lti::MSTClustering::parameters Class Reference

the parameters for the class MSTClustering More...

#include <ltiMSTClustering.h>

Inheritance diagram for lti::MSTClustering::parameters:

[legend]

Collaboration diagram for lti::MSTClustering::parameters:

[legend]

List of all members.

Public Types
enum	eProbabilityFromDistanceTypes
Public Member Functions
	parameters ()
	parameters (const parameters &other)
	~parameters ()
const char *	getTypeName () const
parameters &	copy (const parameters &other)
parameters &	operator= (const parameters &other)
virtual classifier::parameters *	clone () const
bool	write (ioHandler &handler, const bool complete=true) const
bool	read (ioHandler &handler, const bool complete=true)
Public Attributes
float	devFac
int	nbMaxSteps
double	variance
eProbabilityFromDistanceTypes	probabilityFromDistanceMode

Detailed Description

template<class U = l2SquareDistantor<dvector>>
class lti::MSTClustering::parameters

the parameters for the class MSTClustering

Member Enumeration Documentation

template<class U = l2SquareDistantor<dvector>>

enum lti::MSTClustering::parameters::eProbabilityFromDistanceTypes

The different methods that can be used to compute a probability from a distance.

Constructor & Destructor Documentation

template<class U = l2SquareDistantor<dvector>>

lti::MSTClustering::parameters::parameters ( void ) [inline]

default constructor

Reimplemented from lti::clustering::parameters.

References lti::MSTClustering::parameters::devFac, lti::MSTClustering::parameters::nbMaxSteps, lti::MSTClustering::parameters::probabilityFromDistanceMode, and lti::MSTClustering::parameters::variance.

Referenced by lti::MSTClustering::parameters::clone().

template<class U = l2SquareDistantor<dvector>>

lti::MSTClustering::parameters::parameters ( const parameters & other ) [inline]

copy constructor

Parameters:

other

the parameters object to be copied

Reimplemented from lti::clustering::parameters.

References lti::MSTClustering::parameters::copy().

template<class U = l2SquareDistantor<dvector>>

lti::MSTClustering::parameters::~parameters ( ) [inline, virtual]

destructor

Reimplemented from lti::clustering::parameters.

Member Function Documentation

template<class U = l2SquareDistantor<dvector>>

virtual classifier::parameters* lti::MSTClustering::parameters::clone ( ) const [inline, virtual]

returns a pointer to a clone of the parameters

Reimplemented from lti::clustering::parameters.

References lti::MSTClustering::parameters::parameters().

template<class U = l2SquareDistantor<dvector>>

parameters& lti::MSTClustering::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::clustering::parameters.

References lti::clustering::parameters::copy(), lti::MSTClustering::parameters::devFac, lti::MSTClustering::parameters::nbMaxSteps, lti::MSTClustering::parameters::probabilityFromDistanceMode, and lti::MSTClustering::parameters::variance.

Referenced by lti::MSTClustering::parameters::operator=(), and lti::MSTClustering::parameters::parameters().

template<class U = l2SquareDistantor<dvector>>

const char* lti::MSTClustering::parameters::getTypeName ( void ) const [inline, virtual]

returns name of this type

Reimplemented from lti::clustering::parameters.

template<class U = l2SquareDistantor<dvector>>

parameters& lti::MSTClustering::parameters::operator= ( 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::unsupervisedClassifier::parameters.

References lti::MSTClustering::parameters::copy().

template<class U = l2SquareDistantor<dvector>>

bool lti::MSTClustering< U >::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 written, otherwise only the data block will be written.

Returns:: true if write was successful

Reimplemented from lti::clustering::parameters.

References lti::MSTClustering::parameters::devFac, lti::MSTClustering::parameters::nbMaxSteps, lti::MSTClustering::parameters::probabilityFromDistanceMode, lti::ioHandler::readBegin(), lti::ioHandler::readEnd(), and lti::MSTClustering::parameters::variance.

template<class U = l2SquareDistantor<dvector>>

bool lti::MSTClustering< U >::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::clustering::parameters.

References lti::MSTClustering::parameters::devFac, lti::MSTClustering::parameters::nbMaxSteps, lti::MSTClustering::parameters::probabilityFromDistanceMode, lti::MSTClustering::parameters::variance, lti::ioHandler::writeBegin(), and lti::ioHandler::writeEnd().

Member Data Documentation

template<class U = l2SquareDistantor<dvector>>

float lti::MSTClustering::parameters::devFac

Factor that is used to decide if an edge is inconsistent.

First all edges are searched that lie at most nbMaxSteps away from the current node. Then m is the mean of the weights of all these edges and sigma is the standard deviation of these edges. An Edge is considered to be in consistent if is weight is bigger than mean+devFac*sigma. Default is 2.0

Referenced by lti::MSTClustering::parameters::copy(), lti::MSTClustering::parameters::parameters(), lti::MSTClustering::parameters::read(), and lti::MSTClustering::parameters::write().

template<class U = l2SquareDistantor<dvector>>

int lti::MSTClustering::parameters::nbMaxSteps

The maximum number of edges that can be between the starting point and the last edge that is considered when computed inconsistent edges.

Default is 4.

Referenced by lti::MSTClustering::parameters::copy(), lti::MSTClustering::parameters::parameters(), lti::MSTClustering::parameters::read(), and lti::MSTClustering::parameters::write().

template<class U = l2SquareDistantor<dvector>>

eProbabilityFromDistanceTypes lti::MSTClustering::parameters::probabilityFromDistanceMode

The method used to compute a probabilty from the distance between the feature that is classified and the clusters.

Valid valued are: InterClusterDistanceDependend and InterClusterDistanceIndependend. See parameter variance

Referenced by lti::MSTClustering::parameters::copy(), lti::MSTClustering::parameters::parameters(), lti::MSTClustering::parameters::read(), and lti::MSTClustering::parameters::write().

template<class U = l2SquareDistantor<dvector>>

double lti::MSTClustering::parameters::variance

Variance used to compute a possibility from the distances for classification.

While classification, the membership of feature to all clusters is computed. This done by the following: Say d is the minimal distance between the feature that is classified and one of the clusters. Say also w is the the minimal distance between the current cluster the nearest neighboring cluster, then the membership is defined either $e^{-\frac{d^2}{\sigma^2w^2}}$ or $e^{-\frac{d^2}{\sigma^2}}$ depending on the value of probabilityFromDistanceMode Default is 0.33333.