Code for our paper: "You Only Propagate Once: Painless Adversarial Training Using Maximal Principle" by Dinghuai Zhang, Tianyuan Zhang, Yiping Lu, Zhanxing Zhu, Bin Dong.

• Pytorch==1.0.1, torchvision
• Python 3.5
• tensorboardX
• easydict
• tqdm

git clone https://github.com/a1600012888/YOPO-You-Only-Propagate-Once.git
cd YOPO-You-Only-Propagate-Once
pip3 install -r requirements.txt --user

• normal training: experiments/CIFAR10/wide34.natural
• PGD adversarial training: experiments/CIFAR10/wide34.pgd10 run python train.py -d <whcih_gpu>

You can change all the hyper-parameters in config.py. And change network in network.py Actually code in above mentioned director is very flexible and can be easiliy modified. It can be used as a template.

Go to directory experiments/CIFAR10/wide34.yopo-5-3 run python train.py -d <whcih_gpu>

You can change all the hyper-parameters in config.py. And change network in network.py Runing this code for the first time will dowload the dataset in ./experiments/CIFAR10/data/, you can modify the path in dataset.py