KMeansClusteringCluster(Int32, KMeansClusteringStart) Method

Clusters the data into the specified number of clusters, using the specified method of choosing the initial cluster centers.

Namespace: CenterSpace.NMath.Core
Assembly: NMath (in NMath.dll) Version: 7.4

Syntax

C++

Copy

public ClusterSet Cluster(
	int k,
	KMeansClusteringStart start
)

Public Function Cluster ( 
	k As Integer,
	start As KMeansClusteringStart
) As ClusterSet

public:
ClusterSet^ Cluster(
	int k, 
	KMeansClusteringStart start
)

member Cluster : 
        k : int * 
        start : KMeansClusteringStart -> ClusterSet

Parameters

k Int32: The number of clusters to form.
start KMeansClusteringStart: The method to use for choosing the initial cluster centers.

Return Value

ClusterSet
A ClusterSet that identifies the cluster into which each object was grouped. Cluster numbers are based on row indices in this.Data and this.Centers.

Exceptions

Exception	Condition
InvalidArgumentException	Thrown if k is less than one or greater than the number of objects (this.N).
NMathException	Thrown if an empty cluster is encountered. Re-cluster with different starting centers.

Reference

KMeansClustering Class

Cluster Overload

CenterSpace.NMath.Core Namespace