oversample_smote: Oversample a dataset by SMOTE.

In scutr: Balancing Multiclass Datasets for Classification Tasks

Oversample a dataset by SMOTE.

Description

Oversample a dataset by SMOTE.

Usage

oversample_smote(data, cls, cls_col, m, k = NA)

Arguments

data	Dataset to be oversampled.
cls	Class to be oversampled.
cls_col	Column containing class information.
m	Desired number of samples in the oversampled data.
k	Number of neighbors used in SMOTE() to generate synthetic minority instances. This value must be smaller than the number of minority instances already present for a given class. If NA, min(5, n-1) is chosen, where n is the number of instances of the minority class.

Value

The oversampled dataset.

Examples

table(iris$Species)
smoted <- oversample_smote(iris, "setosa", "Species", 100)
nrow(smoted)

scutr documentation built on Nov. 18, 2023, 1:08 a.m.