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.
40 Dec 21, 2022
133 Dec 21, 2022
34 Jan 04, 2023
58 Dec 17, 2022
64 Dec 13, 2022
89 Jan 02, 2023
113 Dec 20, 2022
51 Nov 29, 2022
615 Dec 25, 2022
70 Jan 01, 2023
20 Jun 16, 2021
3 Jan 09, 2022
21 Jan 02, 2022
10 Sep 26, 2022
58 Dec 23, 2022
23 Oct 19, 2022
69 Dec 17, 2022
1 Nov 09, 2021
0 Oct 19, 2021
724 Dec 26, 2022