OPTPP Namespace Reference

Simple Bounds Class The standard form representation is x >= a. More...

Classes

class  Appl_Data_NPSOL
class  BoolVector
 BoolVector defines a vector of bools. More...
class  FDNLF1
 FDNLF1 is a derived class of NLP1. More...
class  FDNLF1APP
 These classes and typedefs are used for Application Launching where an AppLauncher Object is also required so that the launcher specific data can be used for the function evaluation. More...
class  LSQNLF
 LSQNLF is a derived class of NLP2. More...
class  NLF0
 NLF0 is a derived class of NLP0, a nonlinear problem without analytic derivative information. More...
class  NLF0APP
 These classes and typedefs are used for Application Launching where an AppLauncher Object is also required so that the launcher specific data can be used for the function evaluation. More...
class  NLF1
 NLF1 is a derived class of NLP1, a nonlinear problem with analytic first derivatives. More...
class  NLF2
 NLF2 is a derived class of NLP2, a nonlinear problem with analytic first and second derivatives. More...
class  NLP
 NLP is a handle class for NLPBase. More...
class  NLP0
 Base Class for NonLinear Programming Problem For NLP0 the only assumption on the objective function is that it be continuous. More...
class  NLPBase
 NLPBase is the Base Class for NonLinear Programming Problem. More...
class  OptBaNewton
 OptBaNewton implements a bound constrained Newton method with a logarithmic barrier term. More...
class  OptBaQNewton
 OptBaQNewton implements a barrier Quasi-Newton method. More...
class  OptBCEllipsoid
 Bound Constrained Newton abstract data classes. More...
class  OptBCFDNewton
 OptBCFDNewton is a derived class of OptBCNewtonLike. More...
class  OptBCNewton
 OptBCNewton is a derived class of OptBCNewtonLike. More...
class  OptBCNewton1Deriv
 Bound constrained Newton class that will take either an NLP1 or NLP2. More...
class  OptBCNewton2Deriv
 Bound constrained Newton class that requires an NLP2. More...
class  OptBCNewtonLike
 Bound Constrained Newton abstract data classes OptBCNewtonLike OptBCNewton1Deriv OptBCNewton2Deriv OptBCNewtonLike provides common data and functionality for the OptBCQNewton, OptBCFDNewton, and OptBCNewton methods. More...
class  OptBCQNewton
 OptBCQNewton is a derived class of OptBCNewtonLike. More...
class  OptCG
 CG-Like Methods OptCG is a derived class from OptCGLike, which implements a nonlinear conjugate gradient method. More...
class  OptCGLike
 CG-Like Methods OptCG is a derived class from OptCGLike, which implements a nonlinear conjugate gradient method. More...
class  OptConstrFDNewton
 OptConstrFDNewton is a derived class of OptConstrNewtonLike. More...
class  OptConstrNewton
 OptConstrNewton is a derived class of OptConstrNewtonLike. More...
class  OptConstrNewton1Deriv
 Constrained Newton classes that will accept either an NLP1 or NLP2. More...
class  OptConstrNewton2Deriv
 Constrained Newton classes that require an NLP2. More...
class  OptConstrNewtonLike
 Constrained Newton abstract data classes OptConstrNewtonLike OptConstrNewton1Deriv OptConstrNewton2Deriv. More...
class  OptConstrQNewton
 OptConstrQNewton is a derived class of OptConstrNewtonLike. More...
class  OptDHNIPS
 OptDHNIPS is a derived class of OptNIPSLike. More...
class  OptDirect
 OptDirect is a derived class of OptimizeClass and the base class for direct search methods. More...
class  OptFDNewton
 OptFDNewton is a derived class of OptNewtonLike. More...
class  OptFDNIPS
 OptFDNIPS is a derived class of OptNIPSLike. More...
class  OptimizeClass
 Opt is the Base Optimization Class All other Optimization classes are derived from this one. More...
class  OptLBFGS
 The Limited Memory BFGS Method for Large Scale Optimization. More...
class  OptLBFGSLike
 LBFGS-Like Methods OptLBFGS is a derived class of OptLBFGSLike that implements the LBFGS method of J. More...
class  OptNewton
 OptNewton is a derived class of OptNewtonLike. More...
class  OptNewton1Deriv
 Unconstrained Newton class that accepts either an NLP or NLP2. More...
class  OptNewton2Deriv
 Unconstrained Newton class that requires an NLP2. More...
class  OptNewtonLike
 OptNewtonLike is the base class for Newton Methods. More...
class  OptNIPS
 OptNIPS is a derived class of OptNIPSLike. More...
class  OptNIPSLike
 OptNIPSLike is a derived class of OptConstrNewtonLike. More...
class  OptNPSOL
 The interface to NPSOL software package for nonlinear programming. More...
class  OptPDS
 OptPDS is an implementation of a derivative-free algorithm for unconstrained optimization. More...
class  OptppArray
 Simple array class. More...
class  OptppExceptions
 OptppExceptions is the base class for OptppMemoryError, OptppRangeError, OptppMathError, OptppDomainError, OptppZeroDivide. More...
class  OptQNewton
 OptQNewton is a derived class of OptNewtonLike. More...
class  OptQNIPS
 OptQNIPS is a derived class of OptNIPSLike. More...
class  Problem
 Class Problem is the parent for all the different types of solvers. More...
class  SmartPtr
 Howard Hinnant's reference counting handle class. More...
class  TOLS
 TOLS is the Base Class for Tolerances which will be used in the optimization methods. More...

