TrainingResults Struct Reference

This structure contains the optimization algorithm results. More...

#include <optimization_algorithm.h>

Public Member Functions

 TrainingResults ()
 Default constructor. More...
 
 TrainingResults (const Index &epochs_number)
 
virtual ~TrainingResults ()
 Destructor. More...
 
string write_stopping_condition () const
 Return a string with the stopping condition of the Results. More...
 
type get_training_error ()
 
type get_selection_error ()
 
void save (const string &) const
 Returns a string representation of the results structure. More...
 
void print (const string &message=string())
 
Tensor< string, 2 > write_final_results (const Index &=3) const
 Writes final results of the training. More...
 
void resize_training_error_history (const Index &)
 Resizes the training error history keeping the values. More...
 
void resize_selection_error_history (const Index &)
 Resizes the selection error history keeping the values. More...
 

Public Attributes

OptimizationAlgorithm::StoppingCondition stopping_condition = OptimizationAlgorithm::StoppingCondition::MaximumTime
 Stopping condition of the algorithm. More...
 
Tensor< type, 1 > training_error_history
 History of the loss function loss over the training iterations. More...
 
Tensor< type, 1 > selection_error_history
 History of the selection error over the training iterations. More...
 
string elapsed_time
 Elapsed time of the training process. More...
 

Detailed Description

This structure contains the optimization algorithm results.

Definition at line 197 of file optimization_algorithm.h.

Constructor & Destructor Documentation

◆ TrainingResults() [1/2]

TrainingResults ( )
inlineexplicit

Default constructor.

Definition at line 201 of file optimization_algorithm.h.

◆ TrainingResults() [2/2]

TrainingResults ( const Index &  epochs_number)
inlineexplicit

Definition at line 205 of file optimization_algorithm.h.

◆ ~TrainingResults()

virtual ~TrainingResults ( )
inlinevirtual

Destructor.

Definition at line 216 of file optimization_algorithm.h.

Member Function Documentation

◆ get_selection_error()

type get_selection_error ( )
inline

Definition at line 227 of file optimization_algorithm.h.

◆ get_training_error()

type get_training_error ( )
inline

Definition at line 220 of file optimization_algorithm.h.

◆ print()

void print ( const string &  message = string())
inline

Definition at line 238 of file optimization_algorithm.h.

◆ resize_selection_error_history()

void resize_selection_error_history ( const Index &  new_size)

Resizes the selection error history keeping the values.

Resizes the training error history keeping the values.

Parameters
new_sizeSize of training history variables.

Definition at line 461 of file optimization_algorithm.cpp.

◆ resize_training_error_history()

void resize_training_error_history ( const Index &  new_size)

Resizes the training error history keeping the values.

Resizes the training error history keeping the values.

Parameters
new_sizeSize of training history variables.

Definition at line 438 of file optimization_algorithm.cpp.

◆ save()

void save ( const string &  ) const

Returns a string representation of the results structure.

Todo:

Definition at line 528 of file optimization_algorithm.cpp.

◆ write_final_results()

Tensor< string, 2 > write_final_results ( const Index &  precision = 3) const

Writes final results of the training.

Definition at line 534 of file optimization_algorithm.cpp.

◆ write_stopping_condition()

string write_stopping_condition ( ) const

Return a string with the stopping condition of the Results.

Definition at line 411 of file optimization_algorithm.cpp.

Member Data Documentation

◆ elapsed_time

string elapsed_time

Elapsed time of the training process.

Definition at line 283 of file optimization_algorithm.h.

◆ selection_error_history

Tensor<type, 1> selection_error_history

History of the selection error over the training iterations.

Definition at line 279 of file optimization_algorithm.h.

◆ stopping_condition

OptimizationAlgorithm::StoppingCondition stopping_condition = OptimizationAlgorithm::StoppingCondition::MaximumTime

Stopping condition of the algorithm.

Definition at line 257 of file optimization_algorithm.h.

◆ training_error_history

Tensor<type, 1> training_error_history

History of the loss function loss over the training iterations.

Definition at line 275 of file optimization_algorithm.h.


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