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: expects: ExampleSet
Output
- model:
- example set:
Parameters
- relative attribute: Select the attribute which should be used as a base for the relative comparison.