SelectPercentile<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Filter.Univariate

Selects features according to a percentile of the highest scores.

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

Implements: ISupervisedFeatureSelector<T>

For Beginners

Instead of specifying an exact number of features:

  • SelectPercentile(50) keeps the top 50% of features
  • The actual number depends on your original feature count
  • Useful when you want a proportion, not an absolute count

How It Works

SelectPercentile selects the top percentile of features based on a scoring function. For example, selecting the top 10% of features ranked by F-score.

This is similar to SelectKBest but uses a relative threshold (percentile) instead of an absolute number of features.

Constructors

Constructor Summary
SelectPercentile(double percentile, SelectKBestScoreFunc scoringFunction, int[] columnIndices) Creates a new instance of SelectPercentile.

Properties

Property Summary
SelectKBestScoreFunc ScoringFunction { get; } Gets the scoring function used.
Vector<T> PValues { get; } Gets the p-values for each feature, or null when the scoring function has none (mutual information).
Vector<T> Scores { get; } Gets the scores for each feature.
bool SupportsInverseTransform { get; } Gets whether this transformer supports inverse transformation.
double Percentile { get; } Gets the percentile of features to select (0-100).
int[] SelectedIndices { get; } Gets the indices of selected features.

Methods

Method Summary
Matrix<T> FitTransform(Matrix<T> data, Vector<T> target) Fits and transforms the data.
bool[] GetSupportMask() Gets the support mask indicating which features are selected.
string[] GetFeatureNamesOut(string[] inputFeatureNames) Gets the output feature names after transformation.
void Fit(Matrix<T> data, Vector<T> target) Fits the selector by computing feature scores.