BootstrapSelection<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Stability

Bootstrap-based Feature Selection for robust feature importance estimation.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Bootstrap is like asking the same question many times to slightly different versions of your data (by random sampling with replacement). This helps you understand which features are reliably important and which ones might just be important by chance in one particular sample.

How It Works

Uses bootstrap resampling (sampling with replacement) to estimate the variability of feature importance scores. Features with consistently high importance across bootstrap samples are selected.

Constructors

Constructor Summary
BootstrapSelection(int nFeaturesToSelect, int nBootstraps, double confidenceLevel, int? randomState, int[] columnIndices)

Properties

Property Summary
Vector<T> MeanImportances { get; }
Vector<T> StdImportances { get; }
bool SupportsInverseTransform { get; }
double ConfidenceLevel { 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)