itk::GradientDescentOptimizer2 Class Reference

#include <itkGradientDescentOptimizer2.h>

Detailed Description

Implement a gradient descent optimizer.

GradientDescentOptimizer2 implements a simple gradient descent optimizer. At each iteration the current position is updated according to

The learning rate is a fixed scalar defined via SetLearningRate(). The optimizer steps through a user defined number of iterations; no convergence checking is done.

Additionally, user can scale each component of the but setting a scaling vector using method SetScale().

The difference of this class with the itk::GradientDescentOptimizer is that it's based on the ScaledSingleValuedNonLinearOptimizer

See also

ScaledSingleValuedNonLinearOptimizer

Definition at line 54 of file itkGradientDescentOptimizer2.h.

Inheritance diagram for itk::GradientDescentOptimizer2:

Public Types
using	ConstPointer = SmartPointer<const Self>
using	Pointer = SmartPointer<Self>
using	ScaledCostFunctionPointer
using	ScaledCostFunctionType
using	ScalesType
using	Self = GradientDescentOptimizer2
enum	StopConditionType { MaximumNumberOfIterations , MetricError , MinimumStepSize }
using	Superclass = ScaledSingleValuedNonLinearOptimizer
Public Types inherited from itk::ScaledSingleValuedNonLinearOptimizer
using	ConstPointer = SmartPointer<const Self>
using	Pointer = SmartPointer<Self>
using	ScaledCostFunctionPointer = ScaledCostFunctionType::Pointer
using	ScaledCostFunctionType = ScaledSingleValuedCostFunction
using	ScalesType = NonLinearOptimizer::ScalesType
using	Self = ScaledSingleValuedNonLinearOptimizer
using	Superclass = SingleValuedNonLinearOptimizer

Public Member Functions
virtual void	AdvanceOneStep ()
virtual unsigned int	GetCurrentIteration () const
virtual const DerivativeType &	GetGradient ()
virtual const double &	GetLearningRate ()
virtual const unsigned long &	GetNumberOfIterations ()
virtual const DerivativeType &	GetSearchDirection ()
virtual const StopConditionType &	GetStopCondition ()
virtual const double &	GetValue ()
	ITK_DISALLOW_COPY_AND_MOVE (GradientDescentOptimizer2)
	itkOverrideGetNameOfClassMacro (GradientDescentOptimizer2)
virtual void	MetricErrorResponse (ExceptionObject &err)
virtual void	ResumeOptimization ()
virtual void	SetLearningRate (double _arg)
virtual void	SetNumberOfIterations (unsigned long _arg)
void	StartOptimization () override
virtual void	StopOptimization ()
Public Member Functions inherited from itk::ScaledSingleValuedNonLinearOptimizer
const ParametersType &	GetCurrentPosition () const override
virtual bool	GetMaximize () const
virtual const ScaledCostFunctionType *	GetScaledCostFunction ()
virtual const ParametersType &	GetScaledCurrentPosition ()
bool	GetUseScales () const
virtual void	InitializeScales ()
	ITK_DISALLOW_COPY_AND_MOVE (ScaledSingleValuedNonLinearOptimizer)
	itkOverrideGetNameOfClassMacro (ScaledSingleValuedNonLinearOptimizer)
virtual void	MaximizeOff ()
virtual void	MaximizeOn ()
void	SetCostFunction (CostFunctionType *costFunction) override
virtual void	SetMaximize (bool _arg)
virtual void	SetUseScales (bool arg)

Static Public Member Functions
static Pointer	New ()
Static Public Member Functions inherited from itk::ScaledSingleValuedNonLinearOptimizer
static Pointer	New ()

