ITK  5.4.0
Insight Toolkit
Public Types | Public Member Functions | Static Public Member Functions | Protected Types | Protected Member Functions | Private Attributes | List of all members

#include <itkLBFGSOptimizer.h>

Detailed Description

Wrap of the vnl_lbfgs algorithm for use in ITKv4 registration framework. The vnl_lbfgs is a wrapper for the NETLIB fortran code by Nocedal [1].

LBFGS is a quasi-Newton method. Quasi-Newton methods use 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 with 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) The default step length, i.e. starting step length for the line search, is set through SetDefaultStepLength() (default 1.0).

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 SetGradientConvergenceTolerance() (default 1e-5) and the maximum number of function evaluations is set through SetMaximumNumberOfFunctionEvaluations() (default 2000).

Note: The scales set through SetScales should be set or left at one. Otherwise the Hessian approximation will be disturbed and the optimizer is unlikely to find a minima.

References:

[1] NETLIB lbfgs

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

[3] 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.

[4] 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.

Examples
SphinxExamples/src/Core/Transform/GlobalRegistrationTwoImagesBSpline/Code.cxx.

Definition at line 86 of file itkLBFGSOptimizer.h.

+ Inheritance diagram for itk::LBFGSOptimizer:
+ Collaboration diagram for itk::LBFGSOptimizer:

Public Types

using ConstPointer = SmartPointer< const Self >
 
using InternalOptimizerType = vnl_lbfgs
 
using InternalParametersType = vnl_vector< double >
 
using Pointer = SmartPointer< Self >
 
using Self = LBFGSOptimizer
 
using Superclass = SingleValuedNonLinearVnlOptimizer
 
- Public Types inherited from itk::SingleValuedNonLinearVnlOptimizer
using CommandType = ReceptorMemberCommand< Self >
 
using ConstPointer = SmartPointer< const Self >
 
using Pointer = SmartPointer< Self >
 
using Self = SingleValuedNonLinearVnlOptimizer
 
using Superclass = SingleValuedNonLinearOptimizer
 
- Public Types inherited from itk::SingleValuedNonLinearOptimizer
using ConstPointer = SmartPointer< const Self >
 
using CostFunctionPointer = CostFunctionType::Pointer
 
using CostFunctionType = SingleValuedCostFunction
 
using DerivativeType = CostFunctionType::DerivativeType
 
using MeasureType = CostFunctionType::MeasureType
 
using ParametersType = Superclass::ParametersType
 
using Pointer = SmartPointer< Self >
 
using Self = SingleValuedNonLinearOptimizer
 
using Superclass = NonLinearOptimizer
 
- Public Types inherited from itk::NonLinearOptimizer
using ConstPointer = SmartPointer< const Self >
 
using ParametersType = Superclass::ParametersType
 
using Pointer = SmartPointer< Self >
 
using ScalesType = Superclass::ScalesType
 
using Self = NonLinearOptimizer
 
using Superclass = Optimizer
 
- Public Types inherited from itk::Optimizer
using ConstPointer = SmartPointer< const Self >
 
using ParametersType = OptimizerParameters< double >
 
using Pointer = SmartPointer< Self >
 
using ScalesType = Array< double >
 
using Self = Optimizer
 
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 double GetDefaultStepLength ()
 
virtual double GetGradientConvergenceTolerance ()
 
virtual double GetLineSearchAccuracy ()
 
virtual unsigned int GetMaximumNumberOfFunctionEvaluations ()
 
const char * GetNameOfClass () const override
 
vnl_lbfgs * GetOptimizer ()
 
const std::string GetStopConditionDescription () const override
 
virtual bool GetTrace ()
 
MeasureType GetValue () const
 
void SetCostFunction (SingleValuedCostFunction *costFunction) override
 
virtual void SetDefaultStepLength (double f)
 
virtual void SetGradientConvergenceTolerance (double f)
 
virtual void SetLineSearchAccuracy (double f)
 
virtual void SetMaximumNumberOfFunctionEvaluations (unsigned int n)
 
virtual void SetTrace (bool flag)
 
void StartOptimization () override
 
virtual void TraceOn ()
 
- Public Member Functions inherited from itk::SingleValuedNonLinearVnlOptimizer
virtual bool CanUseScales () const
 
const char * GetNameOfClass () const override
 
virtual const bool & GetMaximize () const
 
virtual void SetMaximize (bool _arg)
 
virtual void MaximizeOn ()
 
bool GetMinimize () const
 
void SetMinimize (bool v)
 
void MinimizeOn ()
 
void MinimizeOff ()
 
virtual const MeasureTypeGetCachedValue () const
 
virtual const DerivativeTypeGetCachedDerivative () const
 
virtual const ParametersTypeGetCachedCurrentPosition () const
 
- Public Member Functions inherited from itk::SingleValuedNonLinearOptimizer
virtual CostFunctionTypeGetModifiableCostFunction ()
 
MeasureType GetValue (const ParametersType &parameters) const
 
virtual void SetCostFunction (CostFunctionType *costFunction)
 
