GaussianitySelector<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Distribution
Gaussianity (Normality) based Feature Selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → GaussianitySelector<T>
Implements: IFixedCountFeatureSelector<T>
For Beginners
Many statistical methods assume data follows a bell curve (normal distribution). This selector measures how “normal” each feature is. You can select features that are most normal (for parametric methods) or least normal (to find interesting non-standard patterns).
How It Works
Selects features based on how closely their distributions match a Gaussian (normal) distribution, using the Jarque-Bera test statistic.
Constructors
| Constructor | Summary |
|---|---|
GaussianitySelector(int nFeaturesToSelect, bool preferGaussian, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> JBStatistics { get; } |
|
bool PreferGaussian { get; } |
|
bool SupportsInverseTransform { get; } |
|
int NFeaturesToSelect { get; } |
|
int[] SelectedIndices { get; } |
Methods
| Method | Summary |
|---|---|
Matrix<T> FitTransform(Matrix<T> data) |
|
bool[] GetSupportMask() |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |