Neuron Net
Public Member Functions | Static Public Attributes | Protected Member Functions | List of all members
CNet Class Reference

The main class of the neural network. Contains basic methods for the functioning of a neural network. More...

Public Member Functions

 CNet (CArrayObj *Description)
 Constructor. More...
 
 ~CNet (void)
 Destructor. More...
 
bool feedForward (CArrayDouble *inputVals)
 Feed Forward method. More...
 
void backProp (CArrayDouble *targetVals)
 Back propagation method. More...
 
void getResults (CArrayDouble *&resultVals)
 Method to get results of feed forward process. More...
 
double getRecentAverageError ()
 Method to check quality of study. More...
 
bool Save (string file_name, double error, double undefine, double forecast, datetime time, bool common=true)
 Save method. More...
 
bool Load (string file_name, double &error, double &undefine, double &forecast, datetime &time, bool common=true)
 Load method. More...
 
virtual int Type (void) const
 Identificator of class. More...
 

Static Public Attributes

static double recentAverageSmoothingFactor =10000.0
 Smoothing factor of average error. More...
 

Protected Member Functions

void backPropOCL (CArrayDouble *targetVals)
 Back propagation method for GPU calculation. More...
 

Detailed Description

The main class of the neural network. Contains basic methods for the functioning of a neural network.

Detailed description on the link.

Definition at line 1353 of file NeuroNet.mqh.

Constructor & Destructor Documentation

◆ CNet()

CNet::CNet ( CArrayObj *  Description)

Constructor.

Definition at line 1384 of file NeuroNet.mqh.

◆ ~CNet()

CNet::~CNet ( void  )

Destructor.

Definition at line 2755 of file NeuroNet.mqh.

Member Function Documentation

◆ backProp()

void CNet::backProp ( CArrayDouble *  targetVals)

Back propagation method.

Parameters
[in]targetValsTarget values

Definition at line 1738 of file NeuroNet.mqh.

◆ backPropOCL()

void CNet::backPropOCL ( CArrayDouble *  targetVals)
protected

Back propagation method for GPU calculation.

Parameters
[in]targetValsTarget values

Definition at line 1846 of file NeuroNet.mqh.

◆ feedForward()

bool CNet::feedForward ( CArrayDouble *  inputVals)

Feed Forward method.

Parameters
[in]prevLayerPointer to previos layer.

Definition at line 1633 of file NeuroNet.mqh.

◆ getRecentAverageError()

double CNet::getRecentAverageError ( )
inline

Method to check quality of study.

Returns
Average error

Definition at line 1363 of file NeuroNet.mqh.

◆ getResults()

void CNet::getResults ( CArrayDouble *&  resultVals)

Method to get results of feed forward process.

Parameters
[out]resultValsArray of result values

Definition at line 1897 of file NeuroNet.mqh.

◆ Load()

bool CNet::Load ( string  file_name,
double &  error,
double &  undefine,
double &  forecast,
datetime &  time,
bool  common = true 
)

Load method.

Parameters
[in]file_nameFile name to save
[out]errorAverage error
[out]undefineUndefined percent
[out]Foecastpercent
[out]timeLast study time
[in]commonCommon flag

Definition at line 1975 of file NeuroNet.mqh.

◆ Save()

bool CNet::Save ( string  file_name,
double  error,
double  undefine,
double  forecast,
datetime  time,
bool  common = true 
)

Save method.

Parameters
[in]file_nameFile name to save
[in]errorAverage error
[in]undefineUndefined percent
[in]Foecastpercent
[in]timeLast study time
[in]commonCommon flag

Definition at line 1950 of file NeuroNet.mqh.

◆ Type()

virtual int CNet::Type ( void  ) const
inlinevirtual

Identificator of class.

Returns
Type of class

Definition at line 1370 of file NeuroNet.mqh.

Member Data Documentation

◆ recentAverageSmoothingFactor

double CNet::recentAverageSmoothingFactor =10000.0
static

Smoothing factor of average error.

Definition at line 1369 of file NeuroNet.mqh.


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