Relative Regression

Synopsis

Learns a regression model for predictions relative to another attribute value.

Description

This meta regression learner transforms the label on-the-fly relative to the value of the specified attribute. This is done right before the inner regression learner is applied. This can be useful in order to allow time series predictions on data sets with large trends.

Input

training set

Output

model

example set

This is an example set output port

Parameters

relative attribute

Select the attribute which should be used as a base for the relative comparison.