OpenNN  2.2
Open Neural Networks Library
Public Member Functions | Public Attributes | List of all members
OpenNN::TrainingRateAlgorithm::Triplet Struct Reference

#include <training_rate_algorithm.h>

Public Member Functions

 Triplet (void)
 
virtual ~Triplet (void)
 
bool operator== (const Triplet &other_triplet) const
 
bool has_length_zero (void) const
 
std::string to_string (void) const
 
void print (void) const
 
void check (void) const
 

Public Attributes

Vector< double > A
 
Vector< double > U
 
Vector< double > B
 

Detailed Description

Defines a set of three points (A, U, B) for bracketing a directional minimum.

Definition at line 73 of file training_rate_algorithm.h.

Member Function Documentation

void OpenNN::TrainingRateAlgorithm::Triplet::check ( void  ) const
inline

Checks that the points A, U and B define a minimum. That is, a < u < b, fa > fu and fu < fb. If some of that conditions is not satisfied, an exception is thrown.

Definition at line 148 of file training_rate_algorithm.h.

bool OpenNN::TrainingRateAlgorithm::Triplet::has_length_zero ( void  ) const
inline

Returns true if the length of the interval (A,B) is zero, and false otherwise.

Definition at line 112 of file training_rate_algorithm.h.

bool OpenNN::TrainingRateAlgorithm::Triplet::operator== ( const Triplet other_triplet) const
inline

Equal to operator. It compares this triplet with another triplet. It returns true if both triplets have the same points A, U and B, and false otherwise. @ param other_triplet Triplet to be compared with.

Definition at line 95 of file training_rate_algorithm.h.


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