Dropout2d

Dropout2d(p=0.5, inplace=False)

Zero whole channels of a channels-last 2-d feature map, (N, H, W, C) or (H, W, C), with probability p and scale the rest by 1 / (1 - p) while training; the identity in evaluation mode.

forward

forward(self, input: 'Tensor') -> 'Tensor'

forward(input) -> Tensor