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
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
void	endTestCycle () override
	action to be done when the training cycle is ended (e.g.
void	endTrainCycle (double) override
	action to be done when the training cycle is ended (e.g.
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)
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)
void	startTestCycle () override
	action to be done when the test cycle is started (e.g.
void	startTrainCycle () override
	action to be done when the training cycle is started (e.g.
virtual void	startTraining ()
void	testIteration () override
	callback for monitoring and loggging
size_t	testRepetitions () const
	how often is the test data tested
void	testSample (double error, double output, double target, double weight) override
	action to be done after the computation of a test sample (e.g.
bool	useMultithreading () const
	is multithreading turned on?

Public Attributes
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
std::vector< double >	m_ams
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
double	m_cutValue
std::vector< double >	m_dropOut
double	m_dropRepetitions
double	m_factorWeightDecay
std::string	m_fileNameNetConfig
std::string	m_fileNameResult
std::vector< double >	m_input
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
std::vector< double >	m_output
std::vector< Pattern > *	m_pResultPatternContainer
EnumRegularization	m_regularization
size_t	m_scaleToNumEvents
std::vector< double >	m_significances
double	m_sumOfBkgWeights
double	m_sumOfSigWeights
std::vector< double >	m_targets
size_t	m_testRepetitions
Timer	m_timer
	timer for monitoring
std::vector< double >	m_weights

Protected Attributes
std::shared_ptr< Monitoring >	fMonitoring
bool	m_useMultithreading

#include <TMVA/NeuralNet.h>

Inheritance diagram for TMVA::DNN::ClassificationSettings:

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

◆ addPoint() [1/2]

void TMVA::DNN::Settings::addPoint	(	std::string	histoName,
		double	x )

inlineinherited

for monitoring

Definition at line 821 of file NeuralNet.h.

◆ addPoint() [2/2]

void TMVA::DNN::Settings::addPoint	(	std::string	histoName,
		double	x,
		double	y )

inlineinherited

for monitoring

Definition at line 822 of file NeuralNet.h.

◆ batchSize()

size_t TMVA::DNN::Settings::batchSize ( ) const

inlineinherited

mini-batch size

Definition at line 767 of file NeuralNet.h.

◆ clear()

void TMVA::DNN::Settings::clear ( std::string histoName )

inlineinherited

for monitoring

Definition at line 824 of file NeuralNet.h.

◆ computeResult()

virtual void TMVA::DNN::Settings::computeResult	(	const Net &	,
		std::vector< double > &	)

inlinevirtualinherited

callback for monitoring and logging

Definition at line 809 of file NeuralNet.h.

◆ convergenceCount()

size_t TMVA::DNN::Settings::convergenceCount ( ) const

inlineinherited

returns the current convergence count

Definition at line 827 of file NeuralNet.h.

◆ convergenceSteps()

size_t TMVA::DNN::Settings::convergenceSteps ( ) const

inlineinherited

how many steps until training is deemed to have converged

Definition at line 766 of file NeuralNet.h.

◆ create() [1/2]

void TMVA::DNN::Settings::create	(	std::string	histoName,
		int	bins,
		double	min,
		double	max )

inlineinherited

for monitoring

Definition at line 819 of file NeuralNet.h.

◆ create() [2/2]

void TMVA::DNN::Settings::create	(	std::string	histoName,
		int	bins,
		double	min,
		double	max,
		int	bins2,
		double	min2,
		double	max2 )

inlineinherited

for monitoring

Definition at line 820 of file NeuralNet.h.

◆ cycle()

virtual void TMVA::DNN::Settings::cycle	(	double	progress,
		TString	text )

inlinevirtualinherited

Parameters

text	advance on the progress bar
progress	the new value
text	a label

Definition at line 799 of file NeuralNet.h.

◆ drawSample()

virtual void TMVA::DNN::Settings::drawSample	(	const std::vector< double > &	,
		const std::vector< double > &	,
		const std::vector< double > &	,
		double	)

inlinevirtualinherited

callback for monitoring and logging

Definition at line 807 of file NeuralNet.h.

◆ dropFractions()

const std::vector< double > & TMVA::DNN::Settings::dropFractions ( ) const

inlineinherited

Definition at line 762 of file NeuralNet.h.

◆ dropRepetitions()

size_t TMVA::DNN::Settings::dropRepetitions ( ) const

inlineinherited

Definition at line 761 of file NeuralNet.h.

◆ endTestCycle()

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

overridevirtual

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 )

overridevirtual

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.

◆ exists()

bool TMVA::DNN::Settings::exists ( std::string histoName )

inlineinherited

for monitoring

Definition at line 825 of file NeuralNet.h.

◆ factorWeightDecay()

double TMVA::DNN::Settings::factorWeightDecay ( ) const

inlineinherited

get the weight-decay factor

Definition at line 769 of file NeuralNet.h.

