This class implements the termination tests for the primal-dual system. More...

#include <IpInexactPDTerminationTester.hpp>

Inheritance diagram for Ipopt::InexactPDTerminationTester:

Public Member Functions
virtual bool	InitializeImpl (const OptionsList &options, const std::string &prefix)
	Implementation of the initialization method that has to be overloaded by for each derived class.

virtual bool	InitializeSolve ()
	Method for initializing for the next iterative solve.

virtual ETerminationTest	TestTermination (Index ndim, const Number sol, const Number resid, Index iter, Number norm2_rhs)
	This method checks if the current solution of the iterative linear solver is good enough (by returning the corresponding satisfied termination test), or if the Hessian should be modified.

virtual void	Clear ()
	This method can be called after the Solve is over and we can delete anything that has been allocated to free memory.

virtual Index	GetSolverIterations () const
	Return the number of iterative solver iteration from the most recent solve.

Constructor/Destructor
	InexactPDTerminationTester ()
	Default constructor.

virtual	~InexactPDTerminationTester ()
	Destructor.

Public Member Functions inherited from Ipopt::IterativeSolverTerminationTester
const Journalist &	GetJnlst () const
	An easy way to get the journalist if accessed from the outside.

	IterativeSolverTerminationTester ()
	Default constructor.

virtual	~IterativeSolverTerminationTester ()
	Destructor.

Public Member Functions inherited from Ipopt::AlgorithmStrategyObject
bool	Initialize (const Journalist &jnlst, IpoptNLP &ip_nlp, IpoptData &ip_data, IpoptCalculatedQuantities &ip_cq, const OptionsList &options, const std::string &prefix)
	This method is called every time the algorithm starts again - it is used to reset any internal state.

bool	ReducedInitialize (const Journalist &jnlst, const OptionsList &options, const std::string &prefix)
	Reduced version of the Initialize method, which does not require special Ipopt information.

	AlgorithmStrategyObject ()
	Default Constructor.

virtual	~AlgorithmStrategyObject ()
	Destructor.

Public Member Functions inherited from Ipopt::ReferencedObject
	ReferencedObject ()

virtual	~ReferencedObject ()

Index	ReferenceCount () const

void	AddRef (const Referencer *referencer) const

void	ReleaseRef (const Referencer *referencer) const

Static Public Member Functions
static void	RegisterOptions (SmartPtr< RegisteredOptions > roptions)

Private Member Functions
Default Compiler Generated Methods
(Hidden to avoid implicit creation/calling). These methods are not implemented and we do not want the compiler to implement them for us, so we declare them private and do not define them. This ensures that they will not be implicitly created/called.
InexactPDTerminationTester &	operator= (const InexactPDTerminationTester &)
	Overloaded Assignment Operator.

Private Attributes
Index	last_iter_
	Last iterative solver iteration counter.

Algorithmic options
Number	tcc_psi_
	Psi factor in the tangential component condition.

Number	tcc_theta_
	theta factor in the tangential component condition

Number	tcc_theta_mu_exponent_
	mu exponent when multiplied to theta in the tangential component condition

Number	tcc_zeta_
	zeta factor in the tangential component condition

Number	tt_kappa1_
	kappa_1 factor in termination test 1

Number	tt_kappa2_
	kappa_2 factor in termination test 2

Number	tt_eps2_
	eps_2 constant in termination test 2

Number	tt_eps3_
	eps_3 constant in termination test 3

Number	rho_
	rho constant from penalty parameter update.

Number	inexact_desired_pd_residual_
	Desired reduction of residual.

Index	inexact_desired_pd_residual_iter_
	Number of iterations allowed for desired pd residual.

bool	requires_scaling_
	Is set to true if the linear system is scaled via slacks.

Quantities that are identical for all tests and can be
precomputed.
SmartPtr< const Vector >	curr_Av_c_

SmartPtr< const Vector >	curr_Av_d_

Number	c_norm_

Number	c_plus_Av_norm_

Number	v_norm_scaled_

SmartPtr< const Vector >	curr_grad_barrier_obj_x_

SmartPtr< const Vector >	curr_grad_barrier_obj_s_

SmartPtr< const Matrix >	curr_jac_c_

SmartPtr< const Matrix >	curr_jac_d_

SmartPtr< const Vector >	curr_scaling_slacks_

SmartPtr< Vector >	curr_nabla_phi_plus_ATy_x_

SmartPtr< Vector >	curr_nabla_phi_plus_ATy_s_

Number	curr_Av_norm_

Number	curr_tt1_norm_

Number	curr_tt2_norm_

SmartPtr< const Vector >	curr_Wv_x_

SmartPtr< const Vector >	curr_Wv_s_

bool	try_tt2_

Quantities from previous iteration required in the tests
Number	last_Av_norm_

Number	last_tt1_norm_

Additional Inherited Members
Public Types inherited from Ipopt::IterativeSolverTerminationTester
enum	ETerminationTest { CONTINUE , TEST_1_SATISFIED , TEST_2_SATISFIED , TEST_3_SATISFIED , MODIFY_HESSIAN , OTHER_SATISFIED }
	Enum to report result of termination test. More...

Protected Member Functions inherited from Ipopt::IterativeSolverTerminationTester
void	GetVectors (Index ndim, const Number *array, SmartPtr< const Vector > &comp_x, SmartPtr< const Vector > &comp_s, SmartPtr< const Vector > &comp_c, SmartPtr< const Vector > &comp_d)
	Method for copying a long augmented system array into Vectors in Ipopt notation.

InexactData &	InexData ()
	Method to easily access Inexact data.

InexactCq &	InexCq ()
	Method to easily access Inexact calculated quantities.

Protected Member Functions inherited from Ipopt::AlgorithmStrategyObject
const Journalist &	Jnlst () const

IpoptNLP &	IpNLP () const

IpoptData &	IpData () const

IpoptCalculatedQuantities &	IpCq () const

bool	HaveIpData () const

Detailed Description

This class implements the termination tests for the primal-dual system.

Definition at line 18 of file IpInexactPDTerminationTester.hpp.

Constructor & Destructor Documentation

◆ InexactPDTerminationTester()

Ipopt::InexactPDTerminationTester::InexactPDTerminationTester ( )

Default constructor.

◆ ~InexactPDTerminationTester()

virtual Ipopt::InexactPDTerminationTester::~InexactPDTerminationTester ( )

virtual

Destructor.

Member Function Documentation

◆ InitializeImpl()

virtual bool Ipopt::InexactPDTerminationTester::InitializeImpl	(	const OptionsList &	options,
		const std::string &	prefix
	)

virtual

Implementation of the initialization method that has to be overloaded by for each derived class.

Implements Ipopt::IterativeSolverTerminationTester.

◆ RegisterOptions()

static void Ipopt::InexactPDTerminationTester::RegisterOptions ( SmartPtr< RegisteredOptions > roptions )

static

◆ InitializeSolve()

virtual bool Ipopt::InexactPDTerminationTester::InitializeSolve ( )

virtual

Method for initializing for the next iterative solve.

This must be call before the test methods are called.

Implements Ipopt::IterativeSolverTerminationTester.

◆ TestTermination()

virtual ETerminationTest Ipopt::InexactPDTerminationTester::TestTermination	(	Index	ndim,
		const Number *	sol,
		const Number *	resid,
		Index	iter,
		Number	norm2_rhs
	)

virtual

This method checks if the current solution of the iterative linear solver is good enough (by returning the corresponding satisfied termination test), or if the Hessian should be modified.

The input is the dimension of the augmented system, the current solution vector of the augmented system, the current residual vector.

Implements Ipopt::IterativeSolverTerminationTester.