# nn_hardswish: Hardswish module In torch: Tensors and Neural Networks with 'GPU' Acceleration

## Description

Applies the hardswish function, element-wise, as described in the paper: Searching for MobileNetV3

## Usage

 1 nn_hardswish() 

## Details

\mbox{Hardswish}(x) = ≤ft\{ \begin{array}{ll} 0 & \mbox{if } x ≤ -3, \\ x & \mbox{if } x ≥ +3, \\ x \cdot (x + 3)/6 & \mbox{otherwise} \end{array} \right.

## Shape

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

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

## Examples

 1 2 3 4 5 6 7 8 9 if (torch_is_installed()) { ## Not run: m <- nn_hardswish() input <- torch_randn(2) output <- m(input) ## End(Not run) } 

torch documentation built on Oct. 7, 2021, 9:22 a.m.