lti::gaussianMixtureModel< T >::parameters Class Reference

The parameters for the class gaussianMixtureModel. More...

#include <ltiGaussianMixtureModel.h>

Inheritance diagram for lti::gaussianMixtureModel< T >::parameters:

[legend]

Collaboration diagram for lti::gaussianMixtureModel< T >::parameters:

[legend]

List of all members.

Public Member Functions
	parameters ()
	parameters (const parameters &other)
	~parameters ()
const char *	getTypeName () const
parameters &	copy (const parameters &other)
virtual bool	write (ioHandler &handler, const bool complete=true) const
virtual bool	read (ioHandler &handler, const bool complete=true)
virtual functor::parameters *	clone () const
Public Attributes
int	numberOfComponents
int	iterations
bool	forceIterations
int	partialIterations
bool	useSM
T	lambda
T	emergencyLambda
int	cMax
T	epsilon
bool	reportProgress

---

Detailed Description

template<class T>
class lti::gaussianMixtureModel< T >::parameters

The parameters for the class gaussianMixtureModel.

---

Constructor & Destructor Documentation

template<class T >

lti::gaussianMixtureModel< T >::parameters::parameters

(

void

)

[inline]

Default constructor.

Reimplemented from lti::statisticsFunctor::parameters.

References lti::gaussianMixtureModel< T >::parameters::cMax, lti::gaussianMixtureModel< T >::parameters::emergencyLambda, lti::gaussianMixtureModel< T >::parameters::epsilon, lti::gaussianMixtureModel< T >::parameters::forceIterations, lti::gaussianMixtureModel< T >::parameters::iterations, lti::gaussianMixtureModel< T >::parameters::lambda, lti::gaussianMixtureModel< T >::parameters::numberOfComponents, lti::gaussianMixtureModel< T >::parameters::partialIterations, lti::gaussianMixtureModel< T >::parameters::reportProgress, and lti::gaussianMixtureModel< T >::parameters::useSM.

Referenced by lti::gaussianMixtureModel< T >::parameters::clone().

template<class T >

lti::gaussianMixtureModel< T >::parameters::parameters

(

const parameters &

other

)

[inline]

copy constructor

Parameters:

other

the parameters object to be copied

Reimplemented from lti::statisticsFunctor::parameters.

References lti::gaussianMixtureModel< T >::parameters::copy().

template<class T >

lti::gaussianMixtureModel< T >::parameters::~parameters

(

)

[inline, virtual]

destructor

Reimplemented from lti::statisticsFunctor::parameters.

---

Member Function Documentation

template<class T >

virtual functor::parameters* lti::gaussianMixtureModel< T >::parameters::clone

(

)

const [inline, virtual]

returns a pointer to a clone of the parameters

Reimplemented from lti::statisticsFunctor::parameters.

References lti::gaussianMixtureModel< T >::parameters::parameters().

template<class T >

parameters& lti::gaussianMixtureModel< T >::parameters::copy

(

const parameters &

other

)

[inline]

copy the contents of a parameters object

Parameters:

other

the parameters object to be copied

Returns:

a reference to this parameters object

Reimplemented from lti::statisticsFunctor::parameters.

References lti::gaussianMixtureModel< T >::parameters::cMax, lti::statisticsFunctor::parameters::copy(), lti::gaussianMixtureModel< T >::parameters::emergencyLambda, lti::gaussianMixtureModel< T >::parameters::epsilon, lti::gaussianMixtureModel< T >::parameters::forceIterations, lti::gaussianMixtureModel< T >::parameters::iterations, lti::gaussianMixtureModel< T >::parameters::lambda, lti::gaussianMixtureModel< T >::parameters::numberOfComponents, lti::gaussianMixtureModel< T >::parameters::partialIterations, lti::gaussianMixtureModel< T >::parameters::reportProgress, and lti::gaussianMixtureModel< T >::parameters::useSM.

Referenced by lti::gaussianMixtureModel< T >::parameters::parameters().

template<class T >

const char* lti::gaussianMixtureModel< T >::parameters::getTypeName

(

void

)

const [inline, virtual]

returns name of this type

Reimplemented from lti::statisticsFunctor::parameters.

template<class T >

virtual bool lti::gaussianMixtureModel< T >::parameters::read	(	ioHandler &	handler,
		const bool	complete = true
	)			[inline, virtual]

read the parameters from the given ioHandler

Parameters:

	handler	the ioHandler to be used
	complete	if true (the default) the enclosing begin/end will be also read, otherwise only the data block will be read.

Returns:

true if write was successful

Reimplemented from lti::functor::parameters.

References lti::gaussianMixtureModel< T >::parameters::cMax, lti::gaussianMixtureModel< T >::parameters::emergencyLambda, lti::gaussianMixtureModel< T >::parameters::epsilon, lti::gaussianMixtureModel< T >::parameters::iterations, lti::gaussianMixtureModel< T >::parameters::lambda, lti::gaussianMixtureModel< T >::parameters::numberOfComponents, lti::gaussianMixtureModel< T >::parameters::partialIterations, lti::ioHandler::readBegin(), lti::ioHandler::readEnd(), lti::gaussianMixtureModel< T >::parameters::reportProgress, and lti::gaussianMixtureModel< T >::parameters::useSM.

template<class T >

virtual bool lti::gaussianMixtureModel< T >::parameters::write	(	ioHandler &	handler,
		const bool	complete = true
	)			const [inline, virtual]

write the parameters in the given ioHandler

Parameters:

	handler	the ioHandler to be used
	complete	if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written.

