Description of the class TPolynomialChaos. More...

#include <TPolynomialChaos.h>

Inheritance diagram for URANIE::Modeler::TPolynomialChaos:

Collaboration diagram for URANIE::Modeler::TPolynomialChaos:

[legend]

Public Member Functions
Constructor and Destructor
	TPolynomialChaos (TDataServer tds, TNisp nisp, TString soutput="")
	default constructor from a set of stochastic variables

	TPolynomialChaos (const char *ct)

	TPolynomialChaos (const TPolynomialChaos &TPC)
	copy constructor

virtual	~TPolynomialChaos ()
	destructor

void	setDegree (Int_t degree)
	Set the degree.

void	setAutoDegreeFactor (double autodeg)
	Change the factor that links the number of samples, coefficient and the highest possible degree.

void	setAutoDegreeBoundaries (int amin, int amax=-1)
	Change the min and max degree tested for automatise degree determination.

TTree *	getAutoDegreeResults ()
	Return a pointer to the TTree that contains the degree optimisation results.

int	getBestAutoDegree ()
	Return the bet estimated degree when optimisation is done with regression.

void	computeOutput (double *input)
	Computation of output in relation with input : input[0...nx-1].

Double_t	getOutput (TString jname="")
	Get an output.

Double_t	getOutput (Int_t j)
	Get an output.

Double_t	getEqmLoo (Int_t rank=0)
	Get the Mean squared uncertainty for the output with rank.

Double_t	getErrLoo (Int_t input, Int_t rank=0)
	Get the Leave-One-Out uncertainty for the input (input) and the output (rank)

double	factorial (int n)
	Dummy factorial methods.

void	automatisedDegree (Option_t *option="")
	Computation of coefficients.

void	computeChaosExpansion (TString type, Option_t *option="")
	Computation of coefficients.

Double_t	getMean (Int_t j)
	Get the mean.

Double_t	getMean (TString name="")
	Get the mean.

Double_t	getVariance (Int_t j)
	Get the variance.

Double_t	getVariance (TString name="")
	Get the variance.

Double_t	getCovariance (Int_t i, Int_t j)
	Get the Covariance.

Double_t	getCovariance (TString xname, TString yname)
	Get the Covariance.

Double_t	getCorrelation (Int_t i, Int_t j)
	Get the correlation.

Double_t	getCorrelation (TString xname, TString yname)
	Get the correlation.

Double_t	getIndexFirstOrder (Int_t i, Int_t j=0)
	First index of sensitivity.

Double_t	getIndexFirstOrder (TString xname, TString yname="")
	First index of sensitivity.

Double_t	getIndexTotalOrder (Int_t i, Int_t j=0)
	Total index of sensitivity.

Double_t	getIndexTotalOrder (TString xname, TString yname="")
	Total index of sensitivity.

Double_t	getIndex (TString sinput="", TString yname="")
	Index of sensitivity of a set of variables.

Double_t	getIndexInteraction (TString sinput="", TString yname="")
	Index of interaction sensitivity of a set of variables.

Int_t	getDimensionInput ()
	Number of input.

Int_t	getDimensionOutput ()
	Number of output.

Int_t	getDimensionExpansion ()
	Number of the coefficients of the polynomial chaos.

Int_t	getDegree ()
	Get the degree.

void	exportFunction (const char file="nisp", const char name="nisp_fct")
	Export the model in C++ langage in a file.

Double_t	getCoefficient (Int_t k, TString jname="")
	Get coefficents value.

No accessible - In building
void	generateSample (TString type, Int_t np, Int_t order=1)
	Build a sample for statitical analysis (quantile). Build a sample of size "np" by the method "type".

Double_t	getSample (Int_t k, Int_t j)

Double_t	getSample (Int_t k, TString yname)
	The value of the output yname for the example k.

void	realisation ()
	Polynomial chaos is a random variable : a realisation of this random variable (GetOutput()).

void	getAnova (Int_t nt)
	Edition of the ANOVA decomposition.

void	getAnovaOrdered (Double_t seuil, Int_t j)

void	getAnovaOrdered (Double_t seuil, TString yname="")
	Return the ANOVA ordered decomposition.

void	getAnovaOrderedCoefficients (Double_t seuil, Int_t j)

void	getAnovaOrderedCoefficients (Double_t seuil, TString yname="")
	Edition of the ANOVA ordered decomposition / coefficients.

Double_t	getQuantile (Double_t alpha, Int_t j)
	Return the quantile of order "alpha".

Double_t	getQuantile (Double_t alpha, TString yname)
	Return the quantile of order "alpha".

Double_t	getQuantileWilks (Double_t alpha, Double_t beta, Int_t j=1)
	Return the Wilks' quantile of order "alpha" with confidence "beta".

Double_t	getQuantileWilks (Double_t alpha, Double_t beta, TString yname)
	Return the Wilks' quantile of order "alpha" with confidence "beta".

Double_t	getInvQuantile (Double_t seuil, Int_t j=1)
	Return the probability of a value overrun.

Double_t	getInvQuantile (Double_t seuil, TString yname)
	Return the probability of a value overrun.

void	readTarget (Char_t *file)
	Read a set of target from a file.

void	setInput (Int_t nt, Double_t dt)
	Set an input.

void	propagateInput ()
	Propagation of input which has been specified SetInput()

void	propagateInput (Double_t *dt)
	Propagation of dt[1...nx].

void	save (Char_t *file)
	Saving of the PC in a file.

void	save (string file)

void	setAnova ()
	Preparation on the ANOVA decomposition.

Public Attributes
TNisp *	_nisp
	Message logger.

Int_t	_degree
	Degree of the Chaos polynomial.

PolynomialChaos *	_pc
	Object of type polynomialChaos (library Nisp)

Int_t	_nx
	Number of stochastic variables i.e number of input.

Int_t	_np
	Number of simulation.

Int_t	_ny

Bool_t	_bStoreYHat
	Boolean to specify if we add the \hat{} attribute in the TDS [default kTRUE].

TTree *	_degreeResults
	TDSNtupleD used to store the results of the automatisedDegree method.

double	_autoDegreeFactor
	Factor used to scale the maximum degree knowing _nx and _np.

int	_minAutoDeg
	Minimal value for automatic degree scan.

int	_maxAutoDeg
	Maximal value for automatic degree scan.

int	_bestAutoDeg
	Best value for automatic degree scan.

vector< double > *	_verrloo
	vector to contains the err loo and store them in the tree

int	_degval
	memory arry for the degree in the tree

double	_eqmval
	memory arry for the eqm in the tree

Protected Attributes
URANIE::DataServer::UMessageLogger *	_fLogger

Private Attributes
TDataSpecification *	_listAttOut
	Object of type TDataSpecification used to index the name of the variables.

Bool_t	_blog
	Boolean for edit the log.

Printing Log
void	setLog ()

void	unsetLog ()

void	changeLog ()

Bool_t	getLog ()

virtual void	printLog (Option_t *option="")

void	setCoefficient (Int_t noutput, Int_t num, Double_t coef)
	Set the coefficient beta of _pc.

Double_t	getCoefficient (Int_t noutput, Int_t num)
	Get the coefficient beta of _pc.

void	getTarget ()
	Get the target.

void	setGroupAddVar (Int_t i)
	Add a random variable.

void	writeCodeCToDenormalizeInput (std::ofstream *sourcefile)
	Write set Denormalization.

void	setGroupEmpty ()
	Freeing the set.

Detailed Description

Description of the class TPolynomialChaos.

The class TPolynomialChaos is associated to the classe TNisp. It's a mirror of the class PolynomialChaos, a class of the NISP library. It's used to access to Chaos Polynomial functionalities of the NIPS library in URANIE.

Constructor & Destructor Documentation

◆ TPolynomialChaos() [1/3]

URANIE::Modeler::TPolynomialChaos::TPolynomialChaos	(	TDataServer *	tds,
		TNisp *	nisp,
		TString	soutput = `""`
	)

default constructor from a set of stochastic variables

The constructor builds a object TPolynomialChaos thanks to information which come from a object TDaaServer and TNisp. The elements which constitue the object TPolynomialChaos are determinated in the following way

_nisp : It's the TNisp object given in parameter;
_pc : Ths polynomialchaos object is built from data with come from the stochastic variables of the TNisp object;
_nx : The number of random variable. It's the number of attribute of the TDataServer;
_np : The number of simulation (size of the sample) is given by TNisp
_ny : The number of output.
_degree : It's the degree of the polynomial chaos.

Parameters

tds	: TDataServer object.
nisp	: TNisp object.
soutput	: list of output which be use to build the PC. By default all the output of the tds are used.

Referenced by ClassImp().

◆ TPolynomialChaos() [2/3]

URANIE::Modeler::TPolynomialChaos::TPolynomialChaos ( const char * ct )

Constructor from a file in "nisp" format

Parameters

ct	: the name of the "nisp" file

◆ TPolynomialChaos() [3/3]

URANIE::Modeler::TPolynomialChaos::TPolynomialChaos ( const TPolynomialChaos & TPC )

copy constructor

The copy constructor

Parameters

TPC	: TPolynomialChaos object.

◆ ~TPolynomialChaos()

virtual URANIE::Modeler::TPolynomialChaos::~TPolynomialChaos ( )

virtual

destructor

Referenced by ClassImp().

Member Function Documentation

◆ automatisedDegree()

void URANIE::Modeler::TPolynomialChaos::automatisedDegree ( Option_t * option = "" )

Computation of coefficients.

Calculated coefficients of the polynomial chaos by the methode "type".

Parameters

type	: Name of the method ( Integration or Regression only accepted).

Referenced by ClassImp().

◆ changeLog()

void URANIE::Modeler::TPolynomialChaos::changeLog ( )

inline

References _blog.

◆ computeChaosExpansion()

void URANIE::Modeler::TPolynomialChaos::computeChaosExpansion	(	TString	type,
		Option_t *	option = `""`
	)

Computation of coefficients.

Calculated coefficients of the polynomial chaos by the methode "type".

Parameters

type	: Name of the method ( Integration or Regression only accepted).

Referenced by ClassImp().

◆ computeOutput()

void URANIE::Modeler::TPolynomialChaos::computeOutput ( double * input )

Computation of output in relation with input : input[0...nx-1].

Compute the ouput Y with the values Xi of the input (i=0 to nx-1).

Parameters

input : table of input value.

Referenced by ClassImp().

◆ exportFunction()

void URANIE::Modeler::TPolynomialChaos::exportFunction	(	const char *	file = `"nisp"`,
		const char *	name = `"nisp_fct"`
	)

Export the model in C++ langage in a file.

Writing in a C++ programm, the model of the polynomial chaos.

Parameters

file	(const char *) the file to export the modeler. If empty, use the name of the model, plus the extension of the langage;
name	(const char *) the name of the function in the export file. If empty, use the name of the model.

Referenced by ClassImp().

◆ factorial()

double URANIE::Modeler::TPolynomialChaos::factorial ( int n )

Dummy factorial methods.

Referenced by ClassImp().

◆ generateSample()

void URANIE::Modeler::TPolynomialChaos::generateSample	(	TString	type,
		Int_t	np,
		Int_t	order = `1`
	)

Build a sample for statitical analysis (quantile). Build a sample of size "np" by the method "type".

Todo:

Parameters

type	: method used;
np	: number of simulation;
order	:

Referenced by ClassImp().

◆ getAnova()

void URANIE::Modeler::TPolynomialChaos::getAnova ( Int_t nt )

Edition of the ANOVA decomposition.

Todo:

Referenced by ClassImp().

◆ getAnovaOrdered() [1/2]

void URANIE::Modeler::TPolynomialChaos::getAnovaOrdered	(	Double_t	seuil,
		Int_t	j
	)

Todo:

Referenced by ClassImp().

◆ getAnovaOrdered() [2/2]

void URANIE::Modeler::TPolynomialChaos::getAnovaOrdered	(	Double_t	seuil,
		TString	yname = `""`
	)

Return the ANOVA ordered decomposition.

Parameters

seuil	: value of the seuil;
yname	: name of the output.

refer to URANIE:Modeler::TPolynomialChaos::getAnovaOrdered(Double_t seuil,Int_t j)

◆ getAnovaOrderedCoefficients() [1/2]

void URANIE::Modeler::TPolynomialChaos::getAnovaOrderedCoefficients	(	Double_t	seuil,
		Int_t	j
	)

Todo:

Referenced by ClassImp().

◆ getAnovaOrderedCoefficients() [2/2]

void URANIE::Modeler::TPolynomialChaos::getAnovaOrderedCoefficients	(	Double_t	seuil,
		TString	yname = `""`
	)

Edition of the ANOVA ordered decomposition / coefficients.

Parameters

seuil	: value of the seuil;
yname	: name of the output.

refer to URANIE:Modeler::TPolynomialChaos::getAnovaOrderedCoefficients(Double_t seuil,Int_t j)

◆ getAutoDegreeResults()

TTree * URANIE::Modeler::TPolynomialChaos::getAutoDegreeResults ( )

inline

Return a pointer to the TTree that contains the degree optimisation results.

References _degreeResults.

◆ getBestAutoDegree()

int URANIE::Modeler::TPolynomialChaos::getBestAutoDegree ( )

inline

Return the bet estimated degree when optimisation is done with regression.

References _bestAutoDeg.

◆ getCoefficient() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getCoefficient	(	Int_t	k,
		TString	jname = `""`
	)

Get coefficents value.

Get the coefficeint k of the PC for the output janme

Parameters

k	number of the coefficient (<p)
jname	output name

Referenced by ClassImp().

◆ getCoefficient() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getCoefficient	(	Int_t	noutput,
		Int_t	num
	)

Get the coefficient beta of _pc.

Get the coefficient beta of _pc

Parameters

noutput	: number of output
num	: number of Polynomial Chaos

◆ getCorrelation() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getCorrelation	(	Int_t	i,
		Int_t	j
	)

Get the correlation.

Return the correlation.

Parameters

i	: index of the output;
j	: index of the second output.

Referenced by ClassImp().

◆ getCorrelation() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getCorrelation	(	TString	xname,
		TString	yname
	)

Get the correlation.

Return the correlation

Parameters

xname	: name of the output;
yname	: name of the second ouput.

refer to URANIE:Modeler::TPolynomialChaos:getCorrelation(Int_t i, Int_t j)

◆ getCovariance() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getCovariance	(	Int_t	i,
		Int_t	j
	)

Get the Covariance.

Return the covariance.

Parameters

i	: index of the output;
j	: index of the second output.

Referenced by ClassImp().

◆ getCovariance() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getCovariance	(	TString	xname,
		TString	yname
	)

Get the Covariance.

Return the covariance.

Parameters

xname	: name of the output;
yname	: name of the second output.

refer to URANIE:Modeler::TPolynomialChaos:getCovariance(Int_t i, Int_t j)

◆ getDegree()

Int_t URANIE::Modeler::TPolynomialChaos::getDegree ( )

Get the degree.

Return the degree maximum of the PC

Referenced by ClassImp().

◆ getDimensionExpansion()

Int_t URANIE::Modeler::TPolynomialChaos::getDimensionExpansion ( )

Number of the coefficients of the polynomial chaos.

Return the number of coefficients of the PC.

Referenced by ClassImp().

◆ getDimensionInput()

Int_t URANIE::Modeler::TPolynomialChaos::getDimensionInput ( )

Number of input.

Return the number of input i.e the number of input random variable.

Referenced by ClassImp().

◆ getDimensionOutput()

Int_t URANIE::Modeler::TPolynomialChaos::getDimensionOutput ( )

Number of output.

Return the number of output i.e the number of output variable

Referenced by ClassImp().

◆ getEqmLoo()

Double_t URANIE::Modeler::TPolynomialChaos::getEqmLoo ( Int_t rank = 0 )

Get the Mean squared uncertainty for the output with rank.

Parameters

rank	the rank of the considered output

Referenced by ClassImp().

◆ getErrLoo()

Double_t URANIE::Modeler::TPolynomialChaos::getErrLoo	(	Int_t	input,
		Int_t	rank = `0`
	)

Get the Leave-One-Out uncertainty for the input (input) and the output (rank)

Parameters

input	the input indexes of the doe
rank	the rank of the considered output

Referenced by ClassImp().

◆ getIndex()

Double_t URANIE::Modeler::TPolynomialChaos::getIndex	(	TString	sinput = `""`,
		TString	yname = `""`
	)

Index of sensitivity of a set of variables.

Return the index of sensitivity of the output yname.

Parameters

sinput	: list of input. By default all the input;
yname	: name of the output. By default the first ouput.

Referenced by ClassImp().

◆ getIndexFirstOrder() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getIndexFirstOrder	(	Int_t	i,
		Int_t	j = `0`
	)

First index of sensitivity.

Return the index od sensitivity.

Parameters

i	: index of the input (<_nx);
j	: index of the output (<_ny).

Referenced by ClassImp().

◆ getIndexFirstOrder() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getIndexFirstOrder	(	TString	xname,
		TString	yname = `""`
	)

First index of sensitivity.

Return the index od sensitivity.

Parameters

xname	: name of the input;
yname	: name of the output.

refer to URANIE:Modeler::TPolynomialChaos:getIndexFirstOrder(Int_t i, Int_t j)

◆ getIndexInteraction()

Double_t URANIE::Modeler::TPolynomialChaos::getIndexInteraction	(	TString	sinput = `""`,
		TString	yname = `""`
	)

Index of interaction sensitivity of a set of variables.

Return the interaction sensitivity index of the ouput j.

Parameters

sinput	: list of input. By default all the input;
yname	: name of the output. By default the first ouput.

Referenced by ClassImp().

◆ getIndexTotalOrder() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getIndexTotalOrder	(	Int_t	i,
		Int_t	j = `0`
	)

Total index of sensitivity.

Return the total index of sensitivity.

Parameters

j	: index of the output (<_ny);
i	: index of the variable (<_nx).

Referenced by ClassImp().

◆ getIndexTotalOrder() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getIndexTotalOrder	(	TString	xname,
		TString	yname = `""`
	)

Total index of sensitivity.

Return the total index of sensitivity.

Parameters

xname	: name of the input;
yname	: name of the output.

refer to URANIE:Modeler::TPolynomialChaos:getIndexTotalOrder(Int_t i, Int_t j)

◆ getInvQuantile() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getInvQuantile	(	Double_t	seuil,
		Int_t	j = `1`
	)

Return the probability of a value overrun.

Parameters

seuil	: ;
j	: index of the output.

Referenced by ClassImp().

◆ getInvQuantile() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getInvQuantile	(	Double_t	seuil,
		TString	yname
	)

Return the probability of a value overrun.

Parameters

seuil	: seuil;
yname	: name of the output.

refer to URANIE:Modeler::TPolynomialChaos::getInvQuantile(Double_t seuil, Int_t j)

◆ getLog()

Bool_t URANIE::Modeler::TPolynomialChaos::getLog ( )

inline

References _blog.

◆ getMean() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getMean ( Int_t j )

Get the mean.

Return the mean of the attribute given by the index.

Parameters

j	: index of the attribut.

Referenced by ClassImp().

◆ getMean() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getMean ( TString name = "" )

Get the mean.

Return the mean of the attribute given by the name.

Parameters

name	: name of the attribut. By defalut the forst output.

refer to URANIE:Modeler::TPolynomialChaos:getMean(Int_t j)

◆ getOutput() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getOutput ( Int_t j )

Get an output.

Return the value of a ouput.

Parameters

j	: index od the ouput.

◆ getOutput() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getOutput ( TString jname = "" )

Get an output.

Return the value of a ouput.

Parameters

jname : name of the ouput.

refer to URANIE:Modeler::TPolynomialChaos:getOutput(Int_t j)

Referenced by ClassImp().

◆ getQuantile() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getQuantile	(	Double_t	alpha,
		Int_t	j
	)

Return the quantile of order "alpha".

Parameters

alpha	: order;
j	: index of the output.

Referenced by ClassImp().

