RooMinimizer is a wrapper class around ROOT::Fit:Fitter that provides a seamless interface between the minimizer functionality and the native RooFit interface.

By default the Minimizer is MINUIT for classic FcnMode and MINUIT2 for gradient FcnMode. RooMinimizer can minimize any RooAbsReal function with respect to its parameters. Usual choices for minimization are RooNLLVar and RooChi2Var RooMinimizer has methods corresponding to MINUIT functions like hesse(), migrad(), minos() etc. In each of these function calls the state of the MINUIT engine is synchronized with the state of the RooFit variables: any change in variables, change in the constant status etc is forwarded to MINUIT prior to execution of the MINUIT call. Afterwards the RooFit objects are resynchronized with the output state of MINUIT: changes parameter values, errors are propagated. Various methods are available to control verbosity, profiling, automatic PDF optimization.

Definition at line 51 of file RooMinimizer.h.

Public Types
enum class	FcnMode { classic , gradient , generic_wrapper }

enum	PrintLevel { None =-1 , Reduced =0 , Normal =1 , ExtraForProblem =2 , Maximum =3 }

enum	Strategy { Speed =0 , Balance =1 , Robustness =2 }

Public Types inherited from TObject
enum	{ kIsOnHeap = 0x01000000 , kNotDeleted = 0x02000000 , kZombie = 0x04000000 , kInconsistent = 0x08000000 , kBitMask = 0x00ffffff }

enum	{ kSingleKey = BIT(0) , kOverwrite = BIT(1) , kWriteDelete = BIT(2) }

enum	EDeprecatedStatusBits { kObjInCanvas = BIT(3) }

enum	EStatusBits { kCanDelete = BIT(0) , kMustCleanup = BIT(3) , kIsReferenced = BIT(4) , kHasUUID = BIT(5) , kCannotPick = BIT(6) , kNoContextMenu = BIT(8) , kInvalidObject = BIT(13) }

Public Member Functions
	RooMinimizer (RooAbsReal &function, FcnMode fcnMode=FcnMode::classic)
	Construct MINUIT interface to given function.

	RooMinimizer (std::shared_ptr< RooFit::TestStatistics::RooAbsL > likelihood, RooFit::TestStatistics::LikelihoodMode likelihoodMode=RooFit::TestStatistics::LikelihoodMode::serial, RooFit::TestStatistics::LikelihoodGradientMode likelihoodGradientMode=RooFit::TestStatistics::LikelihoodGradientMode::multiprocess)

	~RooMinimizer () override
	Destructor.

RooPlot *	contour (RooRealVar &var1, RooRealVar &var2, Double_t n1=1, Double_t n2=2, Double_t n3=0, Double_t n4=0, Double_t n5=0, Double_t n6=0, unsigned int npoints=50)
	Create and draw a TH2 with the error contours in the parameters `var1` and `var2`.

Int_t	evalCounter () const

RooFitResult *	fit (const char *options) R__DEPRECATED(6
	Parse traditional RooAbsPdf::fitTo driver options.

ROOT::Fit::Fitter *	fitter ()
	Return underlying ROOT fitter object.

const ROOT::Fit::Fitter *	fitter () const
	Return underlying ROOT fitter object.

virtual RooFitResult *	fitTo (RooAbsData &data, const RooCmdArg &arg1=RooCmdArg::none(), const RooCmdArg &arg2=RooCmdArg::none(), const RooCmdArg &arg3=RooCmdArg::none(), const RooCmdArg &arg4=RooCmdArg::none(), const RooCmdArg &arg5=RooCmdArg::none(), const RooCmdArg &arg6=RooCmdArg::none(), const RooCmdArg &arg7=RooCmdArg::none(), const RooCmdArg &arg8=RooCmdArg::none())
	Fit PDF to given dataset.

virtual RooFitResult *	fitTo (RooAbsData &data, const RooLinkedList &cmdList)
	Fit PDF to given dataset.

ROOT::Math::IMultiGenFunction *	getFitterMultiGenFcn () const

ROOT::Math::IMultiGenFunction *	getMultiGenFcn () const

Int_t	getNPar () const

Int_t	getPrintLevel () const

Int_t	hesse ()
	Execute HESSE.

Int_t	improve ()
	Execute IMPROVE.

Int_t	migrad ()
	Execute MIGRAD.

Int_t	minimize (const char type, const char alg=0)
	Minimise the function passed in the constructor.

Int_t	minos ()
	Execute MINOS.

Int_t	minos (const RooArgSet &minosParamList)
	Execute MINOS for given list of parameters.

void	optimizeConst (Int_t flag)
	If flag is true, perform constant term optimization on function being minimized.

RooFitResult *	save (const char name=0, const char title=0)
	Save and return a RooFitResult snapshot of current minimizer status.

void	saveStatus (const char *label, Int_t status)

Int_t	seek ()
	Execute SEEK.

void	setEps (Double_t eps)
	Change MINUIT epsilon.

void	setErrorLevel (Double_t level)
	Set the level for MINUIT error analysis to the given value.

void	setEvalErrorWall (Bool_t flag)

Bool_t	setLogFile (const char *logf=0)

void	setMaxFunctionCalls (Int_t n)
	Change maximum number of likelihood function calss from MINUIT (RooMinimizer default 500 * #parameters)

void	setMaxIterations (Int_t n)
	Change maximum number of MINUIT iterations (RooMinimizer default 500 * #parameters)

void	setMinimizerType (const char *type)
	Choose the minimizer algorithm.

void	setOffsetting (Bool_t flag)
	Enable internal likelihood offsetting for enhanced numeric precision.

void	setPrintEvalErrors (Int_t numEvalErrors)

Int_t	setPrintLevel (Int_t newLevel)
	Change the MINUIT internal printing level.

void	setProfile (Bool_t flag=kTRUE)

void	setRecoverFromNaNStrength (double strength)
	Try to recover from invalid function values.

void	setStrategy (Int_t strat)
	Change MINUIT strategy to istrat.

void	setVerbose (Bool_t flag=kTRUE)

Int_t	simplex ()
	Execute SIMPLEX.

void	zeroEvalCount ()

Public Member Functions inherited from TObject
	TObject ()
	TObject constructor.