◆ hasConverged()

bool TMVA::DNN::Settings::hasConverged ( double testError )

virtualinherited

has this training converged already?

check for convergence

Definition at line 485 of file NeuralNet.cxx.

◆ learningRate()

double TMVA::DNN::Settings::learningRate ( ) const

inlineinherited

get the learning rate

Definition at line 771 of file NeuralNet.h.

◆ maxConvergenceCount()

size_t TMVA::DNN::Settings::maxConvergenceCount ( ) const

inlineinherited

returns the max convergence count so far

Definition at line 828 of file NeuralNet.h.

◆ minError()

size_t TMVA::DNN::Settings::minError ( ) const

inlineinherited

returns the smallest error so far

Definition at line 829 of file NeuralNet.h.

◆ minimizerType()

MinimizerType TMVA::DNN::Settings::minimizerType ( ) const

inlineinherited

which minimizer shall be used (e.g. SGD)

Definition at line 774 of file NeuralNet.h.

◆ momentum()

double TMVA::DNN::Settings::momentum ( ) const

inlineinherited

get the momentum (e.g. for SGD)

Definition at line 772 of file NeuralNet.h.

◆ pads()

void TMVA::DNN::Settings::pads ( int numPads )

inlineinherited

preparation for monitoring

Definition at line 818 of file NeuralNet.h.

◆ plot()

void TMVA::DNN::Settings::plot	(	std::string	histoName,
		std::string	options,
		int	pad,
		EColor	color )

inlineinherited

for monitoring

Definition at line 823 of file NeuralNet.h.

◆ regularization()

EnumRegularization TMVA::DNN::Settings::regularization ( ) const

inlineinherited

some regularization of the DNN is turned on?

Definition at line 813 of file NeuralNet.h.

◆ repetitions()

int TMVA::DNN::Settings::repetitions ( ) const

inlineinherited

how many steps have to be gone until the batch is changed

Definition at line 773 of file NeuralNet.h.

◆ setDropOut()

template<typename Iterator>

void TMVA::DNN::Settings::setDropOut	(	Iterator	begin,
		Iterator	end,
		size_t	_dropRepetitions )

inlineinherited

set the drop-out configuration (layer-wise)

Parameters

begin	begin of an array or vector denoting the drop-out probabilities for each layer
end	end of an array or vector denoting the drop-out probabilities for each layer
_dropRepetitions	denotes after how many repetitions the drop-out setting (which nodes are dropped out exactly) is changed

Definition at line 759 of file NeuralNet.h.

◆ setMonitoring()

void TMVA::DNN::Settings::setMonitoring ( std::shared_ptr< Monitoring > ptrMonitoring )

inlineinherited

prepared for monitoring

Definition at line 764 of file NeuralNet.h.

◆ setProgressLimits()

virtual void TMVA::DNN::Settings::setProgressLimits	(	double	minProgress = 0,
		double	maxProgress = 100 )

inlinevirtualinherited

Parameters

maxProgress	for monitoring and logging (set the current "progress" limits for the display of the progress)
minProgress	minimum value
maxProgress	maximum value

Definition at line 790 of file NeuralNet.h.

◆ 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 ( )

overridevirtual

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 ( )

overridevirtual

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.

◆ startTraining()

virtual void TMVA::DNN::Settings::startTraining ( )

inlinevirtualinherited

Definition at line 795 of file NeuralNet.h.

◆ testIteration()

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

inlineoverridevirtual

callback for monitoring and loggging

Reimplemented from TMVA::DNN::Settings.

Definition at line 930 of file NeuralNet.h.

◆ testRepetitions()

size_t TMVA::DNN::Settings::testRepetitions ( ) const

inlineinherited

how often is the test data tested

Definition at line 768 of file NeuralNet.h.

◆ testSample()

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

overridevirtual

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.

◆ useMultithreading()

bool TMVA::DNN::Settings::useMultithreading ( ) const

inlineinherited

is multithreading turned on?

Definition at line 815 of file NeuralNet.h.

Member Data Documentation

◆ count_dE

size_t TMVA::DNN::Settings::count_dE

inherited

Definition at line 843 of file NeuralNet.h.

◆ count_E

size_t TMVA::DNN::Settings::count_E

inherited

Definition at line 842 of file NeuralNet.h.

◆ count_mb_dE

size_t TMVA::DNN::Settings::count_mb_dE

inherited

Definition at line 845 of file NeuralNet.h.

◆ count_mb_E

size_t TMVA::DNN::Settings::count_mb_E

inherited

Definition at line 844 of file NeuralNet.h.

◆ fLearningRate

double TMVA::DNN::Settings::fLearningRate

inherited

Definition at line 852 of file NeuralNet.h.

◆ fMinimizerType

MinimizerType TMVA::DNN::Settings::fMinimizerType

inherited

Definition at line 855 of file NeuralNet.h.

◆ fMomentum

double TMVA::DNN::Settings::fMomentum

inherited

Definition at line 853 of file NeuralNet.h.

