#include <itkLBFGS2Optimizerv4.h>

Inheritance diagram for itk::LBFGS2Optimizerv4:

Collaboration diagram for itk::LBFGS2Optimizerv4:

Public Types
using	ConstPointer = SmartPointer< const Self >

using	LineSearchMethodEnum = LBFGS2Optimizerv4Enums::LineSearchMethod

using	MetricType = Superclass::MetricType

using	ParametersType = Superclass::ParametersType

using	Pointer = SmartPointer< Self >

using	PrecisionType = double

using	ScalesType = Superclass::ScalesType

using	Self = LBFGS2Optimizerv4

using	Superclass = GradientDescentOptimizerv4Template< double >

Public Types inherited from itk::GradientDescentOptimizerv4Template< double >
using	ConstPointer = SmartPointer< const Self >

using	InternalComputationValueType = double

using	Pointer = SmartPointer< Self >

using	Self = GradientDescentOptimizerv4Template

using	Superclass = GradientDescentOptimizerBasev4Template< double >

Public Types inherited from itk::GradientDescentOptimizerBasev4Template< double >
using	ConstPointer = SmartPointer< const Self >

using	ConvergenceMonitoringType = itk::Function::WindowConvergenceMonitoringFunction< double >

using	InternalComputationValueType = double

using	MetricTypePointer = typename MetricType::Pointer

using	Pointer = SmartPointer< Self >

using	Self = GradientDescentOptimizerBasev4Template

using	Superclass = ObjectToObjectOptimizerBaseTemplate< double >

using	IndexRangeType = ThreadedIndexedContainerPartitioner::IndexRangeType

Public Types inherited from itk::ObjectToObjectOptimizerBaseTemplate< double >
using	ConstPointer = SmartPointer< const Self >

using	DerivativeType = typename MetricType::DerivativeType

using	MeasureType = typename MetricType::MeasureType

using	MetricType = ObjectToObjectMetricBaseTemplate< double >

using	MetricTypePointer = typename MetricType::Pointer

using	NumberOfParametersType = typename MetricType::NumberOfParametersType

using	ParametersType = OptimizerParameters< double >

using	Pointer = SmartPointer< Self >

using	ScalesEstimatorType = OptimizerParameterScalesEstimatorTemplate< double >

using	ScalesType = OptimizerParameters< double >

using	Self = ObjectToObjectOptimizerBaseTemplate

using	StopConditionDescriptionType = std::ostringstream

using	StopConditionReturnStringType = std::string

using	Superclass = Object

Public Types inherited from itk::Object
using	ConstPointer = SmartPointer< const Self >

using	Pointer = SmartPointer< Self >

using	Self = Object

using	Superclass = LightObject

Public Types inherited from itk::LightObject
using	ConstPointer = SmartPointer< const Self >

using	Pointer = SmartPointer< Self >

using	Self = LightObject

Public Member Functions
virtual ::itk::LightObject::Pointer	CreateAnother () const

virtual const char *	GetNameOfClass () const

virtual const StopConditionReturnStringType	GetStopConditionDescription () const override

void	ResumeOptimization () override

void	StartOptimization (bool doOnlyInitialization=false) override

Public Member Functions inherited from itk::GradientDescentOptimizerv4Template< double >
virtual ::itk::LightObject::Pointer	CreateAnother () const

virtual void	SetLearningRate (double _arg)

virtual const double &	GetLearningRate () const

virtual void	SetMaximumStepSizeInPhysicalUnits (double _arg)

virtual const double &	GetMaximumStepSizeInPhysicalUnits () const

virtual void	SetDoEstimateLearningRateAtEachIteration (bool _arg)

virtual const bool &	GetDoEstimateLearningRateAtEachIteration () const

virtual void	DoEstimateLearningRateAtEachIterationOn ()

virtual void	DoEstimateLearningRateAtEachIterationOff ()

virtual void	SetDoEstimateLearningRateOnce (bool _arg)

virtual const bool &	GetDoEstimateLearningRateOnce () const

virtual void	DoEstimateLearningRateOnceOn ()

virtual void	DoEstimateLearningRateOnceOff ()

virtual void	SetReturnBestParametersAndValue (bool _arg)

virtual const bool &	GetReturnBestParametersAndValue () const

virtual void	ReturnBestParametersAndValueOn ()

virtual void	ReturnBestParametersAndValueOff ()

void	StartOptimization (bool doOnlyInitialization=false) override

void	StopOptimization () override

void	ResumeOptimization () override

virtual void	EstimateLearningRate ()

Public Member Functions inherited from itk::GradientDescentOptimizerBasev4Template< double >
virtual const DerivativeType &	GetGradient () const

virtual const StopConditionObjectToObjectOptimizerEnum &	GetStopCondition () const

const StopConditionReturnStringType	GetStopConditionDescription () const override

virtual void	ModifyGradientByScales ()

virtual void	ModifyGradientByLearningRate ()

