OPTPP::OptNewton Class Reference

#include <OptNewton.h>

Inheritance diagram for OPTPP::OptNewton:

OPTPP::OptNewton2Deriv OPTPP::OptNewtonLike OPTPP::OptimizeClass List of all members.

Public Member Functions

 OptNewton ()
 OptNewton (NLP2 *p)
 OptNewton (NLP2 *p, UPDATEFCN u)
 OptNewton (NLP2 *p, TOLS t)
virtual ~OptNewton ()
void initHessian ()
 Compute the analytic Hessian at the initial point.
NEWMAT::SymmetricMatrix updateH (NEWMAT::SymmetricMatrix &H, int k)
 Returns the analytic Hessian.
int checkDeriv ()
 Compare the analytic gradient with the finite difference gradient.
void printStatus (char *)
 Print status of Newton's method.
real stepTolNorm () const
 Compute length of the step direction.

Detailed Description

OptNewton is a derived class of OptNewtonLike. This class implements an unconstrained Newton Method with analytic Hessian information. The user can select from the following globalization strategies: Linesearch, trust-region, and trustpds.

Copyright (c) 2001, Sandia Corporation.

Author:
J.C. Meza, Sandia National Laboratories,meza@ca.sandia.gov
Note:
Modified by P.J. Williams
Examples:

tstnewton.C.


Constructor & Destructor Documentation

OPTPP::OptNewton::OptNewton (  )  [inline]

Default Constructor

See also:
OptNewton(NLP2* p)

OptNewton(NLP2* p, UPDATEFCN u)

OptNewton(NLP2* p, TOLS t)

OPTPP::OptNewton::OptNewton ( NLP2 p  )  [inline]

Parameters:
p a pointer to an NLP2.

OPTPP::OptNewton::OptNewton ( NLP2 p,
UPDATEFCN  u 
) [inline]

Parameters:
p a pointer to an NLP2.
u a function pointer.

OPTPP::OptNewton::OptNewton ( NLP2 p,
TOLS  t 
) [inline]

Parameters:
p a pointer to an NLP2.
t tolerance class reference.

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

Destructor


Member Function Documentation

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

Compare the analytic gradient with the finite difference gradient.

Reimplemented from OPTPP::OptNewtonLike.

void OPTPP::OptNewton::initHessian (  )  [virtual]

Compute the analytic Hessian at the initial point.

Reimplemented from OPTPP::OptNewtonLike.

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

Print status of Newton's method.

Reimplemented from OPTPP::OptNewtonLike.

real OPTPP::OptNewton::stepTolNorm (  )  const [virtual]

Compute length of the step direction.

Reimplemented from OPTPP::OptimizeClass.

NEWMAT::SymmetricMatrix OPTPP::OptNewton::updateH ( NEWMAT::SymmetricMatrix &  H,
int  k 
) [virtual]

Returns the analytic Hessian.

Implements OPTPP::OptNewtonLike.


The documentation for this class was generated from the following files:
Bug Reports    OPT++ Developers    Copyright Information    GNU Lesser General Public License
Documentation, generated by , last revised August 30, 2006.