ITK
4.6.0
Insight Segmentation and Registration Toolkit
|
#include <itkGradientDescentOptimizerv4.h>
Gradient descent optimizer.
GradientDescentOptimizer implements a simple gradient descent optimizer. At each iteration the current position is updated according to
Optionally, the best metric value and matching parameters can be stored and retried via GetValue() and GetCurrentPosition(). See SetReturnBestParametersAndValue().
Gradient scales can be manually set or automatically estimated, as documented in the base class. The learing rate defaults to 1.0, and can be set in two ways: 1) manually, via SetLearningRate()
. Or, 2) automatically, either at each iteration or only at the first iteration, by assigning a ScalesEstimator via SetScalesEstimator(). When a ScalesEstimator is assigned, the optimizer is enabled by default to estimate learning rate only once, during the first iteration. This behavior can be changed via SetDoEstimateLearningRateAtEveryIteration() and SetDoEstimateLearningRateOnce(). For learning rate to be estimated at each iteration, the user must call SetDoEstimateLearningRateAtEveryIteration(true) and SetDoEstimateLearningRateOnce(false). When enabled, the optimizer computes learning rate(s) such that at each step, each voxel's change in physical space will be less than m_MaximumStepSizeInPhysicalUnits.
m_LearningRate = m_MaximumStepSizeInPhysicalUnits / m_ScalesEstimator->EstimateStepScale(scaledGradient)
where m_MaximumStepSizeInPhysicalUnits defaults to the voxel spacing returned by m_ScalesEstimator->EstimateMaximumStepSize() (which is typically 1 voxel), and can be set by the user via SetMaximumStepSizeInPhysicalUnits(). When SetDoEstimateLearningRateOnce is enabled, the voxel change may become being greater than m_MaximumStepSizeInPhysicalUnits in later iterations.
Definition at line 77 of file itkGradientDescentOptimizerv4.h.
Public Member Functions | |
virtual ::itk::LightObject::Pointer | CreateAnother (void) const |
virtual void | EstimateLearningRate () |
virtual const TInternalComputationValueType & | GetConvergenceValue () const |
virtual const TInternalComputationValueType & | GetLearningRate () const |
virtual const TInternalComputationValueType & | GetMaximumStepSizeInPhysicalUnits () const |
virtual const char * | GetNameOfClass () const |
virtual void | ResumeOptimization () ITK_OVERRIDE |
virtual void | SetConvergenceWindowSize (SizeValueType _arg) |
virtual void | SetLearningRate (TInternalComputationValueType _arg) |
virtual void | SetMaximumStepSizeInPhysicalUnits (TInternalComputationValueType _arg) |
virtual void | SetMinimumConvergenceValue (TInternalComputationValueType _arg) |
virtual void | StartOptimization (bool doOnlyInitialization=false) ITK_OVERRIDE |
virtual void | StopOptimization (void) ITK_OVERRIDE |
virtual void | SetDoEstimateLearningRateAtEachIteration (bool _arg) |
virtual const bool & | GetDoEstimateLearningRateAtEachIteration () const |
virtual void | DoEstimateLearningRateAtEachIterationOn () |
virtual void | DoEstimateLearningRateAtEachIterationOff () |
virtual void | SetDoEstimateLearningRateOnce (bool _arg) |
virtual const bool & | GetDoEstimateLearningRateOnce () const |
virtual void | DoEstimateLearningRateOnceOn () |
virtual void | DoEstimateLearningRateOnceOff () |
virtual void | SetReturnBestParametersAndValue (bool _arg) |
virtual const bool & | GetReturnBestParametersAndValue () const |
virtual void | ReturnBestParametersAndValueOn () |
virtual void | ReturnBestParametersAndValueOff () |
Public Member Functions inherited from itk::GradientDescentOptimizerBasev4Template< TInternalComputationValueType > | |
virtual SizeValueType | GetCurrentIteration () const |
virtual const DerivativeType & | GetGradient () const |
virtual SizeValueType | GetNumberOfIterations () const |
virtual const StopConditionType & | GetStopCondition () const |
virtual const StopConditionReturnStringType | GetStopConditionDescription () const ITK_OVERRIDE |
virtual void | SetNumberOfIterations (SizeValueType _arg) |
virtual void | ModifyGradientByScales () |
virtual void | ModifyGradientByLearningRate () |
Public Member Functions inherited from itk::ObjectToObjectOptimizerBaseTemplate< TInternalComputationValueType > | |
virtual const MeasureType & | GetCurrentMetricValue () const |
virtual const ParametersType & | GetCurrentPosition () const |
virtual const ThreadIdType & | GetNumberOfThreads () const |
virtual const ScalesType & | GetScales () const |
virtual const bool & | GetScalesAreIdentity () const |
bool | GetScalesInitialized (void) const |
virtual const MeasureType & | GetValue () const |
virtual const ScalesType & | GetWeights () const |
virtual const bool & | GetWeightsAreIdentity () const |
virtual void | SetNumberOfThreads (ThreadIdType number) |
virtual void | SetScalesEstimator (ScalesEstimatorType *_arg) |
virtual void | SetWeights (ScalesType _arg) |
virtual void | SetMetric (MetricType *_arg) |
virtual MetricType * | GetModifiableMetric () |
virtual const MetricType * | GetMetric () const |
virtual void | SetScales (const ScalesType &scales) |
virtual void | SetDoEstimateScales (bool _arg) |
virtual const bool & | GetDoEstimateScales () const |
virtual void | DoEstimateScalesOn () |
virtual void | DoEstimateScalesOff () |
Public Member Functions inherited from itk::Object | |
unsigned long | AddObserver (const EventObject &event, Command *) |
unsigned long | AddObserver (const EventObject &event, Command *) const |
virtual void | DebugOff () const |
virtual void | DebugOn () const |
Command * | GetCommand (unsigned long tag) |
bool | GetDebug () const |
MetaDataDictionary & | GetMetaDataDictionary (void) |
const MetaDataDictionary & | GetMetaDataDictionary (void) const |
virtual ModifiedTimeType | GetMTime () const |
virtual const TimeStamp & | GetTimeStamp () const |
bool | HasObserver (const EventObject &event) const |
void | InvokeEvent (const EventObject &) |
void | InvokeEvent (const EventObject &) const |
virtual void | Modified () const |
virtual void | Register () const ITK_OVERRIDE |
void | RemoveAllObservers () |
void | RemoveObserver (unsigned long tag) |
void | SetDebug (bool debugFlag) const |
void | SetMetaDataDictionary (const MetaDataDictionary &rhs) |
virtual void | SetReferenceCount (int) ITK_OVERRIDE |
virtual void | UnRegister () const ITK_OVERRIDE |
virtual void | SetObjectName (std::string _arg) |
virtual const std::string & | GetObjectName () const |
Public Member Functions inherited from itk::LightObject | |
virtual void | Delete () |
virtual int | GetReferenceCount () const |
itkCloneMacro (Self) | |
void | Print (std::ostream &os, Indent indent=0) const |
Static Public Member Functions | |
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 flag) |
Static Public Member Functions inherited from itk::LightObject | |
static void | BreakOnError () |
static Pointer | New () |
Private Member Functions | |
GradientDescentOptimizerv4Template (const Self &) | |
void | operator= (const Self &) |
Additional Inherited Members | |
Protected Types inherited from itk::LightObject | |
typedef int | InternalReferenceCountType |
typedef SmartPointer< const Self > itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::ConstPointer |
Definition at line 85 of file itkGradientDescentOptimizerv4.h.
typedef Superclass::DerivativeType itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::DerivativeType |
Derivative type
Definition at line 98 of file itkGradientDescentOptimizerv4.h.
typedef Superclass::IndexRangeType itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::IndexRangeType |
Definition at line 102 of file itkGradientDescentOptimizerv4.h.
typedef TInternalComputationValueType itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::InternalComputationValueType |
It should be possible to derive the internal computation type from the class object.
Definition at line 91 of file itkGradientDescentOptimizerv4.h.
typedef Superclass::MeasureType itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::MeasureType |
Metric type over which this class is templated
Definition at line 101 of file itkGradientDescentOptimizerv4.h.
typedef Superclass::ParametersType itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::ParametersType |
Definition at line 104 of file itkGradientDescentOptimizerv4.h.
typedef SmartPointer< Self > itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::Pointer |
Definition at line 84 of file itkGradientDescentOptimizerv4.h.
typedef Superclass::ScalesType itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::ScalesType |
Definition at line 103 of file itkGradientDescentOptimizerv4.h.
typedef GradientDescentOptimizerv4Template itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::Self |
Standard class typedefs.
Definition at line 82 of file itkGradientDescentOptimizerv4.h.
typedef Superclass::StopConditionType itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::StopConditionType |
Definition at line 105 of file itkGradientDescentOptimizerv4.h.
typedef GradientDescentOptimizerBasev4Template<TInternalComputationValueType> itk::GradientDescentOptimizerv4Template< TInternalComputationValueType >::Superclass |
Definition at line 83 of file itkGradientDescentOptimizerv4.h.
|
protected |
Default constructor
|
protectedvirtual |
Destructor
|
private |
|
protectedvirtual |
Advance one Step following the gradient direction. Includes transform update.
