567)

* Fix links in README

* Localized READMEs

* Fix copy script

* Fix find doc script

* Update README_ko.md

Co-authored-by: Julien Chaumond <julien@huggingface.co>

Co-authored-by: Julien Chaumond <julien@huggingface.co>

* Adapt build command to new CLI tools (#14578)

* Fix typo

* Fix doc interlinks (#14589)

* Convert PretrainedConfig doc to Markdown

* Use syntax

* Rm pattern <[a-z]+(.html).*>

* Rm huggingface.co/transformers/master

* Rm .html

* Rm .html from index.mdx

* Rm .html from model_summary.rst

* Update index.mdx rm html

* Update remove .html

* Fix inner doc links

* Fix interlink in preprocssing.rst

* Update pr_checks

Co-authored-by: Sylvain Gugger <sylvain.gugger@gmail.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Convert PretrainedConfig doc to Markdown

* Use syntax

* Add necessary doc files (#14496)

* Doc fixes (#14499)

* Fixes for the new front

* Convert DETR file for table

* Title is needed

* Simplify a bit

* Even simpler

* Remove imports

* Fix checkpoints badge

* Fix typo in toctree (#14516)

* Update versions.yml format (#14517)

* Doc new front github actions (#14512)

* Doc new front github actions

* Fix docstring

* Fix feature extraction utils import (#14515)

* Address Julien's comments

* Push to doc-builder

* Ready for merge

* Remove old build and deploy

* Doc misc fixes (#14583)

* Rm versions.yml from doc

* Fix converting.rst

* Rm pretrained_models from toctree

* Fix index links (#14567)

* Fix links in README

* Localized READMEs

* Fix copy script

* Fix find doc script

* Update README_ko.md

Co-authored-by: Julien Chaumond <julien@huggingface.co>

Co-authored-by: Julien Chaumond <julien@huggingface.co>

* Adapt build command to new CLI tools (#14578)

* Fix typo

* Fix doc interlinks (#14589)

* Convert PretrainedConfig doc to Markdown

* Use syntax

* Rm pattern <[a-z]+(.html).*>

* Rm huggingface.co/transformers/master

* Rm .html

* Rm .html from index.mdx

* Rm .html from model_summary.rst

* Update index.mdx rm html

* Update remove .html

* Fix inner doc links

* Fix interlink in preprocssing.rst

* Update pr_checks

Co-authored-by: Sylvain Gugger <sylvain.gugger@gmail.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Styling

Co-authored-by: Mishig Davaadorj <mishig.davaadorj@coloradocollege.edu>
Co-authored-by: Lysandre Debut <lysandre@huggingface.co>
Co-authored-by: Julien Chaumond <julien@huggingface.co>
This commit is contained in:
Sylvain Gugger
2021-12-01 14:13:02 -05:00
committed by GitHub
parent 14cc50d081
commit 4df7d05a87
43 changed files with 1467 additions and 3185 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
docs/source/model_doc/deit.rst
View File

@@ -25,12 +25,12 @@ Overview
The DeiT model was proposed in `Training data-efficient image transformers & distillation through attention
<https://arxiv.org/abs/2012.12877>`__ by Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre
Sablayrolles, Hervé Jégou. The `Vision Transformer (ViT) <https://huggingface.co/transformers/model_doc/vit.html>`__
introduced in `Dosovitskiy et al., 2020 <https://arxiv.org/abs/2010.11929>`__ has shown that one can match or even
outperform existing convolutional neural networks using a Transformer encoder (BERT-like). However, the ViT models
introduced in that paper required training on expensive infrastructure for multiple weeks, using external data. DeiT
(data-efficient image transformers) are more efficiently trained transformers for image classification, requiring far
less data and far less computing resources compared to the original ViT models.
Sablayrolles, Hervé Jégou. The `Vision Transformer (ViT) <vit>`__ introduced in `Dosovitskiy et al., 2020
<https://arxiv.org/abs/2010.11929>`__ has shown that one can match or even outperform existing convolutional neural
networks using a Transformer encoder (BERT-like). However, the ViT models introduced in that paper required training on
expensive infrastructure for multiple weeks, using external data. DeiT (data-efficient image transformers) are more
efficiently trained transformers for image classification, requiring far less data and far less computing resources
compared to the original ViT models.
The abstract from the paper is the following: