ClusterMetrics<T>

Models & Types · AiDotNet.Clustering.Evaluation

Convenience class for computing multiple cluster evaluation metrics at once.

For Beginners

Use this class to evaluate your clustering results.

Example usage:

If you have ground truth labels:

How It Works

This class provides a simple interface to compute multiple clustering metrics with a single call. It returns a ClusteringScores object containing all computed values.

Constructors

Constructor Summary
ClusterMetrics(IDistanceMetric<T> distanceMetric) Initializes a new ClusterMetrics instance.

Methods

Method Summary
ClusteringScores Evaluate(Matrix<T> data, Vector<T> labels) Evaluates clustering using internal metrics only.
ClusteringScores Evaluate(Matrix<T> data, Vector<T> predictedLabels, Vector<T> trueLabels) Evaluates clustering using both internal and external metrics.
ClusteringScores EvaluateExternal(Vector<T> predictedLabels, Vector<T> trueLabels) Computes only external metrics (when data is not available).