Class ClusterCount

Methods to determine the number of clusters (discrete states of a latent variable).

Inheritance

System.Object

ClusterCount

Inherited Members

System.Object.Equals(System.Object)

System.Object.Equals(System.Object, System.Object)

System.Object.GetHashCode()

System.Object.GetType()

System.Object.MemberwiseClone()

System.Object.ReferenceEquals(System.Object, System.Object)

System.Object.ToString()

Namespace: BayesServer.Analysis

Assembly: BayesServer.Analysis.dll

Syntax

public static class ClusterCount

Methods

Detect(Network, Variable, IList<Int32>, Func<Network, DataPartitioning, IEvidenceReaderCommand>, Action<Network, IEvidenceReaderCommand>, ClusterCountOptions)

Determine the number of clusters (discrete states of a latent variable) using cross validation.

Declaration

public static ClusterCountOutput Detect(Network network, Variable cluster, IList<int> clusterCounts, Func<Network, DataPartitioning, IEvidenceReaderCommand> createEvidenceReaderCommand, Action<Network, IEvidenceReaderCommand> learn, ClusterCountOptions options)

Parameters

Type	Name	Description
Network	network	The Bayesian network.
Variable	cluster	The cluster (discrete latent variable) to test.
System.Collections.Generic.IList<System.Int32>	clusterCounts	Different cluster counts to test. Include 1 to test for no cluster variable.
System.Func<Network, DataPartitioning, IEvidenceReaderCommand>	createEvidenceReaderCommand	Requires a method that can create an evidence reader command for a copy of network
System.Action<Network, IEvidenceReaderCommand>	learn	Requires a method that can learn the parameters of a copy of network.
ClusterCountOptions	options	Options affecting how the tests are performed.

Returns

Type	Description
ClusterCountOutput	The number of clusters, and supplementary information.