LearningAlgorithm Class Reference

LearningAlgorithm object. More...

Inheritance diagram for LearningAlgorithm:

Public Member Functions
	LearningAlgorithm (NeuralNet *net)
	Constructor.
	LearningAlgorithm ()
	Constructor.
virtual	~LearningAlgorithm ()
	Destructor.
virtual double	calculateMSE (const Pattern &)=0
	Calculate the Mean Square Error respect to Pattern passed.
virtual double	calculateMSEOnSet (const PatternSet &set)
	Calculate the Mean Square Error respect to all Patterns passed.
double	calculateRMSD (const Pattern &p)
	Calculate the Root Mean Square Deviation, i.e.
double	calculateRMSDOnSet (const PatternSet &p)
	Calculate the Root Mean Square Deviation, i.e.
virtual void	learn ()=0
	a single step of learning algorithm
virtual void	learn (const Pattern &)=0
	Modify the object tring to learn the pattern passed.
virtual void	learnOnSet (const PatternSet &set)
	Modify the object tring to learn all patterns present into PatternSet passed.
PatternSet	loadPatternSet (ConfigurationParameters &params, QString path, QString prefix)
	Utility function for loading a PatternSet from a ConfigurationParameters.
NeuralNet *	neuralNet ()
	Return the NeuralNet setted.
void	savePatternSet (PatternSet &set, ConfigurationParameters &params, QString prefix)
	Utility function for saving a PatternSet to a ConfigurationParameters.
void	setNeuralNet (NeuralNet *net)
	Set the NeuralNet to learn.
Public Member Functions inherited from ParameterSettableWithConfigureFunction
void	addObserver (RuntimeParameterObserver *obs)
virtual void	configure (ConfigurationParameters &params, QString prefix)=0
T	getRuntimeParameter (QString paramName)
virtual ParameterSettableUI *	getUIManager ()
	ParameterSettable ()
virtual void	postConfigureInitialization ()
void	removeObserver (RuntimeParameterObserver *obs)
virtual void	save (ConfigurationParameters &params, QString prefix)=0
void	setRuntimeParameter (QString paramName, T newvalue)
QString	typeName () const
Public Member Functions inherited from ParameterSettable
void	addObserver (RuntimeParameterObserver *obs)
T	getRuntimeParameter (QString paramName)
void	removeObserver (RuntimeParameterObserver *obs)
void	setRuntimeParameter (QString paramName, T newvalue)
QString	typeName () const

Protected Member Functions
virtual void	neuralNetChanged ()=0
	Implemented by subclasses for configuring internal structure when the NeuralNet has been setted.
Protected Member Functions inherited from ParameterSettableWithConfigureFunction
void	notifyChangesToParam (QString paramName)
Protected Member Functions inherited from ParameterSettable
void	notifyChangesToParam (QString paramName)

Additional Inherited Members
Public Types inherited from ParameterSettableWithConfigureFunction
enum	Property
Static Public Member Functions inherited from ParameterSettableWithConfigureFunction
static void	describe (QString type)
static QString	fullParameterDescriptionPath (QString type, QString param)
static QString	fullSubgroupDescriptionPath (QString type, QString sub)
Public Attributes inherited from ParameterSettableWithConfigureFunction
	AllowMultiple
	Default
	IsList
	IsMandatory
Static Public Attributes inherited from ParameterSettableWithConfigureFunction
static const double	Infinity
static const int	MaxInteger
static const int	MinInteger
Static Protected Member Functions inherited from ParameterSettableWithConfigureFunction
static Descriptor	addTypeDescription (QString type, QString shortHelp, QString longHelp=QString(""))
static void	setGraphicalEditor (QString type)

Detailed Description

LearningAlgorithm object.

The LearningAlgorithm object is a the abstract class from which to implement learning algorithms

Definition at line 143 of file learningalgorithm.h.

Constructor & Destructor Documentation

LearningAlgorithm

(

NeuralNet *

net

)

Constructor.

Definition at line 128 of file learningalgorithm.cpp.

LearningAlgorithm

(

)

Constructor.

Definition at line 133 of file learningalgorithm.cpp.

~LearningAlgorithm ( )

virtual

Destructor.

Definition at line 138 of file learningalgorithm.cpp.

Member Function Documentation

virtual double calculateMSE ( const Pattern &  )

pure virtual

Calculate the Mean Square Error respect to Pattern passed.

Implemented in BackPropagationAlgo.

virtual double calculateMSEOnSet ( const PatternSet &  set )

inlinevirtual

Calculate the Mean Square Error respect to all Patterns passed.

Definition at line 173 of file learningalgorithm.h.

double calculateRMSD ( const Pattern &  p )

inline

Calculate the Root Mean Square Deviation, i.e.

the square root of MSE

Definition at line 182 of file learningalgorithm.h.

double calculateRMSDOnSet ( const PatternSet &  p )

inline

Calculate the Root Mean Square Deviation, i.e.

the square root of MSE

Definition at line 186 of file learningalgorithm.h.

virtual void learn ( )

pure virtual

a single step of learning algorithm

Implemented in BackPropagationAlgo.

virtual void learn ( const Pattern &  )

pure virtual

Modify the object tring to learn the pattern passed.

Implemented in BackPropagationAlgo.

virtual void learnOnSet ( const PatternSet &  set )

inlinevirtual

Modify the object tring to learn all patterns present into PatternSet passed.

Definition at line 165 of file learningalgorithm.h.

PatternSet loadPatternSet	(	ConfigurationParameters &	params,
		QString	path,
		QString	prefix
	)

Utility function for loading a PatternSet from a ConfigurationParameters.

Definition at line 141 of file learningalgorithm.cpp.

References ConfigurationParameters::getGroupsWithPrefixList(), ConfigurationParameters::getObjectFromGroup(), ConfigurationParameters::startRememberingGroupObjectAssociations(), and ConfigurationParameters::stopRememberingGroupObjectAssociations().

NeuralNet* neuralNet ( )

inline

Return the NeuralNet setted.

Definition at line 157 of file learningalgorithm.h.

Referenced by BackPropagationAlgo::calculateMSE(), BackPropagationAlgo::learn(), BackPropagationAlgo::neuralNetChanged(), and BackPropagationAlgo::save().

virtual void neuralNetChanged ( )

protectedpure virtual

Implemented by subclasses for configuring internal structure when the NeuralNet has been setted.

Implemented in BackPropagationAlgo.

void savePatternSet	(	PatternSet &	set,
		ConfigurationParameters &	params,
		QString	prefix
	)

Utility function for saving a PatternSet to a ConfigurationParameters.

Definition at line 155 of file learningalgorithm.cpp.

References ConfigurationParameters::createGroup().

void setNeuralNet ( NeuralNet *  net )

inline

Set the NeuralNet to learn.

Definition at line 152 of file learningalgorithm.h.

Referenced by BackPropagationAlgo::configure().

---

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

• nnfw/include/learningalgorithm.h
• nnfw/src/learningalgorithm.cpp