- Public Member Functions inherited from itk::Optimizer
virtual const ParametersTypeGetCurrentPosition () const
 
virtual const ParametersTypeGetInitialPosition () const
 
virtual void SetInitialPosition (const ParametersType &param)
 
void SetScales (const ScalesType &scales)
 
virtual const ScalesTypeGetScales () const
 
virtual const ScalesTypeGetInverseScales () 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
 
LightObject::Pointer CreateAnother () const override
 
virtual void DebugOff () const
 
virtual void DebugOn () const
 
CommandGetCommand (unsigned long tag)
 
bool GetDebug () const
 
MetaDataDictionaryGetMetaDataDictionary ()
 
const MetaDataDictionaryGetMetaDataDictionary () const
 
virtual ModifiedTimeType GetMTime () const
 
virtual const TimeStampGetTimeStamp () 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::SingleValuedNonLinearOptimizer
static Pointer New ()
 
- Static Public Member Functions inherited from itk::NonLinearOptimizer
static Pointer New ()
 
- Static Public Member Functions inherited from itk::Optimizer
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 ()
 

Protected Types

using CostFunctionAdaptorType = Superclass::CostFunctionAdaptorType
 
- Protected Types inherited from itk::SingleValuedNonLinearVnlOptimizer
using CostFunctionAdaptorType = SingleValuedVnlCostFunctionAdaptor
 

Protected Member Functions

 LBFGSOptimizer ()
 
void PrintSelf (std::ostream &os, Indent indent) const override
 
 ~LBFGSOptimizer () override
 
- Protected Member Functions inherited from itk::SingleValuedNonLinearVnlOptimizer
CostFunctionAdaptorTypeGetCostFunctionAdaptor ()
 
const CostFunctionAdaptorTypeGetCostFunctionAdaptor () const
 
CostFunctionAdaptorTypeGetNonConstCostFunctionAdaptor () const
 
void PrintSelf (std::ostream &os, Indent indent) const override
 
void SetCostFunctionAdaptor (CostFunctionAdaptorType *adaptor)
 
 SingleValuedNonLinearVnlOptimizer ()
 
 ~SingleValuedNonLinearVnlOptimizer () override
 
- Protected Member Functions inherited from itk::SingleValuedNonLinearOptimizer
 SingleValuedNonLinearOptimizer ()
 
 ~SingleValuedNonLinearOptimizer () override=default
 
- Protected Member Functions inherited from itk::NonLinearOptimizer
 NonLinearOptimizer ()=default
 
 ~NonLinearOptimizer () override
 
- Protected Member Functions inherited from itk::Optimizer
 Optimizer ()
 
virtual void SetCurrentPosition (const ParametersType &param)
 
 ~Optimizer () override=default
 
- Protected Member Functions inherited from itk::Object
 Object ()
 
bool PrintObservers (std::ostream &os, Indent indent) const
 
virtual void SetTimeStamp (const TimeStamp &timeStamp)
 
 ~Object () override
 
- 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 ()
 

Private Attributes

double m_DefaultStepLength {}
 
double m_GradientConvergenceTolerance {}
 
double m_LineSearchAccuracy {}
 
unsigned int m_MaximumNumberOfFunctionEvaluations {}
 
bool m_OptimizerInitialized {}
 
std::ostringstream m_StopConditionDescription {}
 
bool m_Trace {}
 
std::unique_ptr< InternalOptimizerTypem_VnlOptimizer
 

Additional Inherited Members

- Protected Attributes inherited from itk::SingleValuedNonLinearOptimizer
CostFunctionPointer m_CostFunction {}
 
- Protected Attributes inherited from itk::Optimizer
ParametersType m_CurrentPosition {}
 
bool m_ScalesInitialized { false }
 
- Protected Attributes inherited from itk::LightObject
std::atomic< int > m_ReferenceCount {}
 

Member Typedef Documentation

◆ ConstPointer

Definition at line 95 of file itkLBFGSOptimizer.h.

◆ CostFunctionAdaptorType

using itk::LBFGSOptimizer::CostFunctionAdaptorType = Superclass::CostFunctionAdaptorType
protected

Definition at line 180 of file itkLBFGSOptimizer.h.

◆ InternalOptimizerType

Internal optimizer type.

Definition at line 107 of file itkLBFGSOptimizer.h.

◆ InternalParametersType

using itk::LBFGSOptimizer::InternalParametersType = vnl_vector<double>

InternalParameters type alias.

Definition at line 104 of file itkLBFGSOptimizer.h.

◆ Pointer

Definition at line 94 of file itkLBFGSOptimizer.h.

◆ Self

Standard "Self" type alias.

Definition at line 92 of file itkLBFGSOptimizer.h.

◆ Superclass

Definition at line 93 of file itkLBFGSOptimizer.h.

