MADThreshold<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter
Median Absolute Deviation (MAD) threshold feature selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → MADThreshold<T>
Implements: IFeatureSelector<T>
For Beginners
Instead of using variance (which can be skewed by extreme values), MAD looks at how far typical values are from the middle value. This makes it better for data with outliers. Features with low MAD are nearly constant and don’t provide useful information.
How It Works
Removes features whose Median Absolute Deviation is below a threshold. MAD is a robust measure of spread that is less sensitive to outliers than variance.
Constructors
| Constructor | Summary |
|---|---|
MADThreshold(double threshold, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> MADValues { get; } |
|
bool SupportsInverseTransform { get; } |
|
double Threshold { get; } |
|
int[] SelectedIndices { get; } |
Methods
| Method | Summary |
|---|---|
bool[] GetSupportMask() |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |