grad_relu: Compute the gradient of the relu activation function

View source: R/netzuko.R

grad_reluR Documentation

Compute the gradient of the relu activation function

Description

Compute the gradient of the relu activation function

Usage

grad_relu(s)

Arguments

s

The linear predictors

Value

The gradient of logistic activation evaluated at s

Note

For Internal Use. We set gradient at 0 to be 0.


billyhw/netzuko documentation built on March 23, 2022, 4:26 p.m.