Public Member Functions inherited from itk::ObjectToObjectOptimizerBaseTemplate< double >
virtual void	SetMetric (MetricType *_arg)

virtual MetricType *	GetModifiableMetric ()

virtual const MetricType *	GetMetric () const

virtual const MeasureType &	GetCurrentMetricValue () const

virtual const MeasureType &	GetValue () const

virtual void	SetScales (const ScalesType &scales)

virtual const ScalesType &	GetScales () const

virtual const bool &	GetScalesAreIdentity () const

virtual void	SetWeights (ScalesType _arg)

virtual const ScalesType &	GetWeights () const

virtual const bool &	GetWeightsAreIdentity () const

bool	GetScalesInitialized () const

virtual void	SetScalesEstimator (ScalesEstimatorType *_arg)

virtual void	SetDoEstimateScales (bool _arg)

virtual const bool &	GetDoEstimateScales () const

virtual void	DoEstimateScalesOn ()

virtual void	DoEstimateScalesOff ()

virtual void	SetNumberOfWorkUnits (ThreadIdType number)

virtual const ThreadIdType &	GetNumberOfWorkUnits () const

virtual SizeValueType	GetCurrentIteration () const

virtual const ParametersType &	GetCurrentPosition () const

Public Member Functions inherited from itk::Object
unsigned long	AddObserver (const EventObject &event, Command *)

unsigned long	AddObserver (const EventObject &event, Command *) const

unsigned long	AddObserver (const EventObject &event, std::function< void(const EventObject &)> function) const

virtual void	DebugOff () const

virtual void	DebugOn () const

Command *	GetCommand (unsigned long tag)

bool	GetDebug () const

MetaDataDictionary &	GetMetaDataDictionary ()

const MetaDataDictionary &	GetMetaDataDictionary () const

virtual ModifiedTimeType	GetMTime () const

virtual const TimeStamp &	GetTimeStamp () const

bool	HasObserver (const EventObject &event) const

void	InvokeEvent (const EventObject &)

void	InvokeEvent (const EventObject &) const

virtual void	Modified () const

void	Register () const override

void	RemoveAllObservers ()

void	RemoveObserver (unsigned long tag)

void	SetDebug (bool debugFlag) const

void	SetReferenceCount (int) override

void	UnRegister () const noexcept override

void	SetMetaDataDictionary (const MetaDataDictionary &rhs)

void	SetMetaDataDictionary (MetaDataDictionary &&rrhs)

virtual void	SetObjectName (std::string _arg)

virtual const std::string &	GetObjectName () const

Public Member Functions inherited from itk::LightObject
Pointer	Clone () const

virtual void	Delete ()

virtual int	GetReferenceCount () const

void	Print (std::ostream &os, Indent indent=0) const

Static Public Member Functions
static Pointer	New ()

Static Public Member Functions inherited from itk::GradientDescentOptimizerv4Template< double >
static Pointer	New ()

Static Public Member Functions inherited from itk::Object
static bool	GetGlobalWarningDisplay ()

static void	GlobalWarningDisplayOff ()

static void	GlobalWarningDisplayOn ()

static Pointer	New ()

static void	SetGlobalWarningDisplay (bool val)

Static Public Member Functions inherited from itk::LightObject
static void	BreakOnError ()

static Pointer	New ()

std::unique_ptr< PrivateImplementationHolder >	m_Pimpl

bool	m_EstimateScalesAtEachIteration

double	m_CurrentStepSize

double	m_CurrentParameterNorm

double	m_CurrentGradientNorm

int	m_CurrentNumberOfEvaluations

int	m_StatusCode

void	SetHessianApproximationAccuracy (int m)

int	GetHessianApproximationAccuracy () const

void	SetSolutionAccuracy (PrecisionType epsilon)

PrecisionType	GetSolutionAccuracy () const

void	SetDeltaConvergenceDistance (int nPast)

int	GetDeltaConvergenceDistance () const

void	SetDeltaConvergenceTolerance (PrecisionType tol)

PrecisionType	GetDeltaConvergenceTolerance () const

void	SetMaximumIterations (int maxIterations)

int	GetMaximumIterations () const

SizeValueType	GetNumberOfIterations () const override

void	SetNumberOfIterations (const SizeValueType _arg) override

void	SetLineSearch (const LineSearchMethodEnum &linesearch)

LineSearchMethodEnum	GetLineSearch () const

void	SetMaximumLineSearchEvaluations (int n)

int	GetMaximumLineSearchEvaluations () const

void	SetMinimumLineSearchStep (PrecisionType step)

PrecisionType	GetMinimumLineSearchStep () const

void	SetMaximumLineSearchStep (PrecisionType step)

PrecisionType	GetMaximumLineSearchStep () const

void	SetLineSearchAccuracy (PrecisionType ftol)

PrecisionType	GetLineSearchAccuracy () const

void	SetWolfeCoefficient (PrecisionType wc)

PrecisionType	GetWolfeCoefficient () const

void	SetLineSearchGradientAccuracy (PrecisionType gtol)

PrecisionType	GetLineSearchGradientAccuracy () const

void	SetMachinePrecisionTolerance (PrecisionType xtol)

PrecisionType	GetMachinePrecisionTolerance () const

void	SetOrthantwiseCoefficient (PrecisionType orthant_c)

