#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

$p_{n+1} = p_n + \mbox{learningRate} \, \frac{\partial f(p_n) }{\partial p_n}$

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 $\partial f / \partial p$ 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:

[legend]

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 const char *	GetClassName () const

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)

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)

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.

◆ GetClassName()

virtual const char * itk::GradientDescentOptimizer2::GetClassName ( ) const

virtual

Run-time type information (and related methods).

Reimplemented from itk::ScaledSingleValuedNonLinearOptimizer.

Reimplemented in elastix::AdaGrad< TElastix >, elastix::AdaptiveStochasticGradientDescent< TElastix >, elastix::PreconditionedStochasticGradientDescent< TElastix >, elastix::StandardGradientDescent< TElastix >, itk::AdaptiveStepsizeOptimizer, itk::AdaptiveStochasticGradientDescentOptimizer, itk::PreconditionedASGDOptimizer, 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 )

◆ MetricErrorResponse()

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

virtual

Stop optimization and pass on exception.

◆ New()

static 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

Field 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.

Detailed Description

Public Types

Public Member Functions

Static Public Member Functions

Protected Member Functions

Protected Attributes

Private Attributes

Member Typedef Documentation

◆ ConstPointer

◆ Pointer

◆ ScaledCostFunctionPointer

◆ ScaledCostFunctionType

◆ ScalesType

◆ Self

◆ Superclass

Member Enumeration Documentation

◆ StopConditionType

Constructor & Destructor Documentation

◆ GradientDescentOptimizer2()

◆ ~GradientDescentOptimizer2()

Member Function Documentation

◆ AdvanceOneStep()

◆ GetClassName()

◆ GetCurrentIteration()

◆ GetGradient()

◆ GetLearningRate()

◆ GetNumberOfIterations()

◆ GetSearchDirection()

◆ GetStopCondition()

◆ GetValue()

◆ ITK_DISALLOW_COPY_AND_MOVE()

◆ MetricErrorResponse()

◆ New()

◆ PrintSelf()

◆ ResumeOptimization()

◆ SetLearningRate()

◆ SetNumberOfIterations()

◆ StartOptimization()

◆ StopOptimization()

Field Documentation

◆ m_CurrentIteration

◆ m_Gradient

◆ m_LearningRate

◆ m_NumberOfIterations

◆ m_SearchDirection

◆ m_Stop

◆ m_StopCondition

◆ m_Value