AnalysisBySynthesis
AnalysisBySynthesis copied to clipboard
Adversarially Robust Neural Network on MNIST.
am getting the following error while running \scripts\attacks.py": fmodel = foolbox.models.PyTorchModel(model, # return logits in shape (bs, n_classes) TypeError: __init__() got an unexpected keyword argument 'num_classes'
Hi, I'm trying to reproduce your experiment results recently, but encountered several technical problems(mostly from foolbox), I think it would be really helpful if you could share the evaluatation code....
@lukas-schott , thank you for your implementation, can you also show the training code in order to me to evaluate it on other datasets? Thanks
I'm trying to run the demo in attacks.ipynb, but it is really slow, taking over 30 minutes to run the [PointwiseAttack](https://github.com/bethgelab/AnalysisBySynthesis/blob/458ee76af68778fd9a0ea7c4b411fa4c7eeb8682/scripts/attacks.py#L73). I'm using GPU and running the [classification](https://github.com/bethgelab/AnalysisBySynthesis/blob/458ee76af68778fd9a0ea7c4b411fa4c7eeb8682/scripts/attacks.py#L66) was quite...