◆ getQuantile() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getQuantile	(	Double_t	alpha,
		TString	yname
	)

Return the quantile of order "alpha".

Parameters

alpha	: order;
yname	: name of the output.

refer to URANIE:Modeler::TPolynomialChaos::getQuantile(Double_t alpha, Int_t j)

◆ getQuantileWilks() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getQuantileWilks	(	Double_t	alpha,
		Double_t	beta,
		Int_t	j = `1`
	)

Return the Wilks' quantile of order "alpha" with confidence "beta".

Parameters

alpha	: order of the quantile;
beta	: confidence
j	: index of the output. By default j=1.

Referenced by ClassImp().

◆ getQuantileWilks() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getQuantileWilks	(	Double_t	alpha,
		Double_t	beta,
		TString	yname
	)

Return the Wilks' quantile of order "alpha" with confidence "beta".

Parameters

alpha	: order of the quantile;
beta	: confidence
yname	: name of the output

refer to URANIE:Modeler::TPolynomialChaos::getQuantileWilks(Double_t alpha, Double_t beta, Int_t j)

◆ getSample() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getSample	(	Int_t	k,
		Int_t	j
	)

Todo:

Referenced by ClassImp().

◆ getSample() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getSample	(	Int_t	k,
		TString	yname
	)

The value of the output yname for the example k.

refer to URANIE:Modeler::TPolynomialChaos::getSample(Int_t k,Int_t j)

◆ getTarget()

void URANIE::Modeler::TPolynomialChaos::getTarget ( )

private

Get the target.

Return

Referenced by ClassImp().

◆ getVariance() [1/2]

Double_t URANIE::Modeler::TPolynomialChaos::getVariance ( Int_t j )

Get the variance.

Return the variance of the attribute given by the index.

Parameters

j	: index of the attribut.

Referenced by ClassImp().

◆ getVariance() [2/2]

Double_t URANIE::Modeler::TPolynomialChaos::getVariance ( TString name = "" )

Get the variance.

Return the variance of the attribute given by the name.

Parameters

name	: name of the attribut. By defalut the forst output.

refer to URANIE:Modeler::TPolynomialChaos:getVariance(Int_t j)

◆ printLog()

virtual void URANIE::Modeler::TPolynomialChaos::printLog ( Option_t * option = "" )

virtual

Referenced by ClassImp().

◆ propagateInput() [1/2]

void URANIE::Modeler::TPolynomialChaos::propagateInput ( )

Propagation of input which has been specified SetInput()

Referenced by ClassImp().

◆ propagateInput() [2/2]

void URANIE::Modeler::TPolynomialChaos::propagateInput ( Double_t * dt )

Propagation of dt[1...nx].

◆ readTarget()

void URANIE::Modeler::TPolynomialChaos::readTarget ( Char_t * file )

Read a set of target from a file.

Referenced by ClassImp().

◆ realisation()

void URANIE::Modeler::TPolynomialChaos::realisation ( )

Polynomial chaos is a random variable : a realisation of this random variable (GetOutput()).

Todo:

Referenced by ClassImp().

◆ save() [1/2]

void URANIE::Modeler::TPolynomialChaos::save ( Char_t * file )

Saving of the PC in a file.

Parameters

file	: name of the file;

Referenced by ClassImp().

◆ save() [2/2]

void URANIE::Modeler::TPolynomialChaos::save ( string file )

◆ setAnova()

void URANIE::Modeler::TPolynomialChaos::setAnova ( )

Preparation on the ANOVA decomposition.

Referenced by ClassImp().

◆ setAutoDegreeBoundaries()

void URANIE::Modeler::TPolynomialChaos::setAutoDegreeBoundaries	(	int	amin,
		int	amax = `-1`
	)

Change the min and max degree tested for automatise degree determination.

If no maximum is specified, the automatic specification will be done using the _autoDegreeFactor.

Parameters

amin	minimum to be tested (>=1)
amax	max to be tested. Should be > amin. If empty, automatic determination is done

Referenced by ClassImp().

