te::cl::KMeans< TTRAIN, TCLASSIFY > Class Template Reference

KMeans strategy for classification. Step-by-step: More...

#include <KMeans.h>

Classes
class	Parameters
	Classifier Parameters. More...

Public Member Functions
bool	classify (TCLASSIFY &itBegin, TCLASSIFY &itEnd, const std::vector< unsigned int > &attributesIndices, std::vector< unsigned int > &classification, const bool enableProgressInterface)
bool	initialize (const Parameters &params)
	KMeans ()
bool	train (TTRAIN &itBegin, TTRAIN &itEnd, const std::vector< unsigned int > &attributesIndices, const std::vector< unsigned int > &labels, const bool enableProgressInterface)
	~KMeans ()

Protected Member Functions
unsigned int	getClassification (std::vector< double > values)

Protected Attributes
bool	m_isInitialized
	True if this instance is initialized.
std::vector< std::vector< double > >	m_KMeans
	The vector of K means.
Parameters	m_parameters
	Internal execution parameters.

Detailed Description

template<class TTRAIN, class TCLASSIFY>
class te::cl::KMeans< TTRAIN, TCLASSIFY >

KMeans strategy for classification. Step-by-step:

1. Define a random solution, creating K values of mean.
2. Classify each input value according to the closest mean.
3. Calculate new means using the classified pixels.
4. Go back to step 2, or stop depending on the following checks: 4.1. Check if the new means are different from the previous iteration, using the convergence threshold. 4.2. Check if maximum number of iterations has achieved.

   Note

   Output classification IDs starting at 1.

Definition at line 63 of file KMeans.h.

Constructor & Destructor Documentation

KMeans()

template<class TTRAIN, class TCLASSIFY>

te::cl::KMeans< TTRAIN, TCLASSIFY >::KMeans

(

)

Definition at line 164 of file KMeans.h.

References m_isInitialized, and m_KMeans.

~KMeans()

template<class TTRAIN, class TCLASSIFY>

te::cl::KMeans< TTRAIN, TCLASSIFY >::~KMeans

(

)

Definition at line 171 of file KMeans.h.

Member Function Documentation

classify()

template<class TTRAIN, class TCLASSIFY>

bool te::cl::KMeans< TTRAIN, TCLASSIFY >::classify	(	TCLASSIFY &	itBegin,
		TCLASSIFY &	itEnd,
		const std::vector< unsigned int > &	attributesIndices,
		std::vector< unsigned int > &	classification,
		const bool	enableProgressInterface )

Definition at line 317 of file KMeans.h.

References getClassification(), te::common::TaskProgress::pulse(), TE_TR, and te::common::TaskProgress::UNDEFINED.

getClassification()

template<class TTRAIN, class TCLASSIFY>

unsigned int te::cl::KMeans< TTRAIN, TCLASSIFY >::getClassification ( std::vector< double > values )

protected

Definition at line 353 of file KMeans.h.

References te::cl::GetEuclideanDistance(), m_KMeans, and m_parameters.

Referenced by classify(), and train().

initialize()

template<class TTRAIN, class TCLASSIFY>

bool te::cl::KMeans< TTRAIN, TCLASSIFY >::initialize

(

const Parameters &

params

)

Definition at line 176 of file KMeans.h.

References m_isInitialized, m_KMeans, m_parameters, and TE_TR.

train()

template<class TTRAIN, class TCLASSIFY>

bool te::cl::KMeans< TTRAIN, TCLASSIFY >::train	(	TTRAIN &	itBegin,
		TTRAIN &	itEnd,
		const std::vector< unsigned int > &	attributesIndices,
		const std::vector< unsigned int > &	labels,
		const bool	enableProgressInterface )

Definition at line 199 of file KMeans.h.

References getClassification(), m_KMeans, m_parameters, te::common::TaskProgress::pulse(), TE_TR, and te::common::TaskProgress::UNDEFINED.

Member Data Documentation

m_isInitialized

template<class TTRAIN, class TCLASSIFY>

bool te::cl::KMeans< TTRAIN, TCLASSIFY >::m_isInitialized

protected

True if this instance is initialized.

Definition at line 116 of file KMeans.h.

Referenced by initialize(), and KMeans().

m_KMeans

template<class TTRAIN, class TCLASSIFY>

std::vector<std::vector<double> > te::cl::KMeans< TTRAIN, TCLASSIFY >::m_KMeans

protected

The vector of K means.

Definition at line 117 of file KMeans.h.

Referenced by getClassification(), initialize(), KMeans(), and train().

m_parameters

template<class TTRAIN, class TCLASSIFY>

Parameters te::cl::KMeans< TTRAIN, TCLASSIFY >::m_parameters

protected

Internal execution parameters.

Definition at line 118 of file KMeans.h.

Referenced by getClassification(), initialize(), and train().

---

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

• /home/castejon/castejon_files/develop/terralib5/git_release/src/terralib/classification/KMeans.h