	TObject (const TObject &object)
	TObject copy ctor.

virtual	~TObject ()
	TObject destructor.

void	AbstractMethod (const char *method) const
	Use this method to implement an "abstract" method that you don't want to leave purely abstract.

virtual void	AppendPad (Option_t *option="")
	Append graphics object to current pad.

virtual void	Browse (TBrowser *b)
	Browse object. May be overridden for another default action.

ULong_t	CheckedHash ()
	Check and record whether this class has a consistent Hash/RecursiveRemove setup (*) and then return the regular Hash value for this object.

virtual const char *	ClassName () const
	Returns name of class to which the object belongs.

virtual void	Clear (Option_t *="")

virtual TObject *	Clone (const char *newname="") const
	Make a clone of an object using the Streamer facility.

virtual Int_t	Compare (const TObject *obj) const
	Compare abstract method.

virtual void	Copy (TObject &object) const
	Copy this to obj.

virtual void	Delete (Option_t *option="")
	Delete this object.

virtual Int_t	DistancetoPrimitive (Int_t px, Int_t py)
	Computes distance from point (px,py) to the object.

virtual void	Draw (Option_t *option="")
	Default Draw method for all objects.

virtual void	DrawClass () const
	Draw class inheritance tree of the class to which this object belongs.

virtual TObject *	DrawClone (Option_t *option="") const
	Draw a clone of this object in the current selected pad for instance with: `gROOT->SetSelectedPad(gPad)`.

virtual void	Dump () const
	Dump contents of object on stdout.

virtual void	Error (const char method, const char msgfmt,...) const
	Issue error message.

virtual void	Execute (const char method, const char params, Int_t *error=0)
	Execute method on this object with the given parameter string, e.g.

virtual void	Execute (TMethod method, TObjArray params, Int_t *error=0)
	Execute method on this object with parameters stored in the TObjArray.

virtual void	ExecuteEvent (Int_t event, Int_t px, Int_t py)
	Execute action corresponding to an event at (px,py).

virtual void	Fatal (const char method, const char msgfmt,...) const
	Issue fatal error message.

virtual TObject *	FindObject (const char *name) const
	Must be redefined in derived classes.

virtual TObject *	FindObject (const TObject *obj) const
	Must be redefined in derived classes.

virtual Option_t *	GetDrawOption () const
	Get option used by the graphics system to draw this object.

virtual const char *	GetIconName () const
	Returns mime type name of object.

virtual const char *	GetName () const
	Returns name of object.

virtual char *	GetObjectInfo (Int_t px, Int_t py) const
	Returns string containing info about the object at position (px,py).

virtual Option_t *	GetOption () const

virtual const char *	GetTitle () const
	Returns title of object.

virtual UInt_t	GetUniqueID () const
	Return the unique object id.

virtual Bool_t	HandleTimer (TTimer *timer)
	Execute action in response of a timer timing out.

virtual ULong_t	Hash () const
	Return hash value for this object.

Bool_t	HasInconsistentHash () const
	Return true is the type of this object is known to have an inconsistent setup for Hash and RecursiveRemove (i.e.

virtual void	Info (const char method, const char msgfmt,...) const
	Issue info message.

virtual Bool_t	InheritsFrom (const char *classname) const
	Returns kTRUE if object inherits from class "classname".

virtual Bool_t	InheritsFrom (const TClass *cl) const
	Returns kTRUE if object inherits from TClass cl.

virtual void	Inspect () const
	Dump contents of this object in a graphics canvas.

void	InvertBit (UInt_t f)

Bool_t	IsDestructed () const
	IsDestructed.

virtual Bool_t	IsEqual (const TObject *obj) const
	Default equal comparison (objects are equal if they have the same address in memory).

virtual Bool_t	IsFolder () const
	Returns kTRUE in case object contains browsable objects (like containers or lists of other objects).

R__ALWAYS_INLINE Bool_t	IsOnHeap () const

virtual Bool_t	IsSortable () const

R__ALWAYS_INLINE Bool_t	IsZombie () const

virtual void	ls (Option_t *option="") const
	The ls function lists the contents of a class on stdout.

void	MayNotUse (const char *method) const
	Use this method to signal that a method (defined in a base class) may not be called in a derived class (in principle against good design since a child class should not provide less functionality than its parent, however, sometimes it is necessary).

virtual Bool_t	Notify ()
	This method must be overridden to handle object notification.

void	Obsolete (const char method, const char asOfVers, const char *removedFromVers) const
	Use this method to declare a method obsolete.

void	operator delete (void *ptr)
	Operator delete.

void	operator delete[] (void *ptr)
	Operator delete [].

void *	operator new (size_t sz)

void *	operator new (size_t sz, void *vp)

void *	operator new[] (size_t sz)

void *	operator new[] (size_t sz, void *vp)

TObject &	operator= (const TObject &rhs)
	TObject assignment operator.

virtual void	Paint (Option_t *option="")
	This method must be overridden if a class wants to paint itself.

virtual void	Pop ()
	Pop on object drawn in a pad to the top of the display list.

virtual void	Print (Option_t *option="") const
	This method must be overridden when a class wants to print itself.

virtual Int_t	Read (const char *name)
	Read contents of object with specified name from the current directory.

virtual void	RecursiveRemove (TObject *obj)
	Recursively remove this object from a list.

void	ResetBit (UInt_t f)

virtual void	SaveAs (const char filename="", Option_t option="") const
	Save this object in the file specified by filename.

virtual void	SavePrimitive (std::ostream &out, Option_t *option="")
	Save a primitive as a C++ statement(s) on output stream "out".

void	SetBit (UInt_t f)

void	SetBit (UInt_t f, Bool_t set)
	Set or unset the user status bits as specified in f.

virtual void	SetDrawOption (Option_t *option="")
	Set drawing option for object.

virtual void	SetUniqueID (UInt_t uid)
	Set the unique object id.

virtual void	SysError (const char method, const char msgfmt,...) const
	Issue system error message.

R__ALWAYS_INLINE Bool_t	TestBit (UInt_t f) const

Int_t	TestBits (UInt_t f) const

virtual void	UseCurrentStyle ()
	Set current style settings in this object This function is called when either TCanvas::UseCurrentStyle or TROOT::ForceStyle have been invoked.

virtual void	Warning (const char method, const char msgfmt,...) const
	Issue warning message.

virtual Int_t	Write (const char *name=0, Int_t option=0, Int_t bufsize=0)
	Write this object to the current directory.

virtual Int_t	Write (const char *name=0, Int_t option=0, Int_t bufsize=0) const
	Write this object to the current directory.

Static Public Member Functions
static void	cleanup ()
	Cleanup method called by atexit handler installed by RooSentinel to delete all global heap objects when the program is terminated.

static RooFitResult *	lastMinuitFit (const RooArgList &varList=RooArgList())

Static Public Member Functions inherited from TObject
static Longptr_t	GetDtorOnly ()
	Return destructor only flag.

static Bool_t	GetObjectStat ()
	Get status of object stat flag.

static void	SetDtorOnly (void *obj)
	Set destructor only flag.

static void	SetObjectStat (Bool_t stat)
	Turn on/off tracking of objects in the TObjectTable.

Protected Member Functions
void	applyCovarianceMatrix (TMatrixDSym &V)
	Apply results of given external covariance matrix.

bool	fitFcn () const

RooAbsMinimizerFcn *	fitterFcn ()

const RooAbsMinimizerFcn *	fitterFcn () const

std::ofstream *	logfile ()

Double_t &	maxFCN ()

void	profileStart ()
	Start profiling timer.

void	profileStop ()
	Stop profiling timer and report results of last session.

Protected Member Functions inherited from TObject
virtual void	DoError (int level, const char location, const char fmt, va_list va) const
	Interface to ErrorHandler (protected).

void	MakeZombie ()

Private Member Functions
	RooMinimizer (const RooMinimizer &)

void	initMinimizerFcnDependentPart (double defaultErrorLevel)
	Initialize the part of the minimizer that is dependent on the function to be minimized.

void	initMinimizerFirstPart ()
	Initialize the part of the minimizer that is independent of the function to be minimized.

Private Attributes
TStopwatch	_cumulTimer

TMatrixDSym *	_extV = 0

RooAbsMinimizerFcn *	_fcn

FcnMode	_fcnMode

std::string	_minimizerType = "Minuit"

Int_t	_printLevel = 1

Bool_t	_profile = kFALSE

Bool_t	_profileStart = kFALSE

Int_t	_status = -99

std::vector< std::pair< std::string, int > >	_statusHistory

TStopwatch	_timer

Bool_t	_verbose = kFALSE

Static Private Attributes
static ROOT::Fit::Fitter *	_theFitter = 0

Friends
class	RooAbsPdf

Additional Inherited Members
Protected Types inherited from TObject
enum	{ kOnlyPrepStep = BIT(3) }

#include <RooMinimizer.h>

Inheritance diagram for RooMinimizer:

[legend]

Member Enumeration Documentation

◆ FcnMode

enum class RooMinimizer::FcnMode

strong

Enumerator
classic
gradient
generic_wrapper

Definition at line 53 of file RooMinimizer.h.

◆ PrintLevel

enum RooMinimizer::PrintLevel

Enumerator
None
Reduced
Normal
ExtraForProblem
Maximum

Definition at line 65 of file RooMinimizer.h.

◆ Strategy

enum RooMinimizer::Strategy

Enumerator
Speed
Balance
Robustness

Definition at line 64 of file RooMinimizer.h.

Constructor & Destructor Documentation

◆ RooMinimizer() [1/3]

RooMinimizer::RooMinimizer	(	RooAbsReal &	function,
		FcnMode	fcnMode = `FcnMode::classic`
	)

explicit

Construct MINUIT interface to given function.

Function can be anything, but is typically a -log(likelihood) implemented by RooNLLVar or a chi^2 (implemented by RooChi2Var). Other frequent use cases are a RooAddition of a RooNLLVar plus a penalty or constraint term. This class propagates all RooFit information (floating parameters, their values and errors) to MINUIT before each MINUIT call and propagates all MINUIT information back to the RooFit object at the end of each call (updated parameter values, their (asymmetric errors) etc. The default MINUIT error level for HESSE and MINOS error analysis is taken from the defaultErrorLevel() value of the input function.

