This class has been renamed. Please use TOATDesign instead. More...

#include <TOATSamplingMultiNominal.h>

Inheritance diagram for URANIE::Sampler::TOATSamplingMultiNominal:

[legend]

Collaboration diagram for URANIE::Sampler::TOATSamplingMultiNominal:

[legend]

Public Member Functions
Constructor and Destructor
	TOATSamplingMultiNominal (URANIE::DataServer::TDataServer *tds, TString mode="regular", Int_t number=2)
	Constructor of the OAT sampler.

virtual	~TOATSamplingMultiNominal ()
	Default destructor.

Getting and setting parameters
virtual Bool_t	getHasData ()
	Returns TRUE if the internal data server contains data.

virtual Double_t	getNumberOfModifications ()
	Get the number of modifications per attribute and nominal values.

virtual TString	getSamplingMode ()
	Get OAT sampling mode.

virtual TString	getGenerationMode ()
	Get random generation method.

virtual TList *	getChangingAttList ()
	Get the list of changing attributes.

virtual TMap *	getRangeInformation ()
	Get the list mapping the names of the changing attributes and their corresponding range information.

virtual URANIE::DataServer::TDSNtupleD *	getNominalValueTuple ()
	Get the tuple storing the nominal values of the attributes.

virtual void	setOption (TString mode, Int_t number)
	Set the sampling mode and number of modifications.

void	setRange (TString changingAttName, Double_t rangeValue, Bool_t is_percent=kFALSE)
	Set the range of the sampling for a changing attribute.

void	setRange (TString changingAttName, TString rangeAttName, Bool_t is_percent=kFALSE)
	Set the name of the attribute holding the range of the sampling for a changing attribute.

Generation of the sample
virtual void	generateSample (Option_t *option="")
	Generates the sample.

Printing Log
virtual void	printLog (Option_t *option="")
	Prints the log.

Public Member Functions inherited from URANIE::Sampler::TSamplerStochastic
	TSamplerStochastic (URANIE::DataServer::TDataServer tds, Option_t option, Int_t nCalcul)
	Constructor with a dataserver.

virtual	~TSamplerStochastic ()
	Default destructor.

void	setSeed (Int_t ind=0)
	Init the seed.

Int_t	getSeed ()
	Returns the seed value.

Public Member Functions inherited from URANIE::Sampler::TSampler
	TSampler (URANIE::DataServer::TDataServer tds, Option_t option, Int_t nCalcul)
	Constructor with a TDataServer, the options and the size of the sample.

virtual	~TSampler ()
	Default destructor.

Int_t	GetID ()
	Returns the ID of the class.

void	setMethodName (TString str)
	Sets the method name in a global variable.

TString	getMethodName ()
	Gets the method name.

virtual URANIE::DataServer::TDataServer *	getTDS ()
	Return the TDS filling by the sampling algorithm.

void	parseOption (Option_t *option)
	Parse the option.

virtual void	createListOfAttributes ()
	Creates the List of attributes to simulate.

virtual void	createTuple ()
	Creates the TDSNtupleD of data with only the TStochasticAttributes.

URANIE::DataServer::TDSNtupleD *	getTuple ()
	Returns the TDSNtupleD of data.

virtual void	fillOtherAttributes ()
	Fills the TDSNtupleD of data with other TFormulaAttributes.

void	setLog ()

void	unsetLog ()

void	changeLog ()

Bool_t	getLog ()

Protected Member Functions
State machine
For the stochastic methods, two states can be created init and terminate which are executed just before the generation of the sample for the first one, and after the generation for the second one. They can be used for pre and post treatment purpose.
virtual void	init ()
	Initialise the sampling procedure.

virtual void	terminate ()
	The post-processing step. Nothing is done.

State machine
For the stochastic methods, two2 states can be created init and terminate which are executed just before the generation of the sample for the first one, and after the generation for the second one. They can be used for pre and post treatment purpose.

