ITK  4.1.0
Insight Segmentation and Registration Toolkit
Classes | Public Types | Public Member Functions | Protected Member Functions | Protected Attributes
itk::ParticleSwarmOptimizerBase Class Reference

#include <itkParticleSwarmOptimizerBase.h>

+ Inheritance diagram for itk::ParticleSwarmOptimizerBase:
+ Collaboration diagram for itk::ParticleSwarmOptimizerBase:

List of all members.

Classes

struct  ParticleData

Public Types

typedef SmartPointer< const SelfConstPointer
typedef
CostFunctionType::MeasureType 
MeasureType
typedef unsigned int NumberOfGenerationsType
typedef unsigned int NumberOfIterationsType
typedef unsigned int NumberOfParticlesType
typedef std::vector< std::pair
< ParametersType::ValueType,
ParametersType::ValueType > > 
ParameterBoundsType
typedef SmartPointer< SelfPointer
typedef
Statistics::MersenneTwisterRandomVariateGenerator 
RandomVariateGeneratorType
typedef ParticleSwarmOptimizerBase Self
typedef
SingleValuedNonLinearOptimizer 
Superclass
typedef std::vector< ParticleDataSwarmType
typedef ParametersType::ValueType ValueType

Public Member Functions

virtual MeasureType GetFunctionConvergenceTolerance ()
virtual bool GetInitializeNormalDistribution ()
virtual NumberOfIterationsType GetMaximalNumberOfIterations ()
virtual const char * GetNameOfClass () const
virtual NumberOfGenerationsType GetNumberOfGenerationsWithMinimalImprovement ()
virtual NumberOfParticlesType GetNumberOfParticles ()
ParameterBoundsType GetParameterBounds () const
virtual ParametersType GetParametersConvergenceTolerance ()
virtual double GetPercentageParticlesConverged ()
virtual bool GetPrintSwarm ()
virtual
RandomVariateGeneratorType::IntegerType 
GetSeed ()
virtual const std::string GetStopConditionDescription () const
virtual bool GetUseSeed ()
MeasureType GetValue () const
virtual void InitializeNormalDistributionOff ()
virtual void InitializeNormalDistributionOn ()
void PrintSwarm (std::ostream &os, Indent indent) const
virtual void PrintSwarmOff ()
virtual void PrintSwarmOn ()
virtual void SetFunctionConvergenceTolerance (MeasureType _arg)
virtual void SetInitializeNormalDistribution (bool _arg)
virtual void SetMaximalNumberOfIterations (NumberOfIterationsType _arg)
virtual void SetNumberOfGenerationsWithMinimalImprovement (NumberOfGenerationsType _arg)
void SetNumberOfParticles (NumberOfParticlesType n)
void SetParametersConvergenceTolerance (ValueType convergenceTolerance, unsigned int sz)
virtual void SetParametersConvergenceTolerance (ParametersType _arg)
virtual void SetPercentageParticlesConverged (double _arg)
virtual void SetPrintSwarm (bool _arg)
virtual void SetSeed (RandomVariateGeneratorType::IntegerType _arg)
virtual void SetUseSeed (bool _arg)
void StartOptimization (void)
virtual void UseSeedOff ()
virtual void UseSeedOn ()
void SetInitialSwarm (const SwarmType &initialSwarm)
void ClearSwarm ()
virtual void SetParameterBounds (ParameterBoundsType &bounds)
void SetParameterBounds (std::pair< ParametersType::ValueType, ParametersType::ValueType > &bounds, unsigned int n)

Protected Member Functions

void FileInitialization ()
virtual void Initialize ()
void operator= (const Self &)
 ParticleSwarmOptimizerBase ()
 ParticleSwarmOptimizerBase (const Self &)
void PrintParamtersType (const ParametersType &x, std::ostream &os) const
void PrintSelf (std::ostream &os, Indent indent) const
void RandomInitialization ()
virtual void UpdateSwarm ()=0
virtual void ValidateSettings ()
virtual ~ParticleSwarmOptimizerBase ()

Protected Attributes

CostFunctionType::MeasureType m_FunctionBestValue
std::vector< MeasureTypem_FunctionBestValueMemory
CostFunctionType::MeasureType m_FunctionConvergenceTolerance
bool m_InitializeNormalDistribution
NumberOfIterationsType m_IterationIndex
NumberOfIterationsType m_MaximalNumberOfIterations
NumberOfGenerationsType m_NumberOfGenerationsWithMinimalImprovement
NumberOfParticlesType m_NumberOfParticles
ParameterBoundsType m_ParameterBounds
ParametersType m_ParametersBestValue
ParametersType m_ParametersConvergenceTolerance
std::vector< ParticleDatam_Particles
double m_PercentageParticlesConverged
bool m_PrintSwarm
RandomVariateGeneratorType::IntegerType m_Seed
std::ostringstream m_StopConditionDescription
bool m_UseSeed

Detailed Description

Abstract implementation of a Particle Swarm Optimization (PSO) algorithm.