Protected Member Functions
	GradientDescentOptimizer2 ()
void	PrintSelf (std::ostream &os, Indent indent) const override
	~GradientDescentOptimizer2 () override=default
Protected Member Functions inherited from itk::ScaledSingleValuedNonLinearOptimizer
virtual void	GetScaledDerivative (const ParametersType &parameters, DerivativeType &derivative) const
virtual MeasureType	GetScaledValue (const ParametersType &parameters) const
virtual void	GetScaledValueAndDerivative (const ParametersType &parameters, MeasureType &value, DerivativeType &derivative) const
void	PrintSelf (std::ostream &os, Indent indent) const override
	ScaledSingleValuedNonLinearOptimizer ()
void	SetCurrentPosition (const ParametersType &param) override
virtual void	SetScaledCurrentPosition (const ParametersType &parameters)
	~ScaledSingleValuedNonLinearOptimizer () override=default

Protected Attributes
DerivativeType	m_Gradient {}
DerivativeType	m_SearchDirection {}
StopConditionType	m_StopCondition { MaximumNumberOfIterations }
Protected Attributes inherited from itk::ScaledSingleValuedNonLinearOptimizer
ScaledCostFunctionPointer	m_ScaledCostFunction {}
ParametersType	m_ScaledCurrentPosition {}

Private Attributes
unsigned long	m_CurrentIteration { 0 }
double	m_LearningRate { 1.0 }
unsigned long	m_NumberOfIterations { 100 }
bool	m_Stop { false }
double	m_Value { 0.0 }

Member Typedef Documentation

ConstPointer

using itk::GradientDescentOptimizer2::ConstPointer = SmartPointer<const Self>

Definition at line 63 of file itkGradientDescentOptimizer2.h.

Pointer

using itk::GradientDescentOptimizer2::Pointer = SmartPointer<Self>

Definition at line 62 of file itkGradientDescentOptimizer2.h.

ScaledCostFunctionPointer

using itk::ScaledSingleValuedNonLinearOptimizer::ScaledCostFunctionPointer

Definition at line 87 of file itkScaledSingleValuedNonLinearOptimizer.h.

ScaledCostFunctionType

using itk::ScaledSingleValuedNonLinearOptimizer::ScaledCostFunctionType

Definition at line 86 of file itkScaledSingleValuedNonLinearOptimizer.h.

ScalesType

using itk::ScaledSingleValuedNonLinearOptimizer::ScalesType

Definition at line 85 of file itkScaledSingleValuedNonLinearOptimizer.h.

Self

using itk::GradientDescentOptimizer2::Self = GradientDescentOptimizer2

Standard class typedefs.

Definition at line 60 of file itkGradientDescentOptimizer2.h.

Superclass

using itk::GradientDescentOptimizer2::Superclass = ScaledSingleValuedNonLinearOptimizer

Definition at line 61 of file itkGradientDescentOptimizer2.h.

Member Enumeration Documentation

StopConditionType

enum itk::GradientDescentOptimizer2::StopConditionType

Codes of stopping conditions The MinimumStepSize stopcondition never occurs, but may be implemented in inheriting classes

Enumerator
MaximumNumberOfIterations
MetricError
MinimumStepSize

Definition at line 83 of file itkGradientDescentOptimizer2.h.

Constructor & Destructor Documentation

GradientDescentOptimizer2()

itk::GradientDescentOptimizer2::GradientDescentOptimizer2 ( )

protected

~GradientDescentOptimizer2()

itk::GradientDescentOptimizer2::~GradientDescentOptimizer2 ( )

overrideprotecteddefault

Member Function Documentation

AdvanceOneStep()

virtual void itk::GradientDescentOptimizer2::AdvanceOneStep ( )

virtual

Advance one step following the gradient direction.

Reimplemented in elastix::AdaGrad< TElastix >, elastix::PreconditionedStochasticGradientDescent< TElastix >, and itk::StandardGradientDescentOptimizer.

GetCurrentIteration()

virtual unsigned int itk::GradientDescentOptimizer2::GetCurrentIteration ( ) const

virtual

Get the current iteration number.

GetGradient()

virtual const DerivativeType & itk::GradientDescentOptimizer2::GetGradient ( )

virtual

Get current gradient.

