pytorch-cifar10
pytorch-cifar10 copied to clipboard
questions about shakedrop
In the original paper, shakedrop is given as G(x) = x + (bl + a + bl*a)F(x), but your application makes out = (drop_factor + alpha + alpha * (1 - drop_factor)) * input. Is that a slip of a pen?
@stupiding I have confused about that. I will fix it. Thank you!