#include <normalized_squared_error.h>

Inheritance diagram for OpenNN::NormalizedSquaredError:

Public Member Functions
	NormalizedSquaredError (NeuralNetwork , DataSet )

	NormalizedSquaredError (NeuralNetwork *)

	NormalizedSquaredError (DataSet *)

	NormalizedSquaredError (void)

	NormalizedSquaredError (const tinyxml2::XMLDocument &)

virtual	~NormalizedSquaredError (void)

double	calculate_normalization_coefficient (const Matrix< double > &, const Vector< double > &) const

void	check (void) const

double	calculate_performance (void) const

Vector< double >	calculate_gradient (void) const

Matrix< double >	calculate_Hessian (void) const

double	calculate_performance (const Vector< double > &) const

double	calculate_generalization_performance (void) const

Vector< double >	calculate_terms (void) const

Vector< double >	calculate_terms (const Vector< double > &) const

Matrix< double >	calculate_terms_Jacobian (void) const

PerformanceTerm::FirstOrderTerms	calculate_first_order_terms (void) const

Vector< double >	calculate_squared_errors (void) const

Vector< size_t >	calculate_maximal_errors (const size_t &=10) const

std::string	write_performance_term_type (void) const

tinyxml2::XMLDocument *	to_XML (void) const

void	from_XML (const tinyxml2::XMLDocument &)

std::string	write_information (void) const

Public Member Functions inherited from OpenNN::PerformanceTerm
	PerformanceTerm (void)

	PerformanceTerm (NeuralNetwork *)

	PerformanceTerm (DataSet *)

	PerformanceTerm (MathematicalModel *)

	PerformanceTerm (NeuralNetwork , DataSet )

	PerformanceTerm (NeuralNetwork , MathematicalModel )

	PerformanceTerm (NeuralNetwork , MathematicalModel , DataSet *)

	PerformanceTerm (const tinyxml2::XMLDocument &)

	PerformanceTerm (const PerformanceTerm &)

virtual	~PerformanceTerm (void)

virtual PerformanceTerm &	operator= (const PerformanceTerm &)

virtual bool	operator== (const PerformanceTerm &) const

NeuralNetwork *	get_neural_network_pointer (void) const

MathematicalModel *	get_mathemtaical_model_pointer (void) const

DataSet *	get_data_set_pointer (void) const

NumericalDifferentiation *	get_numerical_differentiation_pointer (void) const

const bool &	get_display (void) const

bool	has_neural_network (void) const

bool	has_mathematical_model (void) const

bool	has_data_set (void) const

bool	has_numerical_differentiation (void) const

virtual void	set (void)

virtual void	set (NeuralNetwork *)

virtual void	set (DataSet *)

virtual void	set (MathematicalModel *)

virtual void	set (NeuralNetwork , DataSet )

virtual void	set (NeuralNetwork , MathematicalModel )

virtual void	set (NeuralNetwork , MathematicalModel , DataSet *)

void	set (const PerformanceTerm &)

virtual void	set_neural_network_pointer (NeuralNetwork *)

virtual void	set_mathematical_model_pointer (MathematicalModel *)

virtual void	set_data_set_pointer (DataSet *)

void	set_numerical_differentiation_pointer (NumericalDifferentiation *)

virtual void	set_default (void)

void	set_display (const bool &)

void	construct_numerical_differentiation (void)

void	delete_numerical_differentiation_pointer (void)

Vector< Vector< double > >	calculate_layers_delta (const Vector< Vector< double > > &, const Vector< double > &) const

Vector< Vector< double > >	calculate_layers_delta (const Vector< Vector< double > > &, const Vector< double > &, const Vector< double > &) const

Matrix< Matrix< double > >	calculate_interlayers_Delta (const Vector< Vector< double > > &, const Vector< Vector< double > > &, const Matrix< Matrix< double > > &, const Vector< double > &, const Matrix< double > &, const Vector< Vector< double > > &) const

