SFFS<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Wrapper

Sequential Floating Forward Selection (SFFS) for feature selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Regular forward selection adds features one at a time without looking back. SFFS is smarter - after adding a new feature, it reconsiders whether any previously added features are now redundant. It’s like building a team and periodically reconsidering earlier hires as the team evolves.

How It Works

SFFS is an advanced sequential search that combines forward selection with conditional backward steps. After adding a feature, it tries removing previously added features to see if removing them improves performance. This allows correction of earlier suboptimal decisions.

Constructors

Constructor Summary
SFFS(int nFeaturesToSelect, Func<Matrix<T>, Vector<T>, double> scorer, int[] columnIndices)

Properties

Property Summary
bool SupportsInverseTransform { get; }
double[] FeatureImportances { 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)