lti::classifier Class Reference

Abstract parent class for all classifiers. More...

#include <ltiClassifier.h>

Inheritance diagram for lti::classifier:

[legend]

Collaboration diagram for lti::classifier:

[legend]

List of all members.

Classes
class	invalidMethodException
	Exception thrown when a method of a functor is not implemented for a specific parameter set. More...
class	invalidOutputException
	Exception thrown when the output objects has not been set yet. More...
class	invalidParametersException
	Exception thrown when the parameters are not set. More...
class	outputTemplate
	The outputTemplate stores the relation between the different positions (sometimes called internal ids) of a classification result and the ids. More...
class	outputVector
	An output vector is the result of classifying data with any classifier. More...
class	parameters
	the parameters for the class classifier More...
class	parametersOutOfRangeException
	Exception thrown when a parameter is out of range. More...
Public Member Functions
	classifier ()
	classifier (const classifier &other)
virtual	~classifier ()
virtual const char *	getTypeName () const
classifier &	copy (const classifier &other)
classifier &	operator= (const classifier &other)
virtual classifier *	clone () const =0
virtual bool	validParameters () const
const parameters &	getParameters () const
virtual bool	setParameters (const parameters &theParam)
virtual void	setOutputTemplate (const outputTemplate &theOutputTemplate)
const outputTemplate &	getOutputTemplate () const
virtual bool	write (ioHandler &handler, const bool complete=true) const
virtual bool	read (ioHandler &handler, const bool complete=true)
Progress Info
Methods used to plug-in and retrieve progress visualization objects.
void	setProgressObject (const progressInfo &progBox)
void	removeProgressObject ()
bool	validProgressObject () const
progressInfo &	getProgressObject ()
const progressInfo &	getProgressObject () const
Static Public Attributes
static const int	unknownObj
static const int	noObject
static const double	nullProbability
Protected Attributes
parameters *	params
progressInfo *	progressBox
outputTemplate	outTemplate

Detailed Description

Abstract parent class for all classifiers.

The classifier class encloses a number of other important classes:

lti::classifier::parameters: similar to the functor::parameters, this is used to specify the characteristics of a classifier like learning rate, etc.
lti::classifier::outputVector: This type is used to return the classification results.
lti::classifier::outputTemplate: Contains information on how the outputVector is to be constructed from a resulting vector of the classification.

Classifiers can be diveded into two large groups: supervised and unsupervised. These two groups have the base classes supervisedClassifier and unsupervisedClassifier which are subclasses of this class. Examples for the first type are RBF networks, multi layer perceptron, maximum likelihood classifiers etc. The second type comprises statistic and neural clustering methods.

All classifiers return an outputVector when classify is called. The outputVector is created by the classifier by applying its outputTemplate to the regular dvector which results from the classification process.

Constructor & Destructor Documentation

lti::classifier::classifier ( )

default constructor

lti::classifier::classifier ( const classifier & other )

copy constructor

Parameters:

other

the object to be copied

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

destructor

Member Function Documentation

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

returns a pointer to a clone of this functor.

Implemented in lti::adaptiveKMeans, lti::bagging, lti::competitiveAgglomeration, lti::DBScan< T >, lti::fuzzyCMeans, lti::hmmClassifier, lti::hmmOnlineClassifier, lti::kMeansClustering, lti::kNNClassifier, lti::lvq, lti::manualCrispDecisionTree, lti::MLP, lti::MSTClustering, lti::rbf, lti::shClassifier, lti::SOFM, lti::SOFM2D, and lti::svm.

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

return the last message set with setStatusString().

This will never return 0. If no status-string has been set yet an empty string (pointer to a string with only the char(0)) will be returned. set a status string.

Parameters:

msg

the const string to be reported next time by getStatusString(). The given string will be copied This message will be usually set within the apply methods to indicate an error cause.

Note that the change of the status string is not considered as a change in the functor status. copy data of "other" functor.