The PSO algorithm was originally presented in:
J. Kennedy, R. Eberhart, "Particle Swarm Optimization", Proc. IEEE Int. Neural Networks, 1995.

The algorithm is a stochastic global search optimization approach. Optimization is performed by maintaining a swarm of particles that traverse the parameter space, searching for the optimal function value. Associated with each particle are its location and speed, in parameter space.

Swarm initialization is performed within the user supplied parameter bounds using either a uniform distribution or a normal distribution centered on the initial parameter values supplied by the user. The search terminates when the maximal number of iterations has been reached or when the change in the best value in the past $g$ generations is below a threshold and the swarm has collapsed (i.e. best personal particle locations are close to the swarm's best location in parameter space).

The actual optimization procedure, updating the swarm, is performed in the subclasses, required to implement the UpdateSwarm() method.

NOTE: This implementation only performs minimization.

Definition at line 55 of file itkParticleSwarmOptimizerBase.h.


Member Typedef Documentation

Measure type. It defines a type used to return the cost function value.

Reimplemented from itk::SingleValuedNonLinearOptimizer.

Definition at line 84 of file itkParticleSwarmOptimizerBase.h.

Definition at line 83 of file itkParticleSwarmOptimizerBase.h.

Definition at line 81 of file itkParticleSwarmOptimizerBase.h.

Definition at line 82 of file itkParticleSwarmOptimizerBase.h.

Definition at line 69 of file itkParticleSwarmOptimizerBase.h.

Definition at line 87 of file itkParticleSwarmOptimizerBase.h.

Standard "Self" typedef.

Reimplemented from itk::SingleValuedNonLinearOptimizer.

Reimplemented in itk::InitializationBiasedParticleSwarmOptimizer, and itk::ParticleSwarmOptimizer.

Definition at line 60 of file itkParticleSwarmOptimizerBase.h.

Definition at line 80 of file itkParticleSwarmOptimizerBase.h.

Definition at line 85 of file itkParticleSwarmOptimizerBase.h.


Constructor & Destructor Documentation


Member Function Documentation

Specify the initial swarm. Useful for evaluating PSO variants. If the initial swarm is set it will be used. To revert to random initialization (uniform or normal particle distributions) set using an empty swarm.

virtual const char* itk::ParticleSwarmOptimizerBase::GetNameOfClass ( ) const [virtual]

Run-time type information (and related methods).

Reimplemented from itk::SingleValuedNonLinearOptimizer.

Reimplemented in itk::InitializationBiasedParticleSwarmOptimizer, and itk::ParticleSwarmOptimizer.

virtual const std::string itk::ParticleSwarmOptimizerBase::GetStopConditionDescription ( ) const [virtual]

Get the reason for termination

Reimplemented from itk::Optimizer.

virtual bool itk::ParticleSwarmOptimizerBase::GetUseSeed ( ) [virtual]

Get the function value for the current position. NOTE: This value is only valid during and after the execution of the StartOptimization() method.

virtual void itk::ParticleSwarmOptimizerBase::Initialize ( ) [protected, virtual]

Initialize the particle swarm, and seed the random number generator.

void itk::ParticleSwarmOptimizerBase::operator= ( const Self ) [protected]

Types inherited from the superclass

Reimplemented from itk::SingleValuedNonLinearOptimizer.

Reimplemented in itk::InitializationBiasedParticleSwarmOptimizer, and itk::ParticleSwarmOptimizer.

void itk::ParticleSwarmOptimizerBase::PrintParamtersType ( const ParametersType x,
std::ostream &  os 
) const [protected]
void itk::ParticleSwarmOptimizerBase::PrintSelf ( std::ostream &  os,
Indent  indent 
) const [protected, virtual]

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::SingleValuedNonLinearOptimizer.

