OpenNN  2.2
Open Neural Networks Library
Classes | Public Member Functions | Private Attributes | List of all members
OpenNN::PruningInputs Class Reference

#include <pruning_inputs.h>

Inheritance diagram for OpenNN::PruningInputs:


struct  PruningInputsResults

Public Member Functions

 PruningInputs (void)
 PruningInputs (TrainingStrategy *)
 PruningInputs (const tinyxml2::XMLDocument &)
 PruningInputs (const std::string &)
virtual ~PruningInputs (void)
const size_t & get_minimum_inputs_number (void) const
const size_t & get_maximum_selection_failures (void) const
void set_default (void)
void set_minimum_inputs_number (const size_t &)
void set_maximum_selection_failures (const size_t &)
PruningInputsResultsperform_inputs_selection (void)
Matrix< std::string > to_string_matrix (void) const
tinyxml2::XMLDocument * to_XML (void) const
void from_XML (const tinyxml2::XMLDocument &)
void write_XML (tinyxml2::XMLPrinter &) const
void save (const std::string &) const
void load (const std::string &)
- Public Member Functions inherited from OpenNN::InputsSelectionAlgorithm
 InputsSelectionAlgorithm (void)
 InputsSelectionAlgorithm (TrainingStrategy *)
 InputsSelectionAlgorithm (const std::string &)
 InputsSelectionAlgorithm (const tinyxml2::XMLDocument &)
virtual ~InputsSelectionAlgorithm (void)
const bool & get_approximation (void) const
TrainingStrategyget_training_strategy_pointer (void) const
bool has_training_strategy (void) const
const size_t & get_trials_number (void) const
const bool & get_reserve_parameters_data (void) const
const bool & get_reserve_loss_data (void) const
const bool & get_reserve_selection_loss_data (void) const
const bool & get_reserve_minimal_parameters (void) const
const PerformanceCalculationMethodget_loss_calculation_method (void) const
const bool & get_display (void) const
const double & get_selection_loss_goal (void) const
const size_t & get_maximum_iterations_number (void) const
const double & get_maximum_time (void) const
const double & get_maximum_correlation (void) const
const double & get_minimum_correlation (void) const
const double & get_tolerance (void) const
std::string write_loss_calculation_method (void) const
void set_approximation (const bool &)
void set_training_strategy_pointer (TrainingStrategy *)
void set_default (void)
void set_trials_number (const size_t &)
void set_reserve_parameters_data (const bool &)
void set_reserve_loss_data (const bool &)
void set_reserve_selection_loss_data (const bool &)
void set_reserve_minimal_parameters (const bool &)
void set_loss_calculation_method (const PerformanceCalculationMethod &)
void set_loss_calculation_method (const std::string &)
void set_display (const bool &)
void set_selection_loss_goal (const double &)
void set_maximum_iterations_number (const size_t &)
void set_maximum_time (const double &)
void set_maximum_correlation (const double &)
void set_minimum_correlation (const double &)
void set_tolerance (const double &)
Matrix< double > calculate_logistic_correlations (void) const
Vector< double > calculate_final_correlations (void) const
void set_neural_inputs (const Vector< bool > &)
Vector< double > perform_minimum_model_evaluation (const Vector< bool > &)
Vector< double > perform_maximum_model_evaluation (const Vector< bool > &)
Vector< double > perform_mean_model_evaluation (const Vector< bool > &)
Vector< double > get_final_losss (const TrainingStrategy::Results &) const
Vector< double > perform_model_evaluation (const Vector< bool > &)
Vector< double > get_parameters_inputs (const Vector< bool > &) const
void delete_selection_history (void)
void delete_loss_history (void)
void delete_parameters_history (void)
void check (void) const
size_t get_input_index (const Vector< Variables::Use >, const size_t)