Parameters:

other

the functor to be copied

Returns:: a reference to this functor object

Reimplemented from lti::ioObject.

Reimplemented in lti::adaptiveKMeans, lti::bagging, lti::centroidClustering, lti::clustering, lti::competitiveAgglomeration, lti::crispDecisionTree, lti::DBScan< T >, lti::decisionTree, lti::fuzzyCMeans, lti::hmmClassifier, lti::hmmOnlineClassifier, lti::kMeansClustering, lti::kNNClassifier, lti::lvq, lti::manualCrispDecisionTree, lti::MLP, lti::MSTClustering, lti::rbf, lti::shClassifier, lti::SOFM, lti::SOFM2D, lti::supervisedInstanceClassifier, lti::supervisedSequenceClassifier, lti::svm, and lti::unsupervisedClassifier.

Referenced by operator=().

const outputTemplate& lti::classifier::getOutputTemplate ( ) const

get a const reference to the outputTemplate

Referenced by lti::kNNClassifier::getColumnId().

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

const progressInfo& lti::classifier::getProgressObject ( ) const

get progress object

progressInfo& lti::classifier::getProgressObject ( )

get progress object

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

returns the name of this type ("classifier")

Reimplemented from lti::ioObject.

Reimplemented in lti::adaptiveKMeans, lti::bagging, lti::centroidClustering, lti::clustering, lti::competitiveAgglomeration, lti::crispDecisionTree, lti::DBScan< T >, lti::decisionTree, lti::fuzzyCMeans, lti::hmmClassifier, lti::hmmOnlineClassifier, lti::kMeansClustering, lti::kNNClassifier, lti::manualCrispDecisionTree, lti::MLP, lti::MSTClustering, lti::rbf, lti::shClassifier, lti::SOFM, lti::SOFM2D, lti::supervisedInstanceClassifier, lti::supervisedSequenceClassifier, lti::svm, and lti::unsupervisedClassifier.

classifier& lti::classifier::operator= ( const classifier & other ) [inline]

Alias for "copy".

Reimplemented from lti::ioObject.

Reimplemented in lti::adaptiveKMeans, lti::bagging, lti::competitiveAgglomeration, lti::crispDecisionTree, lti::DBScan< T >, lti::decisionTree, lti::fuzzyCMeans, lti::hmmClassifier, lti::hmmOnlineClassifier, lti::kNNClassifier, lti::lvq, lti::manualCrispDecisionTree, lti::MLP, lti::MSTClustering, lti::rbf, lti::shClassifier, lti::SOFM, lti::SOFM2D, lti::supervisedInstanceClassifier, lti::supervisedSequenceClassifier, and lti::svm.

References copy().

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

read the classifier 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::ioObject.

Reimplemented in lti::bagging, lti::centroidClustering, lti::crispDecisionTree, lti::decisionTree, lti::hmmClassifier, lti::hmmOnlineClassifier, lti::kNNClassifier, lti::MLP, lti::MSTClustering, lti::rbf, lti::shClassifier, lti::SOFM, lti::SOFM2D, and lti::svm.

void lti::classifier::removeProgressObject ( )

remove the active progress object

virtual void lti::classifier::setOutputTemplate ( const outputTemplate & theOutputTemplate ) [virtual]

sets the classifier's outputTemplate This member makes a copy of theOutputTemplate: the classifier will keep its own copy of the outputTemplate

virtual bool lti::classifier::setParameters ( const parameters & theParam ) [virtual]

sets the classifier's parameters.

This member makes a copy of theParam: the classifier will keep its own copy of the parameters!

Reimplemented in lti::fuzzyCMeans.

void lti::classifier::setProgressObject ( const progressInfo & progBox )

set a progress object

A clone of the given object will be generated.

virtual bool lti::classifier::validParameters ( ) const [inline, virtual]

returns true if the parameters are valid

References params.

bool lti::classifier::validProgressObject ( ) const

return true if a valid progressInfo object has already been setted

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

write the classifier 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.