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.
3 Apr 23, 2022
29 Sep 04, 2022
1 Oct 10, 2021
123 Jan 05, 2023
1 Sep 05, 2022
12 Jul 08, 2022
159 Dec 12, 2022
6 Oct 26, 2022
6.7k Dec 31, 2022
143 Sep 13, 2022
40 Dec 04, 2022
140 Jan 01, 2023
27 Sep 25, 2022
0 Jan 06, 2022
72 Dec 10, 2022
33 Nov 14, 2022
34 Jan 05, 2023
3 Jul 14, 2021
126 Nov 22, 2022
0 Jan 07, 2022