FloatingSearch<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Wrapper

Sequential Floating Forward Selection (SFFS) for feature selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Regular forward selection adds features one at a time but never reconsiders its choices. SFFS can “float” - after adding a feature, it checks if any previously added features are now unnecessary and removes them. This leads to better final selections.

How It Works

SFFS is an advanced wrapper method that performs forward selection with conditional backward elimination. After each forward step, it attempts to remove features that may have become redundant.

Constructors

Constructor Summary
FloatingSearch(int nFeaturesToSelect, int maxBacktrack, int[] columnIndices)

Properties

Property Summary
Vector<T> FeatureScores { get; }
bool SupportsInverseTransform { get; }
int MaxBacktrack { 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)