PrecisionType	GetOrthantwiseCoefficient () const

void	SetOrthantwiseStart (int start)

int	GetOrthantwiseStart () const

void	SetOrthantwiseEnd (int end)

int	GetOrthantwiseEnd () const

virtual PrecisionType	GetCurrentParameterNorm () const

virtual PrecisionType	GetCurrentGradientNorm () const

virtual PrecisionType	GetCurrentStepSize () const

virtual PrecisionType	GetCurrentNumberOfEvaluations () const

virtual void	SetEstimateScalesAtEachIteration (bool _arg)

virtual const bool &	GetEstimateScalesAtEachIteration () const

virtual void	EstimateScalesAtEachIterationOn ()

virtual void	EstimateScalesAtEachIterationOff ()

	LBFGS2Optimizerv4 ()

	~LBFGS2Optimizerv4 () override

void	PrintSelf (std::ostream &os, Indent indent) const override

int	UpdateProgress (const PrecisionType x, const PrecisionType g, const PrecisionType fx, const PrecisionType xnorm, const PrecisionType gnorm, const PrecisionType step, int n, int k, int ls)

PrecisionType	EvaluateCost (const PrecisionType x, PrecisionType g, const int n, const PrecisionType step)

static int	UpdateProgressCallback (void instance, const PrecisionType x, const PrecisionType *g, const PrecisionType fx, const PrecisionType xnorm, const PrecisionType gnorm, const PrecisionType step, int n, int k, int ls)

static PrecisionType	EvaluateCostCallback (void instance, const PrecisionType x, PrecisionType *g, const int n, const PrecisionType step)

void	SetMinimumConvergenceValue (double) override

void	SetConvergenceWindowSize (SizeValueType) override

const double &	GetConvergenceValue () const override

void	AdvanceOneStep () override

Additional Inherited Members
Protected Member Functions inherited from itk::GradientDescentOptimizerv4Template< double >
void	ModifyGradientByScalesOverSubRange (const IndexRangeType &subrange) override

void	ModifyGradientByLearningRateOverSubRange (const IndexRangeType &subrange) override

	GradientDescentOptimizerv4Template ()

	~GradientDescentOptimizerv4Template () override=default

void	PrintSelf (std::ostream &os, Indent indent) const override

Protected Member Functions inherited from itk::GradientDescentOptimizerBasev4Template< double >
	GradientDescentOptimizerBasev4Template ()

	~GradientDescentOptimizerBasev4Template () override=default

Protected Member Functions inherited from itk::ObjectToObjectOptimizerBaseTemplate< double >
	ObjectToObjectOptimizerBaseTemplate ()

	~ObjectToObjectOptimizerBaseTemplate () override

void	PrintSelf (std::ostream &os, Indent indent) const override

Protected Member Functions inherited from itk::Object
	Object ()

	~Object () override

bool	PrintObservers (std::ostream &os, Indent indent) const

virtual void	SetTimeStamp (const TimeStamp &timeStamp)

Protected Member Functions inherited from itk::LightObject
virtual LightObject::Pointer	InternalClone () const

	LightObject ()

virtual void	PrintHeader (std::ostream &os, Indent indent) const

virtual void	PrintTrailer (std::ostream &os, Indent indent) const

virtual	~LightObject ()

Protected Attributes inherited from itk::GradientDescentOptimizerv4Template< double >
double	m_LearningRate

double	m_MinimumConvergenceValue

double	m_ConvergenceValue

MeasureType	m_CurrentBestValue

ParametersType	m_BestParameters

bool	m_ReturnBestParametersAndValue

DerivativeType	m_PreviousGradient

Protected Attributes inherited from itk::GradientDescentOptimizerBasev4Template< double >
bool	m_DoEstimateLearningRateAtEachIteration

bool	m_DoEstimateLearningRateOnce

double	m_MaximumStepSizeInPhysicalUnits

bool	m_UseConvergenceMonitoring

SizeValueType	m_ConvergenceWindowSize

ConvergenceMonitoringType::Pointer	m_ConvergenceMonitoring

DomainThreader< ThreadedIndexedContainerPartitioner, Self >::Pointer	m_ModifyGradientByScalesThreader

DomainThreader< ThreadedIndexedContainerPartitioner, Self >::Pointer	m_ModifyGradientByLearningRateThreader

bool	m_Stop

StopConditionObjectToObjectOptimizerEnum	m_StopCondition

StopConditionDescriptionType	m_StopConditionDescription

DerivativeType	m_Gradient

Protected Attributes inherited from itk::ObjectToObjectOptimizerBaseTemplate< double >
MetricTypePointer	m_Metric

ThreadIdType	m_NumberOfWorkUnits

SizeValueType	m_CurrentIteration

SizeValueType	m_NumberOfIterations

MeasureType	m_CurrentMetricValue

ScalesType	m_Scales

ScalesType	m_Weights

bool	m_ScalesAreIdentity

ScalesEstimatorType::Pointer	m_ScalesEstimator

bool	m_WeightsAreIdentity

bool	m_DoEstimateScales

Protected Attributes inherited from itk::LightObject
std::atomic< int >	m_ReferenceCount

Detailed Description

Wrap of the libLBFGS[1] algorithm for use in ITKv4 registration framework. LibLBFGS is a translation of LBFGS code by Nocedal [2] and adds the orthantwise limited-memmory Quais-Newton method [3] for optimization with L1-norm on the parameters.

