fra31

Hi, thanks for the submission! I ran the evaluation with Linf-bound `eps=8/255` and got CIFAR-10 clean accuracy: 85.32% robust accuracy 51.12% CIFAR-100 clean accuracy: 62.15% robust accuracy 26.94% which seem...

Added, thanks again for the submissions!

Hi, I think a solution could be to use a wrapper like ```python class NewModel(): def __init__(self, model): self.model = model def __call__(self, x): return self.model(x)[0] ``` on your model,...

Hi, thanks for sharing your new model! I've updated the entry. I'll also change the title on RobustBench.

Hi, you can find [here](https://github.com/fra31/auto-attack#how-to-use-autoattack) a quick example of how to use AA. If you want your model to be added to the list, you can check [this](https://github.com/fra31/auto-attack/issues/new/choose). Let me...

Hi, thanks for the interest in our work! Your code looks fine to me, just I wouldn't use any interpolation when plotting the images (i.e. `interpolation='none'`). If you mean that...

Looks fine to me. As sanity check you can run `apgd-ce` as last attack by modifying the order in the `self.attacks_to_run` list.

Hi, it's a bit hard to say with little context, but I guess it might be because the flag `is_tf_model=True` is not set (see [here](https://github.com/fra31/auto-attack#tensorflow-models)).

Hi, could you please point to the exact points where this happens? I think there should always be an `if` statement to distinguish whether the model is a TF one...

Sorry for the late reply. This https://github.com/fra31/auto-attack/pull/92 should fix it.