Definition at line 109 of file RooMinimizer.cxx.

◆ RooMinimizer() [2/3]

RooMinimizer::RooMinimizer	(	std::shared_ptr< RooFit::TestStatistics::RooAbsL >	likelihood,
		RooFit::TestStatistics::LikelihoodMode	likelihoodMode = `RooFit::TestStatistics::LikelihoodMode::serial`,
		RooFit::TestStatistics::LikelihoodGradientMode	likelihoodGradientMode = `RooFit::TestStatistics::LikelihoodGradientMode::multiprocess`
	)

explicit

Definition at line 135 of file RooMinimizer.cxx.

◆ ~RooMinimizer()

RooMinimizer::~RooMinimizer ( )

override

Destructor.

Definition at line 189 of file RooMinimizer.cxx.

◆ RooMinimizer() [3/3]

RooMinimizer::RooMinimizer ( const RooMinimizer & )

private

Member Function Documentation

◆ applyCovarianceMatrix()

void RooMinimizer::applyCovarianceMatrix ( TMatrixDSym & V )

protected

Apply results of given external covariance matrix.

i.e. propagate its errors to all RRV parameter representations and give this matrix instead of the HESSE matrix at the next save() call

Definition at line 944 of file RooMinimizer.cxx.

◆ cleanup()

void RooMinimizer::cleanup ( )

static

Cleanup method called by atexit handler installed by RooSentinel to delete all global heap objects when the program is terminated.

Definition at line 87 of file RooMinimizer.cxx.

◆ contour()

RooPlot * RooMinimizer::contour	(	RooRealVar &	var1,
		RooRealVar &	var2,
		Double_t	n1 = `1`,
		Double_t	n2 = `2`,
		Double_t	n3 = `0`,
		Double_t	n4 = `0`,
		Double_t	n5 = `0`,
		Double_t	n6 = `0`,
		unsigned int	npoints = `50`
	)

Create and draw a TH2 with the error contours in the parameters var1 and var2.

Parameters

[in]	var1	The first parameter (x axis).
[in]	var2	The second parameter (y axis).
[in]	n1	First contour.
[in]	n2	Optional contour. 0 means don't draw.
[in]	n3	Optional contour. 0 means don't draw.
[in]	n4	Optional contour. 0 means don't draw.
[in]	n5	Optional contour. 0 means don't draw.
[in]	n6	Optional contour. 0 means don't draw.
[in]	npoints	Number of points for evaluating the contour.

Up to six contours can be drawn using the arguments n1 to n6 to request the desired coverage in units of \( \sigma = n^2 \cdot \mathrm{ErrorDef} \). See ROOT::Math::Minimizer::ErrorDef().

Definition at line 758 of file RooMinimizer.cxx.

◆ evalCounter()

Int_t RooMinimizer::evalCounter ( ) const

inline

Definition at line 110 of file RooMinimizer.h.

◆ fit()

RooFitResult * RooMinimizer::fit ( const char * options )

Parse traditional RooAbsPdf::fitTo driver options.

m - Run Migrad only h - Run Hesse to estimate errors v - Verbose mode l - Log parameters after each Minuit steps to file t - Activate profile timer r - Save fit result 0 - Run Migrad with strategy 0

Definition at line 318 of file RooMinimizer.cxx.

◆ fitFcn()

bool RooMinimizer::fitFcn ( ) const

protected

Definition at line 340 of file RooMinimizer.cxx.

◆ fitter() [1/2]

ROOT::Fit::Fitter * RooMinimizer::fitter ( )

Return underlying ROOT fitter object.

Definition at line 291 of file RooMinimizer.cxx.

◆ fitter() [2/2]

const ROOT::Fit::Fitter * RooMinimizer::fitter ( ) const

Return underlying ROOT fitter object.

Definition at line 300 of file RooMinimizer.cxx.

◆ fitterFcn() [1/2]

RooAbsMinimizerFcn * RooMinimizer::fitterFcn ( )

protected

Definition at line 932 of file RooMinimizer.cxx.

◆ fitterFcn() [2/2]

const RooAbsMinimizerFcn * RooMinimizer::fitterFcn ( ) const

protected

Definition at line 910 of file RooMinimizer.cxx.

◆ fitTo() [1/2]

RooAbsPdf::fitTo	(	RooAbsData &	data,
		const RooCmdArg &	arg1 = `RooCmdArg::none()`,
		const RooCmdArg &	arg2 = `RooCmdArg::none()`,
		const RooCmdArg &	arg3 = `RooCmdArg::none()`,
		const RooCmdArg &	arg4 = `RooCmdArg::none()`,
		const RooCmdArg &	arg5 = `RooCmdArg::none()`,
		const RooCmdArg &	arg6 = `RooCmdArg::none()`,
		const RooCmdArg &	arg7 = `RooCmdArg::none()`,
		const RooCmdArg &	arg8 = `RooCmdArg::none()`
	)

Fit PDF to given dataset.

If dataset is unbinned, an unbinned maximum likelihood is performed. If the dataset is binned, a binned maximum likelihood is performed. By default the fit is executed through the MINUIT commands MIGRAD, HESSE in succession.

Parameters

[in]	data	Data to fit the PDF to
[in]	arg1	One or more arguments to control the behaviour of the fit

Returns: RooFitResult with fit status and parameters if option Save() is used, nullptr otherwise. The user takes ownership of the fit result.

