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

Add ConvNeXT (#15277)

Browse Source 
* First draft

* Add conversion script

* Improve conversion script

* Improve docs and implement tests

* Define model output class

* Fix tests

* Fix more tests

* Add model to README

* Apply suggestions from code review

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

* Apply more suggestions from code review

* Apply suggestions from code review

* Rename dims to hidden_sizes

* Fix equivalence test

* Rename gamma to gamma_parameter

* Clean up conversion script

* Add ConvNextFeatureExtractor

* Add corresponding tests

* Implement feature extractor correctly

* Make implementation cleaner

* Add ConvNextStem class

* Improve design

* Update design to also include encoder

* Fix gamma parameter

* Use sample docstrings

* Finish conversion, add center cropping

* Replace nielsr by facebook, make feature extractor tests smaller

* Fix integration test

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
This commit is contained in:
NielsRogge
2022-02-07 16:11:37 +01:00
committed by GitHub
parent c47d259241
commit 84eec9e6ba
21 changed files with 1732 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/_toctree.yml
View File

@@ -150,6 +150,8 @@
title: CamemBERT
- local: model_doc/canine
title: CANINE
- local: model_doc/convnext
title: ConvNeXT
- local: model_doc/clip
title: CLIP
- local: model_doc/convbert

2
docs/source/index.mdx
View File

@@ -70,6 +70,7 @@ conversion utilities for the following models.
1. **[ByT5](model_doc/byt5)** (from Google Research) released with the paper [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://arxiv.org/abs/2105.13626) by Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel.
1. **[CamemBERT](model_doc/camembert)** (from Inria/Facebook/Sorbonne) released with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Louis Martin*, Benjamin Muller*, Pedro Javier Ortiz Suárez*, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah and Benoît Sagot.
1. **[CANINE](model_doc/canine)** (from Google Research) released with the paper [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://arxiv.org/abs/2103.06874) by Jonathan H. Clark, Dan Garrette, Iulia Turc, John Wieting.
1. **[ConvNeXT](model_doc/convnext)** (from Facebook AI) released with the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, Saining Xie.
1. **[CLIP](model_doc/clip)** (from OpenAI) released with the paper [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020) by Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, Ilya Sutskever.
1. **[ConvBERT](model_doc/convbert)** (from YituTech) released with the paper [ConvBERT: Improving BERT with Span-based Dynamic Convolution](https://arxiv.org/abs/2008.02496) by Zihang Jiang, Weihao Yu, Daquan Zhou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan.
1. **[CPM](model_doc/cpm)** (from Tsinghua University) released with the paper [CPM: A Large-scale Generative Chinese Pre-trained Language Model](https://arxiv.org/abs/2012.00413) by Zhengyan Zhang, Xu Han, Hao Zhou, Pei Ke, Yuxian Gu, Deming Ye, Yujia Qin, Yusheng Su, Haozhe Ji, Jian Guan, Fanchao Qi, Xiaozhi Wang, Yanan Zheng, Guoyang Zeng, Huanqi Cao, Shengqi Chen, Daixuan Li, Zhenbo Sun, Zhiyuan Liu, Minlie Huang, Wentao Han, Jie Tang, Juanzi Li, Xiaoyan Zhu, Maosong Sun.
@@ -176,6 +177,7 @@ Flax), PyTorch, and/or TensorFlow.
| Canine | ✅ | ❌ | ✅ | ❌ | ❌ |
| CLIP | ✅ | ✅ | ✅ | ✅ | ✅ |
| ConvBERT | ✅ | ✅ | ✅ | ✅ | ❌ |
| ConvNext | ❌ | ❌ | ✅ | ❌ | ❌ |
| CTRL | ✅ | ❌ | ✅ | ✅ | ❌ |
| DeBERTa | ✅ | ✅ | ✅ | ✅ | ❌ |
| DeBERTa-v2 | ✅ | ❌ | ✅ | ✅ | ❌ |

66
docs/source/model_doc/convnext.mdx Normal file
View File

@@ -0,0 +1,66 @@
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
-->
# ConvNeXT
## Overview
The ConvNeXT model was proposed in [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, Saining Xie.
ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them.
The abstract from the paper is the following:
*The "Roaring 20s" of visual recognition began with the introduction of Vision Transformers (ViTs), which quickly superseded ConvNets as the state-of-the-art image classification model.
A vanilla ViT, on the other hand, faces difficulties when applied to general computer vision tasks such as object detection and semantic segmentation. It is the hierarchical Transformers
(e.g., Swin Transformers) that reintroduced several ConvNet priors, making Transformers practically viable as a generic vision backbone and demonstrating remarkable performance on a wide
variety of vision tasks. However, the effectiveness of such hybrid approaches is still largely credited to the intrinsic superiority of Transformers, rather than the inherent inductive
biases of convolutions. In this work, we reexamine the design spaces and test the limits of what a pure ConvNet can achieve. We gradually "modernize" a standard ResNet toward the design
of a vision Transformer, and discover several key components that contribute to the performance difference along the way. The outcome of this exploration is a family of pure ConvNet models
dubbed ConvNeXt. Constructed entirely from standard ConvNet modules, ConvNeXts compete favorably with Transformers in terms of accuracy and scalability, achieving 87.8% ImageNet top-1 accuracy
and outperforming Swin Transformers on COCO detection and ADE20K segmentation, while maintaining the simplicity and efficiency of standard ConvNets.*
Tips:
- See the code examples below each model regarding usage.
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/convnext_architecture.jpg"
alt="drawing" width="600"/>
<small> ConvNeXT architecture. Taken from the <a href="https://arxiv.org/abs/2201.03545">original paper</a>.</small>
This model was contributed by [nielsr](https://huggingface.co/nielsr). The original code can be found [here](https://github.com/facebookresearch/ConvNeXt).
## ConvNeXT specific outputs
[[autodoc]] models.convnext.modeling_convnext.ConvNextModelOutput
## ConvNextConfig
[[autodoc]] ConvNextConfig
## ConvNextFeatureExtractor
[[autodoc]] ConvNextFeatureExtractor
## ConvNextModel
[[autodoc]] ConvNextModel
- forward
## ConvNextForImageClassification
[[autodoc]] ConvNextForImageClassification
- forward