RobustFeatureSelector<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Robust
Robust Feature Selection resistant to outliers and noise.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → RobustFeatureSelector<T>
Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>
For Beginners
Regular feature selection can be fooled by extreme values (outliers) in the data. This method uses statistics that are “robust”
- they don’t change much when a few data points are extreme. This gives
more reliable feature selection for messy real-world data.
How It Works
Uses robust statistics (median, MAD, rank-based measures) to identify important features while being resistant to outliers and noise in the data.
Constructors
| Constructor | Summary |
|---|---|
RobustFeatureSelector(int nFeaturesToSelect, double outlierThreshold, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> RobustScores { 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) |