The following named arguments are supported

Type of CmdArg Options to control construction of -log(L)

ConditionalObservables(Args_t &&... argsOrArgSet) Do not normalize PDF over listed observables.

Extended(Bool_t flag) Add extended likelihood term, off by default

Range(const char* name)

Fit only data inside range with given name. Multiple comma-separated range names can be specified. In this case, the unnormalized PDF \(f(x)\) is normalized by the integral over all ranges \(r_i\):

\[ p(x) = \frac{f(x)}{\sum_i \int_{r_i} f(x) dx}. \]

Range(Double_t lo, Double_t hi) Fit only data inside given range. A range named "fit" is created on the fly on all observables.

SumCoefRange(const char* name) Set the range in which to interpret the coefficients of RooAddPdf components

NumCPU(int num, int strat)

Parallelize NLL calculation on num CPUs

Strategy	Effect
0 = RooFit::BulkPartition (Default)	Divide events in N equal chunks
1 = RooFit::Interleave	Process event iN in process N. Recommended for binned data with a substantial number of zero-bins, which will be distributed across processes more equitably in this strategy
2 = RooFit::SimComponents	Process each component likelihood of a RooSimultaneous fully in a single process and distribute components over processes. This approach can be benificial if normalization calculation time dominates the total computation time of a component (since the normalization calculation must be performed in each process in strategies 0 and 1. However beware that if the RooSimultaneous components do not share many parameters this strategy is inefficient: as most minuit-induced likelihood calculations involve changing a single parameter, only 1 of the N processes will be active most of the time if RooSimultaneous components do not share many parameters
3 = RooFit::Hybrid	Follow strategy 0 for all RooSimultaneous components, except those with less than 30 dataset entries, for which strategy 2 is followed.

SplitRange(Bool_t flag)

Use separate fit ranges in a simultaneous fit. Actual range name for each subsample is assumed to by rangeName_indexState where indexState is the state of the master index category of the simultaneous fit. Using Range("range"), SplitRange() as switches, different ranges could be set like this:

myVariable.setRange("range_pi0", 135, 210);

myVariable.setRange("range_gamma", 50, 210);

Constrain(const RooArgSet&pars) For p.d.f.s that contain internal parameter constraint terms (that is usually product PDFs, where one term of the product depends on parameters but not on the observable(s),), only apply constraints to the given subset of parameters.

ExternalConstraints(const RooArgSet& ) Include given external constraints to likelihood by multiplying them with the original likelihood.

GlobalObservables(const RooArgSet&) Define the set of normalization observables to be used for the constraint terms. If none are specified the constrained parameters are used.

Offset(Bool_t) Offset likelihood by initial value (so that starting value of FCN in minuit is zero). This can improve numeric stability in simultaneously fits with components with large likelihood values

BatchMode(bool on) Experimental batch evaluation mode. This computes a batch of likelihood values at a time, uses faster math functions and possibly auto vectorisation (this depends on the compiler flags). Depending on hardware capabilities, the compiler flags and whether a batch evaluation function was implemented for the PDFs of the model, likelihood computations are 2x to 10x faster. The relative difference of the single log-likelihoods w.r.t. the legacy mode is usually better than 1.E-12, and fit parameters usually agree to better than 1.E-6.

IntegrateBins(double precision)

In binned fits, integrate the PDF over the bins instead of using the probability density at the bin centre. This can reduce the bias observed when fitting functions with high curvature to binned data.

precision > 0: Activate bin integration everywhere. Use precision between 0.01 and 1.E-6, depending on binning. Note that a low precision such as 0.01 might yield identical results to 1.E-4, since the integrator might reach 1.E-4 already in its first integration step. If lower precision is desired (more speed), a RooBinSamplingPdf has to be created manually, and its integrator has to be manipulated directly.
precision = 0: Activate bin integration only for continuous PDFs fit to a RooDataHist.
precision < 0: Deactivate.
See also
RooBinSamplingPdf

Options to control flow of fit procedure

Minimizer("<type>", "<algo>")

Choose minimization package and optionally the algorithm to use. Default is MINUIT/MIGRAD through the RooMinimizer interface, but others can be specified (through RooMinimizer interface).

Type	Algorithm
Minuit	migrad, simplex, minimize (=migrad+simplex), migradimproved (=migrad+improve)
Minuit2	migrad, simplex, minimize, scan
GSLMultiMin	conjugatefr, conjugatepr, bfgs, bfgs2, steepestdescent
GSLSimAn	-

InitialHesse(Bool_t flag) Flag controls if HESSE before MIGRAD as well, off by default

Optimize(Bool_t flag) Activate constant term optimization of test statistic during minimization (on by default)

Hesse(Bool_t flag) Flag controls if HESSE is run after MIGRAD, on by default

Minos(Bool_t flag) Flag controls if MINOS is run after HESSE, off by default

Minos(const RooArgSet& set) Only run MINOS on given subset of arguments

Save(Bool_t flag) Flag controls if RooFitResult object is produced and returned, off by default

Strategy(Int_t flag) Set Minuit strategy (0 to 2, default is 1)

EvalErrorWall(bool flag=true) When parameters are in disallowed regions (e.g. PDF is negative), return very high value to fitter to force it out of that region. This can, however, mean that the fitter gets lost in this region. If this happens, try switching it off.

RecoverFromUndefinedRegions(double strength) When PDF is invalid (e.g. parameter in undefined region), try to direct minimiser away from that region. strength controls the magnitude of the penalty term. Leaving out this argument defaults to 10. Switch off with strength = 0..

