go home Home | Main Page | Modules | Namespace List | Class Hierarchy | Alphabetical List | Data Structures | File List | Namespace Members | Data Fields | Globals | Related Pages
Public Types | Public Member Functions | Static Public Member Functions | Protected Member Functions | Protected Attributes | Private Member Functions | Private Attributes
elastix::ConjugateGradientFRPR< TElastix > Class Template Reference

#include <elxConjugateGradientFRPR.h>

Inheritance diagram for elastix::ConjugateGradientFRPR< TElastix >:
Inheritance graph
[legend]
Collaboration diagram for elastix::ConjugateGradientFRPR< TElastix >:
Collaboration graph
[legend]

Public Types

typedef
Superclass2::ConfigurationPointer 
ConfigurationPointer
typedef
Superclass2::ConfigurationType 
ConfigurationType
typedef SmartPointer< const SelfConstPointer
typedef
Superclass1::CostFunctionPointer 
CostFunctionPointer
typedef
Superclass1::CostFunctionType 
CostFunctionType
typedef
SingleValuedNonLinearOptimizer::DerivativeType 
DerivativeType
typedef Superclass2::ElastixPointer ElastixPointer
typedef Superclass2::ElastixType ElastixType
typedef Superclass2::ITKBaseType ITKBaseType
typedef Superclass1::ParametersType ParametersType
typedef SmartPointer< SelfPointer
typedef
Superclass2::RegistrationPointer 
RegistrationPointer
typedef
Superclass2::RegistrationType 
RegistrationType
typedef ConjugateGradientFRPR Self
typedef FRPROptimizer Superclass1
typedef OptimizerBase< TElastix > Superclass2

Public Member Functions

virtual void AfterEachIteration (void)
virtual void AfterEachResolution (void)
virtual void AfterRegistration (void)
virtual void BeforeEachResolution (void)
virtual void BeforeRegistration (void)
 elxClassNameMacro ("ConjugateGradientFRPR")
virtual const char * GetClassName () const
virtual const doubleGetCurrentDerivativeMagnitude ()
virtual const doubleGetCurrentSearchDirectionMagnitude ()
virtual const doubleGetCurrentStepLength ()
virtual bool GetLineBracketing () const
virtual bool GetLineOptimizing () const
virtual void SetInitialPosition (const ParametersType &param)

Static Public Member Functions

static Pointer New ()

Protected Member Functions

virtual void BracketedLineOptimize (double ax, double bx, double cx, double fa, double fb, double fc, double *extX, double *extVal)
 ConjugateGradientFRPR ()
virtual void GetValueAndDerivative (ParametersType p, double *val, ParametersType *xi)
virtual void LineBracket (double *ax, double *bx, double *cx, double *fa, double *fb, double *fc)
virtual void LineOptimize (ParametersType *p, ParametersType xi, double *val)
virtual void SetLineBracketing (bool _arg)
virtual void SetLineOptimizing (bool _arg)
virtual ~ConjugateGradientFRPR ()

Protected Attributes

double m_CurrentDerivativeMagnitude
double m_CurrentSearchDirectionMagnitude
double m_CurrentStepLength

Private Member Functions

 ConjugateGradientFRPR (const Self &)
const char * DeterminePhase (void) const
void operator= (const Self &)

Private Attributes

bool m_LineBracketing
bool m_LineOptimizing

Detailed Description

template<class TElastix>
class elastix::ConjugateGradientFRPR< TElastix >

The ConjugateGradientFRPR class.

This component is based on the itkFRPROptimizer. This is a Fletcher-Reeves conjugate gradient optimizer, in combination with an exact (dBrent) line search, based on the description in Numerical Recipes in C++

This optimizer support the NewSamplesEveryIteration option. It requests new samples upon every derivative evaluation, but actually this makes no sense for a conjugate gradient optimizer. So, think twice before using it.

Note:
It prints out no stop conditions, since the itk superclass does not generate them.
It considers line search iterations as elastix iterations.
Parameters:
Optimizer: Select this optimizer as follows:
(Optimizer "ConjugateGradientFRPR")
Parameters:
MaximumNumberOfIterations: The maximum number of iterations in each resolution.
example: (MaximumNumberOfIterations 100 100 50)
Default value: 100.
Parameters:
MaximumNumberOfLineSearchIterations: The maximum number of iterations in each resolution.
example: (MaximumNumberOfIterations 10 10 5)
Default value: 10.
Parameters:
StepLength: Set the length of the initial step tried by the line seach, used to bracket the minimum.
example: (StepLength 2.0 1.0 0.5)
Default value: 1.0.
Parameters:
ValueTolerance: Convergence is declared if:

\[ 2.0 * | f_2 - f_1 | \le ValueTolerance * ( | f_1 | + | f_2 | ) \]

example: (ValueTolerance 0.001 0.00001 0.000001)
Default value: 0.00001.
Parameters:
LineSearchStepTolerance: Convergence of the line search is declared if:

\[ | x - x_m | \le tol * |x| - ( b - a ) / 2, \]

where:
$x$ = current mininum of the gain
$a, b$ = current brackets around the minimum
$x_m = (a+b)/2 $
example: (LineSearchStepTolerance 0.001 0.00001 0.000001)
Default value: 0.00001.