Private Attributes
TString	_sMode
	Sampling mode. So far, can be "random" or "regular".

TString	_sGenMode
	Data generation mode. Only used in the random mode context, can be "lhs", or "srs".

Int_t	_nNbModif
	Number of modification for each nominal value of each changing attribute.

TList	_changingAttList
	List of changing attributes.

TMap	_infoRangeMap
	A list mapping the names of the changing attributes and their corresponding range information.

URANIE::DataServer::TDSNtupleD *	_nominalValueTuple
	A tuple storing the nominal values of the attributes.

Bool_t	_bHasData
	Boolean value set to TRUE if the data server _tds contains data.

Additional Inherited Members
Public Attributes inherited from URANIE::Sampler::TSampler
Int_t	_nS
	The size of the sample.

Int_t	_nX
	The size of attributes to sample.

URANIE::DataServer::TDSNtupleD *	_ntsample
	the tntuple of data

TString	_sMethod
	The title of the sampler method.

Bool_t	_blog
	Log Printing.

Bool_t	_bupdateFile
	Update the back up file when generating the attributeformula if there is some.

URANIE::DataServer::TDataServer *	_tds
	Pointer to a TDS.

TList *	_lstOfAttributesToSample
	The list of Stochastic Attributes to sample.

Protected Attributes inherited from URANIE::Sampler::TSamplerStochastic
Int_t	_nSeed

Detailed Description

This class has been renamed. Please use TOATDesign instead.

The principle of the OAT design of experiment is to modify only one attribute at a time, all the other factors being kept at a nominal value. This is the simplest way to compute partial derivatives or sensitivities. For the latter application however, it might not be the best solution as this approach does not explore multi-factorial and non-linear effects.

The present implementation is an improvement of TOATSampling. The user has now two ways of defining the nominal values of the attributes: by setting the default value for each attribute or by providing a tds with a set of nominal values. In the latter case, more than one set of nominal values can be provided, and an OAT sampling will be generated for each one of them.

The generated values are sampled inside an interval, centered on the nominal value. The width of this interval is defined as the "range". For each changing attribute, the user has to define the range, either as a numerical value, or as the name of an attribute of the tds. In both cases, an option allows the resulting value of the range to be interpreted as a percentage of the nominal value.

Two modes of data generation are available. The "regular" mode will generate data regularily spaced over the interval. The "random" mode will generate random values inside the interval. In the "random" mode, the changing attributes must be TStochasticAttribute objects. This means that (at the moment) this mode can't be used on attributes loaded from a file.

The number of generated values is not limited. It just needs to be greater or equal to 2. And if the sampling is "regular", it must also be even. But if an odd value is given, it will be reduced to an even one.

Here is a small example script, showing a regular OAT sampling on 2 attributes whose nominal values are read from a file:

--— file myNominalValues.dat --—

#COLUMN_NAMES: x1 | rx1 | x2

3.0  0.4  2.0
5.0  0.2  3.0

--— data generation script --—

{
// Construction of the data server
URANIE::DataServer::TDataServer tds("myTDS","OAT sampling TDS");
// Load the nominal values
tds.fileDataRead("myNominalValues.dat");
 
// Create the sampler
URANIE::Sampler::TOATSamplingMultiNominal s(&tds, "regular", 4);
 
// Set the range of attribute x1. It will be the value of attribute rx1.
s.setRange("x1","rx1"); 
// Set the range of attribute x2. It will be 10.0% of x2's nominal value.
s.setRange("x2",10.0,kTRUE); 
// Generate an OAT sampling where only x1 and x2 will vary
s.generateSample();
 
