pipe_feature_interactions: Generates permutation interaction effects between sets of...

In jeroenvdhoven/datapiper: datapiper

Description

Generates permutation interaction effects between sets of numeric variables

Usage

pipe_feature_interactions(train, response, columns = 10L,
  max_interactions = 2)

Arguments

train	The train dataset, as a data.frame or data.table. Data.tables may be changed by reference.
response	The column containing the response variable.
columns	Columns to use for interaction effects. Can be a character vector referencing numeric columns, or an integer larger than 2 denoting the minimum number of unique values in any column except response for that column to be considered for interaction effects. This is intended to exclude ordinal columns using numeric representations.
max_interactions	The maximum number of columns that will be considered for interaction effects per variable. E.g. for the value 3, this function will generate ALL 2-way interactions between variables and ALL 3-way interactions between variables. Caution is advised to not set this value too high. Defaults to 2.

Details

Thanks Eduardo.

Value

A list containing the transformed train dataset and a trained pipe.

jeroenvdhoven/datapiper documentation built on July 14, 2019, 9:34 p.m.