MADBasedSelector<T>
Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Robust
Median Absolute Deviation (MAD) based Feature Selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> โ MADBasedSelector<T>
Implements: IFixedCountFeatureSelector<T>
For Beginners
Regular variance can be heavily influenced by outliers (extreme values). MAD uses the median instead of the mean, making it much more robust. Features with higher MAD have more genuine variability, not just variability caused by a few extreme values.
How It Works
Uses Median Absolute Deviation as a robust measure of feature spread that is less sensitive to outliers than variance-based methods.
Constructors
| Constructor | Summary |
|---|---|
MADBasedSelector(int nFeaturesToSelect, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> MADScores { get; } |
|
bool SupportsInverseTransform { get; } |
|
int NFeaturesToSelect { get; } |
|
int[] SelectedIndices { get; } |
Methods
| Method | Summary |
|---|---|
bool[] GetSupportMask() |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |