ITK  4.4.0
Insight Segmentation and Registration Toolkit
Public Types | Public Member Functions | Protected Types | Protected Member Functions | Private Member Functions | Private Attributes | List of all members
itk::SingleValuedNonLinearVnlOptimizer Class Referenceabstract

#include <itkSingleValuedNonLinearVnlOptimizer.h>

+ Inheritance diagram for itk::SingleValuedNonLinearVnlOptimizer:
+ Collaboration diagram for itk::SingleValuedNonLinearVnlOptimizer:

Detailed Description

This class is a base for the Optimization methods that optimize a single valued function.

It is an Adaptor class for optimizers provided by the vnl library

Definition at line 36 of file itkSingleValuedNonLinearVnlOptimizer.h.

Public Types

typedef ReceptorMemberCommand
< Self
CommandType
 
typedef SmartPointer< const SelfConstPointer
 
typedef SmartPointer< SelfPointer
 
typedef
SingleValuedNonLinearVnlOptimizer 
Self
 
typedef
SingleValuedNonLinearOptimizer 
Superclass
 
- Public Types inherited from itk::SingleValuedNonLinearOptimizer
typedef SmartPointer< const SelfConstPointer
 
typedef CostFunctionType::Pointer CostFunctionPointer
 
typedef SingleValuedCostFunction CostFunctionType
 
typedef
CostFunctionType::DerivativeType 
DerivativeType
 
typedef
CostFunctionType::MeasureType 
MeasureType
 
typedef Superclass::ParametersType ParametersType
 
typedef SmartPointer< SelfPointer
 
typedef
SingleValuedNonLinearOptimizer 
Self
 
typedef NonLinearOptimizer Superclass
 
- Public Types inherited from itk::NonLinearOptimizer
typedef SmartPointer< const SelfConstPointer
 
typedef Superclass::ParametersType ParametersType
 
typedef SmartPointer< SelfPointer
 
typedef Superclass::ScalesType ScalesType
 
typedef NonLinearOptimizer Self
 
typedef Optimizer Superclass
 
- Public Types inherited from itk::Optimizer
typedef SmartPointer< const SelfConstPointer
 
typedef OptimizerParameters
< double > 
ParametersType
 
typedef SmartPointer< SelfPointer
 
typedef Array< double > ScalesType
 
typedef Optimizer Self
 
typedef Object Superclass
 
- Public Types inherited from itk::Object
typedef SmartPointer< const SelfConstPointer
 
typedef SmartPointer< SelfPointer
 
typedef Object Self
 
typedef LightObject Superclass
 
- Public Types inherited from itk::LightObject
typedef SmartPointer< const SelfConstPointer
 
typedef SmartPointer< SelfPointer
 
typedef LightObject Self
 

Public Member Functions

virtual const char * GetNameOfClass () const
 
virtual void SetCostFunction (SingleValuedCostFunction *costFunction)=0
 
virtual const bool & GetMaximize ()
 
virtual void SetMaximize (bool _arg)
 
virtual void MaximizeOn ()
 
virtual void MaximizeOff ()
 
bool GetMinimize () const
 
void SetMinimize (bool v)
 
void MinimizeOn ()
 
void MinimizeOff ()
 
virtual const MeasureTypeGetCachedValue ()
 
virtual const DerivativeTypeGetCachedDerivative ()
 
virtual const ParametersTypeGetCachedCurrentPosition ()
 
- Public Member Functions inherited from itk::SingleValuedNonLinearOptimizer
virtual ::itk::LightObject::Pointer CreateAnother (void) const
 
virtual const CostFunctionTypeGetCostFunction () const
 
virtual CostFunctionTypeGetModifiableCostFunction ()
 
MeasureType GetValue (const ParametersType &parameters) const
 
virtual void SetCostFunction (CostFunctionType *costFunction)
 

Protected Types

typedef
SingleValuedVnlCostFunctionAdaptor 
CostFunctionAdaptorType
 

Protected Member Functions

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

void IterationReport (const EventObject &event)
 
void operator= (const Self &)
 
 SingleValuedNonLinearVnlOptimizer (const Self &)
 

Private Attributes

ParametersType m_CachedCurrentPosition
 
DerivativeType m_CachedDerivative
 
MeasureType m_CachedValue
 
CommandType::Pointer m_Command
 
CostFunctionAdaptorTypem_CostFunctionAdaptor
 
bool m_Maximize
 

Additional Inherited Members

- Static Public Member Functions inherited from itk::SingleValuedNonLinearOptimizer
static Pointer New ()
 
- Protected Attributes inherited from itk::SingleValuedNonLinearOptimizer
CostFunctionPointer m_CostFunction
 

Member Typedef Documentation

Command observer that will interact with the ITKVNL cost-function adaptor in order to generate iteration events. This will allow to overcome the limitation of VNL optimizers not offering callbacks for every iteration

Definition at line 48 of file itkSingleValuedNonLinearVnlOptimizer.h.

