OPTPP::OptLBFGS Class Reference

The Limited Memory BFGS Method for Large Scale Optimization. More...

Inheritance diagram for OPTPP::OptLBFGS:

[legend]

Collaboration diagram for OPTPP::OptLBFGS:

[legend]

List of all members.

Public Member Functions
virtual void	acceptStep (int k, int step_type)
virtual int	checkDeriv ()
virtual NEWMAT::ColumnVector	computeSearch (NEWMAT::SymmetricMatrix &)
virtual int	computeStep (NEWMAT::ColumnVector &sk, double stp=1.0)
	Compare analytic vs finite-difference gradient.
virtual void	initOpt ()
	Reset the parameters.
virtual void	optimize ()
	Initialize the optimization.
	OptLBFGS (NLP1 *p, TOLS t, int m)
	OptLBFGS (NLP1 *p, TOLS t)
	OptLBFGS (NLP1 *p, int m)
	OptLBFGS (NLP1 *p)
	OptLBFGS ()
	Default Constructor.
virtual void	printIter (int, double, double, double, double, int)
virtual void	printStatus (char *c)
virtual void	reset ()
	Compute the step direction.
void	setPrintFinalX (bool b)
virtual real	stepTolNorm () const
	Run the optimization.
virtual void	updateModel (int k, int ndim, NEWMAT::ColumnVector x)
virtual	~OptLBFGS ()
	Destructor.
Protected Member Functions
void	initMem (int n)
NLP1 *	nlprob () const

---

Detailed Description

The Limited Memory BFGS Method for Large Scale Optimization.

Solves the unconstrained minimization problem

min F(x) x = (x_1, x_2, ..., x_N),

using the limited memory BFGS method, where N can be large.

The inverse Hessian approximation Hk is computed via BFGS updates to a diagonal matrix H0. The number of updates depend on the previous m steps kept in memory, as set by the user. H0 can be any symmetric positive definite matrix specified by the user (else a default one is constructed).

References:

D. Liu and J. Nocedal, "On the limited memory BFGS method for large scale optimization" Mathematical Programming B 45 (1989), 503-528

Author:

R.A.Oliva, Lawrence Berkely National Laboratories, raoliva@lbl.gov

Definition at line 99 of file OptLBFGS.h.

---

Constructor & Destructor Documentation

OPTPP::OptLBFGS::OptLBFGS

(

)

[inline]

Default Constructor.

See also:

OptLBFGS(NLP1* p)

OptLBFGS(NLP1* p, TOLS t)

Definition at line 130 of file OptLBFGS.h.

References OPTPP::OptimizeClass::method.

OPTPP::OptLBFGS::OptLBFGS

(

NLP1 *

p

)

[inline]

Parameters:

p

a pointer to an NLP1 object

See also:

OptLBFGS(NLP1* p, TOLS t)

Definition at line 136 of file OptLBFGS.h.

References OPTPP::OptimizeClass::method.

OPTPP::OptLBFGS::OptLBFGS	(	NLP1 *	p,
		int	m
	)			[inline]

Parameters:

	p	a pointer to an NLP1 object
	m	integer specifying number of memory vectors

See also:

OptLBFGS(NLP1* p)

OptLBFGS(NLP1* p, TOLS t)

OptLBFGS(NLP1* p, TOLS t, int m)

Definition at line 146 of file OptLBFGS.h.

References OPTPP::OptimizeClass::method.

OPTPP::OptLBFGS::OptLBFGS	(	NLP1 *	p,
		TOLS	t
	)			[inline]

Parameters:

	p	a pointer to an NLP1 object
	t	a TOLS object

See also:

OptLBFGS(NLP1* p, TOLS t, int m)

Definition at line 156 of file OptLBFGS.h.

References OPTPP::OptimizeClass::method.

OPTPP::OptLBFGS::OptLBFGS	(	NLP1 *	p,
		TOLS	t,
		int	m
	)			[inline]

Parameters:

	p	a pointer to an NLP1 object
	t	a TOLS object
	m	integer specifying number of memory vectors

See also:

OptLBFGS(NLP1* p, TOLS t)

Definition at line 165 of file OptLBFGS.h.

References OPTPP::OptimizeClass::method.

virtual OPTPP::OptLBFGS::~OptLBFGS

(

)

[inline, virtual]

Destructor.

Definition at line 173 of file OptLBFGS.h.

---

Member Function Documentation

virtual void OPTPP::OptLBFGS::acceptStep	(	int	,
		int
	)			[inline, virtual]

Note:

Pure virtual functions

Each derived class must define these functions for themselves

Implements OPTPP::OptLBFGSLike.

Definition at line 178 of file OptLBFGS.h.

References OPTPP::OptimizeClass::defaultAcceptStep().

virtual int OPTPP::OptLBFGS::computeStep	(	NEWMAT::ColumnVector &	sk,
		double	stp = 1.0
	)			[virtual]

Compare analytic vs finite-difference gradient.

virtual void OPTPP::OptLBFGS::initOpt

(

)

[virtual]

Reset the parameters.

NLP1* OPTPP::OptLBFGS::nlprob

(

)

const [inline, protected, virtual]

Returns:

Pointer to an NLP1 object

Implements OPTPP::OptLBFGSLike.

Definition at line 111 of file OptLBFGS.h.

virtual void OPTPP::OptLBFGS::optimize

(

)

[virtual]

Initialize the optimization.

Reimplemented from OPTPP::OptLBFGSLike.

virtual void OPTPP::OptLBFGS::reset

(

)

[virtual]

Compute the step direction.

Implements OPTPP::OptimizeClass.

virtual real OPTPP::OptLBFGS::stepTolNorm

(

)

const [virtual]

Run the optimization.

Reimplemented from OPTPP::OptimizeClass.

---

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

• include/OptLBFGS.h