LBFGS is a quasi-Newton method uses an approximate estimate of the inverse Hessian $(\nabla^2 f(x) )^-1$ to scale the gradient step:

$x_{n+1} = x_n - s (\nabla^2 f(x_n) )^-1 \nabla f(x)$

with $ s $ the step size.

The inverse Hessian is approximated from the gradients of previous iteration and thus only the gradient of the objective function is required.

The step size $ s $ is determined through line search which defaults to the approach by More and Thuente [4]. This line search approach finds a step size such that

$\lVert \nabla f(x + s (\nabla^2 f(x_n) )^{-1} \nabla f(x) ) \rVert \le \nu \lVert \nabla f(x) \rVert$

The parameter $\nu$ is set through SetLineSearchAccuracy() (default 0.9) and SetGradientLineSearchAccuracy()

Instead of the More-Tunete method, backtracking with three different conditions [7] are available and can be set through SetLineSearch():

LINESEARCH_BACKTRACKING_ARMIJO
LINESEARCH_BACKTRACKING_WOLFE
LINESEARCH_BACKTRACKING_STRONG_WOLFE

The optimization stops when either the gradient satisfies the condition

$\lVert \nabla f(x) \rVert \le \epsilon \max(1, \lVert X \rVert)$

or a maximum number of function evaluations has been reached. The tolerance $\epsilon$ is set through SetSolutionAccuracy() (default 1e-5) and the maximum number of function evaluations is set through SetMaximumIterations() (default 0 = no maximum).

References:

[1] libLBFGS

[2] NETLIB lbfgs

[3] Galen Andrew and Jianfeng Gao. Scalable training of L1-regularized log-linear models. 24th International Conference on Machine Learning, pp. 33-40, 2007.

[4] Jorge Nocedal. Updating Quasi-Newton Matrices with Limited Storage. Mathematics of Computation, Vol. 35, No. 151, pp. 773-782, 1980.

[5] Dong C. Liu and Jorge Nocedal. On the limited memory BFGS method for large scale optimization. Mathematical Programming B, Vol. 45, No. 3, pp. 503-528, 1989.

[6] More, J. J. and D. J. Thuente. Line Search Algorithms with Guaranteed Sufficient Decrease. ACM Transactions on Mathematical Software 20, no. 3 (1994): 286-307.

[7] John E. Dennis and Robert B. Schnabel. Numerical Methods for Unconstrained Optimization and Nonlinear Equations, Englewood Cliffs, 1983.

Examples: Examples/RegistrationITKv4/DeformableRegistration6.cxx.

Definition at line 162 of file itkLBFGS2Optimizerv4.h.

Member Typedef Documentation

◆ ConstPointer

using itk::LBFGS2Optimizerv4::ConstPointer = SmartPointer<const Self>

Definition at line 194 of file itkLBFGS2Optimizerv4.h.

◆ LineSearchMethodEnum

using itk::LBFGS2Optimizerv4::LineSearchMethodEnum = LBFGS2Optimizerv4Enums::LineSearchMethod

Definition at line 169 of file itkLBFGS2Optimizerv4.h.

◆ MetricType

using itk::LBFGS2Optimizerv4::MetricType = Superclass::MetricType

Definition at line 196 of file itkLBFGS2Optimizerv4.h.

◆ ParametersType

using itk::LBFGS2Optimizerv4::ParametersType = Superclass::ParametersType

Definition at line 197 of file itkLBFGS2Optimizerv4.h.

◆ Pointer

using itk::LBFGS2Optimizerv4::Pointer = SmartPointer<Self>

Definition at line 193 of file itkLBFGS2Optimizerv4.h.

◆ PrecisionType

using itk::LBFGS2Optimizerv4::PrecisionType = double

TODO: currently only double is used in lbfgs need to figure out how to make it a template parameter and set the required define so lbfgs.h uses the correct version

Definition at line 188 of file itkLBFGS2Optimizerv4.h.

◆ ScalesType

using itk::LBFGS2Optimizerv4::ScalesType = Superclass::ScalesType

Definition at line 198 of file itkLBFGS2Optimizerv4.h.

◆ Self

using itk::LBFGS2Optimizerv4::Self = LBFGS2Optimizerv4

Standard "Self" type alias.

Definition at line 191 of file itkLBFGS2Optimizerv4.h.

◆ Superclass

using itk::LBFGS2Optimizerv4::Superclass = GradientDescentOptimizerv4Template<double>

Definition at line 192 of file itkLBFGS2Optimizerv4.h.