Constructor & Destructor Documentation

◆ LBFGSOptimizer()

itk::LBFGSOptimizer::LBFGSOptimizer ( )
protected

◆ ~LBFGSOptimizer()

itk::LBFGSOptimizer::~LBFGSOptimizer ( )
overrideprotected

Member Function Documentation

◆ GetDefaultStepLength()

virtual double itk::LBFGSOptimizer::GetDefaultStepLength ( )
virtual

◆ GetGradientConvergenceTolerance()

virtual double itk::LBFGSOptimizer::GetGradientConvergenceTolerance ( )
virtual

◆ GetLineSearchAccuracy()

virtual double itk::LBFGSOptimizer::GetLineSearchAccuracy ( )
virtual

◆ GetMaximumNumberOfFunctionEvaluations()

virtual unsigned int itk::LBFGSOptimizer::GetMaximumNumberOfFunctionEvaluations ( )
virtual

◆ GetNameOfClass()

const char* itk::LBFGSOptimizer::GetNameOfClass ( ) const
overridevirtual
See also
LightObject::GetNameOfClass()

Reimplemented from itk::Object.

◆ GetOptimizer()

vnl_lbfgs* itk::LBFGSOptimizer::GetOptimizer ( )

Method for getting access to the internal optimizer.

◆ GetStopConditionDescription()

const std::string itk::LBFGSOptimizer::GetStopConditionDescription ( ) const
overridevirtual

Get the reason for termination

Reimplemented from itk::Optimizer.

◆ GetTrace()

virtual bool itk::LBFGSOptimizer::GetTrace ( )
virtual

◆ GetValue()

MeasureType itk::LBFGSOptimizer::GetValue ( ) const

Return Current Value

◆ New()

static Pointer itk::LBFGSOptimizer::New ( )
static

Method for creation through the object factory.

◆ PrintSelf()

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

Methods invoked by Print() to print information about the object including superclasses. Typically not called by the user (use Print() instead) but used in the hierarchical print process to combine the output of several classes.

Reimplemented from itk::Object.

◆ SetCostFunction()

void itk::LBFGSOptimizer::SetCostFunction ( SingleValuedCostFunction costFunction)
overridevirtual

Plug in a Cost Function into the optimizer

Implements itk::SingleValuedNonLinearVnlOptimizer.

◆ SetDefaultStepLength()

virtual void itk::LBFGSOptimizer::SetDefaultStepLength ( double  f)
virtual

Set/Get the default step size. This is a positive real number with a default value of 1.0 which determines the step size in the line search.

◆ SetGradientConvergenceTolerance()

virtual void itk::LBFGSOptimizer::SetGradientConvergenceTolerance ( double  f)
virtual

Set/Get the gradient convergence tolerance. This is a positive real number that determines the accuracy with which the solution is to be found. The optimization terminates when: ||G|| < gtol max(1,||X||) where ||.|| denotes the Euclidean norm.

◆ SetLineSearchAccuracy()

virtual void itk::LBFGSOptimizer::SetLineSearchAccuracy ( double  f)
virtual

Set/Get the line search accuracy. This is a positive real number with a default value of 0.9, which controls the accuracy of the line search. If the function and gradient evaluations are inexpensive with respect to the cost of the iterations it may be advantageous to set the value to a small value (say 0.1).

◆ SetMaximumNumberOfFunctionEvaluations()

virtual void itk::LBFGSOptimizer::SetMaximumNumberOfFunctionEvaluations ( unsigned int  n)
virtual

Set/Get the maximum number of function evaluations allowed.

◆ SetTrace()

virtual void itk::LBFGSOptimizer::SetTrace ( bool  flag)
virtual

Set/Get the optimizer trace flag. If set to true, the optimizer prints out information every iteration.

◆ StartOptimization()

void itk::LBFGSOptimizer::StartOptimization ( )
overridevirtual

Start optimization with an initial value.

Reimplemented from itk::Optimizer.

◆ TraceOn()

virtual void itk::LBFGSOptimizer::TraceOn ( )
virtual

Member Data Documentation

◆ m_DefaultStepLength

double itk::LBFGSOptimizer::m_DefaultStepLength {}
private

Definition at line 191 of file itkLBFGSOptimizer.h.

◆ m_GradientConvergenceTolerance

double itk::LBFGSOptimizer::m_GradientConvergenceTolerance {}
private

Definition at line 189 of file itkLBFGSOptimizer.h.

◆ m_LineSearchAccuracy

double itk::LBFGSOptimizer::m_LineSearchAccuracy {}
private

Definition at line 190 of file itkLBFGSOptimizer.h.

◆ m_MaximumNumberOfFunctionEvaluations

unsigned int itk::LBFGSOptimizer::m_MaximumNumberOfFunctionEvaluations {}
private

Definition at line 188 of file itkLBFGSOptimizer.h.

◆ m_OptimizerInitialized

bool itk::LBFGSOptimizer::m_OptimizerInitialized {}
private

Definition at line 183 of file itkLBFGSOptimizer.h.

◆ m_StopConditionDescription

std::ostringstream itk::LBFGSOptimizer::m_StopConditionDescription {}
mutableprivate

Definition at line 185 of file itkLBFGSOptimizer.h.

◆ m_Trace

bool itk::LBFGSOptimizer::m_Trace {}
private

Definition at line 187 of file itkLBFGSOptimizer.h.

◆ m_VnlOptimizer

std::unique_ptr<InternalOptimizerType> itk::LBFGSOptimizer::m_VnlOptimizer
private

Definition at line 184 of file itkLBFGSOptimizer.h.


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