nn_selu function

SELU module

Applied element-wise, as:

nn_selu(inplace = FALSE)

Arguments

• inplace: (bool, optional): can optionally do the operation in-place. Default: FALSE

Details

with $\alpha = 1.6732632423543772848170429916717$ and $\mbox{scale} = 1.0507009873554804934193349852946$.

More details can be found in the paper Self-Normalizing Neural Networks.

Shape

• Input: $(N, *)$ where * means, any number of additional dimensions
• Output: $(N, *)$, same shape as the input

Examples

if (torch_is_installed()) {
m <- nn_selu()
input <- torch_randn(2)
output <- m(input)
}