Constructor & Destructor Documentation

◆ LBFGS2Optimizerv4()

itk::LBFGS2Optimizerv4::LBFGS2Optimizerv4 ( )

protected

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ ~LBFGS2Optimizerv4()

itk::LBFGS2Optimizerv4::~LBFGS2Optimizerv4 ( )

overrideprotected

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Member Function Documentation

◆ AdvanceOneStep()

void itk::LBFGS2Optimizerv4::AdvanceOneStep ( )

inlineoverrideprivatevirtual

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Reimplemented from itk::GradientDescentOptimizerv4Template< double >.

Definition at line 562 of file itkLBFGS2Optimizerv4.h.

◆ CreateAnother()

virtual::itk::LightObject::Pointer itk::LBFGS2Optimizerv4::CreateAnother ( ) const

virtual

Create an object from an instance, potentially deferring to a factory. This method allows you to create an instance of an object that is exactly the same type as the referring object. This is useful in cases where an object has been cast back to a base class.

Reimplemented from itk::Object.

◆ EstimateScalesAtEachIterationOff()

virtual void itk::LBFGS2Optimizerv4::EstimateScalesAtEachIterationOff ( )

virtual

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ EstimateScalesAtEachIterationOn()

virtual void itk::LBFGS2Optimizerv4::EstimateScalesAtEachIterationOn ( )

virtual

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ EvaluateCost()

PrecisionType itk::LBFGS2Optimizerv4::EvaluateCost	(	const PrecisionType *	x,
		PrecisionType *	g,
		const int	n,
		const PrecisionType	step
	)

protected

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ EvaluateCostCallback()

static PrecisionType itk::LBFGS2Optimizerv4::EvaluateCostCallback	(	void *	instance,
		const PrecisionType *	x,
		PrecisionType *	g,
		const int	n,
		const PrecisionType	step
	)

staticprotected

Function evalation callback from libLBFGS frowrad to instance

◆ GetConvergenceValue()

const double& itk::LBFGS2Optimizerv4::GetConvergenceValue ( ) const

inlineoverrideprivatevirtual

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Reimplemented from itk::GradientDescentOptimizerv4Template< double >.

Definition at line 553 of file itkLBFGS2Optimizerv4.h.

◆ GetCurrentGradientNorm()

virtual PrecisionType itk::LBFGS2Optimizerv4::GetCurrentGradientNorm ( ) const

virtual

Get gradient norm of current iteration

◆ GetCurrentNumberOfEvaluations()

virtual PrecisionType itk::LBFGS2Optimizerv4::GetCurrentNumberOfEvaluations ( ) const

virtual

Get number of evaluations for current iteration

◆ GetCurrentParameterNorm()

virtual PrecisionType itk::LBFGS2Optimizerv4::GetCurrentParameterNorm ( ) const

virtual

Get parameter norm of current iteration

◆ GetCurrentStepSize()

virtual PrecisionType itk::LBFGS2Optimizerv4::GetCurrentStepSize ( ) const

virtual

Get stepsize of current iteration

◆ GetDeltaConvergenceDistance()

int itk::LBFGS2Optimizerv4::GetDeltaConvergenceDistance ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetDeltaConvergenceTolerance()

PrecisionType itk::LBFGS2Optimizerv4::GetDeltaConvergenceTolerance ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetEstimateScalesAtEachIteration()

virtual const bool& itk::LBFGS2Optimizerv4::GetEstimateScalesAtEachIteration ( ) const

virtual

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetHessianApproximationAccuracy()

int itk::LBFGS2Optimizerv4::GetHessianApproximationAccuracy ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetLineSearch()

LineSearchMethodEnum itk::LBFGS2Optimizerv4::GetLineSearch ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetLineSearchAccuracy()

PrecisionType itk::LBFGS2Optimizerv4::GetLineSearchAccuracy ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetLineSearchGradientAccuracy()

PrecisionType itk::LBFGS2Optimizerv4::GetLineSearchGradientAccuracy ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetMachinePrecisionTolerance()

PrecisionType itk::LBFGS2Optimizerv4::GetMachinePrecisionTolerance ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetMaximumIterations()

int itk::LBFGS2Optimizerv4::GetMaximumIterations ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetMaximumLineSearchEvaluations()

int itk::LBFGS2Optimizerv4::GetMaximumLineSearchEvaluations ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetMaximumLineSearchStep()

PrecisionType itk::LBFGS2Optimizerv4::GetMaximumLineSearchStep ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetMinimumLineSearchStep()

PrecisionType itk::LBFGS2Optimizerv4::GetMinimumLineSearchStep ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetNameOfClass()

virtual const char* itk::LBFGS2Optimizerv4::GetNameOfClass ( ) const

virtual

Run-time type information (and related methods).

Reimplemented from itk::GradientDescentOptimizerv4Template< double >.

◆ GetNumberOfIterations()

SizeValueType itk::LBFGS2Optimizerv4::GetNumberOfIterations ( ) const

inlineoverridevirtual

Aliased to Set/Get MaximumIterations to match base class interface.

Reimplemented from itk::ObjectToObjectOptimizerBaseTemplate< double >.

Definition at line 296 of file itkLBFGS2Optimizerv4.h.

