StabilitySelection<T>
Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Ensemble
Stability Selection for robust feature selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> โ StabilitySelection<T>
Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>
For Beginners
A single run of feature selection might pick some features by chance. Stability Selection runs many rounds on different parts of the data and counts how often each feature is selected. Features that are picked consistently (say, 70% of the time) are more trustworthy than those picked rarely.
How It Works
Stability Selection runs feature selection multiple times on random subsamples of the data and selects features that are consistently chosen across iterations. This reduces sensitivity to noise and produces more reliable feature sets.
Constructors
| Constructor | Summary |
|---|---|
StabilitySelection(int nFeaturesToSelect, int nIterations, double sampleFraction, double threshold, Func<Matrix<T>, Vector<T>, int, int[]> baseSelector, int? randomState, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> SelectionProbabilities { get; } |
|
bool SupportsInverseTransform { get; } |
|
double SampleFraction { get; } |
|
double Threshold { get; } |
|
int NFeaturesToSelect { get; } |
|
int NIterations { 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) |