RVT: Robust Vision Transformers
This repository contains PyTorch code for Robust Vision Transformers.
For details see Rethinking the Design Principles of Robust Vision Transformer by Xiaofeng Mao, Gege Qi, Yuefeng Chen, Yuan He and Hui Xue.
Usage
First, clone the repository locally:
git clone https://github.com/vtddggg/Robust-Vision-Transformer.git
Then, install PyTorch 1.7.0+ and torchvision 0.8.1+ and pytorch-image-models 0.3.2:
conda install -c pytorch pytorch torchvision
pip install timm==0.3.2
We use 4 nodes with 8 gpus to train RVT-Ti, RVT-S and RVT-B:
Training RVT-Ti
python -m torch.distributed.launch --nproc_per_node=8 --nnodes=4 main.py --model rvt_tiny --data-path /path/to/imagenet --output_dir output --dist-eval
Training RVT-S
python -m torch.distributed.launch --nproc_per_node=8 --nnodes=4 main.py --model rvt_small --data-path /path/to/imagenet --output_dir output --dist-eval
Training RVT-B
python -m torch.distributed.launch --nproc_per_node=8 --nnodes=4 main.py --model rvt_base --data-path /path/to/imagenet --output_dir output --batch-size 32 --dist-eval
If you want to train RVT-Ti*, RVT-S* or RVT-B*, simply add --use_mask and --use_patch_aug to enable positon-aware attention scaling and patch-wise augmentation.
92 Sep 28, 2021
252 Sep 16, 2022
1 Sep 15, 2022
5 Feb 09, 2022
1.7k Dec 29, 2022
22 Apr 15, 2022
1.7k Jan 08, 2023
16 Oct 17, 2022
1 May 24, 2022
2 Nov 10, 2021
84 Nov 02, 2022
24 Oct 22, 2022
81 Dec 25, 2022
1 Oct 24, 2021
24 Oct 01, 2022
61 Dec 28, 2022
5 Jun 28, 2022
10 May 07, 2022
115 Dec 30, 2022
194 Dec 29, 2022