Reimplemented in itk::QuasiNewtonOptimizerv4Template< TInternalComputationValueType >, itk::GradientDescentLineSearchOptimizerv4Template< TInternalComputationValueType >, itk::RegularStepGradientDescentOptimizerv4< TInternalComputationValueType >, and itk::ConjugateGradientLineSearchOptimizerv4Template< TInternalComputationValueType >.
|
virtual |
Create an object from an instance, potentially deferring to a factory. This method allows you to create an instance of an object that is exactly the same type as the referring object. This is useful in cases where an object has been cast back to a base class.
Reimplemented from itk::Object.
Reimplemented in itk::QuasiNewtonOptimizerv4Template< TInternalComputationValueType >, itk::MultiGradientOptimizerv4Template< TInternalComputationValueType >, and itk::RegularStepGradientDescentOptimizerv4< TInternalComputationValueType >.
|
virtual |
Option to use ScalesEstimator for learning rate estimation at each iteration. The estimation overrides the learning rate set by SetLearningRate(). Default is false.
|
virtual |
Option to use ScalesEstimator for learning rate estimation at each iteration. The estimation overrides the learning rate set by SetLearningRate(). Default is false.
|
virtual |
Option to use ScalesEstimator for learning rate estimation only once, during first iteration. The estimation overrides the learning rate set by SetLearningRate(). Default is true.
|
virtual |
Option to use ScalesEstimator for learning rate estimation only once, during first iteration. The estimation overrides the learning rate set by SetLearningRate(). Default is true.
|
virtual |
Estimate the learning rate based on the current gradient.
Reimplemented in itk::RegularStepGradientDescentOptimizerv4< TInternalComputationValueType >.
|
virtual |
Get current convergence value
|
virtual |
Option to use ScalesEstimator for learning rate estimation at each iteration. The estimation overrides the learning rate set by SetLearningRate(). Default is false.
|
virtual |
Option to use ScalesEstimator for learning rate estimation only once, during first iteration. The estimation overrides the learning rate set by SetLearningRate(). Default is true.
|
virtual |
Get the learning rate.
|
virtual |
Get the maximum step size, in physical space units.
|
virtual |
Run-time type information (and related methods).
Reimplemented from itk::GradientDescentOptimizerBasev4Template< TInternalComputationValueType >.
Reimplemented in itk::QuasiNewtonOptimizerv4Template< TInternalComputationValueType >, itk::GradientDescentLineSearchOptimizerv4Template< TInternalComputationValueType >, itk::ConjugateGradientLineSearchOptimizerv4Template< TInternalComputationValueType >, itk::MultiGradientOptimizerv4Template< TInternalComputationValueType >, and itk::RegularStepGradientDescentOptimizerv4< TInternalComputationValueType >.
|
virtual |
Flag. Set to have the optimizer track and return the best best metric value and corresponding best parameters that were calculated during the optimization. This captures the best solution when the optimizer oversteps or osciallates near the end of an optimization. Results are stored in m_CurrentMetricValue and in the assigned metric's parameters, retrievable via optimizer->GetCurrentPosition(). This option requires additional memory to store the best parameters, which can be large when working with high-dimensional transforms such as DisplacementFieldTransform.
|
protectedvirtual |
Modify the gradient over a given index range.
Implements itk::GradientDescentOptimizerBasev4Template< TInternalComputationValueType >.
Reimplemented in itk::RegularStepGradientDescentOptimizerv4< TInternalComputationValueType >.
|
protectedvirtual |
Modify the gradient over a given index range.
Implements itk::GradientDescentOptimizerBasev4Template< TInternalComputationValueType >.
Reimplemented in itk::RegularStepGradientDescentOptimizerv4< TInternalComputationValueType >.
|
static |
New macro for creation of through a Smart Pointer
|
private |
|
protectedvirtual |
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::GradientDescentOptimizerBasev4Template< TInternalComputationValueType >.
Reimplemented in itk::QuasiNewtonOptimizerv4Template< TInternalComputationValueType >, itk::RegularStepGradientDescentOptimizerv4< TInternalComputationValueType >, itk::MultiGradientOptimizerv4Template< TInternalComputationValueType >, itk::GradientDescentLineSearchOptimizerv4Template< TInternalComputationValueType >, and itk::ConjugateGradientLineSearchOptimizerv4Template< TInternalComputationValueType >.
|
virtual |
Resume optimization. This runs the optimization loop, and allows continuation of stopped optimization
Implements itk::GradientDescentOptimizerBasev4Template< TInternalComputationValueType >.
