TrimmedMeanSelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Robust

Trimmed Mean-based Feature Selection.

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

Implements: IFixedCountFeatureSelector<T>

For Beginners

When computing feature statistics, extreme values (outliers) can skew results. Trimmed mean cuts off the highest and lowest values before computing, giving you a more reliable picture of typical feature behavior without outlier influence.

How It Works

Uses trimmed statistics (removing extreme values) to compute robust feature importance scores that are less affected by outliers.

Constructors

Constructor Summary
TrimmedMeanSelector(int nFeaturesToSelect, double trimProportion, int[] columnIndices)

Properties

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

Methods

Method Summary
bool[] GetSupportMask()
string[] GetFeatureNamesOut(string[] inputFeatureNames)