// Save the results to a file 
tds.exportData("myOATSampling.dat");
}

--— file myOATSampling.dat --—

#COLUMN_NAMES: x1| rx1| x2| __nominal_set__| __modified_att__| myTDS__n__iter__

3.00   0.4   2.000   1.0   -1.0    1.0 
2.80   0.4   2.000   1.0    0.0    2.0 
3.20   0.4   2.000   1.0    0.0    3.0 
2.90   0.4   2.000   1.0    0.0    4.0 
3.10   0.4   2.000   1.0    0.0    5.0 
3.00   0.4   1.900   1.0    2.0    6.0 
3.00   0.4   2.100   1.0    2.0    7.0 
3.00   0.4   1.950   1.0    2.0    8.0 
3.00   0.4   2.050   1.0    2.0    9.0 
5.00   0.2   3.000   2.0   -1.0   10.0 
4.90   0.2   3.000   2.0    0.0   11.0 
5.10   0.2   3.000   2.0    0.0   12.0 
4.95   0.2   3.000   2.0    0.0   13.0 
5.05   0.2   3.000   2.0    0.0   14.0 
5.00   0.2   2.850   2.0    2.0   15.0 
5.00   0.2   3.150   2.0    2.0   16.0 
5.00   0.2   2.925   2.0    2.0   17.0 
5.00   0.2   3.075   2.0    2.0   18.0

The attribute nominal_set refers to the set of nominal values used to generate the data. The attribute modified_att gives the index of the attribute whose value has been modified. If it is -1, this means that all the values are nominal.

Constructor & Destructor Documentation

◆ TOATSamplingMultiNominal()

URANIE::Sampler::TOATSamplingMultiNominal::TOATSamplingMultiNominal	(	URANIE::DataServer::TDataServer *	tds,
		TString	mode = `"regular"`,
		Int_t	number = `2`
	)

Constructor of the OAT sampler.

Parameters

tds	(URANIE::DataServer::TDataServer*) Data server containing the attributes.
mode	(TString): the sampling mode. It can be either: "regular" (or "reg"): generate regularily spaced values around the nominal value. "lhs": generate random values using the "lhs" sampling method "srs": generate random values using the "srs" sampling method "random" (or "rand"): same as "srs". (Default = "regular").
number	(Int_t): the number of values to generate for each nominal value of a given attribute. If _sMode is "regular", number must be even and equal or greater than 2. Otherwise, _nNbModif is set to max(2, number-1).

Exceptions

UErrorExceptions	if "number" is negative or null
UErrorExceptions	if "number" is odd and "mode" is "regular".
UErrorExceptions	if "mode" is unknown.

Deprecated:: This class has been renamed. Please use TOATDesign instead.

Referenced by ClassImp().

◆ ~TOATSamplingMultiNominal()

virtual URANIE::Sampler::TOATSamplingMultiNominal::~TOATSamplingMultiNominal ( )

virtual

Default destructor.

Referenced by ClassImp().

Member Function Documentation

◆ generateSample()

virtual void URANIE::Sampler::TOATSamplingMultiNominal::generateSample ( Option_t * option = "" )

virtual

Generates the sample.

This function creates the OAT sampling data.

Parameters

option (Option_t *): this parameter is not used at the moment. (Default = "").

Implements URANIE::Sampler::TSampler.

Referenced by ClassImp().

◆ getChangingAttList()

virtual TList * URANIE::Sampler::TOATSamplingMultiNominal::getChangingAttList ( )

inlinevirtual

Get the list of changing attributes.

References _changingAttList.

◆ getGenerationMode()

virtual TString URANIE::Sampler::TOATSamplingMultiNominal::getGenerationMode ( )

inlinevirtual

Get random generation method.

References _sGenMode.

◆ getHasData()

virtual Bool_t URANIE::Sampler::TOATSamplingMultiNominal::getHasData ( )

inlinevirtual

Returns TRUE if the internal data server contains data.

References _bHasData.

◆ getNominalValueTuple()

virtual URANIE::DataServer::TDSNtupleD * URANIE::Sampler::TOATSamplingMultiNominal::getNominalValueTuple ( )

inlinevirtual

Get the tuple storing the nominal values of the attributes.