Definition at line 44 of file itkSingleValuedNonLinearVnlOptimizer.h.

Definition at line 96 of file itkSingleValuedNonLinearVnlOptimizer.h.

Definition at line 43 of file itkSingleValuedNonLinearVnlOptimizer.h.

Standard class typedefs.

Definition at line 41 of file itkSingleValuedNonLinearVnlOptimizer.h.

Definition at line 42 of file itkSingleValuedNonLinearVnlOptimizer.h.

Constructor & Destructor Documentation

itk::SingleValuedNonLinearVnlOptimizer::SingleValuedNonLinearVnlOptimizer ( )
protected
virtual itk::SingleValuedNonLinearVnlOptimizer::~SingleValuedNonLinearVnlOptimizer ( )
protectedvirtual
itk::SingleValuedNonLinearVnlOptimizer::SingleValuedNonLinearVnlOptimizer ( const Self )
private

Member Function Documentation

virtual const ParametersType& itk::SingleValuedNonLinearVnlOptimizer::GetCachedCurrentPosition ( )
virtual

Return Cached Values. These method have the advantage of not triggering a recomputation of the metric value, but it has the disadvantage of returning a value that may not be the one corresponding to the current parameters. For GUI update purposes, this method is a good option, for mathematical validation you should rather call GetValue().

virtual const DerivativeType& itk::SingleValuedNonLinearVnlOptimizer::GetCachedDerivative ( )
virtual

Return Cached Values. These method have the advantage of not triggering a recomputation of the metric value, but it has the disadvantage of returning a value that may not be the one corresponding to the current parameters. For GUI update purposes, this method is a good option, for mathematical validation you should rather call GetValue().

virtual const MeasureType& itk::SingleValuedNonLinearVnlOptimizer::GetCachedValue ( )
virtual

Return Cached Values. These method have the advantage of not triggering a recomputation of the metric value, but it has the disadvantage of returning a value that may not be the one corresponding to the current parameters. For GUI update purposes, this method is a good option, for mathematical validation you should rather call GetValue().

const CostFunctionAdaptorType* itk::SingleValuedNonLinearVnlOptimizer::GetCostFunctionAdaptor ( void  ) const
protected
CostFunctionAdaptorType* itk::SingleValuedNonLinearVnlOptimizer::GetCostFunctionAdaptor ( void  )
protected
virtual const bool& itk::SingleValuedNonLinearVnlOptimizer::GetMaximize ( )
virtual

Methods to define whether the cost function will be maximized or minimized. By default the VNL amoeba optimizer is only a minimizer. Maximization is implemented here by notifying the CostFunctionAdaptor which in its turn will multiply the function values and its derivative by -1.0.

bool itk::SingleValuedNonLinearVnlOptimizer::GetMinimize ( ) const
inline

Methods to define whether the cost function will be maximized or minimized. By default the VNL amoeba optimizer is only a minimizer. Maximization is implemented here by notifying the CostFunctionAdaptor which in its turn will multiply the function values and its derivative by -1.0.

Definition at line 72 of file itkSingleValuedNonLinearVnlOptimizer.h.

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

Run-time type information (and related methods).

Reimplemented from itk::SingleValuedNonLinearOptimizer.

Reimplemented in itk::AmoebaOptimizer, itk::LBFGSBOptimizer, itk::ConjugateGradientOptimizer, and itk::LBFGSOptimizer.

