Maksym Andriushchenko

Sorry for missing the original message! Yes, exactly, all the statistics are taken over 25 seeds. Regarding what exactly we report, it differs between datasets (we just follow the literature...

@fra31: agreed, throwing an error sounds good as a temporary solution. And I think it's important to preserve backward compatibility as returning directly tensors has been useful to simplify code...

@dedeswim that would be fantastic!

Hi, Thanks for the interest in our paper. 2*eps is there to ensure that we can go from one corner of the Linf cube to the other, i.e. change a...

Sorry, I somehow missed your reply. [-eps, eps] in line 236 is correct since here we modify the values of the deltas **directly** instead of adding a perturbation to an...

closing this issue for now. feel free to reopen if the problem persists!

Hi @ruitian12, I've updated the leaderboard https://robustbench.github.io/#div_imagenet_corruptions_heading with your models. The numbers are close to your latest ones but not exactly the same (~0.4% difference). This doesn't change the ranking...

perfect! then i'll close this issue

hi, if i remember correctly, it's just needed to make sure that we can have more adversarial (`adv`) samples than the batch size (e.g., up to 3x more). note that...