LowCardinalitySelector<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Unsupervised

Low Cardinality Feature Removal.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → LowCardinalitySelector<T>

Implements: IFeatureSelector<T>

For Beginners

Cardinality is the number of unique values a feature has. A feature with only 1 or 2 unique values doesn’t vary much and may not be useful for distinguishing between samples. This selector removes such low-variety features.

How It Works

Removes features with very low cardinality (few unique values) as they may not provide enough discriminative power.

Constructors

Constructor Summary
LowCardinalitySelector(int minCardinality, int[] columnIndices)

Properties

Property Summary
bool SupportsInverseTransform { get; }
int MinCardinality { get; }
int[] Cardinalities { get; }
int[] SelectedIndices { get; }

Methods

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