Returns:

true if write was successful

Reimplemented from lti::functor::parameters.

References lti::gaussianMixtureModel< T >::parameters::cMax, lti::gaussianMixtureModel< T >::parameters::emergencyLambda, lti::gaussianMixtureModel< T >::parameters::epsilon, lti::gaussianMixtureModel< T >::parameters::iterations, lti::gaussianMixtureModel< T >::parameters::lambda, lti::gaussianMixtureModel< T >::parameters::numberOfComponents, lti::gaussianMixtureModel< T >::parameters::partialIterations, lti::gaussianMixtureModel< T >::parameters::reportProgress, lti::gaussianMixtureModel< T >::parameters::useSM, lti::ioHandler::writeBegin(), and lti::ioHandler::writeEnd().

---

Member Data Documentation

template<class T >

int lti::gaussianMixtureModel< T >::parameters::cMax

This is the maximum number of split-and-merge trials that are done in each SM step.

Referenced by lti::gaussianMixtureModel< T >::parameters::copy(), lti::gaussianMixtureModel< T >::parameters::parameters(), lti::gaussianMixtureModel< T >::parameters::read(), and lti::gaussianMixtureModel< T >::parameters::write().

template<class T >

T lti::gaussianMixtureModel< T >::parameters::emergencyLambda

This is a factor for regularizing the covariance matrices in emergencies.

An emergency occurs when the determinant of a covariance matrix is almost zero (which often occurs due to numerical problems). In that case, the diagonal of the matrix will be emphasized by this factor, until the determinant becomes reasonable.

Referenced by lti::gaussianMixtureModel< T >::parameters::copy(), lti::gaussianMixtureModel< T >::parameters::parameters(), lti::gaussianMixtureModel< T >::parameters::read(), and lti::gaussianMixtureModel< T >::parameters::write().

template<class T >

T lti::gaussianMixtureModel< T >::parameters::epsilon

The magnitude of the random perturbation of the split algorithm.

Referenced by lti::gaussianMixtureModel< T >::parameters::copy(), lti::gaussianMixtureModel< T >::parameters::parameters(), lti::gaussianMixtureModel< T >::parameters::read(), and lti::gaussianMixtureModel< T >::parameters::write().

template<class T >

bool lti::gaussianMixtureModel< T >::parameters::forceIterations

If this is true, the EM always executes the maximum number of iterations, no matter if it converges earlier or not.

Referenced by lti::gaussianMixtureModel< T >::parameters::copy(), and lti::gaussianMixtureModel< T >::parameters::parameters().

template<class T >

int lti::gaussianMixtureModel< T >::parameters::iterations

This is the maximum number of iterations that the EM algorithm is supposed to be run.

Referenced by lti::gaussianMixtureModel< T >::parameters::copy(), lti::gaussianMixtureModel< T >::parameters::parameters(), lti::gaussianMixtureModel< T >::parameters::read(), and lti::gaussianMixtureModel< T >::parameters::write().

template<class T >

T lti::gaussianMixtureModel< T >::parameters::lambda

If You use the split-and-merge version, this parameter is used as the regularization constant.

The larger this constant, the more a covariance matrix will be drawn to be the identity matrix.

Referenced by lti::gaussianMixtureModel< T >::parameters::copy(), lti::gaussianMixtureModel< T >::parameters::parameters(), lti::gaussianMixtureModel< T >::parameters::read(), and lti::gaussianMixtureModel< T >::parameters::write().

template<class T >

int lti::gaussianMixtureModel< T >::parameters::numberOfComponents

This is the number of components of the mixture.

Referenced by lti::gaussianMixtureModel< T >::parameters::copy(), lti::gaussianMixtureModel< T >::parameters::parameters(), lti::gaussianMixtureModel< T >::parameters::read(), and lti::gaussianMixtureModel< T >::parameters::write().

template<class T >

int lti::gaussianMixtureModel< T >::parameters::partialIterations

This is the maximum number of partial iterations that the SMEM algorithm is supposed to be run.

Referenced by lti::gaussianMixtureModel< T >::parameters::copy(), lti::gaussianMixtureModel< T >::parameters::parameters(), lti::gaussianMixtureModel< T >::parameters::read(), and lti::gaussianMixtureModel< T >::parameters::write().

template<class T >

bool lti::gaussianMixtureModel< T >::parameters::reportProgress

Is true if the functor should report its result to a progress info object.

Referenced by lti::gaussianMixtureModel< T >::parameters::copy(), lti::gaussianMixtureModel< T >::parameters::parameters(), lti::gaussianMixtureModel< T >::parameters::read(), and lti::gaussianMixtureModel< T >::parameters::write().

template<class T >

bool lti::gaussianMixtureModel< T >::parameters::useSM

This flag is true, if the estimator should use the Split-and-Merge EM algorithm by Ueda, Nakano, Ghahramani, and Hinton (Neural Computation 12(9), pp.

2109-2128).

Referenced by lti::gaussianMixtureModel< T >::parameters::copy(), lti::gaussianMixtureModel< T >::parameters::parameters(), lti::gaussianMixtureModel< T >::parameters::read(), and lti::gaussianMixtureModel< T >::parameters::write().

---

The documentation for this class was generated from the following file:

• ltiGaussianMixtureModel.h