References _nominalValueTuple.

◆ getNumberOfModifications()

virtual Double_t URANIE::Sampler::TOATSamplingMultiNominal::getNumberOfModifications ( )

inlinevirtual

Get the number of modifications per attribute and nominal values.

References _nNbModif.

◆ getRangeInformation()

virtual TMap * URANIE::Sampler::TOATSamplingMultiNominal::getRangeInformation ( )

inlinevirtual

Get the list mapping the names of the changing attributes and their corresponding range information.

References _infoRangeMap.

◆ getSamplingMode()

virtual TString URANIE::Sampler::TOATSamplingMultiNominal::getSamplingMode ( )

inlinevirtual

Get OAT sampling mode.

References _sMode.

◆ init()

virtual void URANIE::Sampler::TOATSamplingMultiNominal::init ( )

protectedvirtual

Initialise the sampling procedure.

This function creates a tuple containing the nominal values of all the attributes of the data server. It also creates two new attributes:

nominal_set: tells the user which set of nominal value is used;
modified_att: tells which attribute has been modified.

Exceptions

UErrorExceptions if the original data server is empty and one of its attributes has no default value.

Reimplemented from URANIE::Sampler::TSamplerStochastic.

Referenced by ClassImp().

◆ printLog()

virtual void URANIE::Sampler::TOATSamplingMultiNominal::printLog ( Option_t * option = "" )

virtual

Prints the log.

Prints some internal informations about the object.

Parameters

option (Option_t *): this parameter is not used at the moment. (Default = "").

Reimplemented from URANIE::Sampler::TSamplerStochastic.

Referenced by ClassImp().

◆ setOption()

virtual void URANIE::Sampler::TOATSamplingMultiNominal::setOption	(	TString	mode,
		Int_t	number
	)

virtual

Set the sampling mode and number of modifications.

This function sets the values of the _sMode and _nNbModif variables.

Parameters

mode	(TString): the sampling mode. It can be either: "regular": generate regularily spaced values around the nominal value. "lhs": generate random values using the "lhs" sampling method "srs": generate random values using the "srs" sampling method "random": same as "srs".
number	(Int_t): the number of values to generate for each nominal value of a given attribute. If _sMode is "regular", number must be even and equal or greater than 2. Otherwise, _nNbModif is set to max(2, number-1).

Exceptions

UErrorExceptions	if "mode" is unknown.
UErrorExceptions	if "number" is odd and "mode" is "regular".

Referenced by ClassImp().

◆ setRange() [1/2]

void URANIE::Sampler::TOATSamplingMultiNominal::setRange	(	TString	changingAttName,
		Double_t	rangeValue,
		Bool_t	is_percent = `kFALSE`
	)

Set the range of the sampling for a changing attribute.

This function sets the value of the range inside which the values of the changing attribute will be sampled. If this value is R, and the nominal value is N, the new values will be sampled in the interval [N-R/2; N+R/2]. The range can also be interpreted as a percentage p of the nominal value, in which case R = p * N / 100.

Note: a function setRange must be called for every changing attribute. There is no default range value.

Parameters

changingAttName	(TString): name of the changing attribute to sample.
rangeValue	(Double_t): the range value, i.e. the width of the interval around the nominal value where the sampling will take place.
is_percent	(Bool_t): if True, the rangeValue parameter will be interpreted as a percentage of the nominal value of the changing attribute. (Default = kFALSE).

Exceptions

UErrorExceptions	if no attribute changingAttName is found in the data server.
UErrorExceptions	if the data server is empty and the attribute has no default value.
UErrorExceptions	if range is negative or null.
UErrorExceptions	if mode is random and the attribute isn't a TStochasticAttribute.
UErrorExceptions	if a range was already set for the attribute.

Referenced by ClassImp().

◆ setRange() [2/2]

void URANIE::Sampler::TOATSamplingMultiNominal::setRange	(	TString	changingAttName,
		TString	rangeAttName,
		Bool_t	is_percent = `kFALSE`
	)

Set the name of the attribute holding the range of the sampling for a changing attribute.

