KendallCorrelation<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter.Correlation
Kendall Tau Correlation-based feature selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → KendallCorrelation<T>
Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>
For Beginners
Kendall’s Tau looks at pairs of data points and asks: “when one goes up, does the other also go up?” It counts how many pairs agree versus disagree. It’s particularly good when you have a small dataset or many tied values, and gives a more intuitive probability interpretation.
How It Works
Kendall Tau measures the ordinal association between features and target by counting concordant and discordant pairs. It’s more robust than Spearman for small samples and handles ties well.
Constructors
| Constructor | Summary |
|---|---|
KendallCorrelation(int nFeaturesToSelect, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> TauScores { 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) |