NeoSunJZ/Canary_Master
CanarySEFI is a framework for evaluating the robustness of deep learning-based image recognition models. It can evaluate model robustness and attack/defense algorithm effectiveness, encompassing 26 metrics, including 15 models pre-trained by CIFAR-10, CIFAR-100, Fashion-MNIST, and ImageNet (with weights), 20+ attack methods, and 10+ defense methods
PythonApache-2.0
Stargazers
- az1mus
- Blamejin
- ChiunHauYouAlibaba
- dengdcomOppo
- e5dese
- emmmmmmmnmn
- eternalldBeijing Institute of Technology
- FabriceYoparMy Home
- flame737
- gisliteBejing, China
- hlmu
- HuaizhulaFUWaymo
- Jump-Wang-111Beijing Institute of Technology
- KevinelloTencent
- leid19
- Lillyyouwu
- luckyxyhaha
- MartinchlerFathom Information Design
- MustafaFiserCloudreach
- NeoSunJZFudan University
- RHLotus
- SepehrSameni
- spencerwooo@tanlab-bit @bitnp
- ST-CharaTeeFun
- summorsmi
- tdlllNo. 5 Zhongguancun South Street, Haidian District, Beijing, China
- whoamTV
- wqh-31
- xbx12138
- xijwnxBeijing
- XuXingbo
- XYchain
- yzzmmわせだだいがく
- Zhou-AYHuazhong University of Science and Technology
- zhu1024Beijing Institute of Technology
- zsaba