This function associate the name of two attributes of the data server, the first being the changing attribute, the second holding the size of the range of the sampling. In this case, the value of the range will be read directly from the data server or the range attribute's default value.

Note: a function setRange must be called for every changing attribute. There is no default range value.

Parameters

changingAttName	(TString): name of the changing attribute to sample.
rangeAttName	(TString): name of the attribute holding the range value.
is_percent	(Bool_t): if True, the rangeValue parameter will be interpreted as a percentage of the nominal value of the changing attribute. (Default = kFALSE).

Exceptions

UErrorExceptions	if no attribute rangeAttName is found in the data server.
UErrorExceptions	if no attribute changingAttName is found in the data server.
UErrorExceptions	if the data server is empty and the attribute has no default value.
UErrorExceptions	if the data server is empty and the range attribute has no default value.
UErrorExceptions	if the data server is empty and the range attribute's default value is negative or null.
UErrorExceptions	if mode is random and the attribute isn't a TStochasticAttribute.
UErrorExceptions	if a range was already set for the attribute.

◆ terminate()

virtual void URANIE::Sampler::TOATSamplingMultiNominal::terminate ( )

inlineprotectedvirtual

The post-processing step. Nothing is done.

Reimplemented from URANIE::Sampler::TSamplerStochastic.

Referenced by ClassImp().

Member Data Documentation

◆ _bHasData

Bool_t URANIE::Sampler::TOATSamplingMultiNominal::_bHasData

private

Boolean value set to TRUE if the data server _tds contains data.

Referenced by ClassImp(), and getHasData().

◆ _changingAttList

TList URANIE::Sampler::TOATSamplingMultiNominal::_changingAttList

private

List of changing attributes.

Referenced by ClassImp(), and getChangingAttList().

◆ _infoRangeMap

TMap URANIE::Sampler::TOATSamplingMultiNominal::_infoRangeMap

private

A list mapping the names of the changing attributes and their corresponding range information.

Referenced by ClassImp(), and getRangeInformation().

◆ _nNbModif

Int_t URANIE::Sampler::TOATSamplingMultiNominal::_nNbModif

private

Number of modification for each nominal value of each changing attribute.

Referenced by ClassImp(), and getNumberOfModifications().

◆ _nominalValueTuple

URANIE::DataServer::TDSNtupleD* URANIE::Sampler::TOATSamplingMultiNominal::_nominalValueTuple

private

A tuple storing the nominal values of the attributes.

Referenced by ClassImp(), and getNominalValueTuple().

◆ _sGenMode

TString URANIE::Sampler::TOATSamplingMultiNominal::_sGenMode

private

Data generation mode. Only used in the random mode context, can be "lhs", or "srs".

Referenced by ClassImp(), and getGenerationMode().

◆ _sMode

TString URANIE::Sampler::TOATSamplingMultiNominal::_sMode

private

Sampling mode. So far, can be "random" or "regular".

Referenced by ClassImp(), and getSamplingMode().

Documentation / Manuel développeur

Modules disponibles

Public Member Functions

Protected Member Functions

Private Attributes

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ TOATSamplingMultiNominal()

◆ ~TOATSamplingMultiNominal()

Member Function Documentation

◆ generateSample()

◆ getChangingAttList()

◆ getGenerationMode()

◆ getHasData()

◆ getNominalValueTuple()

◆ getNumberOfModifications()

◆ getRangeInformation()

◆ getSamplingMode()

◆ init()

◆ printLog()

◆ setOption()

◆ setRange() [1/2]

◆ setRange() [2/2]

◆ terminate()

Member Data Documentation

◆ _bHasData

◆ _changingAttList

◆ _infoRangeMap

◆ _nNbModif

◆ _nominalValueTuple

◆ _sGenMode

◆ _sMode