ClassificationAlgorithm< DSClass > Class Template Reference

#include <base_algorithm.h>

Collaboration diagram for ClassificationAlgorithm< DSClass >:

Public Types
typedef DSClass	DS
typedef DS::TS	TS

Public Member Functions
	ClassificationAlgorithm (const std::string &_ID="")
virtual	~ClassificationAlgorithm ()
virtual void	train (DS &_ds, const TS &_examples) const
virtual void	train (DS &_ds, TS &_examples) const
virtual double	test (const DS &_ds, const TS &_examples) const =0
virtual void	printSelf (std::ostream &_os) const
virtual const std::string &	setID () const
virtual std::string &	setID ()
void	setListener (BaseListener *_listener)

Protected Attributes
std::string	ID
BaseListener *	listener

Detailed Description

template<class DSClass>
class ClassificationAlgorithm< DSClass >

This is the parent class for all algorithms that can be trained and applied in order to classify a set of data. A xmipp Algorithm object must have: 1) a train method that accepts a set of examples 2) an apply method that accepts an example and classifies it

Definition at line 54 of file base_algorithm.h.

Member Typedef Documentation

DS

template<class DSClass>

typedef DSClass ClassificationAlgorithm< DSClass >::DS

Definition at line 58 of file base_algorithm.h.

TS

template<class DSClass>

typedef DS::TS ClassificationAlgorithm< DSClass >::TS

Definition at line 59 of file base_algorithm.h.

Constructor & Destructor Documentation

ClassificationAlgorithm()

template<class DSClass>

ClassificationAlgorithm< DSClass >::ClassificationAlgorithm ( const std::string &  _ID = "" )

inline

Constructor. Parameter: _ID an ID string unique for each algorithm class

Definition at line 65 of file base_algorithm.h.

                                                      : ID(_ID)
    {};

~ClassificationAlgorithm()

template<class DSClass>

virtual ClassificationAlgorithm< DSClass >::~ClassificationAlgorithm ( )

inlinevirtual

Destructor. The default destructor

Definition at line 72 of file base_algorithm.h.

    {};

Member Function Documentation

printSelf()

template<class DSClass>

virtual void ClassificationAlgorithm< DSClass >::printSelf ( std::ostream &  _os ) const

inlinevirtual

Print itself on standard output

Reimplemented in SOM, and FuzzyCMeans.

Definition at line 106 of file base_algorithm.h.

    {
        _os << "xmippBaseAlgo" << std::endl; // to identify it as an algorith
        _os << ID << std::endl;
    };

setID() [1/2]

template<class DSClass>

virtual const std::string& ClassificationAlgorithm< DSClass >::setID ( ) const

inlinevirtual

Set ID (returns a const reference to the ID)

Definition at line 114 of file base_algorithm.h.

    {
        return ID;
    };

setID() [2/2]

template<class DSClass>

virtual std::string& ClassificationAlgorithm< DSClass >::setID ( )

inlinevirtual

Set ID (returns a non-const reference to the ID)

Definition at line 121 of file base_algorithm.h.

    {
        return ID;
    };

setListener()

template<class DSClass>

void ClassificationAlgorithm< DSClass >::setListener ( BaseListener *  _listener )

inline

Defines Listener class

Definition at line 129 of file base_algorithm.h.

    {
        listener = _listener;
    };

test()

template<class DSClass>

virtual double ClassificationAlgorithm< DSClass >::test ( const DS &  _ds, const TS &  _examples  ) const

pure virtual

Method to test the algorithm. Note that this method is pure virtual, so it must be defined in every xmipp descendant class. Parameter: _ds Data structure to train. Const because its not affected by test Parameter: _examples A training set with the training examples.

Returns

The test error, that is, error obtained for the test file. Might be MSE error, or success in classification error, or whatever; just be consistent for each algorithm

train() [1/2]

template<class DSClass>

virtual void ClassificationAlgorithm< DSClass >::train ( DS &  _ds, const TS &  _examples  ) const

inlinevirtual

Method to train the algorithm. Note that this method is pure virtual, so it ust be defined in every xmipp descendant class. Parameter: _ds Data structure to train Parameter: _examples A training set with the training examples.

Definition at line 81 of file base_algorithm.h.

        {};

train() [2/2]

template<class DSClass>

virtual void ClassificationAlgorithm< DSClass >::train ( DS &  _ds, TS &  _examples  ) const

inlinevirtual

Method to train the algorithm. Note that this method is pure virtual, so it ust be defined in every xmipp descendant class. Parameter: _ds Data structure to train Parameter: _examples A training set with the training examples.

Definition at line 90 of file base_algorithm.h.

        {};

Member Data Documentation

ID

template<class DSClass>

std::string ClassificationAlgorithm< DSClass >::ID

protected

Definition at line 132 of file base_algorithm.h.

listener

template<class DSClass>

BaseListener* ClassificationAlgorithm< DSClass >::listener

protected

Definition at line 136 of file base_algorithm.h.

---

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

• xmipp/libraries/classification/base_algorithm.h