FitOptions(const char* optStr)

Deprecated:: Steer fit with classic options string (for backward compatibility).

Attention: Use of this option excludes use of any of the new style steering options.

SumW2Error(Bool_t flag)

Apply correction to errors and covariance matrix. This uses two covariance matrices, one with the weights, the other with squared weights, to obtain the correct errors for weighted likelihood fits. If this option is activated, the corrected covariance matrix is calculated as \( V_\mathrm{corr} = V C^{-1} V \), where \( V \) is the original covariance matrix and \( C \) is the inverse of the covariance matrix calculated using the squared weights. This allows to switch between two interpretations of errors:

SumW2Error

Interpretation

true

The errors reflect the uncertainty of the Monte Carlo simulation. Use this if you want to know how much accuracy you can get from the available Monte Carlo statistics.

Example: Simulation with 1000 events, the average weight is 0.1. The errors are as big as if one fitted to 1000 events.

false

The errors reflect the errors of a dataset, which is as big as the sum of weights. Use this if you want to know what statistical errors you would get if you had a dataset with as many events as the (weighted) Monte Carlo simulation represents.

Example (Data as above): The errors are as big as if one fitted to 100 events.

Note: If the SumW2Error correction is enabled, the covariance matrix quality stored in the RooFitResult object will be the minimum of the original covariance matrix quality and the quality of the covariance matrix calculated with the squared weights.

AsymptoticError() Use the asymptotically correct approach to estimate errors in the presence of weights. This is slower but more accurate than SumW2Error. See also https://arxiv.org/abs/1911.01303).

PrefitDataFraction(double fraction)

Runs a prefit on a small dataset of size fraction*(actual data size). This can speed up fits by finding good starting values for the parameters for the actual fit.

Warning: Prefitting may give bad results when used in binned analysis.

Options to control informational output

Verbose(Bool_t flag) Flag controls if verbose output is printed (NLL, parameter changes during fit).

Timer(Bool_t flag) Time CPU and wall clock consumption of fit steps, off by default.

PrintLevel(Int_t level) Set Minuit print level (-1 to 3, default is 1). At -1 all RooFit informational messages are suppressed as well. See RooMinimizer::PrintLevel for the meaning of the levels.

Warnings(Bool_t flag) Enable or disable MINUIT warnings (enabled by default)

PrintEvalErrors(Int_t numErr) Control number of p.d.f evaluation errors printed per likelihood evaluation. A negative value suppresses output completely, a zero value will only print the error count per p.d.f component, a positive value will print details of each error up to numErr messages per p.d.f component.

PyROOT

The RooAbsPdf::fitTo() function is pythonized with the command argument pythonization. The keywords must correspond to the CmdArgs of the function.

Definition at line 158 of file RooAbsPdf.cxx.

◆ fitTo() [2/2]

RooFitResult * RooAbsPdf::fitTo	(	RooAbsData &	data,
		const RooLinkedList &	cmdList
	)

Fit PDF to given dataset.

If dataset is unbinned, an unbinned maximum likelihood is performed. If the dataset is binned, a binned maximum likelihood is performed. By default the fit is executed through the MINUIT commands MIGRAD, HESSE and MINOS in succession.

See RooAbsPdf::fitTo(RooAbsData&,RooCmdArg&,RooCmdArg&,RooCmdArg&,RooCmdArg&,RooCmdArg&,RooCmdArg&,RooCmdArg&,RooCmdArg&)

for documentation of options

Definition at line 161 of file RooAbsPdf.cxx.

◆ getFitterMultiGenFcn()

ROOT::Math::IMultiGenFunction * RooMinimizer::getFitterMultiGenFcn ( ) const

Definition at line 881 of file RooMinimizer.cxx.

◆ getMultiGenFcn()

ROOT::Math::IMultiGenFunction * RooMinimizer::getMultiGenFcn ( ) const

Definition at line 887 of file RooMinimizer.cxx.

◆ getNPar()

Int_t RooMinimizer::getNPar ( ) const

inline

Definition at line 119 of file RooMinimizer.h.

◆ getPrintLevel()

Int_t RooMinimizer::getPrintLevel ( ) const

Definition at line 1059 of file RooMinimizer.cxx.

◆ hesse()

Int_t RooMinimizer::hesse ( )

Execute HESSE.

Changes in parameter values and calculated errors are automatically propagated back the RooRealVars representing the floating parameters in the MINUIT operation.

Definition at line 431 of file RooMinimizer.cxx.

◆ improve()

Int_t RooMinimizer::improve ( )

Execute IMPROVE.

Changes in parameter values and calculated errors are automatically propagated back the RooRealVars representing the floating parameters in the MINUIT operation.

Definition at line 623 of file RooMinimizer.cxx.

◆ initMinimizerFcnDependentPart()

void RooMinimizer::initMinimizerFcnDependentPart ( double defaultErrorLevel )

private

Initialize the part of the minimizer that is dependent on the function to be minimized.

Definition at line 162 of file RooMinimizer.cxx.

◆ initMinimizerFirstPart()

void RooMinimizer::initMinimizerFirstPart ( )

private

Initialize the part of the minimizer that is independent of the function to be minimized.

Definition at line 149 of file RooMinimizer.cxx.

◆ lastMinuitFit()

RooFitResult * RooMinimizer::lastMinuitFit ( const RooArgList & varList = RooArgList() )

static

Definition at line 953 of file RooMinimizer.cxx.

◆ logfile()

std::ofstream * RooMinimizer::logfile ( )

inlineprotected

Definition at line 129 of file RooMinimizer.h.

