ROL
Public Member Functions | Private Attributes | List of all members
ROL::ProjectedNewtonStep< Real > Class Template Reference
Optimization Steps

Provides the interface to compute optimization steps with projected Newton's method using line search. More...

#include <ROL_ProjectedNewtonStep.hpp>

Inheritance diagram for ROL::ProjectedNewtonStep< Real >:

Public Member Functions

 ProjectedNewtonStep (ROL::ParameterList &parlist, const bool computeObj=true)
 Constructor.
void initialize (Vector< Real > &x, const Vector< Real > &s, const Vector< Real > &g, Objective< Real > &obj, BoundConstraint< Real > &bnd, AlgorithmState< Real > &algo_state)
 Initialize step with bound constraint.
void compute (Vector< Real > &s, const Vector< Real > &x, Objective< Real > &obj, BoundConstraint< Real > &bnd, AlgorithmState< Real > &algo_state)
 Compute step.
void update (Vector< Real > &x, const Vector< Real > &s, Objective< Real > &obj, BoundConstraint< Real > &bnd, AlgorithmState< Real > &algo_state)
 Update step, if successful.
std::string printHeader (void) const
 Print iterate header.
std::string printName (void) const
 Print step name.
std::string print (AlgorithmState< Real > &algo_state, bool print_header=false) const
 Print iterate status.
Public Member Functions inherited from ROL::Step< Real >
virtual ~Step ()
 Step (void)
virtual void initialize (Vector< Real > &x, const Vector< Real > &g, Objective< Real > &obj, BoundConstraint< Real > &con, AlgorithmState< Real > &algo_state)
 Initialize step with bound constraint.
virtual void initialize (Vector< Real > &x, const Vector< Real > &g, Vector< Real > &l, const Vector< Real > &c, Objective< Real > &obj, Constraint< Real > &con, AlgorithmState< Real > &algo_state)
 Initialize step with equality constraint.
virtual void initialize (Vector< Real > &x, const Vector< Real > &g, Vector< Real > &l, const Vector< Real > &c, Objective< Real > &obj, Constraint< Real > &con, BoundConstraint< Real > &bnd, AlgorithmState< Real > &algo_state)
 Initialize step with equality constraint.
virtual void compute (Vector< Real > &s, const Vector< Real > &x, const Vector< Real > &l, Objective< Real > &obj, Constraint< Real > &con, AlgorithmState< Real > &algo_state)
 Compute step (equality constraints).
virtual void update (Vector< Real > &x, Vector< Real > &l, const Vector< Real > &s, Objective< Real > &obj, Constraint< Real > &con, AlgorithmState< Real > &algo_state)
 Update step, if successful (equality constraints).
virtual void compute (Vector< Real > &s, const Vector< Real > &x, const Vector< Real > &l, Objective< Real > &obj, Constraint< Real > &con, BoundConstraint< Real > &bnd, AlgorithmState< Real > &algo_state)
 Compute step (equality constraints).
virtual void update (Vector< Real > &x, Vector< Real > &l, const Vector< Real > &s, Objective< Real > &obj, Constraint< Real > &con, BoundConstraint< Real > &bnd, AlgorithmState< Real > &algo_state)
 Update step, if successful (equality constraints).
const ROL::Ptr< const StepState< Real > > getStepState (void) const
 Get state for step object.
void reset (const Real searchSize=1.0)
 Get state for step object.

Private Attributes

ROL::Ptr< Vector< Real > > gp_
 Additional vector storage.
ROL::Ptr< Vector< Real > > d_
 Additional vector storage.
int verbosity_
 Verbosity level.
const bool computeObj_
bool useProjectedGrad_
 Whether or not to use to the projected gradient criticality measure.

Additional Inherited Members

Protected Member Functions inherited from ROL::Step< Real >
ROL::Ptr< StepState< Real > > getState (void)

Detailed Description

template<class Real>
class ROL::ProjectedNewtonStep< Real >

Provides the interface to compute optimization steps with projected Newton's method using line search.

Definition at line 25 of file ROL_ProjectedNewtonStep.hpp.

Constructor & Destructor Documentation

◆ ProjectedNewtonStep()

template<class Real>
ROL::ProjectedNewtonStep< Real >::ProjectedNewtonStep ( ROL::ParameterList & parlist,
const bool computeObj = true )
inline

Constructor.

Standard constructor to build a ProjectedNewtonStep object. Algorithmic specifications are passed in through a ROL::ParameterList.

Parameters
[in]parlistis a parameter list containing algorithmic specifications

Definition at line 47 of file ROL_ProjectedNewtonStep.hpp.

References computeObj_, d_, gp_, ROL::Step< Real >::Step(), useProjectedGrad_, and verbosity_.

Member Function Documentation

◆ initialize()

template<class Real>
void ROL::ProjectedNewtonStep< Real >::initialize ( Vector< Real > & x,
const Vector< Real > & s,
const Vector< Real > & g,
Objective< Real > & obj,
BoundConstraint< Real > & con,
AlgorithmState< Real > & algo_state )
inlinevirtual

Initialize step with bound constraint.

Reimplemented from ROL::Step< Real >.

Definition at line 56 of file ROL_ProjectedNewtonStep.hpp.