◆ setAutoDegreeFactor()

void URANIE::Modeler::TPolynomialChaos::setAutoDegreeFactor ( double autodeg )

Change the factor that links the number of samples, coefficient and the highest possible degree.

If no maximum is specified for automatic degree optimisation, then the maximal degree to be tested is computed knowing that Ncoeff <= np /autodeg, where np is number of pattern and Ncoeff = (nx+deg)!/(nx!*deg!);

Parameters

autodeg change the value of autodeg (default is 1.5). Should be >= 1.

Referenced by ClassImp().

◆ setCoefficient()

void URANIE::Modeler::TPolynomialChaos::setCoefficient	(	Int_t	noutput,
		Int_t	num,
		Double_t	coef
	)

Set the coefficient beta of _pc.

Set the coefficient beta of _pc

Parameters

noutput	: number of output
num	: number of Polynomial Chaos
coef	: new value of coefficient

Referenced by ClassImp(), and ClassImp().

◆ setDegree()

void URANIE::Modeler::TPolynomialChaos::setDegree ( Int_t degree )

Set the degree.

Set the degree of the polynomial chaos.

Parameters

degree : degree max of the polynomial.

Referenced by ClassImp(), and ClassImp().

◆ setGroupAddVar()

void URANIE::Modeler::TPolynomialChaos::setGroupAddVar ( Int_t i )

private

Add a random variable.

Add a random variable (rank i) in the set

Parameters

i	: index of the input variable added

Referenced by ClassImp().

◆ setGroupEmpty()

void URANIE::Modeler::TPolynomialChaos::setGroupEmpty ( )

private

Freeing the set.

Freeing the set

Referenced by ClassImp().

◆ setInput()

void URANIE::Modeler::TPolynomialChaos::setInput	(	Int_t	nt,
		Double_t	dt
	)

Set an input.

Referenced by ClassImp().

◆ setLog()

void URANIE::Modeler::TPolynomialChaos::setLog ( )

inline

References _blog.

◆ unsetLog()

void URANIE::Modeler::TPolynomialChaos::unsetLog ( )

inline

References _blog.

◆ writeCodeCToDenormalizeInput()

void URANIE::Modeler::TPolynomialChaos::writeCodeCToDenormalizeInput ( std::ofstream * sourcefile )

private

Write set Denormalization.

Write the C code to denormalize the inputs parameters

Referenced by ClassImp().

Member Data Documentation

◆ _autoDegreeFactor

double URANIE::Modeler::TPolynomialChaos::_autoDegreeFactor

Factor used to scale the maximum degree knowing _nx and _np.

Referenced by ClassImp().

◆ _bestAutoDeg

int URANIE::Modeler::TPolynomialChaos::_bestAutoDeg

Best value for automatic degree scan.

Referenced by ClassImp(), and getBestAutoDegree().

◆ _blog

Bool_t URANIE::Modeler::TPolynomialChaos::_blog

private

Boolean for edit the log.

Referenced by changeLog(), ClassImp(), getLog(), setLog(), and unsetLog().

◆ _bStoreYHat

Bool_t URANIE::Modeler::TPolynomialChaos::_bStoreYHat

Boolean to specify if we add the \hat{} attribute in the TDS [default kTRUE].

Referenced by ClassImp().

◆ _degree

Int_t URANIE::Modeler::TPolynomialChaos::_degree

Degree of the Chaos polynomial.

Referenced by ClassImp().

◆ _degreeResults

TTree* URANIE::Modeler::TPolynomialChaos::_degreeResults

TDSNtupleD used to store the results of the automatisedDegree method.

Referenced by ClassImp(), and getAutoDegreeResults().

◆ _degval

int URANIE::Modeler::TPolynomialChaos::_degval

memory arry for the degree in the tree

Referenced by ClassImp().

◆ _eqmval

double URANIE::Modeler::TPolynomialChaos::_eqmval

memory arry for the eqm in the tree

Referenced by ClassImp().

◆ _fLogger

URANIE::DataServer::UMessageLogger* URANIE::Modeler::TPolynomialChaos::_fLogger

protected

Referenced by ClassImp().

