Back to Aurora Vision Library website

You are here: Start » Function Reference » Data Classification » Clustering » ClusterData_KMeans

ClusterData_KMeans


Header: AVL.h
Namespace: avl
Module: FoundationPro

Clusters data using KMeans algorithm.

Syntax

C++
C#
 
void avl::ClusterData_KMeans
(
	const atl::Array<atl::Array<float> >& inData,
	const int inClusters,
	const int inMaxIterations,
	const int inSeed,
	const float inTerminationFactor,
	const avl::KMeansClusteringMethod::Type inClusteringMethod,
	avl::Matrix& outCentroids,
	atl::Array<int>& outPointToClusterAssignment,
	float& outDistanceSum
)

Parameters

Name Type Range Default Description
Input value inData const Array<Array<float> >& Data set, array of examples
Input value inClusters const int 2 - + 2 Number of clusters to extract
Input value inMaxIterations const int 10 - 1000 200 Maximal number of procedure iterations
Input value inSeed const int 0 - 5489 Seed to init random engine
Input value inTerminationFactor const float 1.0 - 2.0 1.5f Additional factor of procedure stop
Input value inClusteringMethod const KMeansClusteringMethod::Type KMeansPlusPlus KMeans variant to use
Output value outCentroids Matrix& Resulting centroid points in feature space
Output value outPointToClusterAssignment Array<int>& Array of input point assignments to generated clusters
Output value outDistanceSum float& Sum of squared distances from points to its respective cluster centroids

Errors

List of possible exceptions:

Error type Description
DomainError Cannot make more clusters than there is data in input dataset in ClusterData_KMeans.
DomainError Empty dataset on input in ClusterData_KMeans.
DomainError Inconsistent number of data coordinates in input dataset in ClusterData_KMeans.