Inheritance diagram for OpenNN::GradientDescent::GradientDescentResults:

Public Member Functions
	GradientDescentResults (void)

	GradientDescentResults (GradientDescent *new_gradient_descent_pointer)

virtual	~GradientDescentResults (void)

void	resize_training_history (const size_t &)

std::string	to_string (void) const

Matrix< std::string >	write_final_results (const size_t &precision=3) const

Public Attributes
GradientDescent *	gradient_descent_pointer

Vector< Vector< double > >	parameters_history

Vector< double >	parameters_norm_history

Vector< double >	performance_history

Vector< double >	generalization_performance_history

Vector< Vector< double > >	gradient_history

Vector< double >	gradient_norm_history

Vector< Vector< double > >	training_direction_history

Vector< double >	training_rate_history

Vector< double >	elapsed_time_history

Vector< double >	final_parameters

double	final_parameters_norm

double	final_performance

double	final_generalization_performance

Vector< double >	final_gradient

double	final_gradient_norm

Vector< double >	final_training_direction

double	final_training_rate

double	elapsed_time

size_t	iterations_number

std::string	stopping_criterion

Detailed Description

This structure contains the training results for the gradient descent.

Definition at line 71 of file gradient_descent.h.

Member Function Documentation

void OpenNN::GradientDescent::GradientDescentResults::resize_training_history ( const size_t & new_size )

Resizes the training history variables which are to be reserved by the training algorithm.

Parameters

new_size Size of training history variables.

Definition at line 1146 of file gradient_descent.cpp.

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