Skip to contents

Randomly zero out entire channels (a channel is a 3D feature map, e.g., the \(j\)-th channel of the \(i\)-th sample in the batched input is a 3D tensor \(\mbox{input}[i, j]\)).


nn_dropout3d(p = 0.5, inplace = FALSE)



(float, optional): probability of an element to be zeroed.


(bool, optional): If set to TRUE, will do this operation in-place


Each channel will be zeroed out independently on every forward call with probability p using samples from a Bernoulli distribution. Usually the input comes from nn_conv2d modules.

As described in the paper Efficient Object Localization Using Convolutional Networks , if adjacent pixels within feature maps are strongly correlated (as is normally the case in early convolution layers) then i.i.d. dropout will not regularize the activations and will otherwise just result in an effective learning rate decrease.

In this case, nn_dropout3d will help promote independence between feature maps and should be used instead.


  • Input: \((N, C, D, H, W)\)

  • Output: \((N, C, D, H, W)\) (same shape as input)


if (torch_is_installed()) {
m <- nn_dropout3d(p = 0.2)
input <- torch_randn(20, 16, 4, 32, 32)
output <- m(input)