Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/ziplab/EcoFormer
[NeurIPS 2022 Spotlight] This is the official PyTorch implementation of "EcoFormer: Energy-Saving Attention with Linear Complexity"
https://github.com/ziplab/EcoFormer
classification efficient-transformers neurips-2022 pytorch vision-transformer
Last synced: 3 months ago
JSON representation
[NeurIPS 2022 Spotlight] This is the official PyTorch implementation of "EcoFormer: Energy-Saving Attention with Linear Complexity"
- Host: GitHub
- URL: https://github.com/ziplab/EcoFormer
- Owner: ziplab
- License: apache-2.0
- Created: 2022-09-15T08:51:52.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2022-11-15T04:08:04.000Z (over 1 year ago)
- Last Synced: 2024-01-20T15:57:29.661Z (6 months ago)
- Topics: classification, efficient-transformers, neurips-2022, pytorch, vision-transformer
- Language: Python
- Homepage:
- Size: 786 KB
- Stars: 66
- Watchers: 5
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Lists
- awesome-stars - ziplab/EcoFormer - [NeurIPS 2022 Spotlight] This is the official PyTorch implementation of "EcoFormer: Energy-Saving Attention with Linear Complexity" (Python)
README
# EcoFormer: Energy-Saving Attention with Linear Complexity（NeurIPS 2022 Spotlight) 🚀
[](https://opensource.org/licenses/Apache-2.0)
This is the official PyTorch implementation of [EcoFormer: Energy-Saving Attention with Linear Complexity](https://arxiv.org/abs/2209.09004) by [Jing Liu](https://scholar.google.com/citations?user=-lHaZH4AAAAJ&hl=en), [Zizheng Pan](https://scholar.google.com.au/citations?user=w_VMopoAAAAJ&hl=en), [Haoyu He](https://scholar.google.com.au/citations?user=aU1zMhUAAAAJ&hl=en), [Jianfei Cai](https://scholar.google.com/citations?user=N6czCoUAAAAJ&hl=en), and [Bohan Zhuang](https://scholar.google.com.au/citations?user=DFuDBBwAAAAJ).
## News
- **11/11/2022.** EcoFormer is selected as Spotlight!
- **08/10/2022.** We release the source code. Any issues are welcomed!
- **15/09/2022.** EcoFormer is accepted by NeurIPS 2022! 🔥🔥🔥
## A Gentle Introduction
We present a novel energy-saving attention mechanism with linear complexity, called EcoFormer, to save the vast majority of multiplications from a new binarization perspective. More details can be found in our [paper](https://arxiv.org/abs/2209.09004).
## Installation
### Requirements
- Python ≥ 3.8
- PyTorch 1.10.1
- CUDA 11.1
- Torchvision 0.11.2
- PyTorch Image Models (timm) 0.4.9
- MMCV 1.3.8
- Einops 0.4.1
- SciPy 1.8.0
### Instructions
Use [Anaconda](https://www.anaconda.com) to create the running environment for the project, kindly run
```bash
git clone https://github.com/ziplab/EcoFormer
cd EcoFormer
conda env create -f environment/environment.yml
conda activate ecoformer
```
**Note**: If the above instructions does not work on your machine, please refer to [environment/README.md](./environment/README.md) for manual installation and trouble shootings.
## Getting Started
For experiments on PVTv2, please refer to [pvt](./pvt).
For experiments on twins, please refer to [twins](./twins).
## Results and Model Zoo
| Model | #Mul. (B) | #Add. (B) | Energy (B pJ) | Throughput (images/s) | Top-1 Acc. (%) | Download |
| ----------- | --------- | --------- | ------------- | --------------------- | -------------- | ---------------------------------------------------------------------------------------------- |
| PVTv2-B0 | 0.54 | 0.56 | 2.5 | 1379 | 70.44 | [Github](https://github.com/ziplab/EcoFormer/releases/download/v1.0/pvtv2_b0_ecoformer.pth) |
| PVTv2-B1 | 2.03 | 2.09 | 9.4 | 874 | 78.38 | [Github](https://github.com/ziplab/EcoFormer/releases/download/v1.0/pvtv2_b1_ecoformer.pth) |
| PVTv2-B2 | 3.85 | 3.97 | 17.8 | 483 | 81.28 | [Github](https://github.com/ziplab/EcoFormer/releases/download/v1.0/pvtv2_b2_ecoformer.pth) |
| PVTv2-B3 | 6.54 | 6.75 | 30.25 | 325 | 81.96 | [Github](https://github.com/ziplab/EcoFormer/releases/download/v1.0/pvtv2_b3_ecoformer.pth) |
| PVTv2-B4 | 9.57 | 9.82 | 44.25 | 249 | 81.90 | [Github](https://github.com/ziplab/EcoFormer/releases/download/v1.0/pvtv2_b4_ecoformer.pth) |
| Twins-SVT-S | 2.72 | 2.81 | 12.6 | 576 | 80.22 | [Github](https://github.com/ziplab/EcoFormer/releases/download/v1.0/twins_svt_s_ecoformer.pth) |
## Citation
If you find ``EcoFormer`` useful in your research, please consider to cite the following related papers:
```BibTeX
@inproceedings{liu2022ecoformer,
title={EcoFormer: Energy-Saving Attention with Linear Complexity},
author={Liu, Jing and Pan, Zizheng and He, Haoyu and Cai, Jianfei and Zhuang, Bohan},
booktitle={NeurIPS},
year={2022}
}
```
## License
This repository is released under the Apache 2.0 license as found in the [LICENSE](./LICENSE) file.
## Acknowledgement
This repository is built upon [PVT](https://github.com/whai362/PVT) and [Twins](https://github.com/Meituan-AutoML/Twins). We thank the authors for their open-sourced code.