nn_hardsigmoid: Hardsigmoid module

In torch: Tensors and Neural Networks with 'GPU' Acceleration

Hardsigmoid module

Description

Applies the element-wise function:

Usage

nn_hardsigmoid()

Details

\mbox{Hardsigmoid}(x) = \left\{ \begin{array}{ll} 0 & \mbox{if~} x \le -3, \\ 1 & \mbox{if~} x \ge +3, \\ x / 6 + 1 / 2 & \mbox{otherwise} \end{array} \right.

Shape

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

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

Examples

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

torch documentation built on June 7, 2023, 6:19 p.m.