Typedefs

typedef CompoundConstraint *(* INITCONFCN )(int)
typedef void(* INITFCN )(int, NEWMAT::ColumnVector &)
typedef void(* INITFCNAPP )(int, NEWMAT::ColumnVector &, AppLauncher *launcher)
typedef double real
typedef void(* UPDATEFCN )(int, int, NEWMAT::ColumnVector)
typedef void(* USERFCN0 )(int, const NEWMAT::ColumnVector &, real &, int &)
typedef void(* USERFCN0APP )(int, const NEWMAT::ColumnVector &, real &, int &, AppLauncher *launcher)
 These classes and typedefs are used for Application Launching where an AppLauncher Object is also required so that the launcher specific data can be used for the function evaluation.
typedef void(* USERFCN0V )(int, const NEWMAT::ColumnVector &, real &, int &, void *)
typedef void(* USERFCN1 )(int, int, const NEWMAT::ColumnVector &, real &, NEWMAT::ColumnVector &, int &)
typedef void(* USERFCN1V )(int, int, const NEWMAT::ColumnVector &, real &, NEWMAT::ColumnVector &, int &, void *)
typedef void(* USERFCN2 )(int, int, const NEWMAT::ColumnVector &, real &, NEWMAT::ColumnVector &, NEWMAT::SymmetricMatrix &, int &)
typedef void(* USERFCN2A )(int, int, int, const NEWMAT::ColumnVector &, real &, NEWMAT::ColumnVector &, NEWMAT::Matrix &, int &)
typedef void(* USERFCN2AV )(int, int, int, const NEWMAT::ColumnVector &, real &, NEWMAT::ColumnVector &, NEWMAT::Matrix &, int &, void *)
typedef void(* USERFCN2V )(int, int, const NEWMAT::ColumnVector &, real &, NEWMAT::ColumnVector &, NEWMAT::SymmetricMatrix &, int &, void *)
typedef void(* USERFCNLSQ0 )(int, const NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, int &)
typedef void(* USERFCNLSQ0V )(int, const NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, int &, void *v)
typedef void(* USERFCNLSQ1 )(int, int, const NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, NEWMAT::Matrix &, int &)
typedef void(* USERFCNLSQ1V )(int, int, const NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, NEWMAT::Matrix &, int &, void *v)
typedef void(* USERNLNCON0 )(int, const NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, int &)
typedef void(* USERNLNCON0APP )(int, int, const NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, int &, AppLauncher *launcher)
typedef void(* USERNLNCON1 )(int, int, const NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, NEWMAT::Matrix &, int &)
typedef void(* USERNLNCON2 )(int, int, const NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, NEWMAT::Matrix &, OptppArray< NEWMAT::SymmetricMatrix > &, int &)

