torch (version 0.8.1)

nn_softmin: Softmin

Description

Applies the Softmin function to an n-dimensional input Tensor rescaling them so that the elements of the n-dimensional output Tensor lie in the range [0, 1] and sum to 1. Softmin is defined as:

Usage

nn_softmin(dim)

Value

a Tensor of the same dimension and shape as the input, with values in the range [0, 1].

Arguments

dim

(int): A dimension along which Softmin will be computed (so every slice along dim will sum to 1).

Shape

• Input: \((*)\) where * means, any number of additional dimensions

• Output: \((*)\), same shape as the input

Details

$$ \mbox{Softmin}(x_{i}) = \frac{\exp(-x_i)}{\sum_j \exp(-x_j)} $$

Examples

if (torch_is_installed()) {
m <- nn_softmin(dim = 1)
input <- torch_randn(2, 2)
output <- m(input)
}