de.lmu.ifi.dbs.elki.algorithm.clustering.em

Class TwoPassMultivariateGaussianModelFactory<V extends NumberVector>

java.lang.Object

de.lmu.ifi.dbs.elki.algorithm.clustering.em.AbstractEMModelFactory<V,EMModel>

de.lmu.ifi.dbs.elki.algorithm.clustering.em.TwoPassMultivariateGaussianModelFactory<V>

Type Parameters:

V - vector type

All Implemented Interfaces:

EMClusterModelFactory<V,EMModel>

public class TwoPassMultivariateGaussianModelFactory<V extends NumberVector>
extends AbstractEMModelFactory<V,EMModel>

Factory for EM with multivariate Gaussian models (with covariance; also known as Gaussian Mixture Modeling, GMM). These models have individual covariance matrixes, so this corresponds to the 'VVV' model in Mclust (R).

Since:

0.7.5

Author:

Erich Schubert

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	TwoPassMultivariateGaussianModelFactory.Parameterizer<V extends NumberVector> Parameterization class

Field Summary

Fields inherited from class de.lmu.ifi.dbs.elki.algorithm.clustering.em.AbstractEMModelFactory

initializer

Constructor Summary

Constructors

Constructor and Description
TwoPassMultivariateGaussianModelFactory(KMeansInitialization initializer) Constructor.

Method Summary

Modifier and Type	Method and Description
java.util.List<TwoPassMultivariateGaussianModel>	buildInitialModels(Database database, Relation<V> relation, int k, NumberVectorDistanceFunction<? super V> df) Build the initial models

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

TwoPassMultivariateGaussianModelFactory

public TwoPassMultivariateGaussianModelFactory(KMeansInitialization initializer)

Constructor.

Parameters:

initializer - Class for choosing the initial seeds.

Method Detail

buildInitialModels

public java.util.List<TwoPassMultivariateGaussianModel> buildInitialModels(Database database,
                                                                           Relation<V> relation,
                                                                           int k,
                                                                           NumberVectorDistanceFunction<? super V> df)

Description copied from interface: EMClusterModelFactory

Build the initial models

Parameters:

database - Database

relation - Relation

k - Number of clusters

df - Distance function

Returns:

Initial models