GainRatioSelection<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter

Gain Ratio Feature Selection (normalized Information Gain).

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Information gain tends to favor features with many unique values (like IDs), even if they’re not truly useful. Gain ratio fixes this by dividing the information gain by how “spread out” the feature values are. This gives fairer scores across features with different numbers of unique values.

How It Works

Gain Ratio normalizes information gain by the intrinsic information of the feature (split information). This corrects for the bias of information gain towards features with many values.

Constructors

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

Properties

Property Summary
Vector<T> GainRatioScores { get; }
Vector<T> InformationGain { get; }
Vector<T> SplitInfo { 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)