torch (version 0.8.1)

nn_log_softmax: LogSoftmax module

Description

Applies the \(\log(\mbox{Softmax}(x))\) function to an n-dimensional input Tensor. The LogSoftmax formulation can be simplified as:

Usage

nn_log_softmax(dim)

Value

a Tensor of the same dimension and shape as the input with values in the range [-inf, 0)

Arguments

dim

(int): A dimension along which LogSoftmax will be computed.

Shape

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

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

Details

$$ \mbox{LogSoftmax}(x_{i}) = \log\left(\frac{\exp(x_i) }{ \sum_j \exp(x_j)} \right) $$

Examples

if (torch_is_installed()) {
m <- nn_log_softmax(1)
input <- torch_randn(2, 3)
output <- m(input)
}