https://github.com/martinrepo/aisafety
https://github.com/martinrepo/aisafety
Last synced: 8 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/martinrepo/aisafety
- Owner: MartinRepo
- Created: 2023-01-17T22:55:48.000Z (over 3 years ago)
- Default Branch: master
- Last Pushed: 2023-07-12T02:59:22.000Z (almost 3 years ago)
- Last Synced: 2025-03-12T16:43:18.646Z (over 1 year ago)
- Language: Python
- Size: 58.3 MB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# AIsafety
## This is an adversarial training on the fashionminst dataset.
The traning process is in Competition.py, it used pytorch and was trained by some algorithms such as FGSM(Fast Gradient Sign Method), PGD(Projected Gradient descent), CW(Carlini and Wagner Attacks).
The model is 1000.pt and 1001.pt, both of them are available. This model defeated over 95% of my classmates.
Cheers 🍻