Private Attributes

size_t minimum_inputs_number
size_t maximum_selection_failures

Additional Inherited Members

- Public Types inherited from OpenNN::InputsSelectionAlgorithm
enum  PerformanceCalculationMethod { Minimum, Maximum, Mean }
enum  StoppingCondition {
  MaximumTime, SelectionLossGoal, MaximumInputs, MinimumInputs,
  MaximumIterations, MaximumSelectionFailures, CorrelationGoal, AlgorithmFinished
- Protected Attributes inherited from OpenNN::InputsSelectionAlgorithm
bool approximation
Vector< Vector< bool > > inputs_history
Vector< double > selection_loss_history
Vector< double > loss_history
Vector< Vector< double > > parameters_history
size_t trials_number
PerformanceCalculationMethod loss_calculation_method
bool reserve_parameters_data
bool reserve_loss_data
bool reserve_selection_loss_data
bool reserve_minimal_parameters
bool display
double selection_loss_goal
size_t maximum_iterations_number
double maximum_correlation
double minimum_correlation
double maximum_time
double tolerance

Detailed Description

This concrete class represents a pruning algorithm for the inputs selection of a neural network.

Definition at line 47 of file pruning_inputs.h.

Constructor & Destructor Documentation

◆ PruningInputs() [1/3]

OpenNN::PruningInputs::PruningInputs ( TrainingStrategy new_training_strategy_pointer)

Training strategy constructor.

new_training_strategy_pointerPointer to a training strategy object.

Definition at line 36 of file pruning_inputs.cpp.

◆ PruningInputs() [2/3]

OpenNN::PruningInputs::PruningInputs ( const tinyxml2::XMLDocument &  pruning_inputs_document)

XML constructor.

pruning_inputs_documentPointer to a TinyXML document containing the pruning inputs data.

Definition at line 60 of file pruning_inputs.cpp.

◆ PruningInputs() [3/3]

OpenNN::PruningInputs::PruningInputs ( const std::string &  file_name)

File constructor.

file_nameName of XML pruning inputs file.

Definition at line 48 of file pruning_inputs.cpp.

Member Function Documentation

◆ from_XML()

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

Deserializes a TinyXML document into this pruning inputs object.

documentTinyXML document containing the member data.

Definition at line 1055 of file pruning_inputs.cpp.

◆ load()

void OpenNN::PruningInputs::load ( const std::string &  file_name)

Loads a pruning inputs object from a XML-type file.

file_nameName of pruning inputs XML-type file.

Definition at line 1396 of file pruning_inputs.cpp.

◆ save()

void OpenNN::PruningInputs::save ( const std::string &  file_name) const

Saves to a XML-type file the members of the pruning inputs object.

file_nameName of pruning inputs XML-type file.

Definition at line 1381 of file pruning_inputs.cpp.

◆ set_maximum_selection_failures()

void OpenNN::PruningInputs::set_maximum_selection_failures ( const size_t &  new_maximum_loss_failures)

Sets the maximum selection failures for the pruning inputs algorithm.

new_maximum_loss_failuresMaximum number of selection failures in the pruning inputs algorithm.

Definition at line 147 of file pruning_inputs.cpp.

◆ set_minimum_inputs_number()

void OpenNN::PruningInputs::set_minimum_inputs_number ( const size_t &  new_minimum_inputs_number)

Sets the minimum inputs for the pruning inputs algorithm.

new_minimum_inputs_numberMinimum number of inputs in the pruning inputs algorithm.

Definition at line 122 of file pruning_inputs.cpp.

◆ to_XML()

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

Prints to the screen the pruning inputs parameters, the stopping criteria and other user stuff concerning the pruning inputs object.

Definition at line 705 of file pruning_inputs.cpp.

◆ write_XML()

void OpenNN::PruningInputs::write_XML ( tinyxml2::XMLPrinter &  file_stream) const

Serializes the pruning inputs object into a XML document of the TinyXML library without keep the DOM tree in memory. See the OpenNN manual for more information about the format of this document.

Definition at line 918 of file pruning_inputs.cpp.

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