KernelPCASelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Kernel

Kernel PCA-based Feature Selection.

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

Implements: IFixedCountFeatureSelector<T>

For Beginners

Regular PCA finds linear patterns. Kernel PCA can find non-linear patterns by implicitly mapping data to a higher dimension. We select features that are most important in this richer representation, capturing complex relationships that linear methods might miss.

How It Works

Uses kernel PCA to map features to a higher-dimensional space and selects features that contribute most to the principal components in kernel space.

Constructors

Constructor Summary
KernelPCASelector(int nFeaturesToSelect, double gamma, int nComponents, int[] columnIndices)

Properties

Property Summary
Vector<T> FeatureImportances { get; }
bool SupportsInverseTransform { get; }
double Gamma { get; }
int NFeaturesToSelect { get; }
int[] SelectedIndices { get; }

Methods

Method Summary
bool[] GetSupportMask()
string[] GetFeatureNamesOut(string[] inputFeatureNames)