GetLearningRate()

virtual const double & itk::GradientDescentOptimizer2::GetLearningRate ( )

virtual

Get the learning rate.

GetNumberOfIterations()

virtual const unsigned long & itk::GradientDescentOptimizer2::GetNumberOfIterations ( )

virtual

Get the number of iterations.

GetSearchDirection()

virtual const DerivativeType & itk::GradientDescentOptimizer2::GetSearchDirection ( )

virtual

Get current search direction

GetStopCondition()

virtual const StopConditionType & itk::GradientDescentOptimizer2::GetStopCondition ( )

virtual

Get Stop condition.

GetValue()

virtual const double & itk::GradientDescentOptimizer2::GetValue ( )

virtual

Get the current value.

ITK_DISALLOW_COPY_AND_MOVE()

itk::GradientDescentOptimizer2::ITK_DISALLOW_COPY_AND_MOVE

(

GradientDescentOptimizer2

)

itkOverrideGetNameOfClassMacro()

itk::GradientDescentOptimizer2::itkOverrideGetNameOfClassMacro

(

GradientDescentOptimizer2

)

Run-time type information (and related methods).

MetricErrorResponse()

virtual void itk::GradientDescentOptimizer2::MetricErrorResponse ( ExceptionObject & err )

virtual

Stop optimization and pass on exception.

New()

Pointer itk::GradientDescentOptimizer2::New ( )

static

Method for creation through the object factory.

PrintSelf()

void itk::GradientDescentOptimizer2::PrintSelf ( std::ostream & os, Indent indent ) const

overrideprotected

ResumeOptimization()

virtual void itk::GradientDescentOptimizer2::ResumeOptimization ( )

virtual

Resume previously stopped optimization with current parameters

See also

StopOptimization.

Reimplemented in elastix::AdaGrad< TElastix >, elastix::AdaptiveStochasticGradientDescent< TElastix >, and elastix::PreconditionedStochasticGradientDescent< TElastix >.

SetLearningRate()

virtual void itk::GradientDescentOptimizer2::SetLearningRate ( double _arg )

virtual

Set the learning rate.

SetNumberOfIterations()

virtual void itk::GradientDescentOptimizer2::SetNumberOfIterations ( unsigned long _arg )

virtual

Set the number of iterations.

StartOptimization()

void itk::GradientDescentOptimizer2::StartOptimization ( )

override

Start optimization.

StopOptimization()

virtual void itk::GradientDescentOptimizer2::StopOptimization ( )

virtual

Stop optimization.

See also

ResumeOptimization

Member Data Documentation

m_CurrentIteration

unsigned long itk::GradientDescentOptimizer2::m_CurrentIteration { 0 }

private

Definition at line 155 of file itkGradientDescentOptimizer2.h.

m_Gradient

DerivativeType itk::GradientDescentOptimizer2::m_Gradient {}

protected

Definition at line 146 of file itkGradientDescentOptimizer2.h.

m_LearningRate

double itk::GradientDescentOptimizer2::m_LearningRate { 1.0 }

private

Definition at line 152 of file itkGradientDescentOptimizer2.h.

m_NumberOfIterations

unsigned long itk::GradientDescentOptimizer2::m_NumberOfIterations { 100 }

private

Definition at line 154 of file itkGradientDescentOptimizer2.h.

m_SearchDirection

DerivativeType itk::GradientDescentOptimizer2::m_SearchDirection {}

protected

Definition at line 147 of file itkGradientDescentOptimizer2.h.

m_Stop

bool itk::GradientDescentOptimizer2::m_Stop { false }

private

Definition at line 153 of file itkGradientDescentOptimizer2.h.

m_StopCondition

StopConditionType itk::GradientDescentOptimizer2::m_StopCondition { MaximumNumberOfIterations }

protected

Definition at line 148 of file itkGradientDescentOptimizer2.h.

m_Value

double itk::GradientDescentOptimizer2::m_Value { 0.0 }

private

Definition at line 151 of file itkGradientDescentOptimizer2.h.