References ROL::Vector< Real >::clone(), d_, gp_, and ROL::Step< Real >::initialize().

◆ compute()

template<class Real>
void ROL::ProjectedNewtonStep< Real >::compute ( Vector< Real > & s,
const Vector< Real > & x,
Objective< Real > & obj,
BoundConstraint< Real > & bnd,
AlgorithmState< Real > & algo_state )
inlinevirtual

Compute step.

Reimplemented from ROL::Step< Real >.

Definition at line 64 of file ROL_ProjectedNewtonStep.hpp.

References ROL::Step< Real >::getState(), ROL::AlgorithmState< Real >::gnorm, gp_, ROL::Objective< Real >::invHessVec(), ROL::Vector< Real >::plus(), ROL::BoundConstraint< Real >::pruneActive(), ROL::BoundConstraint< Real >::pruneInactive(), ROL::ROL_EPSILON(), and ROL::Vector< Real >::scale().

◆ update()

template<class Real>
void ROL::ProjectedNewtonStep< Real >::update ( Vector< Real > & x,
const Vector< Real > & s,
Objective< Real > & obj,
BoundConstraint< Real > & bnd,
AlgorithmState< Real > & algo_state )
inlinevirtual

Update step, if successful.

Reimplemented from ROL::Step< Real >.

Definition at line 83 of file ROL_ProjectedNewtonStep.hpp.

References computeObj_, ROL::BoundConstraint< Real >::computeProjectedGradient(), d_, ROL::Step< Real >::getState(), ROL::AlgorithmState< Real >::gnorm, gp_, ROL::Objective< Real >::gradient(), ROL::AlgorithmState< Real >::iter, ROL::AlgorithmState< Real >::iterateVec, ROL::AlgorithmState< Real >::nfval, ROL::AlgorithmState< Real >::ngrad, ROL::Vector< Real >::norm(), ROL::Vector< Real >::plus(), ROL::BoundConstraint< Real >::project(), ROL::ROL_EPSILON(), ROL::AlgorithmState< Real >::snorm, ROL::Objective< Real >::update(), useProjectedGrad_, ROL::AlgorithmState< Real >::value, and ROL::Objective< Real >::value().

◆ printHeader()

template<class Real>
std::string ROL::ProjectedNewtonStep< Real >::printHeader ( void ) const
inlinevirtual

Print iterate header.

Reimplemented from ROL::Step< Real >.

Definition at line 123 of file ROL_ProjectedNewtonStep.hpp.

References ROL::DESCENT_NEWTON, ROL::EDescentToString(), and verbosity_.

Referenced by print().

◆ printName()

template<class Real>
std::string ROL::ProjectedNewtonStep< Real >::printName ( void ) const
inlinevirtual

Print step name.

Reimplemented from ROL::Step< Real >.

Definition at line 149 of file ROL_ProjectedNewtonStep.hpp.

References ROL::DESCENT_NEWTON, and ROL::EDescentToString().

Referenced by print().

◆ print()

template<class Real>
std::string ROL::ProjectedNewtonStep< Real >::print ( AlgorithmState< Real > & algo_state,
bool printHeader = false ) const
inlinevirtual

Print iterate status.

Reimplemented from ROL::Step< Real >.

Definition at line 154 of file ROL_ProjectedNewtonStep.hpp.

References ROL::AlgorithmState< Real >::gnorm, ROL::AlgorithmState< Real >::iter, ROL::AlgorithmState< Real >::nfval, ROL::AlgorithmState< Real >::ngrad, printHeader(), printName(), ROL::AlgorithmState< Real >::snorm, and ROL::AlgorithmState< Real >::value.

Member Data Documentation

◆ gp_

template<class Real>
ROL::Ptr<Vector<Real> > ROL::ProjectedNewtonStep< Real >::gp_
private

Additional vector storage.

Definition at line 28 of file ROL_ProjectedNewtonStep.hpp.

Referenced by compute(), initialize(), ProjectedNewtonStep(), and update().

◆ d_

template<class Real>
ROL::Ptr<Vector<Real> > ROL::ProjectedNewtonStep< Real >::d_
private

Additional vector storage.

Definition at line 29 of file ROL_ProjectedNewtonStep.hpp.

Referenced by initialize(), ProjectedNewtonStep(), and update().

◆ verbosity_

template<class Real>
int ROL::ProjectedNewtonStep< Real >::verbosity_
private

Verbosity level.

Definition at line 30 of file ROL_ProjectedNewtonStep.hpp.

Referenced by printHeader(), and ProjectedNewtonStep().

◆ computeObj_

template<class Real>
const bool ROL::ProjectedNewtonStep< Real >::computeObj_
private

Definition at line 31 of file ROL_ProjectedNewtonStep.hpp.

Referenced by ProjectedNewtonStep(), and update().

◆ useProjectedGrad_

template<class Real>
bool ROL::ProjectedNewtonStep< Real >::useProjectedGrad_
private

Whether or not to use to the projected gradient criticality measure.

Definition at line 32 of file ROL_ProjectedNewtonStep.hpp.

Referenced by ProjectedNewtonStep(), and update().

The documentation for this class was generated from the following file:
Generated by doxygen 1.15.0