◆ maxFCN()

Double_t & RooMinimizer::maxFCN ( )

inlineprotected

Definition at line 130 of file RooMinimizer.h.

◆ migrad()

Int_t RooMinimizer::migrad ( )

Execute MIGRAD.

Changes in parameter values and calculated errors are automatically propagated back the RooRealVars representing the floating parameters in the MINUIT operation.

Definition at line 402 of file RooMinimizer.cxx.

◆ minimize()

Int_t RooMinimizer::minimize	(	const char *	type,
		const char *	alg = `0`
	)

Minimise the function passed in the constructor.

Parameters

[in] type Type of fitter to use, e.g. "Minuit" "Minuit2".

Attention: This overrides the default fitter of this RooMinimizer.

Parameters

[in] alg Fit algorithm to use. (Optional)

Definition at line 370 of file RooMinimizer.cxx.

◆ minos() [1/2]

Int_t RooMinimizer::minos ( )

Execute MINOS.

Changes in parameter values and calculated errors are automatically propagated back the RooRealVars representing the floating parameters in the MINUIT operation.

Definition at line 468 of file RooMinimizer.cxx.

◆ minos() [2/2]

Int_t RooMinimizer::minos ( const RooArgSet & minosParamList )

Execute MINOS for given list of parameters.

Changes in parameter values and calculated errors are automatically propagated back the RooRealVars representing the floating parameters in the MINUIT operation.

Definition at line 506 of file RooMinimizer.cxx.

◆ optimizeConst()

void RooMinimizer::optimizeConst ( Int_t flag )

If flag is true, perform constant term optimization on function being minimized.

Definition at line 661 of file RooMinimizer.cxx.

◆ profileStart()

void RooMinimizer::profileStart ( )

protected

Start profiling timer.

Definition at line 857 of file RooMinimizer.cxx.

◆ profileStop()

void RooMinimizer::profileStop ( )

protected

Stop profiling timer and report results of last session.

Definition at line 870 of file RooMinimizer.cxx.

◆ save()

RooFitResult * RooMinimizer::save	(	const char *	userName = `0`,
		const char *	userTitle = `0`
	)

Save and return a RooFitResult snapshot of current minimizer status.

This snapshot contains the values of all constant parameters, the value of all floating parameters at RooMinimizer construction and after the last MINUIT operation, the MINUIT status, variance quality, EDM setting, number of calls with evaluation problems, the minimized function value and the full correlation matrix.

Definition at line 676 of file RooMinimizer.cxx.

◆ saveStatus()

void RooMinimizer::saveStatus	(	const char *	label,
		Int_t	status
	)

inline

Definition at line 108 of file RooMinimizer.h.

◆ seek()

Int_t RooMinimizer::seek ( )

Execute SEEK.

Changes in parameter values and calculated errors are automatically propagated back the RooRealVars representing the floating parameters in the MINUIT operation.

Definition at line 565 of file RooMinimizer.cxx.

◆ setEps()

void RooMinimizer::setEps ( Double_t eps )

Change MINUIT epsilon.

Definition at line 258 of file RooMinimizer.cxx.

◆ setErrorLevel()

void RooMinimizer::setErrorLevel ( Double_t level )

Set the level for MINUIT error analysis to the given value.

This function overrides the default value that is taken in the RooMinimizer constructor from the defaultErrorLevel() method of the input function

Definition at line 247 of file RooMinimizer.cxx.

◆ setEvalErrorWall()

void RooMinimizer::setEvalErrorWall ( Bool_t flag )

inline

Definition at line 70 of file RooMinimizer.h.

◆ setLogFile()

Bool_t RooMinimizer::setLogFile ( const char * logf = 0 )

inline

Definition at line 99 of file RooMinimizer.h.

◆ setMaxFunctionCalls()

void RooMinimizer::setMaxFunctionCalls ( Int_t n )

Change maximum number of likelihood function calss from MINUIT (RooMinimizer default 500 * #parameters)

Definition at line 233 of file RooMinimizer.cxx.

◆ setMaxIterations()

void RooMinimizer::setMaxIterations ( Int_t n )

Change maximum number of MINUIT iterations (RooMinimizer default 500 * #parameters)

Definition at line 221 of file RooMinimizer.cxx.

◆ setMinimizerType()

void RooMinimizer::setMinimizerType ( const char * type )

Choose the minimizer algorithm.

Definition at line 277 of file RooMinimizer.cxx.

◆ setOffsetting()

void RooMinimizer::setOffsetting ( Bool_t flag )

Enable internal likelihood offsetting for enhanced numeric precision.

Definition at line 268 of file RooMinimizer.cxx.

◆ setPrintEvalErrors()

void RooMinimizer::setPrintEvalErrors ( Int_t numEvalErrors )

inline

Definition at line 96 of file RooMinimizer.h.

◆ setPrintLevel()

Int_t RooMinimizer::setPrintLevel ( Int_t newLevel )

Change the MINUIT internal printing level.

Definition at line 649 of file RooMinimizer.cxx.

◆ setProfile()

void RooMinimizer::setProfile ( Bool_t flag = kTRUE )

inline

Definition at line 98 of file RooMinimizer.h.

◆ setRecoverFromNaNStrength()

void RooMinimizer::setRecoverFromNaNStrength ( double strength )

inline

Try to recover from invalid function values.

When invalid function values are encountered, a penalty term is returned to the minimiser to make it back off. This sets the strength of this penalty.

