Function which has parameters.

For example, one could define a one-dimensional Gaussian, by considering x as an input coordinate for the evaluation of the function, and the mean and the square root of the variance as parameters.

AS OF NOW PARAMETRICFUNCTION INHERITS FROM FCNBASE INSTEAD OF GENERICFUNCTION. THIS IS ONLY BECAUSE NUMERICAL2PGRADIENTCALCULATOR NEEDS AN FCNBASE OBJECT AND WILL BE CHANGED!!!!!!!!!!!!!!!!

Definition at line 43 of file ParametricFunction.h.

Public Member Functions
	ParametricFunction (int nparams)
	Constructor which initializes the ParametricFunction by setting the number of parameters.
	ParametricFunction (std::span< const double > params)
	Constructor which initializes the ParametricFunction with the parameters given as input.
virtual double	ErrorDef () const
	Error definition of the function.
virtual std::vector< double >	G2 (std::vector< double > const &) const
	Return the diagonal elements of the Hessian (second derivatives).
virtual std::vector< double >	GetGradient (std::vector< double > const &x) const
	Member function returning the Gradient of the function with respect to its variables (but without including gradients with respect to its internal parameters).
virtual const std::vector< double > &	GetParameters () const
	Accessor for the state of the parameters.
virtual std::vector< double >	Gradient (std::vector< double > const &) const
	Return the gradient vector of the function at the given parameter point.
virtual std::vector< double >	GradientWithPrevResult (std::vector< double > const &parameters, double , double , double *) const
virtual GradientParameterSpace	gradParameterSpace () const
virtual bool	HasG2 () const
virtual bool	HasGradient () const
virtual bool	HasHessian () const
virtual std::vector< double >	Hessian (std::vector< double > const &) const
	Return the full Hessian matrix of the function.
virtual unsigned int	NumberOfParameters () const
	Accessor for the number of parameters.
double	operator() (std::vector< double > const &x) const override=0
	Evaluates the function with the given coordinates.
virtual double	operator() (std::vector< double > const &x, std::vector< double > const &params) const
	Evaluates the function with the given coordinates and Parameter values.
virtual void	SetErrorDef (double)
	add interface to set dynamically a new error definition Re-implement this function if needed.
virtual void	SetParameters (std::vector< double > const &params) const
	Sets the parameters of the ParametricFunction.
virtual double	Up () const =0
	Error definition of the function.

Protected Attributes
std::vector< double >	par
	The vector containing the parameters of the function It is mutable for "historical reasons" as in the hierarchy methods and classes are const and all the implications of changing them back to non-const are not clear.

#include <Minuit2/ParametricFunction.h>

Inheritance diagram for ROOT::Minuit2::ParametricFunction:

Constructor & Destructor Documentation

◆ ParametricFunction() [1/2]

ROOT::Minuit2::ParametricFunction::ParametricFunction ( std::span< const double > params )

inline

Constructor which initializes the ParametricFunction with the parameters given as input.

Parameters

params vector containing the initial Parameter values

Definition at line 55 of file ParametricFunction.h.

◆ ParametricFunction() [2/2]

ROOT::Minuit2::ParametricFunction::ParametricFunction ( int nparams )

inline

Constructor which initializes the ParametricFunction by setting the number of parameters.

Parameters

nparams number of parameters of the parametric function

Definition at line 66 of file ParametricFunction.h.

Member Function Documentation

◆ ErrorDef()

virtual double ROOT::Minuit2::FCNBase::ErrorDef ( ) const

inlinevirtualinherited

Error definition of the function.

MINUIT defines Parameter errors as the change in Parameter Value required to change the function Value by up. Normally, for chisquared fits it is 1, and for negative log likelihood, its Value is 0.5. If the user wants instead the 2-sigma errors for chisquared fits, it becomes 4, as Chi2(x+n*sigma) = Chi2(x) + n*n.

Comment a little bit better with links!!!!!!!!!!!!!!!!!

