ICAP<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter.Multivariate

Interaction Capping (ICAP) feature selection.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → ICAP<T>

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

ICAP looks for features that work well together. Sometimes two features alone aren’t very informative, but combined they reveal patterns. ICAP tries to find these synergistic combinations while still avoiding redundancy. It’s like finding teammates who complement each other’s abilities.

How It Works

ICAP extends mRMR by adding interaction information. It considers not just pairwise redundancy but also synergistic effects where combining features provides more information than the sum of individual contributions.

Constructors

Constructor Summary
ICAP(int nFeaturesToSelect, int nBins, int[] columnIndices)

Properties

Property Summary
Vector<T> ICAPScores { get; }
bool SupportsInverseTransform { 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)