Definition at line 73 of file elxConjugateGradientFRPR.h.


Member Typedef Documentation

Reimplemented from elastix::OptimizerBase< TElastix >.

Definition at line 109 of file elxConjugateGradientFRPR.h.

Reimplemented from elastix::OptimizerBase< TElastix >.

Definition at line 108 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef SmartPointer<const Self> elastix::ConjugateGradientFRPR< TElastix >::ConstPointer

Definition at line 86 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef Superclass1::CostFunctionPointer elastix::ConjugateGradientFRPR< TElastix >::CostFunctionPointer

Definition at line 99 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef Superclass1::CostFunctionType elastix::ConjugateGradientFRPR< TElastix >::CostFunctionType

Typedef's inherited from Superclass1.

Definition at line 98 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef SingleValuedNonLinearOptimizer::DerivativeType elastix::ConjugateGradientFRPR< TElastix >::DerivativeType

Definition at line 103 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef Superclass2::ElastixPointer elastix::ConjugateGradientFRPR< TElastix >::ElastixPointer

Reimplemented from elastix::OptimizerBase< TElastix >.

Definition at line 107 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef Superclass2::ElastixType elastix::ConjugateGradientFRPR< TElastix >::ElastixType

Typedef's inherited from Elastix.

Reimplemented from elastix::OptimizerBase< TElastix >.

Definition at line 106 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef Superclass2::ITKBaseType elastix::ConjugateGradientFRPR< TElastix >::ITKBaseType

Reimplemented from elastix::OptimizerBase< TElastix >.

Definition at line 112 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef Superclass1::ParametersType elastix::ConjugateGradientFRPR< TElastix >::ParametersType

Reimplemented from elastix::OptimizerBase< TElastix >.

Definition at line 101 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef SmartPointer<Self> elastix::ConjugateGradientFRPR< TElastix >::Pointer

Definition at line 85 of file elxConjugateGradientFRPR.h.

Reimplemented from elastix::OptimizerBase< TElastix >.

Definition at line 111 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef Superclass2::RegistrationType elastix::ConjugateGradientFRPR< TElastix >::RegistrationType

Reimplemented from elastix::OptimizerBase< TElastix >.

Definition at line 110 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef ConjugateGradientFRPR elastix::ConjugateGradientFRPR< TElastix >::Self

Standard ITK.

Reimplemented from elastix::OptimizerBase< TElastix >.

Definition at line 82 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef FRPROptimizer elastix::ConjugateGradientFRPR< TElastix >::Superclass1

Definition at line 83 of file elxConjugateGradientFRPR.h.

template<class TElastix >
typedef OptimizerBase<TElastix> elastix::ConjugateGradientFRPR< TElastix >::Superclass2

Definition at line 84 of file elxConjugateGradientFRPR.h.


Constructor & Destructor Documentation

template<class TElastix >
elastix::ConjugateGradientFRPR< TElastix >::ConjugateGradientFRPR ( ) [protected]
template<class TElastix >
virtual elastix::ConjugateGradientFRPR< TElastix >::~ConjugateGradientFRPR ( ) [inline, protected, virtual]

Definition at line 142 of file elxConjugateGradientFRPR.h.

template<class TElastix >
elastix::ConjugateGradientFRPR< TElastix >::ConjugateGradientFRPR ( const Self ) [private]

Member Function Documentation

template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::AfterEachIteration ( void  ) [virtual]

Reimplemented from elastix::BaseComponent.

template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::AfterEachResolution ( void  ) [virtual]

Reimplemented from elastix::BaseComponent.

template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::AfterRegistration ( void  ) [virtual]

Reimplemented from elastix::BaseComponent.

template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::BeforeEachResolution ( void  ) [virtual]

Reimplemented from elastix::BaseComponent.

template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::BeforeRegistration ( void  ) [virtual]

Methods to set parameters and print output at different stages in the registration process.

Reimplemented from elastix::BaseComponent.

template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::BracketedLineOptimize ( double  ax,
double  bx,
double  cx,
double  fa,
double  fb,
double  fc,
double extX,
double extVal 
) [protected, virtual]

Given a bracketing triple of points and their function values, returns a bounded extreme. These values are in parameter space, along the current line and wrt the current origin set via SetLine. Optimization terminates based on MaximumIteration, StepTolerance, or ValueTolerance. Implemented as Brent line optimers from NRC.

