Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
sguokaiwei@163.com 06ed213c88 | 2 years ago | |
---|---|---|
.. | ||
metrics | 2 years ago | |
.DS_Store | 2 years ago | |
README.md | 2 years ago | |
requirements.txt | 2 years ago | |
robustness_validation.py | 2 years ago | |
test_example.sh | 2 years ago | |
utils.py | 2 years ago |
This repository contains the robust models trained on ImageNet, and the scripts for robustness evaluation.
The benchmarked results have been contained in ARES-Bench.
First, clone the repository locally:
git clone https://github.com/alibaba/easyrobust.git
cd easyrobust
pip install -r requirements.txt
Then test runing on ImageNet Validation set:
python robustness_validation.py --model=resnet50 --interpolation=3 --imagenet_val_path=/path/to/ILSVRC/Data/CLS-LOC/val
The trained models will be downloaded automaticly. If you want to download the checkpoints manually, check the urls in utils.py.
The code supports Stylized-ImageNet, ImageNet-V2, ImageNet-R, ImageNet-A, ImageNet-Sketch, ObjectNet, ImageNet-C, AutoAttack evaluation. See test_example.sh for details.
18 Adversarially trained models are opened in utils.py
.
We collect some non-adversarially robust models based on resnet50. To test these models, replace the this line with following urls:
No Description
Python Text Cuda C++ Shell other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》