BootstrapSelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.CrossValidation

Bootstrap based Feature Selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Bootstrap sampling creates many random subsets of data (with replacement). Features that are consistently important across all these subsets are more reliable than those that only work on specific subsets.

How It Works

Selects features based on their importance stability across multiple bootstrap samples, ensuring robust feature selection.

Constructors

Constructor Summary
BootstrapSelector(int nFeaturesToSelect, int nBootstraps, double sampleRatio, int? randomState, int[] columnIndices)

Properties

Property Summary
Vector<T> StabilityScores { get; }
bool SupportsInverseTransform { get; }
double SampleRatio { get; }
int NBootstraps { 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)