Vector< double >	calculate_point_gradient (const Vector< double > &, const Vector< Vector< double > > &, const Vector< Vector< double > > &) const

Vector< double >	calculate_point_gradient (const Vector< Matrix< double > > &, const Vector< Vector< double > > &) const

Matrix< double >	calculate_point_Hessian (const Vector< Vector< double > > &, const Vector< Vector< Vector< double > > > &, const Matrix< Matrix< double > > &, const Vector< Vector< double > > &, const Matrix< Matrix< double > > &) const

virtual Vector< double >	calculate_gradient (const Vector< double > &) const

virtual Matrix< double >	calculate_Hessian (const Vector< double > &) const

virtual std::string	to_string (void) const

size_t	calculate_Kronecker_delta (const size_t &, const size_t &) const

Private Attributes
Vector< double >	training_target_mean

Additional Inherited Members
Protected Attributes inherited from OpenNN::PerformanceTerm
NeuralNetwork *	neural_network_pointer

DataSet *	data_set_pointer

MathematicalModel *	mathematical_model_pointer

NumericalDifferentiation *	numerical_differentiation_pointer

bool	display

Detailed Description

This class represents the normalized squared error performance term. This performance term is used in data modeling problems. If it has a value of unity then the neural network is predicting the data "in the mean", A value of zero means perfect prediction of data.

Definition at line 44 of file normalized_squared_error.h.

Constructor & Destructor Documentation

OpenNN::NormalizedSquaredError::NormalizedSquaredError	(	NeuralNetwork *	new_neural_network_pointer,
		DataSet *	new_data_set_pointer
	)

explicit

Neural network and data set constructor. It creates a normalized squared error performance term associated to a neural network and measured on a data set. It also initializes all the rest of class members to their default values.

Parameters

new_neural_network_pointer	Pointer to a neural network object.
new_data_set_pointer	Pointer to a data set object.

Definition at line 68 of file normalized_squared_error.cpp.

OpenNN::NormalizedSquaredError::NormalizedSquaredError ( NeuralNetwork * new_neural_network_pointer )

explicit

Neural network constructor. It creates a normalized squared error performance term associated to a neural network object but not measured on any data set. It also initializes all the rest of class members to their default values.

Parameters

new_neural_network_pointer Pointer to a neural network object.

Definition at line 40 of file normalized_squared_error.cpp.

OpenNN::NormalizedSquaredError::NormalizedSquaredError ( DataSet * new_data_set_pointer )

explicit

Data set constructor. It creates a normalized squared error performance term not associated to any neural network but to be measured on a data set object. It also initializes all the rest of class members to their default values.

Parameters

new_data_set_pointer Pointer to a data set object.

Definition at line 54 of file normalized_squared_error.cpp.

OpenNN::NormalizedSquaredError::NormalizedSquaredError ( void )

explicit

Default constructor. It creates a normalized squared error performance term object not associated to any neural network and not measured on any data set. It also initializes all the rest of class members to their default values.

Definition at line 28 of file normalized_squared_error.cpp.

OpenNN::NormalizedSquaredError::NormalizedSquaredError ( const tinyxml2::XMLDocument & normalized_squared_error_document )

explicit

XML constructor. It creates a normalized squared error not associated to any neural network and not measured on any data set. It also sets all the rest of class members from a TinyXML document->

Parameters

normalized_squared_error_document XML document with the class members.

Definition at line 81 of file normalized_squared_error.cpp.

Member Function Documentation

NormalizedSquaredError::FirstOrderTerms OpenNN::NormalizedSquaredError::calculate_first_order_terms ( void ) const

virtual

Returns a first order performance terms performance structure, which contains the values and the Jacobian of the performance terms function.

Todo:

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 1009 of file normalized_squared_error.cpp.

