LiJinqiao

G is updated to nan

7

I'm attacking the imageNet image. I used the Inceptionv3 pre-training model provided by Pytorch. The following parameters are used for normalization: `mean = torch.Tensor([0.485, 0.456, 0.406]).view(1,3,1,1)` `std = torch.Tensor([0.229, 0.224,...

How do I adjust the hyperparameters or modify the code so that I don't use structural attacks?

Thanks for the code！ I found this method to be slightly slower, taking 300 seconds to attack a CIFAR10 image. In the demo you provided, a batch contains only one...

Thanks for the code！ Can you provide the pre-trained generator model for the CIFAR10 dataset? I only see the model for ImageNet.