NormalizedMutualInformation<T>

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

Normalized Mutual Information for scale-invariant dependency measurement.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Regular mutual information can be higher for features with more distinct values. NMI normalizes this so you can fairly compare features with different numbers of unique values.

How It Works

Normalized Mutual Information (NMI) divides mutual information by entropy to get a value between 0 and 1, making comparisons across features more fair.

Constructors

Constructor Summary
NormalizedMutualInformation(int nFeaturesToSelect, int nBins, NormalizationMethod<T> normalization, int[] columnIndices)

Properties

Property Summary
Vector<T> NMIScores { 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)