Skip to contents

Create Balanced Training Set

Source: R/SyntheticData.R
balance_classes.Rd

Creates a balanced training set using synthetic data augmentation. Designed for use within cross-validation folds.

Usage

balance_classes(task, method = c("smote", "noise", "undersample"), ratio = 1)

Arguments

task

An mlr3 classification task

method

Augmentation method: "smote", "noise", "undersample"

ratio

Target ratio of minority to majority (default: 1.0)

Value

A balanced task