SBFS<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Wrapper

Sequential Backward Floating Selection (SBFS) for feature selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

SBFS starts with all features and removes them one by one, but it’s smart enough to add features back if removing them was a bad idea. It’s like cleaning out a closet - you might throw something away, then realize you need it after throwing other things out.

How It Works

SBFS is a wrapper method that starts with all features and iteratively removes the least useful ones. Unlike simple backward elimination, SBFS can conditionally add features back if their removal was a mistake given later removals.

Constructors

Constructor Summary
SBFS(int nFeaturesToSelect, int maxIterations, double tolerance, int[] columnIndices)

Properties

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