Reimplemented in itk::InitializationBiasedParticleSwarmOptimizer, and itk::ParticleSwarmOptimizer.

void itk::ParticleSwarmOptimizerBase::PrintSwarm ( std::ostream &  os,
Indent  indent 
) const

Print the swarm information to the given output stream. Each line (particle data) is of the form: current_parameters current_velocity current_value best_parameters best_value

The optimization algorithm will terminate when the function improvement in the last m_NumberOfGenerationsWithMinimalImprovement generations is less than m_FunctionConvergenceTolerance and the maximal distance between particles and the best particle in each dimension is less than m_ParametersConvergenceTolerance[i] for the specified percentage of the particles. That is, we haven't improved the best function value for a while and in the parameter space most (m%) of our particles attained their best value close to the swarm's best value. NOTE: The use of different tolerances for each dimension is desired when optimizing over non-commensurate parameters (e.g. rotation and translation). Alternatively, we could use ITK's parameter scaling approach. The current approach seems more intuitive.

Specify whether to initialize the particles using a normal distribution centered on the user supplied initial value or a uniform distribution. If the optimum is expected to be near the initial value it is likely that initializing with a normal distribution will result in faster convergence.

Specify the initial swarm. Useful for evaluating PSO variants. If the initial swarm is set it will be used. To revert to random initialization (uniform or normal particle distributions) set using an empty swarm.

Set/Get maximal number of iterations - the maximal number of function evaluations is m_MaximalNumberOfIterations*m_NumberOfParticles

Set/Get the number of generations to continue with minimal improvement in the function value, |f_best(g_i) - f_best(g_k)|<threshold where k <= i+NumberOfGenerationsWithMinimalImprovement Minimal value is one.

Set/Get number of particles in the swarm - the maximal number of function evaluations is m_MaximalNumberOfIterations*m_NumberOfParticles

Set/Get the parameter bounds. Search for optimal value is inside these bounds. NOTE: It is assumed that the first entry is the minimal value, second is the maximal value.

Set/Get the parameter bounds. Search for optimal value is inside these bounds. NOTE: It is assumed that the first entry is the minimal value, second is the maximal value.

void itk::ParticleSwarmOptimizerBase::SetParametersConvergenceTolerance ( ValueType  convergenceTolerance,
unsigned int  sz 
)

Set parameters convergence tolerance using the same value for all, sz, parameters

virtual void itk::ParticleSwarmOptimizerBase::SetPrintSwarm ( bool  _arg) [virtual]

Indicate whether or not to output the swarm information when printing an object. By default this option is turned off as it generates too much information.

Set the random number seed for the swarm. Use this method to produce reaptible results, typically, for testing.

virtual void itk::ParticleSwarmOptimizerBase::SetUseSeed ( bool  _arg) [virtual]

Use a specific seed to initialize the random number generator. If On, use m_Seed to seed the random number generator. Default is Off.

Start optimization.

Reimplemented from itk::Optimizer.

virtual void itk::ParticleSwarmOptimizerBase::UpdateSwarm ( ) [protected, pure virtual]

Implement your update rule in this function.

Implemented in itk::InitializationBiasedParticleSwarmOptimizer, and itk::ParticleSwarmOptimizer.

virtual void itk::ParticleSwarmOptimizerBase::UseSeedOff ( ) [virtual]
virtual void itk::ParticleSwarmOptimizerBase::UseSeedOn ( ) [virtual]
virtual void itk::ParticleSwarmOptimizerBase::ValidateSettings ( ) [protected, virtual]

Member Data Documentation

Definition at line 235 of file itkParticleSwarmOptimizerBase.h.

Definition at line 236 of file itkParticleSwarmOptimizerBase.h.

Definition at line 233 of file itkParticleSwarmOptimizerBase.h.

Definition at line 226 of file itkParticleSwarmOptimizerBase.h.

Definition at line 238 of file itkParticleSwarmOptimizerBase.h.

Definition at line 228 of file itkParticleSwarmOptimizerBase.h.

Definition at line 229 of file itkParticleSwarmOptimizerBase.h.

Definition at line 227 of file itkParticleSwarmOptimizerBase.h.

Definition at line 230 of file itkParticleSwarmOptimizerBase.h.

Definition at line 237 of file itkParticleSwarmOptimizerBase.h.

Definition at line 231 of file itkParticleSwarmOptimizerBase.h.

Definition at line 234 of file itkParticleSwarmOptimizerBase.h.

Definition at line 232 of file itkParticleSwarmOptimizerBase.h.

Definition at line 224 of file itkParticleSwarmOptimizerBase.h.

Definition at line 239 of file itkParticleSwarmOptimizerBase.h.

Definition at line 225 of file itkParticleSwarmOptimizerBase.h.

Definition at line 240 of file itkParticleSwarmOptimizerBase.h.


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