GeneticAlgorithm Class Reference

Class imlementing the Genetic Algorithm. More...

#include <GeneticAlgorithm.h>

Collaboration diagram for GeneticAlgorithm:

Public Member Functions
	GeneticAlgorithm ()
	~GeneticAlgorithm ()
IModel *	optimize (IPopulation *population)
void	initialize (IPopulation *population)
bool	nextGeneration ()
int	getCurrentGeneration ()
void	setNGenerationsMax (int generationsMax)
void	setPopulationSize (int populationSize)

Private Attributes
int	m_generationsMax
	Stores the maximum number of generations.
int	m_populationSize
	Stores the desired population size.
int	m_currentGeneration
	Stores the number of the current generation.
IPopulation *	m_population
	Stores a pointer to the population being optimized.

Detailed Description

Class imlementing the Genetic Algorithm.

There is no need to derive from this class: the implementation is built on top of the IModel, IFigureOfMerit and IPopulation interfaces.

The algorithm flow is as follows:

• Creates an initial population.
• Rank the population.
• Repeat the following until a solution is found or a maximum numeber of generations is reached:
  • Select parents among the fittest individuals.
  • Cross them over to form a new population.
  • Mutate some individuals.
  • Rank the new population.

Definition at line 23 of file GeneticAlgorithm.h.

Constructor & Destructor Documentation

GeneticAlgorithm()

GeneticAlgorithm::GeneticAlgorithm

(

)

Default Constructor

Definition at line 6 of file GeneticAlgorithm.cxx.

~GeneticAlgorithm()

GeneticAlgorithm::~GeneticAlgorithm

(

)

Destructor

Definition at line 12 of file GeneticAlgorithm.cxx.

Member Function Documentation

optimize()

IModel * GeneticAlgorithm::optimize

(

IPopulation *

population

)

Finds the best solution given a population of models.

Parameters

population

Population of models to optimize.

Returns

Best fitted model after optimization.

Definition at line 20 of file GeneticAlgorithm.cxx.

initialize()

void GeneticAlgorithm::initialize

(

IPopulation *

population

)

Initialize the algorithm before the optimization loop starts.

This function is provided so that the user have the option to control the optimization loop and possibly execute some code before/after each iteration. This can be useful for example to implement tests that need to monitor the progress of the optimization.

Parameters

population

Population of models to optimize.

Definition at line 37 of file GeneticAlgorithm.cxx.

nextGeneration()

bool GeneticAlgorithm::nextGeneration

(

)

Perform one iteration of the optimization loop: creates next generation of models.

This function is provided so that the user have the option to control the optimization loop and possibly execute some code before/after each iteration. This can be useful for example to implement tests that need to monitor the progress of the optimization.

Returns

true if more generations are needed, false if optimal solution has been reached.

Definition at line 53 of file GeneticAlgorithm.cxx.

getCurrentGeneration()

int GeneticAlgorithm::getCurrentGeneration

(

)

Returns the current generation number.

Returns

Number of the current generation.

Definition at line 75 of file GeneticAlgorithm.cxx.

setNGenerationsMax()

void GeneticAlgorithm::setNGenerationsMax

(

int

generationsMax

)

Sets the maximum number of generations before giving up.

Parameters

Desired

maximum number of iterations.

Definition at line 83 of file GeneticAlgorithm.cxx.

setPopulationSize()

void GeneticAlgorithm::setPopulationSize

(

int

populationSize

)

Sets the population size

Parameters

Desired

population size.

Definition at line 92 of file GeneticAlgorithm.cxx.

---

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

• include/GeneticAlgorithm.h
• src/GeneticAlgorithm.cxx