◆ fMonitoring

std::shared_ptr<Monitoring> TMVA::DNN::Settings::fMonitoring

protectedinherited

Definition at line 865 of file NeuralNet.h.

◆ fRepetitions

int TMVA::DNN::Settings::fRepetitions

inherited

Definition at line 854 of file NeuralNet.h.

◆ m_ams

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

Definition at line 1000 of file NeuralNet.h.

◆ m_batchSize

size_t TMVA::DNN::Settings::m_batchSize

inherited

mini-batch size

Definition at line 838 of file NeuralNet.h.

◆ m_convergenceCount

size_t TMVA::DNN::Settings::m_convergenceCount

inherited

Definition at line 857 of file NeuralNet.h.

◆ m_convergenceSteps

size_t TMVA::DNN::Settings::m_convergenceSteps

inherited

number of steps without improvement to consider the DNN to have converged

Definition at line 837 of file NeuralNet.h.

◆ m_cutValue

double TMVA::DNN::ClassificationSettings::m_cutValue

Definition at line 1008 of file NeuralNet.h.

◆ m_dropOut

std::vector<double> TMVA::DNN::Settings::m_dropOut

inherited

Definition at line 850 of file NeuralNet.h.

◆ m_dropRepetitions

double TMVA::DNN::Settings::m_dropRepetitions

inherited

Definition at line 849 of file NeuralNet.h.

◆ m_factorWeightDecay

double TMVA::DNN::Settings::m_factorWeightDecay

inherited

Definition at line 840 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_maxConvergenceCount

size_t TMVA::DNN::Settings::m_maxConvergenceCount

inherited

Definition at line 858 of file NeuralNet.h.

◆ m_maxProgress

double TMVA::DNN::Settings::m_maxProgress

inherited

current limits for the progress bar

Definition at line 834 of file NeuralNet.h.

◆ m_minError

double TMVA::DNN::Settings::m_minError

inherited

Definition at line 859 of file NeuralNet.h.

◆ m_minProgress

double TMVA::DNN::Settings::m_minProgress

inherited

current limits for the progress bar

Definition at line 833 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_regularization

EnumRegularization TMVA::DNN::Settings::m_regularization

inherited

Definition at line 847 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_testRepetitions

size_t TMVA::DNN::Settings::m_testRepetitions

inherited

Definition at line 839 of file NeuralNet.h.

◆ m_timer

Timer TMVA::DNN::Settings::m_timer

inherited

timer for monitoring

Definition at line 832 of file NeuralNet.h.

◆ m_useMultithreading

bool TMVA::DNN::Settings::m_useMultithreading

protectedinherited

Definition at line 863 of file NeuralNet.h.

◆ m_weights

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

Definition at line 998 of file NeuralNet.h.

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

tmva/tmva/inc/TMVA/NeuralNet.h
tmva/tmva/src/NeuralNet.cxx

Public Member Functions

Public Attributes

Protected Attributes

Constructor & Destructor Documentation

◆ ClassificationSettings()

◆ ~ClassificationSettings()

Member Function Documentation

◆ addPoint() [1/2]

◆ addPoint() [2/2]

◆ batchSize()

◆ clear()

◆ computeResult()

◆ convergenceCount()

◆ convergenceSteps()

◆ create() [1/2]

◆ create() [2/2]

◆ cycle()

◆ drawSample()

◆ dropFractions()

◆ dropRepetitions()

◆ endTestCycle()

◆ endTrainCycle()

◆ exists()

◆ factorWeightDecay()

◆ hasConverged()

◆ learningRate()

◆ maxConvergenceCount()

◆ minError()

◆ minimizerType()

◆ momentum()

◆ pads()

◆ plot()

◆ regularization()

◆ repetitions()

◆ setDropOut()

◆ setMonitoring()

◆ setProgressLimits()

◆ setResultComputation()

◆ setWeightSums()

◆ startTestCycle()

◆ startTrainCycle()

◆ startTraining()

◆ testIteration()

◆ testRepetitions()

◆ testSample()

◆ useMultithreading()

Member Data Documentation

◆ count_dE

◆ count_E

◆ count_mb_dE

◆ count_mb_E

◆ fLearningRate

◆ fMinimizerType

◆ fMomentum

◆ fMonitoring

◆ fRepetitions

◆ m_ams

◆ m_batchSize

◆ m_convergenceCount

◆ m_convergenceSteps

◆ m_cutValue

◆ m_dropOut

◆ m_dropRepetitions

◆ m_factorWeightDecay

◆ m_fileNameNetConfig

◆ m_fileNameResult

◆ m_input

◆ m_maxConvergenceCount

◆ m_maxProgress

◆ m_minError

◆ m_minProgress

◆ m_output

◆ m_pResultPatternContainer

◆ m_regularization

◆ m_scaleToNumEvents

◆ m_significances

◆ m_sumOfBkgWeights

◆ m_sumOfSigWeights

◆ m_targets

◆ m_testRepetitions