lti::bagging Class Reference

Implementation of bagging classification. More...

#include <ltiBagging.h>

Inheritance diagram for lti::bagging:

[legend]

Collaboration diagram for lti::bagging:

[legend]

List of all members.

Classes
class	parameters
	the parameters for the class bagging More...
Public Member Functions
	bagging ()
	bagging (const parameters &par)
	bagging (const bagging &other)
virtual	~bagging ()
virtual const char *	getTypeName () const
bool	train (const dmatrix &input, const ivector &ids)
bool	classify (const dvector &feature, classifier::outputVector &result) const
bagging &	copy (const bagging &other)
bagging &	operator= (const bagging &other)
virtual classifier *	clone () const
const parameters &	getParameters () const
bool	read (ioHandler &handler, const bool complete=true)
bool	write (ioHandler &handler, const bool complete=true) const
Protected Attributes
std::list < supervisedInstanceClassifier * > *	ensemble
supervisedInstanceClassifierFactory	factory

---

Detailed Description

Implementation of bagging classification.

For the bagging a set of classifiers, the ensemble, is trained with randomly drawn subsets of the training data. For classication each classifier in the ensemble is classifies the data and the different results of each classifier are combined to a single result.

---

Constructor & Destructor Documentation

lti::bagging::bagging

(

)

default constructor

lti::bagging::bagging

(

const parameters &

par

)

Construct a classifier using the given parameters.

lti::bagging::bagging

(

const bagging &

other

)

copy constructor

Parameters:

other

the object to be copied

virtual lti::bagging::~bagging

(

)

[virtual]

destructor

---

Member Function Documentation

bool lti::bagging::classify	(	const dvector &	feature,
		classifier::outputVector &	result
	)			const

classification and aggregating

virtual classifier* lti::bagging::clone

(

)

const [virtual]

returns a pointer to a clone of this classifier.

Implements lti::classifier.

bagging& lti::bagging::copy

(

const bagging &

other

)

copy data of "other" classifier.

Parameters:

other

the classifier to be copied

Returns:

a reference to this classifier object

Reimplemented from lti::classifier.

const parameters& lti::bagging::getParameters

(

)

const

returns used parameters

Reimplemented from lti::classifier.

virtual const char* lti::bagging::getTypeName

(

)

const [virtual]

returns the name of this type ("bagging")

Reimplemented from lti::classifier.

bagging& lti::bagging::operator=

(

const bagging &

other

)

alias for copy member

Parameters:

other

the classifier to be copied

Returns:

a reference to this classifier object

Reimplemented from lti::classifier.

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

read the bagging classifier from the given ioHandler

Reimplemented from lti::classifier.

bool lti::bagging::train	(	const dmatrix &	input,
		const ivector &	ids
	)

training of the ensembles

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

write to bagging classifier object to the given ioHandler

Reimplemented from lti::classifier.

---

Member Data Documentation

std::list<supervisedInstanceClassifier*>* lti::bagging::ensemble [protected]

list that collects all classifiers that are trained with different training data sets.

supervisedInstanceClassifierFactory lti::bagging::factory [protected]

object factory for constructing the classifiers in ensemble

---

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

• ltiBagging.h