◆ GetOrthantwiseCoefficient()

PrecisionType itk::LBFGS2Optimizerv4::GetOrthantwiseCoefficient ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetOrthantwiseEnd()

int itk::LBFGS2Optimizerv4::GetOrthantwiseEnd ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetOrthantwiseStart()

int itk::LBFGS2Optimizerv4::GetOrthantwiseStart ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetSolutionAccuracy()

PrecisionType itk::LBFGS2Optimizerv4::GetSolutionAccuracy ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ GetStopConditionDescription()

virtual const StopConditionReturnStringType itk::LBFGS2Optimizerv4::GetStopConditionDescription ( ) const

overridevirtual

Stop condition return string type

Implements itk::ObjectToObjectOptimizerBaseTemplate< double >.

◆ GetWolfeCoefficient()

PrecisionType itk::LBFGS2Optimizerv4::GetWolfeCoefficient ( ) const

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ New()

static Pointer itk::LBFGS2Optimizerv4::New ( )

static

Method for creation through the object factory.

◆ PrintSelf()

void itk::LBFGS2Optimizerv4::PrintSelf	(	std::ostream &	os,
		Indent	indent
	)		const

overrideprotectedvirtual

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Reimplemented from itk::Object.

◆ ResumeOptimization()

void itk::LBFGS2Optimizerv4::ResumeOptimization ( )

overridevirtual

Resume optimization. This runs the optimization loop, and allows continuation of stopped optimization

Implements itk::GradientDescentOptimizerBasev4Template< double >.

◆ SetConvergenceWindowSize()

void itk::LBFGS2Optimizerv4::SetConvergenceWindowSize ( SizeValueType )

inlineoverrideprivatevirtual

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Reimplemented from itk::GradientDescentOptimizerv4Template< double >.

Definition at line 548 of file itkLBFGS2Optimizerv4.h.

◆ SetDeltaConvergenceDistance()

void itk::LBFGS2Optimizerv4::SetDeltaConvergenceDistance ( int nPast )

Set/Get distance for delta-based convergence test. This parameter determines the distance, in iterations, to compute the rate of decrease of the objective function. If the value of this parameter is zero, the library does not perform the delta-based convergence test. The default value is 0.

◆ SetDeltaConvergenceTolerance()

void itk::LBFGS2Optimizerv4::SetDeltaConvergenceTolerance ( PrecisionType tol )

Delta for convergence test. This parameter determines the minimum rate of decrease of the objective function. The library stops iterations when the following condition is met: $(f' - f) / f < \delta$ , where f' is the objective value of past iterations ago, and f is the objective value of the current iteration. The default value is 0.

◆ SetEstimateScalesAtEachIteration()

virtual void itk::LBFGS2Optimizerv4::SetEstimateScalesAtEachIteration ( bool _arg )

virtual

Option to use ScalesEstimator for estimating scales at each iteration. The estimation overrides the scales set by SetScales(). Default is true.

◆ SetHessianApproximationAccuracy()

void itk::LBFGS2Optimizerv4::SetHessianApproximationAccuracy ( int m )

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

◆ SetLineSearch()

void itk::LBFGS2Optimizerv4::SetLineSearch ( const LineSearchMethodEnum & linesearch )

The line search algorithm. This parameter specifies a line search algorithm to be used by the L-BFGS routine. See lbfgs.h for enumeration of line search type. Defaults to More-Thuente's method.

◆ SetLineSearchAccuracy()

void itk::LBFGS2Optimizerv4::SetLineSearchAccuracy ( PrecisionType ftol )

A parameter to control the accuracy of the line search routine. The default value is 1e-4. This parameter should be greater than zero and smaller than 0.5.

◆ SetLineSearchGradientAccuracy()

void itk::LBFGS2Optimizerv4::SetLineSearchGradientAccuracy ( PrecisionType gtol )

A parameter to control the gradient accuracy of the More-Thuente line search routine. The default value is 0.9. If the function and gradient evaluations are inexpensive with respect to the cost of the iteration (which is sometimes the case when solving very large problems) it may be advantageous to set this parameter to a small value. A typical small value is 0.1. This parameter shuold be greater than the ftol parameter (1e-4) and smaller than 1.0.

◆ SetMachinePrecisionTolerance()

void itk::LBFGS2Optimizerv4::SetMachinePrecisionTolerance ( PrecisionType xtol )

The machine precision for floating-point values. This parameter must be a positive value set by a client program to estimate the machine precision. The line search routine will terminate with the status code (LBFGSERR_ROUNDING_ERROR) if the relative width of the interval of uncertainty is less than this parameter.

◆ SetMaximumIterations()

void itk::LBFGS2Optimizerv4::SetMaximumIterations ( int maxIterations )

The maximum number of iterations. The lbfgs() function terminates an optimization process with LBFGSERR_MAXIMUMITERATION status code when the iteration count exceeds this parameter. Setting this parameter to zero continues an optimization process until a convergence or error. The default value is 0.

◆ SetMaximumLineSearchEvaluations()

void itk::LBFGS2Optimizerv4::SetMaximumLineSearchEvaluations ( int n )

The maximum number of trials for the line search. This parameter controls the number of function and gradients evaluations per iteration for the line search routine. The default value is 20.

