huber_loss: Huber Loss function (smooth mean absolute error) with...

In tbonza/supml: "Algorithms for Supervised Machine Learning"

Description

Huber Loss function (smooth mean absolute error) with parameter delta

Usage

huber_loss(e, delta)

Arguments

e	vector of error terms Yi - Y^hat_i
delta	const for determining huber loss conditional

Value

huber loss function error

Examples

e <- matrix(data=0, nrow=10,ncol=1)
delta <- 1.0
huber_loss(e, delta)

e <- rnorm(10, 0, 1)
huber_loss(e, delta)

tbonza/supml documentation built on May 17, 2019, 3:14 a.m.