Definition at line 70 of file FCNBase.h.

◆ G2()

virtual std::vector< double > ROOT::Minuit2::FCNBase::G2 ( std::vector< double > const & ) const

inlinevirtualinherited

Return the diagonal elements of the Hessian (second derivatives).

By default, returns an empty vector. Override this method if analytic second derivatives (per-parameter curvature) are available.

Parameters

v	Parameter vector.

Returns: Vector of second derivatives with respect to each parameter.

Reimplemented in ROOT::Minuit2::FCNAdapter.

Definition at line 116 of file FCNBase.h.

◆ GetGradient()

std::vector< double > ROOT::Minuit2::ParametricFunction::GetGradient ( std::vector< double > const & x ) const

virtual

Member function returning the Gradient of the function with respect to its variables (but without including gradients with respect to its internal parameters).

Parameters

x	vector containing the coordinates of the point where the Gradient is to be calculated.

Returns: the Gradient vector of the function at the given point.

Definition at line 22 of file ParametricFunction.cxx.

◆ GetParameters()

virtual const std::vector< double > & ROOT::Minuit2::ParametricFunction::GetParameters ( ) const

inlinevirtual

Accessor for the state of the parameters.

Returns: vector containing the present Parameter settings

Definition at line 91 of file ParametricFunction.h.

◆ Gradient()

virtual std::vector< double > ROOT::Minuit2::FCNBase::Gradient ( std::vector< double > const & ) const

inlinevirtualinherited

Return the gradient vector of the function at the given parameter point.

By default, returns an empty vector (no analytic gradient provided). Override this method if an analytic gradient is available.

Parameters

v	Parameter vector.

Returns: Gradient vector with respect to the parameters.

Reimplemented in ROOT::Minuit2::FCNAdapter, and ROOT::Minuit2::FumiliFCNBase.

Definition at line 95 of file FCNBase.h.

◆ GradientWithPrevResult()

virtual std::vector< double > ROOT::Minuit2::FCNBase::GradientWithPrevResult	(	std::vector< double > const &	parameters,
		double *	,
		double *	,
		double *	) const

inlinevirtualinherited

Warning: Not meant to be overridden! This is a requirement for an internal optimization in RooFit that might go away with any refactoring.

Definition at line 99 of file FCNBase.h.

◆ gradParameterSpace()

virtual GradientParameterSpace ROOT::Minuit2::FCNBase::gradParameterSpace ( ) const

inlinevirtualinherited

Warning: Not meant to be overridden! This is a requirement for an internal optimization in RooFit that might go away with any refactoring.

Definition at line 107 of file FCNBase.h.

◆ HasG2()

virtual bool ROOT::Minuit2::FCNBase::HasG2 ( ) const

inlinevirtualinherited

Reimplemented in ROOT::Minuit2::FCNAdapter.

Definition at line 129 of file FCNBase.h.

◆ HasGradient()

virtual bool ROOT::Minuit2::FCNBase::HasGradient ( ) const

inlinevirtualinherited

Reimplemented in ROOT::Minuit2::FCNAdapter, ROOT::Minuit2::FCNGradientBase, and ROOT::Minuit2::FumiliFCNBase.

Definition at line 86 of file FCNBase.h.

◆ HasHessian()

virtual bool ROOT::Minuit2::FCNBase::HasHessian ( ) const

inlinevirtualinherited

Reimplemented in ROOT::Minuit2::FCNAdapter.

Definition at line 127 of file FCNBase.h.

◆ Hessian()

virtual std::vector< double > ROOT::Minuit2::FCNBase::Hessian ( std::vector< double > const & ) const

inlinevirtualinherited

Return the full Hessian matrix of the function.

By default, returns an empty vector. Override this method if the full analytic Hessian (matrix of second derivatives) is available.

Parameters

v	Parameter vector.

Returns: Flattened Hessian matrix.