◆ SetMaximumLineSearchStep()

void itk::LBFGS2Optimizerv4::SetMaximumLineSearchStep ( PrecisionType step )

The maximum step of the line search. The default value is 1e+20. This value need not be modified unless the exponents are too large for the machine being used, or unless the problem is extremely badly scaled (in which case the exponents should be increased).

◆ SetMinimumConvergenceValue()

void itk::LBFGS2Optimizerv4::SetMinimumConvergenceValue ( double )

inlineoverrideprivatevirtual

itkGradientDecentOptimizerv4Template specific non supported methods.

Reimplemented from itk::GradientDescentOptimizerv4Template< double >.

Definition at line 544 of file itkLBFGS2Optimizerv4.h.

◆ SetMinimumLineSearchStep()

void itk::LBFGS2Optimizerv4::SetMinimumLineSearchStep ( PrecisionType step )

The minimum step of the line search routine. The default value is 1e-20. This value need not be modified unless the exponents are too large for the machine being used, or unless the problem is extremely badly scaled (in which case the exponents should be increased).

◆ SetNumberOfIterations()

void itk::LBFGS2Optimizerv4::SetNumberOfIterations ( const SizeValueType _arg )

inlineoverridevirtual

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Reimplemented from itk::ObjectToObjectOptimizerBaseTemplate< double >.

Definition at line 301 of file itkLBFGS2Optimizerv4.h.

◆ SetOrthantwiseCoefficient()

void itk::LBFGS2Optimizerv4::SetOrthantwiseCoefficient ( PrecisionType orthant_c )

Coeefficient for the L1 norm of variables. This parameter should be set to zero for standard minimization problems. Setting this parameter to a positive value activates Orthant-Wise Limited-memory Quasi-Newton (OWL-QN) method, which minimizes the objective function F(x) combined with the L1 norm |x| of the variables, $F(x) + C |x|}. $ . This parameter is the coefficient for the |x|, i.e., C. As the L1 norm |x| is not differentiable at zero, the library modifies function and gradient evaluations from a client program suitably; a client program thus have only to return the function value F(x) and gradients G(x) as usual. The default value is zero.

◆ SetOrthantwiseEnd()

void itk::LBFGS2Optimizerv4::SetOrthantwiseEnd ( int end )

End index for computing L1 norm of the variables. This parameter is valid only for OWL-QN method (i.e., $ orthantwise_c != 0 $ ). This parameter e, (0 < e <= N) specifies the index number at which the library stops computing the L1 norm of the variables x,

◆ SetOrthantwiseStart()

void itk::LBFGS2Optimizerv4::SetOrthantwiseStart ( int start )

Start index for computing L1 norm of the variables. This parameter is valid only for OWL-QN method (i.e., $ orthantwise_c != 0 $ ). This parameter b (0 <= b < N) specifies the index number from which the library computes the L1 norm of the variables x,

$|x| := |x_{b}| + |x_{b+1}| + ... + |x_{N}| .$

In other words, variables $x_1, ..., x_{b-1}$ are not used for computing the L1 norm. Setting b, (0 < b < N), one can protect variables, $x_1, ..., x_{b-1}$ (e.g., a bias term of logistic regression) from being regularized. The default value is zero.

◆ SetSolutionAccuracy()

void itk::LBFGS2Optimizerv4::SetSolutionAccuracy ( PrecisionType epsilon )

Set/Get epsilon for convergence test. This parameter determines the accuracy with which the solution is to be found. A minimization terminates when $||g|| < \epsilon * max(1, ||x||)$ , where ||.|| denotes the Euclidean (L2) norm. The default value is 1e-5.

◆ SetWolfeCoefficient()

void itk::LBFGS2Optimizerv4::SetWolfeCoefficient ( PrecisionType wc )

A coefficient for the Wolfe condition. This parameter is valid only when the backtracking line-search algorithm is used with the Wolfe condition, LINESEARCH_BACKTRACKING_STRONG_WOLFE or LINESEARCH_BACKTRACKING_WOLFE . The default value is 0.9. This parameter should be greater than the ftol parameter and smaller than 1.0.

◆ StartOptimization()

void itk::LBFGS2Optimizerv4::StartOptimization ( bool doOnlyInitialization = false )

overridevirtual

Start optimization with an initial value.

Reimplemented from itk::ObjectToObjectOptimizerBaseTemplate< double >.

◆ UpdateProgress()

int itk::LBFGS2Optimizerv4::UpdateProgress	(	const PrecisionType *	x,
		const PrecisionType *	g,
		const PrecisionType	fx,
		const PrecisionType	xnorm,
		const PrecisionType	gnorm,
		const PrecisionType	step,
		int	n,
		int	k,
		int	ls
	)

protected

Update the progress as reported from libLBFSG and notify itkObject

◆ UpdateProgressCallback()

static int itk::LBFGS2Optimizerv4::UpdateProgressCallback	(	void *	instance,
		const PrecisionType *	x,
		const PrecisionType *	g,
		const PrecisionType	fx,
		const PrecisionType	xnorm,
		const PrecisionType	gnorm,
		const PrecisionType	step,
		int	n,
		int	k,
		int	ls
	)

staticprotected

Progress callback from libLBFGS forwards it to the specific instance

Member Data Documentation

◆ m_CurrentGradientNorm

double itk::LBFGS2Optimizerv4::m_CurrentGradientNorm

private

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Definition at line 536 of file itkLBFGS2Optimizerv4.h.

