Softmax2d module
Applies SoftMax over features to each spatial location.
When given an image of Channels x Height x Width, it will
apply Softmax to each location (Channels, h_i, w_j)
nn_softmax2d()
a Tensor of the same dimension and shape as the input with
values in the range [0, 1]
Input: (N, C, H, W)
Output: (N, C, H, W) (same shape as input)
if (torch_is_installed()) {
m <- nn_softmax2d()
input <- torch_randn(2, 3, 12, 13)
output <- m(input)
}Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.