error: Error Calculation

In audachang/recommenderlab.test: Lab for Developing and Testing Recommender Algorithms

Description

Calculate the mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE) and for matrices also the Frobenius norm (identical to RMSE).

Usage

MSE(true, predicted, na.rm = TRUE)
RMSE(true, predicted, na.rm = TRUE)
MAE(true, predicted, na.rm = TRUE)
frobenius(true, predicted, na.rm = TRUE)

Arguments

true	true values.
predicted	predicted values
na.rm	ignore missing values.

Details

Frobenius norm requires matrices.

Value

The error value.

Examples

true <- rnorm(10)
predicted <- rnorm(10)

MAE(true, predicted)
MSE(true, predicted)
RMSE(true, predicted)

true <- matrix(rnorm(9), nrow = 3)
predicted <- matrix(rnorm(9), nrow = 3)

frobenius(true, predicted)

audachang/recommenderlab.test documentation built on May 20, 2019, 1:27 p.m.