MADSelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Unsupervised

Median Absolute Deviation (MAD) based feature selection.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> โ†’ MADSelector<T>

Implements: IFixedCountFeatureSelector<T>

For Beginners

Standard variance can be thrown off by outliers. MAD uses the median instead of mean, making it more robust. Features with higher MAD have more consistent spread in their values.

How It Works

Uses MAD as a robust measure of variability. Unlike variance, MAD is resistant to outliers, making it better for selecting features in datasets with extreme values.

Constructors

Constructor Summary
MADSelector(int nFeaturesToSelect, double threshold, int[] columnIndices)

Properties

Property Summary
Vector<T> MADScores { get; }
bool SupportsInverseTransform { get; }
double Threshold { get; }
int NFeaturesToSelect { get; }
int[] SelectedIndices { get; }

Methods

Method Summary
Matrix<T> FitTransform(Matrix<T> data)
bool[] GetSupportMask()
string[] GetFeatureNamesOut(string[] inputFeatureNames)