This is a fork of the ARES source code, adapted for use with our models. A description of the original code, as well as the MIT licence for it can be found here: https://pythonlang.dev/repo/thu-ml-ares/.
A Python library for adversarial machine learning focusing on benchmarking adversarial robustness.
pytorch_ares/
: PyTorch implementation of ares1.0.easyrobust/
: EasyRobust is a tool for training your robust models. Now it support adversarial / non-adversarial training of CNN / ViT models on ImageNet.