◆ _listAttOut

TDataSpecification* URANIE::Modeler::TPolynomialChaos::_listAttOut

private

Object of type TDataSpecification used to index the name of the variables.

Referenced by ClassImp().

◆ _maxAutoDeg

int URANIE::Modeler::TPolynomialChaos::_maxAutoDeg

Maximal value for automatic degree scan.

Referenced by ClassImp().

◆ _minAutoDeg

int URANIE::Modeler::TPolynomialChaos::_minAutoDeg

Minimal value for automatic degree scan.

Referenced by ClassImp().

◆ _nisp

TNisp* URANIE::Modeler::TPolynomialChaos::_nisp

Message logger.

Object of ype TNisp

Referenced by ClassImp().

◆ _np

Int_t URANIE::Modeler::TPolynomialChaos::_np

Number of simulation.

Referenced by ClassImp().

◆ _nx

Int_t URANIE::Modeler::TPolynomialChaos::_nx

Number of stochastic variables i.e number of input.

Referenced by ClassImp().

◆ _ny

Int_t URANIE::Modeler::TPolynomialChaos::_ny

Referenced by ClassImp().

◆ _pc

PolynomialChaos* URANIE::Modeler::TPolynomialChaos::_pc

Object of type polynomialChaos (library Nisp)

Referenced by ClassImp(), and ClassImp().

◆ _verrloo

vector<double>* URANIE::Modeler::TPolynomialChaos::_verrloo

vector to contains the err loo and store them in the tree

Referenced by ClassImp().

Documentation / Developer's manual

Available modules

Public Member Functions

Public Attributes

Protected Attributes

Private Attributes

Printing Log

Detailed Description

Constructor & Destructor Documentation

◆ TPolynomialChaos() [1/3]

◆ TPolynomialChaos() [2/3]

◆ TPolynomialChaos() [3/3]

◆ ~TPolynomialChaos()

Member Function Documentation

◆ automatisedDegree()

◆ changeLog()

◆ computeChaosExpansion()

◆ computeOutput()

◆ exportFunction()

◆ factorial()

◆ generateSample()

◆ getAnova()

◆ getAnovaOrdered() [1/2]

◆ getAnovaOrdered() [2/2]

◆ getAnovaOrderedCoefficients() [1/2]

◆ getAnovaOrderedCoefficients() [2/2]

◆ getAutoDegreeResults()

◆ getBestAutoDegree()

◆ getCoefficient() [1/2]

◆ getCoefficient() [2/2]

◆ getCorrelation() [1/2]

◆ getCorrelation() [2/2]

◆ getCovariance() [1/2]

◆ getCovariance() [2/2]

◆ getDegree()

◆ getDimensionExpansion()

◆ getDimensionInput()

◆ getDimensionOutput()

◆ getEqmLoo()

◆ getErrLoo()

◆ getIndex()

◆ getIndexFirstOrder() [1/2]

◆ getIndexFirstOrder() [2/2]

◆ getIndexInteraction()

◆ getIndexTotalOrder() [1/2]

◆ getIndexTotalOrder() [2/2]

◆ getInvQuantile() [1/2]

◆ getInvQuantile() [2/2]

◆ getLog()

◆ getMean() [1/2]

◆ getMean() [2/2]

◆ getOutput() [1/2]

◆ getOutput() [2/2]

◆ getQuantile() [1/2]

◆ getQuantile() [2/2]

◆ getQuantileWilks() [1/2]

◆ getQuantileWilks() [2/2]

◆ getSample() [1/2]

◆ getSample() [2/2]

◆ getTarget()

◆ getVariance() [1/2]

◆ getVariance() [2/2]

◆ printLog()

◆ propagateInput() [1/2]

◆ propagateInput() [2/2]

◆ readTarget()

◆ realisation()

◆ save() [1/2]

◆ save() [2/2]

◆ setAnova()

◆ setAutoDegreeBoundaries()

◆ setAutoDegreeFactor()

◆ setCoefficient()

◆ setDegree()

◆ setGroupAddVar()

◆ setGroupEmpty()

◆ setInput()

◆ setLog()

◆ unsetLog()

◆ writeCodeCToDenormalizeInput()