Note: A strength of zero is equivalent to a constant penalty (= the gradient vanishes, ROOT < 6.24). Positive values lead to a gradient pointing away from the undefined regions. Use ~10 to force the minimiser away from invalid function values.

Definition at line 72 of file RooMinimizer.h.

◆ setStrategy()

void RooMinimizer::setStrategy ( Int_t istrat )

Change MINUIT strategy to istrat.

Accepted codes are 0,1,2 and represent MINUIT strategies for dealing most efficiently with fast FCNs (0), expensive FCNs (2) and 'intermediate' FCNs (1)

Definition at line 209 of file RooMinimizer.cxx.

◆ setVerbose()

void RooMinimizer::setVerbose ( Bool_t flag = kTRUE )

inline

Definition at line 97 of file RooMinimizer.h.

◆ simplex()

Int_t RooMinimizer::simplex ( )

Execute SIMPLEX.

Changes in parameter values and calculated errors are automatically propagated back the RooRealVars representing the floating parameters in the MINUIT operation.

Definition at line 594 of file RooMinimizer.cxx.

◆ zeroEvalCount()

void RooMinimizer::zeroEvalCount ( )

inline

Definition at line 111 of file RooMinimizer.h.

Friends And Related Symbol Documentation

◆ RooAbsPdf

friend class RooAbsPdf

friend

Definition at line 123 of file RooMinimizer.h.

Member Data Documentation

◆ _cumulTimer

TStopwatch RooMinimizer::_cumulTimer

private

Definition at line 148 of file RooMinimizer.h.

◆ _extV

TMatrixDSym* RooMinimizer::_extV = 0

private

Definition at line 151 of file RooMinimizer.h.

◆ _fcn

RooAbsMinimizerFcn* RooMinimizer::_fcn

private

Definition at line 153 of file RooMinimizer.h.

◆ _fcnMode

FcnMode RooMinimizer::_fcnMode

private

Definition at line 155 of file RooMinimizer.h.

◆ _minimizerType

std::string RooMinimizer::_minimizerType = "Minuit"

private

Definition at line 154 of file RooMinimizer.h.

◆ _printLevel

Int_t RooMinimizer::_printLevel = 1

private

Definition at line 142 of file RooMinimizer.h.

◆ _profile

Bool_t RooMinimizer::_profile = kFALSE

private

Definition at line 144 of file RooMinimizer.h.

◆ _profileStart

Bool_t RooMinimizer::_profileStart = kFALSE

private

Definition at line 149 of file RooMinimizer.h.

◆ _status

Int_t RooMinimizer::_status = -99

private

Definition at line 143 of file RooMinimizer.h.

◆ _statusHistory

std::vector<std::pair<std::string,int> > RooMinimizer::_statusHistory

private

Definition at line 159 of file RooMinimizer.h.

◆ _theFitter

ROOT::Fit::Fitter * RooMinimizer::_theFitter = 0

staticprivate

Definition at line 157 of file RooMinimizer.h.

◆ _timer

TStopwatch RooMinimizer::_timer

private

Definition at line 147 of file RooMinimizer.h.

◆ _verbose

Bool_t RooMinimizer::_verbose = kFALSE

private

Definition at line 146 of file RooMinimizer.h.

Libraries for RooMinimizer:

[legend]

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

roofit/roofitcore/inc/RooMinimizer.h
roofit/roofitcore/src/RooMinimizer.cxx

Public Types

Public Member Functions

Static Public Member Functions

Protected Member Functions

Private Member Functions

Private Attributes

Static Private Attributes

Friends

Additional Inherited Members

Member Enumeration Documentation

◆ FcnMode

◆ PrintLevel

◆ Strategy

Constructor & Destructor Documentation

◆ RooMinimizer() [1/3]

◆ RooMinimizer() [2/3]

◆ ~RooMinimizer()

◆ RooMinimizer() [3/3]

Member Function Documentation

◆ applyCovarianceMatrix()

◆ cleanup()

◆ contour()

◆ evalCounter()

◆ fit()

◆ fitFcn()

◆ fitter() [1/2]

◆ fitter() [2/2]

◆ fitterFcn() [1/2]

◆ fitterFcn() [2/2]

◆ fitTo() [1/2]

PyROOT

◆ fitTo() [2/2]

◆ getFitterMultiGenFcn()

◆ getMultiGenFcn()

◆ getNPar()

◆ getPrintLevel()

◆ hesse()

◆ improve()

◆ initMinimizerFcnDependentPart()

◆ initMinimizerFirstPart()

◆ lastMinuitFit()

◆ logfile()

◆ maxFCN()

◆ migrad()

◆ minimize()

◆ minos() [1/2]

◆ minos() [2/2]

◆ optimizeConst()

◆ profileStart()

◆ profileStop()

◆ save()

◆ saveStatus()

◆ seek()

◆ setEps()

◆ setErrorLevel()

◆ setEvalErrorWall()

◆ setLogFile()

◆ setMaxFunctionCalls()

◆ setMaxIterations()

◆ setMinimizerType()

◆ setOffsetting()

◆ setPrintEvalErrors()

◆ setPrintLevel()

◆ setProfile()

◆ setRecoverFromNaNStrength()

◆ setStrategy()

◆ setVerbose()

◆ simplex()

◆ zeroEvalCount()

Friends And Related Symbol Documentation

◆ RooAbsPdf

Member Data Documentation

◆ _cumulTimer

◆ _extV

◆ _fcn

◆ _fcnMode

◆ _minimizerType

◆ _printLevel

◆ _profile

◆ _profileStart