This implementation sets the LineOptimizing flag to 'true', calls the the superclass's implementation, stores extX as the current step length, and sets the LineOptimizing flag to 'false' again.

template<class TElastix >
const char* elastix::ConjugateGradientFRPR< TElastix >::DeterminePhase ( void  ) const [private]
template<class TElastix >
elastix::ConjugateGradientFRPR< TElastix >::elxClassNameMacro ( "ConjugateGradientFRPR< TElastix >"  )

Name of this class.

template<class TElastix >
virtual const char* elastix::ConjugateGradientFRPR< TElastix >::GetClassName ( ) const [virtual]

Run-time type information (and related methods).

Reimplemented from elastix::OptimizerBase< TElastix >.

template<class TElastix >
virtual const double& elastix::ConjugateGradientFRPR< TElastix >::GetCurrentDerivativeMagnitude ( ) [virtual]

Return the magnitude of the cached derivative

template<class TElastix >
virtual const double& elastix::ConjugateGradientFRPR< TElastix >::GetCurrentSearchDirectionMagnitude ( ) [virtual]

Get the magnitude of the line search direction

template<class TElastix >
virtual const double& elastix::ConjugateGradientFRPR< TElastix >::GetCurrentStepLength ( ) [virtual]

Get the current gain

template<class TElastix >
virtual bool elastix::ConjugateGradientFRPR< TElastix >::GetLineBracketing ( ) const [virtual]
template<class TElastix >
virtual bool elastix::ConjugateGradientFRPR< TElastix >::GetLineOptimizing ( ) const [virtual]

Check if the optimizer is currently Bracketing the minimum, or is optimizing along a line

template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::GetValueAndDerivative ( ParametersType  p,
double val,
ParametersType xi 
) [protected, virtual]

Get the value of the n-dimensional cost function at this scalar step distance along the current line direction from the current line origin. Line origin and distances are set via SetLine.

This implementation calls the Superclass' implementation and caches the computed derivative's magnitude. Besides, it invokes the SelectNewSamples method.

template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::LineBracket ( double ax,
double bx,
double cx,
double fa,
double fb,
double fc 
) [protected, virtual]

The LineBracket routine from NRC. Uses current origin and line direction (from SetLine) to find a triple of points (ax, bx, cx) that bracket the extreme "near" the origin. Search first considers the point StepLength distance from ax. IMPORTANT: The value of ax and the value of the function at ax (i.e., fa), must both be provided to this function.

This implementation sets the LineBracketing flag to 'true', calls the superclass' implementation, stores bx as the current step length, invokes an iteration event, and sets the LineBracketing flag to 'false'

template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::LineOptimize ( ParametersType p,
ParametersType  xi,
double val 
) [protected, virtual]

store the line search direction's (xi) magnitude and call the superclass' implementation.

template<class TElastix >
static Pointer elastix::ConjugateGradientFRPR< TElastix >::New ( ) [static]

Method for creation through the object factory.

template<class TElastix >
void elastix::ConjugateGradientFRPR< TElastix >::operator= ( const Self ) [private]
template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::SetInitialPosition ( const ParametersType param) [virtual]

Override the SetInitialPosition.

template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::SetLineBracketing ( bool  _arg) [protected, virtual]
template<class TElastix >
virtual void elastix::ConjugateGradientFRPR< TElastix >::SetLineOptimizing ( bool  _arg) [protected, virtual]

Set if the optimizer is currently bracketing the minimum, or is optimizing along a line


Field Documentation

template<class TElastix >
double elastix::ConjugateGradientFRPR< TElastix >::m_CurrentDerivativeMagnitude [protected]

To store the latest computed derivative's magnitude

Definition at line 142 of file elxConjugateGradientFRPR.h.

template<class TElastix >
double elastix::ConjugateGradientFRPR< TElastix >::m_CurrentSearchDirectionMagnitude [protected]

Variable to store the line search direction magnitude

Definition at line 148 of file elxConjugateGradientFRPR.h.

template<class TElastix >
double elastix::ConjugateGradientFRPR< TElastix >::m_CurrentStepLength [protected]

the current gain

Definition at line 151 of file elxConjugateGradientFRPR.h.

template<class TElastix >
bool elastix::ConjugateGradientFRPR< TElastix >::m_LineBracketing [private]

Definition at line 208 of file elxConjugateGradientFRPR.h.

template<class TElastix >
bool elastix::ConjugateGradientFRPR< TElastix >::m_LineOptimizing [private]

Definition at line 207 of file elxConjugateGradientFRPR.h.



Generated on 24-05-2012 for elastix by doxygen 1.7.6.1 elastix logo