search

tml-epfl / understanding-fast-adv-training

Understanding and Improving Fast Adversarial Training [NeurIPS 2020]
https://arxiv.org/abs/2007.02617
94 stars 12 forks source link

grad1 in train.py #1

Closed P2333 closed 4 years ago

P2333 commented 4 years ago

Hi Maksym,

Thank you for sharing the code. I think there may lack a definition of grad1 (line 195) in train.py before it is referenced.

max-andr commented 4 years ago

Oh, can be, thanks for telling! Probably, I've refactored the code a bit uncarefully. I will fix it soon!

P2333 commented 4 years ago

Cool! Thank you

max-andr commented 4 years ago

Fixed now (essentially i had to just use grad instead of the undefined grad1).

Thanks a lot for letting me know!