Package org.ddogleg.clustering.gmm

Class SeedFromKMeans_F64

java.lang.Object

org.ddogleg.clustering.gmm.SeedFromKMeans_F64

All Implemented Interfaces:

InitializeGmm_F64

public class SeedFromKMeans_F64
extends Object
implements InitializeGmm_F64

Initializes the mixture models by applying K-Means first. The mean will be the center of the clusters, variance computed from its members, and weight based on the total number of points assigned.

Constructor Summary

Constructors

Constructor	Description
SeedFromKMeans_F64(StandardKMeans<double[]> kmeans)

Method Summary

Modifier and Type	Method	Description
void	init(int pointDimension, long randomSeed)	Initializes internal data structures.
void	selectSeeds(LArrayAccessor<double[]> points, List<GaussianGmm_F64> seeds)
void	setVerbose(boolean verbose)	Turn on verbose output to standard out

Methods inherited from class java.lang.Object

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

Constructor Details

SeedFromKMeans_F64

public SeedFromKMeans_F64(StandardKMeans<double[]> kmeans)

Method Details

init

public void init(int pointDimension,
 long randomSeed)

Description copied from interface: InitializeGmm_F64

Initializes internal data structures. Must be called first.

Specified by:

init in interface InitializeGmm_F64

Parameters:

pointDimension - Number of degrees of freedom in each point.

randomSeed - Seed for any random number generators used internally.

selectSeeds

public void selectSeeds(LArrayAccessor<double[]> points,
 List<GaussianGmm_F64> seeds)

Specified by:

selectSeeds in interface InitializeGmm_F64

Parameters:

points - (input) Set of points which is to be clustered.

seeds - (output) List containing storage for the initial Gaussians.

setVerbose

public void setVerbose(boolean verbose)

Description copied from interface: InitializeGmm_F64

Turn on verbose output to standard out

Specified by:

setVerbose in interface InitializeGmm_F64