Vector< double > OpenNN::NormalizedSquaredError::calculate_gradient ( void ) const

virtual

Returns the normalized squared error function gradient of a multilayer perceptron on a data set. It uses the error back-propagation method.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 523 of file normalized_squared_error.cpp.

Matrix< double > OpenNN::NormalizedSquaredError::calculate_Hessian ( void ) const

virtual

Returns the normalized squared error function Hessian of a multilayer perceptron on a data set. It uses the error back-propagation method.

Todo:

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 673 of file normalized_squared_error.cpp.

Vector< size_t > OpenNN::NormalizedSquaredError::calculate_maximal_errors ( const size_t & maximal_errors_number = 10 ) const

Returns a vector with the indices of the instances which have the maximum error.

Parameters

maximal_errors_number Number of instances required.

Definition at line 1108 of file normalized_squared_error.cpp.

double OpenNN::NormalizedSquaredError::calculate_normalization_coefficient	(	const Matrix< double > &	target_data,
		const Vector< double > &	target_data_mean
	)		const

Returns the normalization coefficient to be used for the performance of the error. This is measured on the training instances of the data set.

Definition at line 103 of file normalized_squared_error.cpp.

double OpenNN::NormalizedSquaredError::calculate_performance ( const Vector< double > & parameters ) const

virtual

Returns which would be the performance of a multilayer perceptron for an hypothetical vector of parameters. It does not set that vector of parameters to the multilayer perceptron.

Parameters

parameters Vector of potential parameters for the multilayer perceptron associated to the performance functional.

Implements OpenNN::PerformanceTerm.

Definition at line 306 of file normalized_squared_error.cpp.

Vector< double > OpenNN::NormalizedSquaredError::calculate_terms ( void ) const

virtual

Returns performance vector of the performance terms function for the normalized squared error. It uses the error back-propagation method.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 686 of file normalized_squared_error.cpp.

Vector< double > OpenNN::NormalizedSquaredError::calculate_terms ( const Vector< double > & network_parameters ) const

virtual

Returns which would be the performance terms performance vector of a multilayer perceptron for an hypothetical vector of multilayer perceptron parameters. It does not set that vector of parameters to the multilayer perceptron.

Parameters

network_parameters Vector of a potential multilayer_perceptron_pointer parameters for the multilayer perceptron associated to the performance functional.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 787 of file normalized_squared_error.cpp.

Matrix< double > OpenNN::NormalizedSquaredError::calculate_terms_Jacobian ( void ) const

virtual

Returns the terms_Jacobian matrix of the sum squared error function, whose elements are given by the derivatives of the squared errors data set with respect to the multilayer perceptron parameters. The terms_Jacobian matrix here is computed using a back-propagation algorithm.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 837 of file normalized_squared_error.cpp.

void OpenNN::NormalizedSquaredError::check ( void ) const

virtual

Checks that there are a neural network and a data set associated to the normalized squared error, and that the numbers of inputs and outputs in the neural network are equal to the numbers of inputs and targets in the data set. If some of the above conditions is not hold, the method throws an exception.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 115 of file normalized_squared_error.cpp.

void OpenNN::NormalizedSquaredError::from_XML ( const tinyxml2::XMLDocument & document )

virtual

Loads a root mean squared error object from a XML document.

Parameters

document Pointer to a TinyXML document with the object data.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 1185 of file normalized_squared_error.cpp.

tinyxml2::XMLDocument * OpenNN::NormalizedSquaredError::to_XML ( void ) const

virtual

Serializes the normalized squared error object into a XML document of the TinyXML library. See the OpenNN manual for more information about the format of this element.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 1152 of file normalized_squared_error.cpp.

std::string OpenNN::NormalizedSquaredError::write_information ( void ) const

virtual

Returns a string with the default information of the performance term. It will be used by the training strategy to monitor the training process. By default this information is empty.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 1214 of file normalized_squared_error.cpp.

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

Public Member Functions

Private Attributes

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation