HarmonySearch<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Metaheuristic

Harmony Search for feature selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Imagine a band trying to write the perfect song. They remember good riffs from past songs (harmony memory) and sometimes try completely new notes (random pitch). The best combinations are kept for future improvisation. In feature selection, we’re looking for the perfect “harmony” of features.

How It Works

Harmony Search is inspired by the improvisation process of musicians. Just as musicians create harmonies by combining notes from memory with random variations, this algorithm creates solutions by mixing previous good solutions with random exploration.

Constructors

Constructor Summary
HarmonySearch(int nFeaturesToSelect, int harmonyMemorySize, int maxIterations, double hmcr, double par, double bw, int? randomState, int[] columnIndices)

Properties

Property Summary
Vector<T> FeatureScores { get; }
bool SupportsInverseTransform { get; }
double HMCR { get; }
double PAR { get; }
int HarmonyMemorySize { get; }
int MaxIterations { 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)