Value Error ("The shape of mean and the standard deviation must be identical") in attacks

[abhinav-vimal-17]

1. adv = ProjectedGradientDescent(classifier, targeted=False, max_iter=10, eps_step=1, eps=5)
2. adv_img = adv.generate(adv_in_img), where adv_in_img is a numpy array

In the second line, I am getting Value Error ("The shape of mean and the standard deviation must be identical")

[beat-buesser]

Hi @abhinav-vimal-17 How did you define the mean and the standard deviation in the definition of the classifier?