Inheritance diagram for Pattern:

Classes
class	PatternInfo

Public Member Functions
	Pattern ()
	Construct an empty Pattern. More...

	~Pattern ()
	Destructor. More...

virtual void	configure (ConfigurationParameters &params, QString prefix)
	Configures the object using a ConfigurationParameters object. More...

const DoubleVector &	inputsOf (Cluster *) const
	return stored information if exists, otherwise it return a zero vector More...

PatternInfo &	operator[] (Cluster *)
	return the stored information More...

const DoubleVector &	outputsOf (Cluster *) const
	return stored information if exists, otherwise it return a zero vector More...

virtual void	save (ConfigurationParameters &params, QString prefix)
	Save the actual status of parameters into the ConfigurationParameters object passed. More...

void	setInputsOf (Cluster *, const DoubleVector &)
	set the inputs associated with Cluster passed More...

void	setInputsOutputsOf (Cluster *, const DoubleVector &inputs, const DoubleVector &outputs)
	set the both inputs and outputs associated with Cluster passed More...

void	setOutputsOf (Cluster *, const DoubleVector &)
	set the outputs associated with Cluster passed More...

Public Member Functions inherited from ParameterSettable
void	addObserver (RuntimeParameterObserver *obs)

T	getRuntimeParameter (QString paramName)

virtual ParameterSettableUI *	getUIManager ()

virtual void	postConfigureInitialization ()

void	removeObserver (RuntimeParameterObserver *obs)

void	setRuntimeParameter (QString paramName, T newvalue)

QString	typeName () const

Static Public Member Functions
static void	describe (QString type)
	Add to Factory::typeDescriptions() the descriptions of all parameters and subgroups. More...

Static Public Member Functions inherited from ParameterSettable
static void	describe (QString type)

static QString	fullParameterDescriptionPath (QString type, QString param)

static QString	fullSubgroupDescriptionPath (QString type, QString sub)

Additional Inherited Members
Public Types inherited from ParameterSettable
enum	Property

Public Attributes inherited from ParameterSettable
	AllowMultiple

	Default

	IsList

	IsMandatory

Static Public Attributes inherited from ParameterSettable
static const double	Infinity

static const int	MaxInteger

static const int	MinInteger

Protected Member Functions inherited from ParameterSettable
void	notifyChangesToParam (QString paramName)

Static Protected Member Functions inherited from ParameterSettable
static Descriptor	addTypeDescription (QString type, QString shortHelp, QString longHelp=QString(""))

static void	setGraphicalEditor (QString type)

Detailed Description

Pattern object.

The Pattern object represent a configuration (partial or complete) of the network that can be used by a learning algorithm as model to learn. A LearningAlgorithm will use Pattern for setup network's inputs, spread network and eventually calculate the error, and modify the network's paramenters on the basis of neurons activities in order to get closer to neuron activities presented by the Pattern

Pattern simply associates inputs/outputs pairs to Clusters:

Pattern pat;
pat->setInputsOf( anInputCluster, Inputs );
pat->setOutputsOf( anOutputCluster, Outputs );
pat->setInputOutputsOf( anHiddenClusterToReset, Inputs, Outputs );
//--- retrieve stored information:
pat[aCluster].inputs;
pat[anotherCluster].outputs;
//--- add/modify informations:
pat[aNewCluster].outputs.assign( aRealVecOfData );
pat[aNewCluster].outputs += 3.0;

Warning: Pay attention when you use operator[] because it silently add new data. Like QMap::operator[]

Definition at line 60 of file learningalgorithm.h.

Constructor & Destructor Documentation

Pattern ( )

inline

Construct an empty Pattern.

Definition at line 68 of file learningalgorithm.h.

~Pattern ( )

inline

Destructor.

Definition at line 70 of file learningalgorithm.h.

Member Function Documentation

void configure	(	ConfigurationParameters &	params,
		QString	prefix
	)

virtual

Configures the object using a ConfigurationParameters object.

The Pattern has to be declared using a sequence of parameters with the following schema:

[aPattern]
cluster:1 = nameOfCluster
inputs:1 = list of the inputs associated to cluster:1
outputs:1 = list of the outputs associated to cluster:1
...
cluster:i = nameOf_ith_Cluster
inputs:i = list of the inputs associated to cluster:i
outputs:i = list of the outputs associated to cluster:i

Essentialy, the parameters are grouped using the identifier string after the ':' char. And the inputs and outputs with a given identifier are associated to the cluster with the same identifier. If a cluster:i is present, then at least inputs:i or outputs:i parameter must be present

Parameters

params	the configuration parameters object with parameters to use
prefix	the prefix to use to access the object configuration parameters. This is guaranteed to end with the separator character when called by the factory, so you don't need to add one

Implements ParameterSettableWithConfigureFunction.

Definition at line 67 of file learningalgorithm.cpp.

References ConfigurationParameters::getObjectFromParameter(), ConfigurationParameters::getParametersWithPrefixList(), ConfigurationParameters::getValue(), Pattern::setInputsOf(), and Pattern::setOutputsOf().

void describe ( QString type )

static

Add to Factory::typeDescriptions() the descriptions of all parameters and subgroups.

Definition at line 127 of file learningalgorithm.cpp.

References ParameterSettable::addTypeDescription(), ParameterSettable::AllowMultiple, ParameterSettable::IsList, and ParameterSettable::IsMandatory.

const DoubleVector & inputsOf ( Cluster * cl ) const

return stored information if exists, otherwise it return a zero vector

Definition at line 43 of file learningalgorithm.cpp.

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