OpenNN::MinkowskiError Class Reference

#include <minkowski_error.h>

Inheritance diagram for OpenNN::MinkowskiError:

OpenNN::PerformanceTerm

List of all members.

Public Member Functions

 MinkowskiError (void)
 MinkowskiError (NeuralNetwork *)
 MinkowskiError (DataSet *)
 MinkowskiError (NeuralNetwork *, DataSet *)
 MinkowskiError (TiXmlElement *)
virtual ~MinkowskiError (void)
double get_Minkowski_parameter (void) const
void set_default (void)
void set_Minkowski_parameter (const double &)
void check (void) const
double calculate_evaluation (void) const
double calculate_evaluation (const Vector< double > &) const
double calculate_generalization_evaluation (void) const
Vector< double > calculate_gradient (void) const
Matrix< double > calculate_Hessian (void) const
std::string write_performance_term_type (void) const
TiXmlElement * to_XML (void) const
void from_XML (TiXmlElement *)
int calculate_sign (const double &) const


Detailed Description

This class represents the Minkowski error performance term. The Minkowski error measures the difference between the outputs of a neural network and the targets in a data set. This performance term is used in data modeling problems. It can be more useful when the data set presents outliers.

Definition at line 32 of file minkowski_error.h.


Constructor & Destructor Documentation

OpenNN::MinkowskiError::MinkowskiError ( void   )  [explicit]

Default constructor. It creates Minkowski error performance term 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 42 of file minkowski_error.cpp.

OpenNN::MinkowskiError::MinkowskiError ( NeuralNetwork new_neural_network_pointer  )  [explicit]

Neural network constructor. It creates a Minkowski error performance term associated to a neural network 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 55 of file minkowski_error.cpp.

OpenNN::MinkowskiError::MinkowskiError ( DataSet new_data_set_pointer  )  [explicit]

Data set constructor. It creates a Minkowski error performance term not associated to any neural network but to be measured on a data set. 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 69 of file minkowski_error.cpp.

OpenNN::MinkowskiError::MinkowskiError ( NeuralNetwork new_neural_network_pointer,
DataSet new_data_set_pointer 
) [explicit]

Neural network and data set constructor. It creates a Minkowski error performance term object 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 84 of file minkowski_error.cpp.

OpenNN::MinkowskiError::MinkowskiError ( TiXmlElement *  mean_squared_error_element  )  [explicit]

XML constructor. It creates a Minkowski error object neither associated to a neural network nor to a data set. The object members are loaded by means of a XML element.

Parameters:
mean_squared_error_element Tiny XML element with the Minkowski error elements.

Definition at line 98 of file minkowski_error.cpp.

OpenNN::MinkowskiError::~MinkowskiError ( void   )  [virtual]

Destructor. It does not delete any pointer.

Definition at line 112 of file minkowski_error.cpp.


Member Function Documentation

double OpenNN::MinkowskiError::get_Minkowski_parameter ( void   )  const

This method returns the Minkowski exponent value used to calculate the error.

Definition at line 123 of file minkowski_error.cpp.

void OpenNN::MinkowskiError::set_default ( void   )  [virtual]

This method sets the default values to a Minkowski error object:

  • Minkowski parameter: 1.5.
  • Display: true.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 137 of file minkowski_error.cpp.

void OpenNN::MinkowskiError::set_Minkowski_parameter ( const double &  new_Minkowski_parameter  ) 

This method sets a new Minkowski exponent value to be used in order to calculate the error. The Minkowski R-value must be comprised between 1 and 2.

Parameters:
new_Minkowski_parameter Minkowski exponent value.

Definition at line 151 of file minkowski_error.cpp.

void OpenNN::MinkowskiError::check ( void   )  const [virtual]

This method checks that there are a neural network and a data set associated to the Minkowski 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 178 of file minkowski_error.cpp.

double OpenNN::MinkowskiError::calculate_evaluation ( void   )  const [virtual]

This method returns the Minkowski error evaluation.

Implements OpenNN::PerformanceTerm.

Definition at line 313 of file minkowski_error.cpp.

double OpenNN::MinkowskiError::calculate_evaluation ( const Vector< double > &  parameters  )  const [virtual]

This method returns which would be the Minkowski error of for an hypothetical vector of parameters. It does not set that vector of parameters to the neural network.

Parameters:
parameters Vector of a potential parameters for the neural network associated to the Minkowski error.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 268 of file minkowski_error.cpp.

double OpenNN::MinkowskiError::calculate_generalization_evaluation ( void   )  const [virtual]

This method returns the Minkowski error of the multilayer perceptron measured on the generalization instances of the data set.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 369 of file minkowski_error.cpp.

Vector< double > OpenNN::MinkowskiError::calculate_gradient ( void   )  const [virtual]

This method returns the Minkowski error function gradient of a multilayer perceptron on an inputs-targets data set. It uses the error back-propagation method.

Todo:

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 423 of file minkowski_error.cpp.

Matrix< double > OpenNN::MinkowskiError::calculate_Hessian ( void   )  const [virtual]

Todo:

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 480 of file minkowski_error.cpp.

std::string OpenNN::MinkowskiError::write_performance_term_type ( void   )  const [virtual]

This method returns a string with the name of the Minkowski error performance type, "MINKOWSKI_ERROR".

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 492 of file minkowski_error.cpp.

TiXmlElement * OpenNN::MinkowskiError::to_XML ( void   )  const [virtual]

This method serializes the Minkowski error object into a XML element of the TinyXML library. See the OpenNN manual for more information about the format of this element.

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 503 of file minkowski_error.cpp.

void OpenNN::MinkowskiError::from_XML ( TiXmlElement *  Minkoski_error_element  )  [virtual]

Todo:

Reimplemented from OpenNN::PerformanceTerm.

Definition at line 544 of file minkowski_error.cpp.

int OpenNN::MinkowskiError::calculate_sign ( const double &  value  )  const

This method returns -1 if some real value is negative and 1 if it is positive.

Parameters:
value Real value.

Definition at line 595 of file minkowski_error.cpp.


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

Generated on Sun Aug 26 11:58:19 2012 for OpenNN by  doxygen 1.5.9