KruskalWallisTest<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter.Univariate

Kruskal-Wallis H test for non-parametric feature selection.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → KruskalWallisTest<T>

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Unlike ANOVA which assumes data follows a normal bell curve, Kruskal-Wallis works with any data distribution. It converts values to ranks and tests whether the ranks differ between groups. This is more robust when your data has outliers or non-normal distributions.

How It Works

A non-parametric alternative to one-way ANOVA that doesn’t assume normal distributions. Tests whether samples from different groups originate from the same distribution based on ranks.

Constructors

Constructor Summary
KruskalWallisTest(int nFeaturesToSelect, double pValueThreshold, int[] columnIndices)

Properties

Property Summary
Vector<T> HStatistics { get; }
Vector<T> PValues { get; }
bool SupportsInverseTransform { get; }
int NFeaturesToSelect { get; }
int[] SelectedIndices { get; }

Methods

Method Summary
Matrix<T> FitTransform(Matrix<T> data, Vector<T> target)
bool[] GetSupportMask()
string[] GetFeatureNamesOut(string[] inputFeatureNames)
void Fit(Matrix<T> data, Vector<T> target)