Logo ROOT  
Reference Guide
 
Loading...
Searching...
No Matches
TMVA::DNN::ClassificationSettings Class Reference

Settings for classificationused to distinguish between different function signatures.

contains additional settings if the DNN problem is classification

Definition at line 894 of file NeuralNet.h.

Public Member Functions

 ClassificationSettings (TString name, size_t _convergenceSteps=15, size_t _batchSize=10, size_t _testRepetitions=7, double _factorWeightDecay=1e-5, EnumRegularization _regularization=EnumRegularization::NONE, size_t _scaleToNumEvents=0, MinimizerType _eMinimizerType=MinimizerType::fSteepest, double _learningRate=1e-5, double _momentum=0.3, int _repetitions=3, bool _useMultithreading=true)
 c'tor
 
virtual ~ClassificationSettings ()
 d'tor
 
virtual void endTestCycle ()
 action to be done when the training cycle is ended (e.g.
 
void endTrainCycle (double)
 action to be done when the training cycle is ended (e.g.
 
void setResultComputation (std::string _fileNameNetConfig, std::string _fileNameResult, std::vector< Pattern > *_resultPatternContainer)
 preparation for monitoring output
 
void setWeightSums (double sumOfSigWeights, double sumOfBkgWeights)
 set the weight sums to be scaled to (preparations for monitoring output)
 
virtual void startTestCycle ()
 action to be done when the test cycle is started (e.g.
 
void startTrainCycle ()
 action to be done when the training cycle is started (e.g.
 
void testIteration ()
 callback for monitoring and loggging
 
void testSample (double error, double output, double target, double weight)
 action to be done after the computation of a test sample (e.g.
 
- Public Member Functions inherited from TMVA::DNN::Settings
 Settings (TString name, size_t _convergenceSteps=15, size_t _batchSize=10, size_t _testRepetitions=7, double _factorWeightDecay=1e-5, TMVA::DNN::EnumRegularization _regularization=TMVA::DNN::EnumRegularization::NONE, MinimizerType _eMinimizerType=MinimizerType::fSteepest, double _learningRate=1e-5, double _momentum=0.3, int _repetitions=3, bool _multithreading=true)
 c'tor
 
virtual ~Settings ()
 d'tor
 
void addPoint (std::string histoName, double x)
 for monitoring
 
void addPoint (std::string histoName, double x, double y)
 for monitoring
 
size_t batchSize () const
 mini-batch size
 
void clear (std::string histoName)
 for monitoring
 
virtual void computeResult (const Net &, std::vector< double > &)
 callback for monitoring and logging
 
size_t convergenceCount () const
 returns the current convergence count
 
size_t convergenceSteps () const
 how many steps until training is deemed to have converged
 
void create (std::string histoName, int bins, double min, double max)
 for monitoring
 
void create (std::string histoName, int bins, double min, double max, int bins2, double min2, double max2)
 for monitoring
 
virtual void cycle (double progress, TString text)
 
virtual void drawSample (const std::vector< double > &, const std::vector< double > &, const std::vector< double > &, double)
 callback for monitoring and logging
 
const std::vector< double > & dropFractions () const
 
size_t dropRepetitions () const
 
bool exists (std::string histoName)
 for monitoring
 
double factorWeightDecay () const
 get the weight-decay factor
 
virtual bool hasConverged (double testError)
 has this training converged already?
 
double learningRate () const
 get the learning rate
 
size_t maxConvergenceCount () const
 returns the max convergence count so far
 
size_t minError () const
 returns the smallest error so far
 
MinimizerType minimizerType () const
 which minimizer shall be used (e.g. SGD)
 
double momentum () const
 get the momentum (e.g. for SGD)
 
void pads (int numPads)
 preparation for monitoring
 
void plot (std::string histoName, std::string options, int pad, EColor color)
 for monitoring
 
EnumRegularization regularization () const
 some regularization of the DNN is turned on?
 
int repetitions () const
 how many steps have to be gone until the batch is changed
 
template<typename Iterator >
void setDropOut (Iterator begin, Iterator end, size_t _dropRepetitions)
 set the drop-out configuration (layer-wise)
 
void setMonitoring (std::shared_ptr< Monitoring > ptrMonitoring)
 prepared for monitoring
 
virtual void setProgressLimits (double minProgress=0, double maxProgress=100)
 
virtual void startTraining ()
 
size_t testRepetitions () const
 how often is the test data tested
 
bool useMultithreading () const
 is multithreading turned on?
 

Public Attributes

std::vector< doublem_ams
 
double m_cutValue
 
std::string m_fileNameNetConfig
 
std::string m_fileNameResult
 
std::vector< doublem_input
 
std::vector< doublem_output
 
std::vector< Pattern > * m_pResultPatternContainer
 
size_t m_scaleToNumEvents
 
std::vector< doublem_significances
 
double m_sumOfBkgWeights
 
double m_sumOfSigWeights
 
std::vector< doublem_targets
 
std::vector< doublem_weights
 
- Public Attributes inherited from TMVA::DNN::Settings
size_t count_dE
 
size_t count_E
 
size_t count_mb_dE
 
size_t count_mb_E
 
double fLearningRate
 
MinimizerType fMinimizerType
 
double fMomentum
 
int fRepetitions
 
size_t m_batchSize
 mini-batch size
 
size_t m_convergenceCount
 
size_t m_convergenceSteps
 number of steps without improvement to consider the DNN to have converged
 
std::vector< doublem_dropOut
 
double m_dropRepetitions
 
double m_factorWeightDecay
 
size_t m_maxConvergenceCount
 
double m_maxProgress
 current limits for the progress bar
 
double m_minError
 
double m_minProgress
 current limits for the progress bar
 
EnumRegularization m_regularization
 
size_t m_testRepetitions
 
Timer m_timer
 timer for monitoring
 

Additional Inherited Members

- Protected Attributes inherited from TMVA::DNN::Settings
std::shared_ptr< MonitoringfMonitoring
 
bool m_useMultithreading
 

#include <TMVA/NeuralNet.h>

Inheritance diagram for TMVA::DNN::ClassificationSettings:
[legend]

