QuantileRatioSelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Quantile

Quantile Ratio based Feature Selection.

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

Implements: IFixedCountFeatureSelector<T>

For Beginners

Quantile ratio compares values at different percentiles. For example, the 90th/10th percentile ratio shows how much larger the high values are compared to low values. This can detect features with skewed distributions or outliers.

How It Works

Selects features based on the ratio between two quantiles, measuring the relative spread at different parts of the distribution.

Constructors

Constructor Summary
QuantileRatioSelector(int nFeaturesToSelect, double upperQuantile, double lowerQuantile, int[] columnIndices)

Properties

Property Summary
Vector<T> RatioValues { get; }
bool SupportsInverseTransform { get; }
double LowerQuantile { get; }
double UpperQuantile { get; }
int NFeaturesToSelect { get; }
int[] SelectedIndices { get; }

Methods

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