Optimizer minimizes a multivariate function. More...

#include <geogram/numerics/optimizer.h>

Inheritance diagram for GEO::Optimizer:

Public Types
typedef void(*	funcgrad_callback) (index_t N, double x, double &f, double g)
	Optimizer callback that evaluates a function and its gradient. More...

typedef void(*	newiteration_callback) (index_t N, const double x, double f, const double g, double gnorm)
	Optimizer callback that is called at each iteration. More...

typedef void(*	evalhessian_callback) (index_t N, double x, double &f, double g, HESSIAN_MATRIX &hessian)
	Optimizer callback that evaluates a function, its gradient and its Hessian. More...

Public Member Functions
virtual void	optimize (double *x)=0
	Minimizes a function, starting from initial value x. More...

void	set_N (index_t N)
	Defines the number of variables.

index_t	get_N () const
	Returns the number of variables.

void	set_M (index_t M)
	Defines the inner number of iterations. More...

index_t	get_M () const
	Returns the inner number of iterations.

void	set_max_iter (index_t maxiter)
	Defines the maximum number of iterations.

index_t	get_max_iter () const
	Returns the maximum number of iterations.

void	set_funcgrad_callback (funcgrad_callback fp)
	Defines the callback that evaluates the function to be minimized and its gradient.

void	set_newiteration_callback (newiteration_callback fp)
	Defines a callback that will be called at each iteration.

void	set_evalhessian_callback (evalhessian_callback fp)
	Defines the callback that evaluates the Hessian of function to be minimized (second order derivatives). More...

void	set_epsg (double eg)
	Defines the stopping criterion in terms of gradient magnitude.

void	set_epsf (double ef)
	Defines the stopping criterion in terms of function value.

void	set_epsx (double ex)
	Defines the stopping criterion in terms of variation of x.

void	set_verbose (bool verb)
	Enables or disables verbose logs.

Public Member Functions inherited from GEO::Counted
void	ref () const
	Increments the reference count. More...

void	unref () const
	Decrements the reference count. More...

bool	is_shared () const
	Check if the object is shared. More...

int	nb_refs () const
	Gets the number of references that point to this object. More...

Static Public Member Functions
static Optimizer *	create (const std::string &name="default")
	Creates an Optimizer. More...

Static Public Member Functions inherited from GEO::Counted
static void	ref (const Counted *counted)
	Increments the reference count. More...

static void	unref (const Counted *counted)
	Decrements the reference count. More...

Protected Member Functions
	Optimizer ()
	Optimizer constructor. More...

	~Optimizer () override
	Optimizer destructor.

Protected Member Functions inherited from GEO::Counted
	Counted ()
	Creates a reference counted object. More...

virtual	~Counted ()
	Destroys a reference counted object. More...

Protected Attributes
index_t	n_

index_t	m_

index_t	max_iter_

funcgrad_callback	funcgrad_callback_

newiteration_callback	newiteration_callback_

evalhessian_callback	evalhessian_callback_

double	epsg_

double	epsf_

double	epsx_

bool	verbose_

Related Functions
(Note that these are not member functions.)
typedef SmartPointer< Optimizer >	Optimizer_var
	Smart pointer that contains an Optimizer object.

typedef Factory0< Optimizer >	OptimizerFactory
	Optimizer Factory. More...

Detailed Description

Optimizer minimizes a multivariate function.

The dimension of the problem is defined by set_N(). The multivariate function to be minimized is defined by a callback that evaluates the function and its gradient, specified by set_funcgrad_callback(). Optimizer implements the numeric part of CentroidalVoronoiTesselation.

Optimizer objects are created using method create() which uses the Factory service. New Optimizers can be implemented and registered to the factory using geo_register_Optimizer_creator().

See also: OptimizerFactory; geo_register_Optimizer_creator

Definition at line 77 of file optimizer.h.

Member Typedef Documentation

◆ evalhessian_callback

typedef void(* GEO::Optimizer::evalhessian_callback) (index_t N, double *x, double &f, double *g, HESSIAN_MATRIX &hessian)

Optimizer callback that evaluates a function, its gradient and its Hessian.

