OPTPP::OptCG Class Reference

#include <OptCG.h>

Inheritance diagram for OPTPP::OptCG:

List of all members.


Public Member Functions
	OptCG ()
	OptCG (NLP1 *p)
	OptCG (NLP1 *p, TOLS t)
virtual	~OptCG ()
virtual NEWMAT::ColumnVector	computeSearch (NEWMAT::SymmetricMatrix &)
virtual void	acceptStep (int k, int step_type)
virtual void	updateModel (int k, int ndim, NEWMAT::ColumnVector x)
virtual int	checkDeriv ()
	Compare the analytic gradient with the finite-difference gradient.
virtual int	computeStep (NEWMAT::ColumnVector sk)
	Compute the step direction based upon specified globalization strategy.
virtual void	reset ()
	Reset the parameters.
virtual void	initOpt ()
	Initialize the optimization method.
virtual void	optimize ()
	Run the optimization method.
virtual real	stepTolNorm () const
	Compute steplength.
virtual void	printStatus (char *)
	Print the status to the optimization method at the current iteration.
Protected Member Functions
NLP1 *	nlprob () const
	Pointer to an NLP1 object.
Private Attributes
NLP1 *	nlp

Detailed Description

CG-Like Methods OptCG is a derived class from OptCGLike, which implements a nonlinear conjugate gradient method. This version uses the Polak-Ribiere formula and a linesearch routine due to More and Thuente as implemented in the routines mcsrch and mcstep.

Author:: J.C. Meza, Sandia National Laboratories, meza@ca.sandia.gov

Constructor & Destructor Documentation

OPTPP::OptCG::OptCG ( ) [inline]

Default Constructor
See also:
OptCG(NLP1* p)
OptCG(NLP1* p, TOLS t)

OPTPP::OptCG::OptCG ( NLP1 * p ) [inline]

Parameters:

p a pointer to an NLP1 object

See also:
OptCG(NLP1* p, TOLS t)

OPTPP::OptCG::OptCG ( NLP1 * p,

TOLS t

) [inline]

Parameters:

p a pointer to an NLP1 object

t a TOLS object

See also:
OptCG(NLP1* p)

virtual OPTPP::OptCG::~OptCG ( ) [inline, virtual]

Destructor

Member Function Documentation

virtual void OPTPP::OptCG::acceptStep ( int k,

int step_type

) [inline, virtual]

Note:
Pure virtual functions
Each derived class must define these functions for themselves

Implements OPTPP::OptCGLike.

int OPTPP::OptCG::checkDeriv ( ) [virtual]

Compare the analytic gradient with the finite-difference gradient.

Reimplemented from OPTPP::OptCGLike.

virtual NEWMAT::ColumnVector OPTPP::OptCG::computeSearch ( NEWMAT::SymmetricMatrix & ) [inline, virtual]

Implements OPTPP::OptimizeClass.

int OPTPP::OptCG::computeStep ( NEWMAT::ColumnVector sk ) [virtual]

Compute the step direction based upon specified globalization strategy.

void OPTPP::OptCG::initOpt ( ) [virtual]

Initialize the optimization method.

NLP1* OPTPP::OptCG::nlprob ( ) const [inline, protected, virtual]

Pointer to an NLP1 object.

Returns:
Pointer to an NLP1 object

Implements OPTPP::OptCGLike.

void OPTPP::OptCG::optimize ( ) [virtual]

Run the optimization method.

Reimplemented from OPTPP::OptCGLike.

void OPTPP::OptCG::printStatus ( char * ) [virtual]

Print the status to the optimization method at the current iteration.

Implements OPTPP::OptimizeClass.