RFECV<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Wrapper

Recursive Feature Elimination with Cross-Validation (RFECV).

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

Implements: ISupervisedFeatureSelector<T>

For Beginners

Regular RFE requires you to specify how many features to keep. RFECV figures out the best number automatically by testing different counts and seeing which gives the best validation score.

How It Works

RFECV uses cross-validation to find the optimal number of features. It performs RFE for different feature counts and selects the count that maximizes cross-validated performance.

Constructors

Constructor Summary
RFECV(Func<Matrix<T>, Vector<T>, T> scoringFunc, Func<Matrix<T>, Vector<T>, Vector<T>> importanceFunc, int minFeatures, int step, int nFolds, int? randomState, int[] columnIndices)

Properties

Property Summary
Vector<T> CVScores { get; }
bool SupportsInverseTransform { get; }
int MinFeatures { get; }
int OptimalFeatureCount { 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)