lti::SOFM Class Reference

This is the base class for all Self Organizing Feature Maps. More...

#include <ltiSOFM.h>

Inheritance diagram for lti::SOFM:

[legend]

Collaboration diagram for lti::SOFM:

[legend]

List of all members.

Classes
class	parameters
	the parameters for the class SOFM More...
Public Member Functions
	SOFM ()
	SOFM (const SOFM &other)
virtual	~SOFM ()
virtual const char *	getTypeName () const
SOFM &	copy (const SOFM &other)
SOFM &	operator= (const SOFM &other)
virtual classifier *	clone () const =0
const parameters &	getParameters () const
virtual bool	train (const dmatrix &input, ivector &ids)
virtual bool	train (const dmatrix &input)=0
virtual bool	classify (const dvector &feature, outputVector &result) const =0
virtual bool	write (ioHandler &handler, const bool complete=true) const
virtual bool	read (ioHandler &handler, const bool complete=true)
const dmatrix &	getPrototypes () const
Protected Member Functions
virtual bool	setLearnRates (const int &nbData)
Protected Attributes
dmatrix	grid
double	lrOrder
double	lrOrderDelta
double	sigma
double	sigmaDelta
double	lrConvergeA
double	lrConvergeB

Detailed Description

This is the base class for all Self Organizing Feature Maps.

They are modelled after Kohonen's book 'Self-Organizing Maps' published by Springer in 1995.

SOFMs have a two phase training: In the first phase the ordering of the neurons is established. The neighborhood should be rather large during this phase. In the second phase the neurons are more finely tuned to the underlying probability distributions with a small neighborhood. This phase is called convergence phase.

Constructor & Destructor Documentation

lti::SOFM::SOFM ( )

default constructor

lti::SOFM::SOFM ( const SOFM & other )

copy constructor

Parameters:

other

the object to be copied

virtual lti::SOFM::~SOFM ( ) [virtual]

destructor

Member Function Documentation

virtual bool lti::SOFM::classify	(	const dvector &	feature,
		outputVector &	result
	)			const `[pure virtual]`

Classification.

Classifies the feature and returns the outputVector with the classification result.

Parameters:

	feature	the vector to be classified
	result	the result of the classification

Returns:: false if an error occurred during classification else true

Implements lti::unsupervisedClassifier.

Implemented in lti::SOFM2D.

virtual classifier* lti::SOFM::clone ( ) const [pure virtual]

returns a pointer to a clone of this classifier.

Implements lti::classifier.

Implemented in lti::SOFM2D.

SOFM& lti::SOFM::copy ( const SOFM & other )

copy data of "other" classifier.

Parameters:

other

the classifier to be copied

Returns:: a reference to this classifier object

Reimplemented from lti::unsupervisedClassifier.

Reimplemented in lti::SOFM2D.

const parameters& lti::SOFM::getParameters ( ) const

returns used parameters

Reimplemented from lti::unsupervisedClassifier.

Reimplemented in lti::SOFM2D.

const dmatrix& lti::SOFM::getPrototypes ( ) const [inline]

returns the prototypes of the SOFM.

Each row of the matrix contains one prototype. Depending on the dimensionality these are ordered in dimensions. For two dimensions it is Y,X giving first the prototype at (0,0) than at (0,1) and so on. In three dimensions Z has the highest order (Z,Y,X): (0,0,0),(0,0,1),...,(0,0,n),(0,1,0),...

Referenced by lti::SOFM2DVisualizer::componentPlane(), lti::SOFM2DVisualizer::drawClasses(), lti::SOFM2DVisualizer::sammonsMapper(), and lti::SOFM2DVisualizer::uMatrix().

virtual const char* lti::SOFM::getTypeName ( ) const [virtual]

returns the name of this type ("SOFM")

Reimplemented from lti::unsupervisedClassifier.

Reimplemented in lti::SOFM2D.

SOFM& lti::SOFM::operator= ( const SOFM & other )

alias for copy member

Parameters:

other

the classifier to be copied

Returns:: a reference to this classifier object

Reimplemented from lti::classifier.

Reimplemented in lti::SOFM2D.

virtual bool lti::SOFM::read	(	ioHandler &	handler,
		const bool	complete = `true`
	)			`[virtual]`

read the SOFM from 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::classifier.

Reimplemented in lti::SOFM2D.

virtual bool lti::SOFM::setLearnRates ( const int & nbData ) [protected, virtual]

calculate learn-rates and deltas from the parameters.

virtual bool lti::SOFM::train ( const dmatrix & input ) [pure virtual]

Unsupervised training.

The row vectors in the input matrix are used for training of the classifier.

Parameters:

input

the matrix with the input vectors (each row is a training vector)

Returns:: true if successful, false otherwise. (if false you can check the error message with getStatusString())

Implements lti::unsupervisedClassifier.

Implemented in lti::SOFM2D.

virtual bool lti::SOFM::train	(	const dmatrix &	input,
		ivector &	ids
	)			`[virtual]`

Unsupervised training.

The vectors in the input matrix will be put into groups according to the training algorithm. Additionally, an integer indicating the class each point belongs to is returned.

By default this method uses the other train method train(const dmatrix&) and then calls classify(const dvector&) to get the ids for each trainvector. These ids are then returned.

Parameters:

	input	the matrix with the input vectors (each row is a training vector)
	ids	vector of class ids for each input point

Returns:: true if successful, false otherwise. (if false you can check the error message with getStatusString())

Reimplemented from lti::unsupervisedClassifier.

Reimplemented in lti::SOFM2D.

virtual bool lti::SOFM::write	(	ioHandler &	handler,
		const bool	complete = `true`
	)			const `[virtual]`

write the SOFM 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.