Direct Graphical Models  v.1.7.0
DirectGraphicalModels::TrainNodeCvANNParams Struct Reference

OpenCV Artificial neural network parameters. More...

#include <TrainNodeCvANN.h>

Public Member Functions

 TrainNodeCvANNParams ()
 
 TrainNodeCvANNParams (word _numLayers, double _weightScale, double _momentumScale, int _maxCount, double _epsilon, int _term_criteria_type, int _maxSamples)
 

Public Attributes

word numLayers
 Number of layers of neurons. More...
 
double weightScale
 Strength of the weight gradient term. The recommended value is about 0.1. Default value is 0.1. More...
 
double momentumScale
 Strength of the momentum term (the difference between weights on the 2 previous iterations). This parameter provides some inertia to smooth the random fluctuations of the weights. It can vary from 0 (the feature is disabled) to 1 and beyond. The value 0.1 or so is good enough. Default value is 0.1. More...
 
int maxCount
 The maximum number of iterations (time / accuracy) More...
 
double epsilon
 The desired accuracy or change in parameters at which the iterative algorithm stops. More...
 
int term_criteria_type
 Termination cirteria type (according the the two previous parameters) More...
 
size_t maxSamples
 Maximum number of samples to be used in training. 0 means using all the samples. More...
 

Detailed Description

OpenCV Artificial neural network parameters.

Definition at line 12 of file TrainNodeCvANN.h.

Constructor & Destructor Documentation

◆ TrainNodeCvANNParams() [1/2]

DirectGraphicalModels::TrainNodeCvANNParams::TrainNodeCvANNParams ( )
inline

Definition at line 21 of file TrainNodeCvANN.h.

◆ TrainNodeCvANNParams() [2/2]

DirectGraphicalModels::TrainNodeCvANNParams::TrainNodeCvANNParams ( word  _numLayers,
double  _weightScale,
double  _momentumScale,
int  _maxCount,
double  _epsilon,
int  _term_criteria_type,
int  _maxSamples 
)
inline

Definition at line 22 of file TrainNodeCvANN.h.

Member Data Documentation

◆ epsilon

double DirectGraphicalModels::TrainNodeCvANNParams::epsilon

The desired accuracy or change in parameters at which the iterative algorithm stops.

Definition at line 17 of file TrainNodeCvANN.h.

◆ maxCount

int DirectGraphicalModels::TrainNodeCvANNParams::maxCount

The maximum number of iterations (time / accuracy)

Definition at line 16 of file TrainNodeCvANN.h.

◆ maxSamples

size_t DirectGraphicalModels::TrainNodeCvANNParams::maxSamples

Maximum number of samples to be used in training. 0 means using all the samples.

Definition at line 19 of file TrainNodeCvANN.h.

◆ momentumScale

double DirectGraphicalModels::TrainNodeCvANNParams::momentumScale

Strength of the momentum term (the difference between weights on the 2 previous iterations). This parameter provides some inertia to smooth the random fluctuations of the weights. It can vary from 0 (the feature is disabled) to 1 and beyond. The value 0.1 or so is good enough. Default value is 0.1.

Definition at line 15 of file TrainNodeCvANN.h.

◆ numLayers

word DirectGraphicalModels::TrainNodeCvANNParams::numLayers

Number of layers of neurons.

Definition at line 13 of file TrainNodeCvANN.h.

◆ term_criteria_type

int DirectGraphicalModels::TrainNodeCvANNParams::term_criteria_type

Termination cirteria type (according the the two previous parameters)

Definition at line 18 of file TrainNodeCvANN.h.

◆ weightScale

double DirectGraphicalModels::TrainNodeCvANNParams::weightScale

Strength of the weight gradient term. The recommended value is about 0.1. Default value is 0.1.

Definition at line 14 of file TrainNodeCvANN.h.


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