org.openimaj.ml.gmm

Enum GaussianMixtureModelEM.CovarianceType

java.lang.Object

java.lang.Enum<GaussianMixtureModelEM.CovarianceType>

org.openimaj.ml.gmm.GaussianMixtureModelEM.CovarianceType

All Implemented Interfaces:

Serializable, Comparable<GaussianMixtureModelEM.CovarianceType>

Enclosing class:

GaussianMixtureModelEM

public static enum GaussianMixtureModelEM.CovarianceType
extends Enum<GaussianMixtureModelEM.CovarianceType>

Different forms of covariance matrix supported by the GaussianMixtureModelEM.

Author:

Jonathon Hare (jsh2@ecs.soton.ac.uk)

Enum Constant Summary

Enum Constants

Enum Constant and Description
Diagonal Gaussians with diagonal covariance matrices.
Full Gaussians with full covariance
Spherical Spherical Gaussians: variance is the same along all axes and zero across-axes.
Tied Gaussians with a tied covariance matrix; the same covariance matrix is shared by all the gaussians.

Method Summary

Modifier and Type	Method and Description
protected abstract MultivariateGaussian[]	createGaussians(int ngauss, int ndims)
protected abstract void	mstep(GaussianMixtureModelEM.EMGMM gmm, GaussianMixtureModelEM learner, Jama.Matrix X, Jama.Matrix responsibilities, Jama.Matrix weightedXsum, double[] inverseWeights) Mode specific maximisation-step.
protected abstract void	setCovariances(MultivariateGaussian[] gaussians, Jama.Matrix cv)
static GaussianMixtureModelEM.CovarianceType	valueOf(String name) Returns the enum constant of this type with the specified name.
static GaussianMixtureModelEM.CovarianceType[]	values() Returns an array containing the constants of this enum type, in the order they are declared.

Methods inherited from class java.lang.Enum

clone, compareTo, equals, finalize, getDeclaringClass, hashCode, name, ordinal, toString, valueOf

Methods inherited from class java.lang.Object

getClass, notify, notifyAll, wait, wait, wait

Enum Constant Detail

Spherical

public static final GaussianMixtureModelEM.CovarianceType Spherical

Spherical Gaussians: variance is the same along all axes and zero across-axes.

Diagonal

public static final GaussianMixtureModelEM.CovarianceType Diagonal

Gaussians with diagonal covariance matrices.

Full

public static final GaussianMixtureModelEM.CovarianceType Full

Gaussians with full covariance

Tied

public static final GaussianMixtureModelEM.CovarianceType Tied

Gaussians with a tied covariance matrix; the same covariance matrix is shared by all the gaussians.

Method Detail

values

public static GaussianMixtureModelEM.CovarianceType[] values()

Returns an array containing the constants of this enum type, in the order they are declared. This method may be used to iterate over the constants as follows:

for (GaussianMixtureModelEM.CovarianceType c : GaussianMixtureModelEM.CovarianceType.values())
    System.out.println(c);

Returns:

an array containing the constants of this enum type, in the order they are declared

valueOf

public static GaussianMixtureModelEM.CovarianceType valueOf(String name)

Returns the enum constant of this type with the specified name. The string must match exactly an identifier used to declare an enum constant in this type. (Extraneous whitespace characters are not permitted.)

Parameters:

name - the name of the enum constant to be returned.

Returns:

the enum constant with the specified name

Throws:

IllegalArgumentException - if this enum type has no constant with the specified name

NullPointerException - if the argument is null

createGaussians

protected abstract MultivariateGaussian[] createGaussians(int ngauss,
                                                          int ndims)

setCovariances

protected abstract void setCovariances(MultivariateGaussian[] gaussians,
                                       Jama.Matrix cv)

mstep

protected abstract void mstep(GaussianMixtureModelEM.EMGMM gmm,
                              GaussianMixtureModelEM learner,
                              Jama.Matrix X,
                              Jama.Matrix responsibilities,
                              Jama.Matrix weightedXsum,
                              double[] inverseWeights)

Mode specific maximisation-step. Implementors should use the state to update the covariance of each of the GaussianMixtureModelEM#gaussians.

Parameters:

gmm - the mixture model being learned

X - the data

responsibilities - matrix with the same number of rows as X where each col is the amount that the data point belongs to each gaussian

weightedXsum - responsibilities.T * X

inverseWeights - 1/weights