EmbeddedWrapper<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Hybrid

Embedded-Wrapper hybrid feature selection.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> โ†’ EmbeddedWrapper<T>

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

This is like having an expert advisor (embedded method) suggest which features might be important, then carefully verifying those suggestions (wrapper). The advisor speeds up the search while the verification ensures accuracy.

How It Works

Embedded-Wrapper combines embedded methods (like regularization) with wrapper search. It uses embedded feature importance to guide the wrapper search, making the search more efficient.

Constructors

Constructor Summary
EmbeddedWrapper(int nFeaturesToSelect, double lambda, int maxWrapperIterations, int[] columnIndices)

Properties

Property Summary
Vector<T> EmbeddedImportance { get; }
Vector<T> WrapperScores { get; }
bool SupportsInverseTransform { get; }
double Lambda { get; }
int MaxWrapperIterations { 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)