JMI<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Filter.InformationTheory

Joint Mutual Information (JMI) for feature selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

JMI looks at how much a feature tells you about the target when combined with previously selected features. A feature might be weak alone but very powerful in combination with others. JMI finds these synergistic feature combinations.

How It Works

JMI selects features that maximize joint mutual information with the target, considering both the individual feature-target relationship and how features work together.

Constructors

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

Properties

Property Summary
Vector<T> JMIScores { get; }
bool SupportsInverseTransform { get; }
int NBins { 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)