HarmonySearchFS<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Wrapper

Harmony Search algorithm for feature selection optimization.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Like musicians improvising together, this algorithm creates new feature combinations by blending elements from known good combinations, with occasional random exploration.

How It Works

Inspired by musical improvisation, Harmony Search maintains a “harmony memory” of good solutions and creates new ones by combining and modifying existing harmonies.

Constructors

Constructor Summary
HarmonySearchFS(int nFeaturesToSelect, int harmonyMemorySize, int nIterations, double harmonyMemoryConsideringRate, double pitchAdjustingRate, Func<Matrix<T>, Vector<T>, bool[], T> fitnessFunc, int? randomState, int[] columnIndices)

Properties

Property Summary
T BestFitness { get; }
Vector<T> FeatureImportances { 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)