Reimplemented in ROOT::Minuit2::FCNAdapter, and ROOT::Minuit2::FumiliFCNBase.

Definition at line 125 of file FCNBase.h.

◆ NumberOfParameters()

virtual unsigned int ROOT::Minuit2::ParametricFunction::NumberOfParameters ( ) const

inlinevirtual

Accessor for the number of parameters.

Returns: the number of function parameters

Definition at line 100 of file ParametricFunction.h.

◆ operator()() [1/2]

double ROOT::Minuit2::ParametricFunction::operator() ( std::vector< double > const & x ) const

overridepure virtual

Evaluates the function with the given coordinates.

Parameters

x	vector containing the input coordinates

Returns: the result of the function evaluation with the given coordinates.

Implements ROOT::Minuit2::FCNBase.

◆ operator()() [2/2]

virtual double ROOT::Minuit2::ParametricFunction::operator()	(	std::vector< double > const &	x,
		std::vector< double > const &	params ) const

inlinevirtual

Evaluates the function with the given coordinates and Parameter values.

This member function is useful to implement when speed is an issue as it is faster to call only one function instead of two (SetParameters and operator()). The default implementation, provided for convenience, does the latter.

Parameters

x	vector containing the input coordinates
params	vector containing the Parameter values

Returns: the result of the function evaluation with the given coordinates and parameters

Definition at line 135 of file ParametricFunction.h.

◆ SetErrorDef()

virtual void ROOT::Minuit2::FCNBase::SetErrorDef ( double )

inlinevirtualinherited

add interface to set dynamically a new error definition Re-implement this function if needed.

Reimplemented in ROOT::Minuit2::FCNAdapter, and ROOT::Minuit2::FumiliFCNAdapter< Function >.

Definition at line 84 of file FCNBase.h.

◆ SetParameters()

virtual void ROOT::Minuit2::ParametricFunction::SetParameters ( std::vector< double > const & params ) const

inlinevirtual

Sets the parameters of the ParametricFunction.

Parameters

params vector containing the Parameter values

Definition at line 76 of file ParametricFunction.h.

◆ Up()

virtual double ROOT::Minuit2::FCNBase::Up ( ) const

pure virtualinherited

Error definition of the function.

MINUIT defines Parameter errors as the change in Parameter Value required to change the function Value by up. Normally, for chisquared fits it is 1, and for negative log likelihood, its Value is 0.5. If the user wants instead the 2-sigma errors for chisquared fits, it becomes 4, as Chi2(x+n*sigma) = Chi2(x) + n*n.

Implemented in ROOT::Minuit2::FCNAdapter, ROOT::Minuit2::FumiliChi2FCN, ROOT::Minuit2::FumiliFCNAdapter< Function >, and ROOT::Minuit2::FumiliMaximumLikelihoodFCN.

Member Data Documentation

◆ par

std::vector<double> ROOT::Minuit2::ParametricFunction::par

mutableprotected

The vector containing the parameters of the function It is mutable for "historical reasons" as in the hierarchy methods and classes are const and all the implications of changing them back to non-const are not clear.

Definition at line 166 of file ParametricFunction.h.

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

math/minuit2/inc/Minuit2/ParametricFunction.h
math/minuit2/src/ParametricFunction.cxx

Public Member Functions

Protected Attributes

Constructor & Destructor Documentation

◆ ParametricFunction() [1/2]

◆ ParametricFunction() [2/2]

Member Function Documentation

◆ ErrorDef()

◆ G2()

◆ GetGradient()

◆ GetParameters()

◆ Gradient()

◆ GradientWithPrevResult()

◆ gradParameterSpace()

◆ HasG2()

◆ HasGradient()

◆ HasHessian()

◆ Hessian()

◆ NumberOfParameters()

◆ operator()() [1/2]

◆ operator()() [2/2]

◆ SetErrorDef()

◆ SetParameters()

◆ Up()

Member Data Documentation

◆ par