The class for a complete MultiLayer Perceptron. More...

#include <aims/neuralnet/multilayerperceptron.h>

Collaboration diagram for AimsMultilayerPerceptron:

[legend]

Public Member Functions
Constructors and Destructor
	AimsMultilayerPerceptron (const AimsData< int32_t > &topology, int ninputs, float learningrate, float momentum, float maxerror, int stabilitytime, int maxtime)
	The constructor needs 7 parameters. More...

	AimsMultilayerPerceptron (const AimsData< int32_t > &topology, const AimsData< float > &weights, int ninputs, float learningrate=0, float momentum=0, float maxerror=0, int stabilitytime=0, int maxtime=0)
	The constructor needs 8 parameters. More...

virtual	~AimsMultilayerPerceptron ()
	Destructor deletes all the layers. More...

Methods
AimsData< int32_t >	topology () const
	Return the topology of the MLP network. More...

int	nLayers () const
	Return the number of layers of the network. More...

int	nInputs () const
	Return the number of inputs of the network. More...

int	nOutputs () const
	Return the number of outputs of the network. More...

float	learningRate () const
	Return the learning rate of the network. More...

float	momentum () const
	Return the momentum of the network. More...

float	maxError () const
	Return the maximum allowed error for convergence. More...

int	stabilityTime () const
	Return the minimum time for the net to be stable. More...

int	maxTime () const
	Return the maximum time for learning step. More...

void	setLearningRate (float learningrate)
	Set the learning rate of the network. More...

void	setMomentum (float momentum)
	Set the momentum of the network. More...

void	setMaxError (float maxerror)
	Set the maximum allowed error for convergence. More...

void	setStabilityTime (int stabilitytime)
	Set the minimum time for the net to be stable. More...

void	setMaxTime (int maxtime)
	Set the maximum time for learning step. More...

void	learn (const AimsData< float > &base, const AimsData< float > &target, int counter)
	Send a learning operation on a data base with a target. More...

void	forward (const AimsData< float > &input)
	Forward propagation. More...

void	backPropagation ()
	Gradient backward propagation. More...

void	adjust (const AimsData< float > &input)
	Modification of the weights. More...

AimsData< float >	test (const AimsData< float > &base)
	Test a data base and return the result. More...

AimsData< float >	weights () const
	Return all the weights of all the neurons in a volume. More...

void	save (const std::string &filetopology, const std::string &fileweight)
	Save the network to both G.I.S. files. More...

Protected Attributes
Data
int	_nLayers
	Number of layers. More...

int	_nInputs
	Number of inputs for each neuron. More...

int	_nOutputs
	Number of outputs of the network. More...

float	_learningRate
	Learning rate. More...

float	_momentum
	Momentum coefficient. More...

float	_maxError
	Maximum error to consider the network learned. More...

int	_stabilityTime
	Minimum number of epochs for the network to be considered stable. More...

int	_maxTime
	Maximum of number of epochs for the learning step. More...

AimsMLPLayer **	_layer
	Pointer to layer pointers. More...

AimsData< int32_t > *	_topology
	Topology of the network, i.e. number of neurons for each layer. More...

Detailed Description

The class for a complete MultiLayer Perceptron.

Definition at line 162 of file multilayerperceptron.h.

Constructor & Destructor Documentation

◆ AimsMultilayerPerceptron() [1/2]

AimsMultilayerPerceptron::AimsMultilayerPerceptron	(	const AimsData< int32_t > &	topology,
		int	ninputs,
		float	learningrate,
		float	momentum,
		float	maxerror,
		int	stabilitytime,
		int	maxtime
	)

The constructor needs 7 parameters.

Parameters

topology	data containing the number of neurons for each layer
ninputs	number of inputs per neuron
learningrate	learning rate
momentum	momentum coefficient
maxerror	maximum error to consider the network learned
stabilitytime	minimum number of epochs for the network to be considered stable
maxtime	maximum of number of epochs for the learning step

◆ AimsMultilayerPerceptron() [2/2]

AimsMultilayerPerceptron::AimsMultilayerPerceptron	(	const AimsData< int32_t > &	topology,
		const AimsData< float > &	weights,
		int	ninputs,
		float	learningrate = `0`,
		float	momentum = `0`,
		float	maxerror = `0`,
		int	stabilitytime = `0`,
		int	maxtime = `0`
	)

The constructor needs 8 parameters.

Parameters

topology	data containing the number of neurons for each layer
weights	neuron weights
ninputs	number of inputs per neuron
learningrate	learning rate [default=0]
momentum	momentum coefficient [default=0]
maxerror	max error to consider the network learned [default=0]
stabilitytime	minimum number of epochs for the network to be considered stable [defaul=0]
maxtime	maximum of number of epochs for the learning step [default=0]

◆ ~AimsMultilayerPerceptron()

virtual AimsMultilayerPerceptron::~AimsMultilayerPerceptron ( )

virtual

Destructor deletes all the layers.

Member Function Documentation

◆ adjust()

void AimsMultilayerPerceptron::adjust ( const AimsData< float > & input )

Modification of the weights.

◆ backPropagation()

void AimsMultilayerPerceptron::backPropagation ( )

Gradient backward propagation.