Enumerations

enum  ConstraintType {
  Leqn, NLeqn, Lineq, NLineq,
  Bound
}
enum  DerivOption { ForwardDiff, BackwardDiff, CentralDiff }
enum  FcnMode {
  NLPNoOp = 0, NLPFunction = 1, NLPGradient = 2, NLPHessian = 4,
  NLPConstraint = 8, NLPCJacobian = 16
}
enum  MeritFcn { NormFmu, ArgaezTapia, VanShanno }
enum  SearchStrategy { LineSearch, TrustRegion, TrustPDS }
enum  SpecOption { NoSpec, Spec1, Spec2 }
enum  Step_type { Cauchy_Step, Dogleg_Step, Newton_Step, Backtrack_Step }

Functions

void abort_handler (int code)
int backtrack (NLP1 *, ostream *, NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, real *, int itnmax=5, real ftol=1.e-4, real stpmax=1.e3, real stpmin=1.e-9)
void bailout (const OptppExceptions &e)
int bump (int start, int to)
string d (int val, int w=0, int p=0, char c=' ', opt_mode f=0)
int dogleg (NLP1 *, ostream *, NEWMAT::SymmetricMatrix &, NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, real &, real &, real)
string e (double val, int w=0, int p=0, char c=' ', opt_mode f=0)
string f (double val, int w=0, int p=0, char c=' ', opt_mode f=0)
string format (int val, oformatstate const &fmt)
string format (double val, oformatstate const &fmt)
void FPrint (ostream *fout, const NEWMAT::DiagonalMatrix &X)
void FPrint (ostream *fout, const NEWMAT::SymmetricMatrix &X)
void FPrint (ostream *fout, const NEWMAT::Matrix &X)
int linesearch (NLP1 *, ostream *, NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, real *, real stpmax=1.e3, real stpmin=1.e-9, int itnmax=5, real ftol=1.e-4, real xtol=2.2e-16, real gtol=0.9)
NEWMAT::ReturnMatrix MCholesky (NEWMAT::SymmetricMatrix &)
int mcsrch (NLP1 *, NEWMAT::ColumnVector &, ostream *, real *, int itnmax=5, real ftol=1.e-4, real xtol=2.2e-16, real gtol=0.9, real stpmax=1.e3, real stpmin=1.e-9)
int mcstep (real *, real *, real *, real *, real *, real *, real *, real, real, bool *, real, real, int *)
template<class T >
ostream & operator<< (ostream &os, const OptppArray< T > &item)
void opt_default_update_model (int k, int dim, NEWMAT::ColumnVector x)
void OptppdomainError (const char *mesg, const double &badValue)
void OptppfatalError (const char *mesg)
void OptppmathError (const char *mesg)
void OptppmemoryError (const char *mesg)
void OptpprangeError (const char *mesg, int i, int low, int high)
void OptppzeroDivide (const char *mesg)
int pdschk (NLP0 *, int, double *, double *, double, double *, int, double)
int pdsinit (NLP0 *, ostream *, int, int, int *, int *, double, double *, double *, double *, int *, double *, double *, double *, double *, double *, char *, double, int, int, double)
int pdsopt (NLP0 *, ostream *, double *, int *, int, char *, int, int, double, int, int, double, double *, double, int, double *, int *, char *, double, double, double *, int, int, int, double)
int pdsstep (NLP1 *, ostream *, NEWMAT::SymmetricMatrix &, NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, real &, real &, real, double &, bool, int)
int pdswork (NLP0 *, ostream *, ofstream *, int, double, int, int, int *, double, int, double *, double *, int *, double *, double *, int *, int, double, double *, char *, double, double, int, int, int, double, FILE *)
NEWMAT::ReturnMatrix PertChol (NEWMAT::SymmetricMatrix &, NEWMAT::Real, NEWMAT::Real &)
void Print (const NEWMAT::LowerTriangularMatrix &X)
void Print (const NEWMAT::SymmetricMatrix &X)
void Print (const NEWMAT::DiagonalMatrix &X)
void Print (const NEWMAT::UpperTriangularMatrix &X)
void Print (const NEWMAT::Matrix &X)
 Print various quantities in newmat classes.
template<class T >
OptppArray< T > sliceOptppArray (const OptppArray< OptppArray< T > > &array, int index)
int trustpds (NLP1 *, ostream *, NEWMAT::SymmetricMatrix &, NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, real &, real &, real stpmax=1.e3, real stpmin=1.e-9, int searchSize=64)
int trustregion (NLP1 *, ostream *, NEWMAT::SymmetricMatrix &, NEWMAT::ColumnVector &, NEWMAT::ColumnVector &, real &, real &, real stpmax=1.e3, real stpmin=1.e-9)
template<class T >
OptppArray< T > tuple (const T &a, const T &b, const T &c, const T &d, const T &e, const T &f)
template<class T >
OptppArray< T > tuple (const T &a, const T &b, const T &c, const T &d, const T &e)
template<class T >
OptppArray< T > tuple (const T &a, const T &b, const T &c, const T &d)
template<class T >
OptppArray< T > tuple (const T &a, const T &b, const T &c)
template<class T >
OptppArray< T > tuple (const T &a, const T &b)
template<class T >
OptppArray< T > tuple (const T &a)