Constructor & Destructor Documentation

◆ ClassificationSettings()

TMVA::DNN::ClassificationSettings::ClassificationSettings ( TString  name,
size_t  _convergenceSteps = 15,
size_t  _batchSize = 10,
size_t  _testRepetitions = 7,
double  _factorWeightDecay = 1e-5,
EnumRegularization  _regularization = EnumRegularization::NONE,
size_t  _scaleToNumEvents = 0,
MinimizerType  _eMinimizerType = MinimizerType::fSteepest,
double  _learningRate = 1e-5,
double  _momentum = 0.3,
int  _repetitions = 3,
bool  _useMultithreading = true 
)
inline

c'tor

Definition at line 901 of file NeuralNet.h.

◆ ~ClassificationSettings()

virtual TMVA::DNN::ClassificationSettings::~ClassificationSettings ( )
inlinevirtual

d'tor

Definition at line 924 of file NeuralNet.h.

Member Function Documentation

◆ endTestCycle()

void TMVA::DNN::ClassificationSettings::endTestCycle ( )
virtual

action to be done when the training cycle is ended (e.g.

update some monitoring output)

Reimplemented from TMVA::DNN::Settings.

Definition at line 326 of file NeuralNet.cxx.

◆ endTrainCycle()

void TMVA::DNN::ClassificationSettings::endTrainCycle ( double  )
virtual

action to be done when the training cycle is ended (e.g.

update some monitoring output)

Reimplemented from TMVA::DNN::Settings.

Definition at line 296 of file NeuralNet.cxx.

◆ setResultComputation()

void TMVA::DNN::ClassificationSettings::setResultComputation ( std::string  _fileNameNetConfig,
std::string  _fileNameResult,
std::vector< Pattern > *  _resultPatternContainer 
)

preparation for monitoring output

Definition at line 520 of file NeuralNet.cxx.

◆ setWeightSums()

void TMVA::DNN::ClassificationSettings::setWeightSums ( double  sumOfSigWeights,
double  sumOfBkgWeights 
)

set the weight sums to be scaled to (preparations for monitoring output)

Definition at line 512 of file NeuralNet.cxx.

◆ startTestCycle()

void TMVA::DNN::ClassificationSettings::startTestCycle ( )
virtual

action to be done when the test cycle is started (e.g.

update some monitoring output)

Reimplemented from TMVA::DNN::Settings.

Definition at line 316 of file NeuralNet.cxx.

◆ startTrainCycle()

void TMVA::DNN::ClassificationSettings::startTrainCycle ( )
virtual

action to be done when the training cycle is started (e.g.

update some monitoring output)

Reimplemented from TMVA::DNN::Settings.

Definition at line 281 of file NeuralNet.cxx.

◆ testIteration()

void TMVA::DNN::ClassificationSettings::testIteration ( )
inlinevirtual

callback for monitoring and loggging

Reimplemented from TMVA::DNN::Settings.

Definition at line 930 of file NeuralNet.h.

◆ testSample()

void TMVA::DNN::ClassificationSettings::testSample ( double  error,
double  output,
double  target,
double  weight 
)
virtual

action to be done after the computation of a test sample (e.g.

update some monitoring output)

Reimplemented from TMVA::DNN::Settings.

Definition at line 304 of file NeuralNet.cxx.

Member Data Documentation

◆ m_ams

std::vector<double> TMVA::DNN::ClassificationSettings::m_ams

Definition at line 1000 of file NeuralNet.h.

◆ m_cutValue

double TMVA::DNN::ClassificationSettings::m_cutValue

Definition at line 1008 of file NeuralNet.h.

◆ m_fileNameNetConfig

std::string TMVA::DNN::ClassificationSettings::m_fileNameNetConfig

Definition at line 1011 of file NeuralNet.h.

◆ m_fileNameResult

std::string TMVA::DNN::ClassificationSettings::m_fileNameResult

Definition at line 1010 of file NeuralNet.h.

◆ m_input

std::vector<double> TMVA::DNN::ClassificationSettings::m_input

Definition at line 995 of file NeuralNet.h.

◆ m_output

std::vector<double> TMVA::DNN::ClassificationSettings::m_output

Definition at line 996 of file NeuralNet.h.

◆ m_pResultPatternContainer

std::vector<Pattern>* TMVA::DNN::ClassificationSettings::m_pResultPatternContainer

Definition at line 1009 of file NeuralNet.h.

◆ m_scaleToNumEvents

size_t TMVA::DNN::ClassificationSettings::m_scaleToNumEvents

Definition at line 1006 of file NeuralNet.h.

◆ m_significances

std::vector<double> TMVA::DNN::ClassificationSettings::m_significances

Definition at line 1001 of file NeuralNet.h.

◆ m_sumOfBkgWeights

double TMVA::DNN::ClassificationSettings::m_sumOfBkgWeights

Definition at line 1005 of file NeuralNet.h.

◆ m_sumOfSigWeights

double TMVA::DNN::ClassificationSettings::m_sumOfSigWeights

Definition at line 1004 of file NeuralNet.h.

◆ m_targets

std::vector<double> TMVA::DNN::ClassificationSettings::m_targets

Definition at line 997 of file NeuralNet.h.

◆ m_weights

std::vector<double> TMVA::DNN::ClassificationSettings::m_weights

Definition at line 998 of file NeuralNet.h.

Libraries for TMVA::DNN::ClassificationSettings:

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