MultiLabelMutualInformation<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.MultiLabel

Multi-Label Mutual Information for multi-label classification feature selection.

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

Implements: IFixedCountFeatureSelector<T>

For Beginners

In multi-label classification, each sample can have multiple labels at once. This method finds features that are informative for predicting multiple labels, not just one.

How It Works

Extends mutual information to handle multiple target labels simultaneously. Aggregates MI scores across all labels to identify features relevant to multiple prediction targets.

Constructors

Constructor Summary
MultiLabelMutualInformation(int nFeaturesToSelect, int nBins, AggregationMethod<T> aggregation, int[] columnIndices)

Properties

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

Methods

Method Summary
Matrix<T> FitTransform(Matrix<T> data, Matrix<T> targets)
bool[] GetSupportMask()
string[] GetFeatureNamesOut(string[] inputFeatureNames)
void Fit(Matrix<T> data, Matrix<T> targets)