Closed shudong-zhang closed 5 years ago
I solved the issue. There is now a sample adversarial image and its corresponding clean image. Accuracy.py can be used to check the model prediction being restored.
thank you very much!I have already solved the problem with my own label file.
Aamir Mustafa notifications@github.com于2019年1月10日 周四18:46写道:
I solved the issue. There is now a sample adversarial image and its corresponding clean image. Accuracy.py can be used to check the model prediction being restored.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/aamir-mustafa/super-resolution-adversarial-defense/issues/2#issuecomment-453052085, or mute the thread https://github.com/notifications/unsubscribe-auth/AWI6-tILpO1o5oLN_wg1lSK7NML1eqy5ks5vBxnygaJpZM4Z2r-f .
Please upload Corresponding files. Thanks!