CostFunctionAdaptorType* itk::SingleValuedNonLinearVnlOptimizer::GetNonConstCostFunctionAdaptor ( void  ) const
protected

The purpose of this method is to get around the lack of const-correctness in VNL cost-functions and optimizers

void itk::SingleValuedNonLinearVnlOptimizer::IterationReport ( const EventObject event)
private

Callback function for the Command Observer

virtual void itk::SingleValuedNonLinearVnlOptimizer::MaximizeOff ( )
virtual

Methods to define whether the cost function will be maximized or minimized. By default the VNL amoeba optimizer is only a minimizer. Maximization is implemented here by notifying the CostFunctionAdaptor which in its turn will multiply the function values and its derivative by -1.0.

virtual void itk::SingleValuedNonLinearVnlOptimizer::MaximizeOn ( )
virtual

Methods to define whether the cost function will be maximized or minimized. By default the VNL amoeba optimizer is only a minimizer. Maximization is implemented here by notifying the CostFunctionAdaptor which in its turn will multiply the function values and its derivative by -1.0.

void itk::SingleValuedNonLinearVnlOptimizer::MinimizeOff ( )
inline

Methods to define whether the cost function will be maximized or minimized. By default the VNL amoeba optimizer is only a minimizer. Maximization is implemented here by notifying the CostFunctionAdaptor which in its turn will multiply the function values and its derivative by -1.0.

Definition at line 78 of file itkSingleValuedNonLinearVnlOptimizer.h.

void itk::SingleValuedNonLinearVnlOptimizer::MinimizeOn ( )
inline

Methods to define whether the cost function will be maximized or minimized. By default the VNL amoeba optimizer is only a minimizer. Maximization is implemented here by notifying the CostFunctionAdaptor which in its turn will multiply the function values and its derivative by -1.0.

Definition at line 76 of file itkSingleValuedNonLinearVnlOptimizer.h.

void itk::SingleValuedNonLinearVnlOptimizer::operator= ( const Self )
private
void itk::SingleValuedNonLinearVnlOptimizer::PrintSelf ( std::ostream &  os,
Indent  indent 
) const
protectedvirtual

Print out internal state

Reimplemented from itk::SingleValuedNonLinearOptimizer.

virtual void itk::SingleValuedNonLinearVnlOptimizer::SetCostFunction ( SingleValuedCostFunction costFunction)
pure virtual

Set the cost Function. This method has to be overloaded by derived classes because the CostFunctionAdaptor requires to know the number of parameters at construction time. This number of parameters is obtained at run-time from the itkCostFunction. As a consequence each derived optimizer should construct its own CostFunctionAdaptor when overloading this method

Implemented in itk::LBFGSBOptimizer, itk::AmoebaOptimizer, itk::ConjugateGradientOptimizer, and itk::LBFGSOptimizer.

void itk::SingleValuedNonLinearVnlOptimizer::SetCostFunctionAdaptor ( CostFunctionAdaptorType adaptor)
protected
virtual void itk::SingleValuedNonLinearVnlOptimizer::SetMaximize ( bool  _arg)
virtual

Methods to define whether the cost function will be maximized or minimized. By default the VNL amoeba optimizer is only a minimizer. Maximization is implemented here by notifying the CostFunctionAdaptor which in its turn will multiply the function values and its derivative by -1.0.

void itk::SingleValuedNonLinearVnlOptimizer::SetMinimize ( bool  v)
inline

Methods to define whether the cost function will be maximized or minimized. By default the VNL amoeba optimizer is only a minimizer. Maximization is implemented here by notifying the CostFunctionAdaptor which in its turn will multiply the function values and its derivative by -1.0.

Definition at line 74 of file itkSingleValuedNonLinearVnlOptimizer.h.

Member Data Documentation

ParametersType itk::SingleValuedNonLinearVnlOptimizer::m_CachedCurrentPosition
mutableprivate

Definition at line 124 of file itkSingleValuedNonLinearVnlOptimizer.h.

DerivativeType itk::SingleValuedNonLinearVnlOptimizer::m_CachedDerivative
mutableprivate

Definition at line 126 of file itkSingleValuedNonLinearVnlOptimizer.h.

MeasureType itk::SingleValuedNonLinearVnlOptimizer::m_CachedValue
mutableprivate

Definition at line 125 of file itkSingleValuedNonLinearVnlOptimizer.h.

CommandType::Pointer itk::SingleValuedNonLinearVnlOptimizer::m_Command
private

Definition at line 122 of file itkSingleValuedNonLinearVnlOptimizer.h.

CostFunctionAdaptorType* itk::SingleValuedNonLinearVnlOptimizer::m_CostFunctionAdaptor
private

Definition at line 118 of file itkSingleValuedNonLinearVnlOptimizer.h.

bool itk::SingleValuedNonLinearVnlOptimizer::m_Maximize
private

Definition at line 120 of file itkSingleValuedNonLinearVnlOptimizer.h.


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