Detailed Description

Simple Bounds Class The standard form representation is x >= a.

Portable error routines: under the hood, they throw an exception on machines where exceptions are supported, or simply print a mesg and then terminate on machines w/o exception support.

NonLinearInequality is a derived class of NonLinearConstraint.

NonLinearEquation is a derived class of NonLinearConstraint.

NonLinearConstraint is a derived class of ConstraintBase.

NLP2: NLP1 + Second derivatives NLP2 is derived from NLP1 by adding the necessary information to compute and store the Hessian.

NLP1: NLP0 + First derivatives NLP1 is derived from the base class NLP0 by adding the necessary information to compute and store the gradient.

LinearInequality is a derived class of LinearConstraint.

LinearEquation is a derived class of LinearConstraint.

LinearConstraint is a derived class of ConstraintBase.

ConstraintBase is an abstract class.

Constraint is a handle class to ConstraintBase.

CompoundConstraint is an array of constraints.

Author:
J. C. Meza, Sandia National Laboratories, meza@ca.sandia.gov
Note:
Modified by P.J. Williams 02/2006
Author:
P.J. Williams, Sandia National Laboratories, pwillia@sandia.gov
Date:
Last modified 02/2007
Author:
P.J. Williams, Sandia National Laboratories, pwillia@sandia.gov
Date:
Last modified 02/2006

All other constraint classes are derived from this one

Author:
P.J. Williams, Sandia National Laboratories, pwillia@sandia.gov
Date:
Last modified 02/2006

LinearConstraint is an abstract class, which provides common data and functionality to LinearEquation and LinearInequality.

Author:
P.J. Williams, Sandia National Laboratories, pwillia@sandia.gov
Date:
Last modified 02/06/2007
Author:
P.J. Williams, Sandia National Laboratories, pwillia@sandia.gov
Date:
modified 02/2006

Linear Inequalities - Ax >= b

Author:
P.J. Williams, Sandia National Laboratories, pwillia@sandia.gov
Date:
Last modified 02/2006
Author:
J.C. Meza, Sandia National Laboratories, meza@ca.sandia.gov
Note:
Modified by P. J. Williams to incorporate namespaces
Sandia National Laboratories, pwillia@sandia.gov
Author:
J.C. Meza, Lawrence Berkeley National Laboratory,
Note:
Modified by P. J. Williams to incorporate namespaces
Date:
02/2006

NonLinearConstraint provides common data and functionality to NonLinearEquation and NonLinearInequality.

Author:
P.J. Williams, Sandia National Laboratories, pwillia@sandia.gov
Date:
Last modified 02/06/2007

Standard Form g(x) >= 0

Author:
P.J. Williams, Sandia National Laboratories, pwillia@sandia.gov
Date:
Last modified 02/2006

to switch compilation of exceptions on/off, toggle the macro USEEXCEPTIONS


Typedef Documentation

typedef void(* OPTPP::USERFCN0APP)(int, const NEWMAT::ColumnVector &, real &, int &, AppLauncher *launcher)

These classes and typedefs are used for Application Launching where an AppLauncher Object is also required so that the launcher specific data can be used for the function evaluation.

Definition at line 18 of file NLFAPP.h.


Function Documentation

void OPTPP::Print ( const NEWMAT::Matrix &  X  ) 

Print various quantities in newmat classes.

Generated on Mon Jan 24 12:04:41 2011 for FASTlib by  doxygen 1.6.3