◆ m_CurrentNumberOfEvaluations

int itk::LBFGS2Optimizerv4::m_CurrentNumberOfEvaluations

private

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Definition at line 537 of file itkLBFGS2Optimizerv4.h.

◆ m_CurrentParameterNorm

double itk::LBFGS2Optimizerv4::m_CurrentParameterNorm

private

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Definition at line 535 of file itkLBFGS2Optimizerv4.h.

◆ m_CurrentStepSize

double itk::LBFGS2Optimizerv4::m_CurrentStepSize

private

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Definition at line 534 of file itkLBFGS2Optimizerv4.h.

◆ m_EstimateScalesAtEachIteration

bool itk::LBFGS2Optimizerv4::m_EstimateScalesAtEachIteration

private

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Definition at line 533 of file itkLBFGS2Optimizerv4.h.

◆ m_Pimpl

std::unique_ptr<PrivateImplementationHolder> itk::LBFGS2Optimizerv4::m_Pimpl

private

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Definition at line 530 of file itkLBFGS2Optimizerv4.h.

◆ m_StatusCode

int itk::LBFGS2Optimizerv4::m_StatusCode

private

Set/Get the number of corrections to approximate the inverse hessian matrix. The L-BFGS routine stores the computation results of previous m iterations to approximate the inverse hessian matrix of the current iteration. This parameter controls the size of the limited memories (corrections). The default value is 6. Values less than 3 are not recommended. Large values will result in excessive computing time.

Definition at line 538 of file itkLBFGS2Optimizerv4.h.

The documentation for this class was generated from the following file:

itkLBFGS2Optimizerv4.h

Public Types

Public Member Functions

Static Public Member Functions

Additional Inherited Members

Detailed Description

Member Typedef Documentation

◆ ConstPointer

◆ LineSearchMethodEnum

◆ MetricType

◆ ParametersType

◆ Pointer

◆ PrecisionType

◆ ScalesType

◆ Self

◆ Superclass

Constructor & Destructor Documentation

◆ LBFGS2Optimizerv4()

◆ ~LBFGS2Optimizerv4()

Member Function Documentation

◆ AdvanceOneStep()

◆ CreateAnother()

◆ EstimateScalesAtEachIterationOff()

◆ EstimateScalesAtEachIterationOn()

◆ EvaluateCost()

◆ EvaluateCostCallback()

◆ GetConvergenceValue()

◆ GetCurrentGradientNorm()

◆ GetCurrentNumberOfEvaluations()

◆ GetCurrentParameterNorm()

◆ GetCurrentStepSize()

◆ GetDeltaConvergenceDistance()

◆ GetDeltaConvergenceTolerance()

◆ GetEstimateScalesAtEachIteration()

◆ GetHessianApproximationAccuracy()

◆ GetLineSearch()

◆ GetLineSearchAccuracy()

◆ GetLineSearchGradientAccuracy()

◆ GetMachinePrecisionTolerance()

◆ GetMaximumIterations()

◆ GetMaximumLineSearchEvaluations()

◆ GetMaximumLineSearchStep()

◆ GetMinimumLineSearchStep()

◆ GetNameOfClass()

◆ GetNumberOfIterations()

◆ GetOrthantwiseCoefficient()

◆ GetOrthantwiseEnd()

◆ GetOrthantwiseStart()

◆ GetSolutionAccuracy()

◆ GetStopConditionDescription()

◆ GetWolfeCoefficient()

◆ New()

◆ PrintSelf()

◆ ResumeOptimization()

◆ SetConvergenceWindowSize()

◆ SetDeltaConvergenceDistance()

◆ SetDeltaConvergenceTolerance()

◆ SetEstimateScalesAtEachIteration()

◆ SetHessianApproximationAccuracy()

◆ SetLineSearch()

◆ SetLineSearchAccuracy()

◆ SetLineSearchGradientAccuracy()

◆ SetMachinePrecisionTolerance()

◆ SetMaximumIterations()

◆ SetMaximumLineSearchEvaluations()

◆ SetMaximumLineSearchStep()

◆ SetMinimumConvergenceValue()

◆ SetMinimumLineSearchStep()

◆ SetNumberOfIterations()

◆ SetOrthantwiseCoefficient()

◆ SetOrthantwiseEnd()

◆ SetOrthantwiseStart()

◆ SetSolutionAccuracy()

◆ SetWolfeCoefficient()

◆ StartOptimization()

◆ UpdateProgress()

◆ UpdateProgressCallback()

Member Data Documentation

◆ m_CurrentGradientNorm

◆ m_CurrentNumberOfEvaluations

◆ m_CurrentParameterNorm