Reimplemented in itk::MultiGradientOptimizerv4Template< TInternalComputationValueType >.
|
virtual |
Flag. Set to have the optimizer track and return the best best metric value and corresponding best parameters that were calculated during the optimization. This captures the best solution when the optimizer oversteps or osciallates near the end of an optimization. Results are stored in m_CurrentMetricValue and in the assigned metric's parameters, retrievable via optimizer->GetCurrentPosition(). This option requires additional memory to store the best parameters, which can be large when working with high-dimensional transforms such as DisplacementFieldTransform.
|
virtual |
Flag. Set to have the optimizer track and return the best best metric value and corresponding best parameters that were calculated during the optimization. This captures the best solution when the optimizer oversteps or osciallates near the end of an optimization. Results are stored in m_CurrentMetricValue and in the assigned metric's parameters, retrievable via optimizer->GetCurrentPosition(). This option requires additional memory to store the best parameters, which can be large when working with high-dimensional transforms such as DisplacementFieldTransform.
|
virtual |
Window size for the convergence checker. The convergence checker calculates convergence value by fitting to a window of the energy (metric value) profile.
The default m_ConvergenceWindowSize is set to 50 to pass all tests. It is suggested to use 10 for less stringent convergence checking.
|
virtual |
Option to use ScalesEstimator for learning rate estimation at each iteration. The estimation overrides the learning rate set by SetLearningRate(). Default is false.
|
virtual |
Option to use ScalesEstimator for learning rate estimation only once, during first iteration. The estimation overrides the learning rate set by SetLearningRate(). Default is true.
|
virtual |
Set the learning rate.
|
virtual |
Set the maximum step size, in physical space units.
Only relevant when m_ScalesEstimator is set by user, and automatic learning rate estimation is enabled. See main documentation.
|
virtual |
Minimum convergence value for convergence checking. The convergence checker calculates convergence value by fitting to a window of the energy profile. When the convergence value reaches a small value, it would be treated as converged.
The default m_MinimumConvergenceValue is set to 1e-8 to pass all tests. It is suggested to use 1e-6 for less stringent convergence checking.
|
virtual |
Flag. Set to have the optimizer track and return the best best metric value and corresponding best parameters that were calculated during the optimization. This captures the best solution when the optimizer oversteps or osciallates near the end of an optimization. Results are stored in m_CurrentMetricValue and in the assigned metric's parameters, retrievable via optimizer->GetCurrentPosition(). This option requires additional memory to store the best parameters, which can be large when working with high-dimensional transforms such as DisplacementFieldTransform.
|
virtual |
Start and run the optimization
Reimplemented from itk::GradientDescentOptimizerBasev4Template< TInternalComputationValueType >.
Reimplemented in itk::RegularStepGradientDescentOptimizerv4< TInternalComputationValueType >, itk::MultiGradientOptimizerv4Template< TInternalComputationValueType >, itk::QuasiNewtonOptimizerv4Template< TInternalComputationValueType >, and itk::ConjugateGradientLineSearchOptimizerv4Template< TInternalComputationValueType >.
|
virtual |
Stop optimization. The object is left in a state so the optimization can be resumed by calling ResumeOptimization.
Reimplemented from itk::GradientDescentOptimizerBasev4Template< TInternalComputationValueType >.
Reimplemented in itk::MultiGradientOptimizerv4Template< TInternalComputationValueType >.
|
protected |
Definition at line 235 of file itkGradientDescentOptimizerv4.h.
|
protected |
Current convergence value.
Definition at line 231 of file itkGradientDescentOptimizerv4.h.
|
protected |
Store the best value and related paramters
Definition at line 234 of file itkGradientDescentOptimizerv4.h.
|
protected |
Manual learning rate to apply. It is overridden by automatic learning rate estimation if enabled. See main documentation.
Definition at line 212 of file itkGradientDescentOptimizerv4.h.
|
protected |
Minimum convergence value for convergence checking. The convergence checker calculates convergence value by fitting to a window of the energy profile. When the convergence value reaches a small value, such as 1e-8, it would be treated as converged.
Definition at line 227 of file itkGradientDescentOptimizerv4.h.
|
protected |
Store the previous gradient value at each iteration, so we can detect the changes in geradient direction. This is needed by the regular step gradient descent and Quasi newton optimizers.
Definition at line 245 of file itkGradientDescentOptimizerv4.h.
|
protected |
Flag to control returning of best value and parameters.
Definition at line 238 of file itkGradientDescentOptimizerv4.h.