See also: set_evalhessian_callback()

Definition at line 101 of file optimizer.h.

◆ funcgrad_callback

typedef void(* GEO::Optimizer::funcgrad_callback) (index_t N, double *x, double &f, double *g)

Optimizer callback that evaluates a function and its gradient.

See also: set_funcgrad_callback()

Definition at line 84 of file optimizer.h.

◆ newiteration_callback

typedef void(* GEO::Optimizer::newiteration_callback) (index_t N, const double *x, double f, const double *g, double gnorm)

Optimizer callback that is called at each iteration.

See also: set_newiteration_callback()

Definition at line 92 of file optimizer.h.

Constructor & Destructor Documentation

◆ Optimizer()

GEO::Optimizer::Optimizer ( )

protected

Optimizer constructor.

Should never be called directly, use create() instead.

Member Function Documentation

◆ create()

static Optimizer* GEO::Optimizer::create ( const std::string & name = "default" )

static

Creates an Optimizer.

Parameters

[in]

name

name of the Optimizer to create:

"HLBFG" - BFGS (quasi-Newton)
"HM1QN3" - for non-smooth functions
"HCG" - non-linear conjugate gradient
"HLBFGS_HESS" - BFGS with Hessian (full Newton)
"default" - equivalent to "HLBFGS"

Return values

nullptr if name is not a valid Optimizer algorithm name

otherwise,a

pointer to an Optimizer object. The returned pointer must be stored in a Optimizer_var that does automatic destruction:

Optimizer_var optimizer = Optimizer::create("HLBFGS") ;

GEO::Optimizer::create

static Optimizer * create(const std::string &name="default")

Creates an Optimizer.

GEO::Optimizer::Optimizer_var

SmartPointer< Optimizer > Optimizer_var

Smart pointer that contains an Optimizer object.

Definition: optimizer.h:267

◆ optimize()

virtual void GEO::Optimizer::optimize ( double * x )

pure virtual

Minimizes a function, starting from initial value x.

Parameters

[in] x is of size n, where n was defined with set_N()

◆ set_evalhessian_callback()

void GEO::Optimizer::set_evalhessian_callback ( evalhessian_callback fp )

inline

Defines the callback that evaluates the Hessian of function to be minimized (second order derivatives).

Only used in "HLBFGS_HESS" mode.

Definition at line 195 of file optimizer.h.

◆ set_M()

void GEO::Optimizer::set_M ( index_t M )

inline

Defines the inner number of iterations.

Used by HLBFGSOptimizer.

Definition at line 147 of file optimizer.h.

Friends And Related Function Documentation

◆ OptimizerFactory

typedef Factory0<Optimizer> OptimizerFactory

This Factory is used to create Optimizer objects. It can also be used to register new Optimizer implementations.

See also: geo_register_Optimizer_creator; Factory

Definition at line 277 of file optimizer.h.

Member Data Documentation

◆ epsg_

double GEO::Optimizer::epsg_

protected

Error tolerance on x, f and g

Definition at line 258 of file optimizer.h.

◆ m_

index_t GEO::Optimizer::m_

protected

Number of corrections in the BFGS scheme of Hessian approximation update

Definition at line 249 of file optimizer.h.

◆ max_iter_

index_t GEO::Optimizer::max_iter_

protected

Max iterations

Definition at line 251 of file optimizer.h.

◆ n_

index_t GEO::Optimizer::n_

protected

Size of the problem

Definition at line 244 of file optimizer.h.

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

geogram/numerics/optimizer.h

Public Types

Public Member Functions

Static Public Member Functions

Protected Member Functions

Protected Attributes

Related Functions

Detailed Description

Member Typedef Documentation

◆ evalhessian_callback

◆ funcgrad_callback

◆ newiteration_callback

Constructor & Destructor Documentation

◆ Optimizer()

Member Function Documentation

◆ create()

◆ optimize()

◆ set_evalhessian_callback()

◆ set_M()

Friends And Related Function Documentation

◆ OptimizerFactory

Member Data Documentation

◆ epsg_

◆ m_

◆ max_iter_

◆ n_