◆ forward()

void AimsMultilayerPerceptron::forward ( const AimsData< float > & input )

Forward propagation.

◆ learn()

void AimsMultilayerPerceptron::learn	(	const AimsData< float > &	base,
		const AimsData< float > &	target,
		int	counter
	)

Send a learning operation on a data base with a target.

◆ learningRate()

float AimsMultilayerPerceptron::learningRate ( ) const

Return the learning rate of the network.

◆ maxError()

float AimsMultilayerPerceptron::maxError ( ) const

Return the maximum allowed error for convergence.

◆ maxTime()

int AimsMultilayerPerceptron::maxTime ( ) const

Return the maximum time for learning step.

◆ momentum()

float AimsMultilayerPerceptron::momentum ( ) const

Return the momentum of the network.

◆ nInputs()

int AimsMultilayerPerceptron::nInputs ( ) const

Return the number of inputs of the network.

◆ nLayers()

int AimsMultilayerPerceptron::nLayers ( ) const

Return the number of layers of the network.

◆ nOutputs()

int AimsMultilayerPerceptron::nOutputs ( ) const

Return the number of outputs of the network.

◆ save()

void AimsMultilayerPerceptron::save	(	const std::string &	filetopology,
		const std::string &	fileweight
	)

Save the network to both G.I.S. files.

◆ setLearningRate()

void AimsMultilayerPerceptron::setLearningRate ( float learningrate )

Set the learning rate of the network.

◆ setMaxError()

void AimsMultilayerPerceptron::setMaxError ( float maxerror )

Set the maximum allowed error for convergence.

◆ setMaxTime()

void AimsMultilayerPerceptron::setMaxTime ( int maxtime )

Set the maximum time for learning step.

◆ setMomentum()

void AimsMultilayerPerceptron::setMomentum ( float momentum )

Set the momentum of the network.

◆ setStabilityTime()

void AimsMultilayerPerceptron::setStabilityTime ( int stabilitytime )

Set the minimum time for the net to be stable.

◆ stabilityTime()

int AimsMultilayerPerceptron::stabilityTime ( ) const

Return the minimum time for the net to be stable.

◆ test()

AimsData<float> AimsMultilayerPerceptron::test ( const AimsData< float > & base )

Test a data base and return the result.

◆ topology()

AimsData<int32_t> AimsMultilayerPerceptron::topology ( ) const

Return the topology of the MLP network.

◆ weights()

AimsData<float> AimsMultilayerPerceptron::weights ( ) const

Return all the weights of all the neurons in a volume.

Member Data Documentation

◆ _layer

AimsMLPLayer** AimsMultilayerPerceptron::_layer

protected

Pointer to layer pointers.

Definition at line 183 of file multilayerperceptron.h.

◆ _learningRate

float AimsMultilayerPerceptron::_learningRate

protected

Learning rate.

Definition at line 173 of file multilayerperceptron.h.

◆ _maxError

float AimsMultilayerPerceptron::_maxError

protected

Maximum error to consider the network learned.

Definition at line 177 of file multilayerperceptron.h.

◆ _maxTime

int AimsMultilayerPerceptron::_maxTime

protected

Maximum of number of epochs for the learning step.

Definition at line 181 of file multilayerperceptron.h.

◆ _momentum

float AimsMultilayerPerceptron::_momentum

protected

Momentum coefficient.

Definition at line 175 of file multilayerperceptron.h.

◆ _nInputs

int AimsMultilayerPerceptron::_nInputs

protected

Number of inputs for each neuron.

Definition at line 169 of file multilayerperceptron.h.

◆ _nLayers

int AimsMultilayerPerceptron::_nLayers

protected

Number of layers.

Definition at line 167 of file multilayerperceptron.h.

◆ _nOutputs

int AimsMultilayerPerceptron::_nOutputs

protected

Number of outputs of the network.

Definition at line 171 of file multilayerperceptron.h.

◆ _stabilityTime

int AimsMultilayerPerceptron::_stabilityTime

protected

Minimum number of epochs for the network to be considered stable.

Definition at line 179 of file multilayerperceptron.h.

◆ _topology

AimsData<int32_t>* AimsMultilayerPerceptron::_topology

protected

Topology of the network, i.e. number of neurons for each layer.

Definition at line 185 of file multilayerperceptron.h.

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

aims/neuralnet/multilayerperceptron.h

Public Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ AimsMultilayerPerceptron() [1/2]

◆ AimsMultilayerPerceptron() [2/2]

◆ ~AimsMultilayerPerceptron()

Member Function Documentation

◆ adjust()

◆ backPropagation()

◆ forward()

◆ learn()

◆ learningRate()

◆ maxError()

◆ maxTime()

◆ momentum()

◆ nInputs()

◆ nLayers()

◆ nOutputs()

◆ save()

◆ setLearningRate()

◆ setMaxError()

◆ setMaxTime()

◆ setMomentum()

◆ setStabilityTime()

◆ stabilityTime()

◆ test()

◆ topology()

◆ weights()

Member Data Documentation

◆ _layer

◆ _learningRate

◆ _maxError

◆ _maxTime

◆ _momentum

◆ _nInputs

◆ _nLayers

◆ _nOutputs

◆ _stabilityTime

◆ _topology