Common FARSA setup of GA. More...

Inheritance diagram for LaralGA:

Classes
class	evaluationThread
Public Member Functions
	LaralGA ()
	Constructor.
virtual	~LaralGA ()
	Destructor.
virtual void	configure (ConfigurationParameters &params, QString prefix)
	Configure the object using the ConfigurationParameters specified.
virtual QVector< Evaluation * >	evaluationPool ()
	Return the Evaluation Pool containing the Evaluations used.
virtual Evaluation *	evaluationPrototype ()
	Returns the Evaluation object used as prototype to eventually generate other evaluators.
virtual void	finalize ()
	Finalize the Genetic Algorithm Process.
MultiTrials *	fitnessFunction ()
	Return the fitness function.
virtual void	gaStep ()
	Execute a single minor step and return.
virtual void	initialize ()
	Initialize the Genetic Algorithm Process: set current replication to zero set current generation to zero.
bool	isElitismEnabled ()
	Return true if elitism is enabled.
Mutation *	mutation ()
	Return the Mutation operator.
int	numElitism ()
	Return the number of elited Genotypes.
int	numOffspring ()
	Return the number of offspring for each selected Genotype.
int	numReproducing ()
	Return the number of selected Genotypes.
int	numThreads ()
	Return the number of thread currently used.
virtual void	save (ConfigurationParameters &params, QString prefix)
	Save the actual status of parameters into the ConfigurationParameters object passed.
void	setFitnessFunction (MultiTrials *fitfunc)
	Set the fitness function to use.
void	setMutation (Mutation *mutate)
	Set the Mutation operator to use.
void	setNumThreads (int numThreads)
	Set the number of thread to use; or in other words the number of parallel evaluation of Genotypes.
void	setReproduceParams (int nreproducing, bool useElitism=false, int nelitism=0)
	Set Reproducing parameters
virtual void	skipEvaluation ()
	Skip the evaluation phase and put the algorithm in the state just after the evaluation.
Static Public Member Functions
static void	describe (QString type)
	Add to Factory::typeDescriptions() the descriptions of all parameters and subgroups.
Protected Types
enum	GAPhases { initEvaluation, evaluating, nextGeneration_pass1, nextGeneration_pass2, endEvolution }
	used for manage GA cycle More...
Protected Attributes
unsigned int	currGenotype
	current Genotype in evaluation
GAPhases	currPhase
	current Phase
bool	elitismEnabled
	true is elitism is enabled
MultiTrials *	fitfunc
	fitness function
bool	isFinalized
	True if LaralGA is finalized.
bool	isInitialized
	True if LaralGA is initialized.
Mutation *	muta
	mutation operator
int	nelitism
	number of elited genotypes
int	noffspring
	number of offspring
int	nreproducing
	number of selected genotypes

Detailed Description

Common FARSA setup of GA.

Description

This GA implements the common characteristics of FARSA setup:

Rank Selection with Truncation
Use MultiTrials mechanism for Evaluating the Fitness of a Genotype
Possibility to set one of Mutations availables (or a custom one)
No Crossover

Warnings

Definition at line 44 of file laralga.h.

Member Enumeration Documentation

enum GAPhases [protected]

used for manage GA cycle

Definition at line 136 of file laralga.h.

Constructor & Destructor Documentation

LaralGA ( )

Constructor.

Definition at line 31 of file laralga.cpp.

References LaralGA::currGenotype, LaralGA::currPhase, LaralGA::elitismEnabled, LaralGA::fitfunc, LaralGA::isFinalized, LaralGA::isInitialized, LaralGA::muta, LaralGA::nelitism, LaralGA::noffspring, LaralGA::nreproducing, and GeneticAlgo::numGens.

~LaralGA ( ) [virtual]

Destructor.

Definition at line 47 of file laralga.cpp.

Member Function Documentation

void configure	(	ConfigurationParameters &	params,
		QString	prefix
	)		`[virtual]`

Configure the object using the ConfigurationParameters specified.

Parameters:

params	the object with configuration parameters
prefix	the group in which parameters are expected to be (this is terminated by a separator character). We expect to receive as prefix the group which contains all GA parameters

Implements ParameterSettableWithConfigureFunction.

Definition at line 51 of file laralga.cpp.

References LaralGA::fitfunc, GeneticAlgo::genome(), ConfigurationParameters::getObjectFromGroup(), ConfigurationParameters::getValue(), ConfigurationParameters::GroupSeparator(), LaralGA::nelitism, GeneticAlgo::numGenerations(), LaralGA::setFitnessFunction(), Evaluation::setGenome(), GeneticAlgo::setGenome(), LaralGA::setMutation(), GeneticAlgo::setNumGenerations(), LaralGA::setNumThreads(), and LaralGA::setReproduceParams().

void describe ( QString type ) [static]

Add to Factory::typeDescriptions() the descriptions of all parameters and subgroups.

Reimplemented from ParameterSettable.

Definition at line 84 of file laralga.cpp.

QVector< Evaluation * > evaluationPool ( ) [virtual]

Return the Evaluation Pool containing the Evaluations used.

Implements GeneticAlgo.

Definition at line 125 of file laralga.cpp.

Evaluation * evaluationPrototype ( ) [virtual]

Returns the Evaluation object used as prototype to eventually generate other evaluators.

This object could be or could be not used to compute the fitness of genotypes (depending on the specific genetic algorithm)

Implements GeneticAlgo.

Definition at line 120 of file laralga.cpp.

References LaralGA::fitfunc.

void finalize ( ) [virtual]

Finalize the Genetic Algorithm Process.

Implements GeneticAlgo.

Definition at line 277 of file laralga.cpp.

References GeneticAlgo::evolutionEnd, LaralGA::isFinalized, and LaralGA::isInitialized.

Referenced by LaralGA::gaStep().

MultiTrials * fitnessFunction ( )

Return the fitness function.

Definition at line 133 of file laralga.cpp.

References LaralGA::fitfunc.

Referenced by LaralGA::save().

void gaStep ( ) [virtual]

Execute a single minor step and return.

Implements GeneticAlgo.

Definition at line 206 of file laralga.cpp.

References LaralGA::currPhase, GeneticAlgo::evaluationDone, LaralGA::finalize(), GeneticAlgo::generation(), GeneticAlgo::genome(), GeneticAlgo::numGens, Genotype::rankGreaterThanComparator(), GeneticAlgo::setGeneration(), Genome::size(), and GeneticAlgo::updateStats().

void initialize ( ) [virtual]

Initialize the Genetic Algorithm Process:

set current replication to zero
set current generation to zero.

Implements GeneticAlgo.

Definition at line 178 of file laralga.cpp.

References LaralGA::currGenotype, LaralGA::currPhase, GeneticAlgo::evaluationDone, GeneticAlgo::evolutionEnd, LaralGA::fitfunc, LaralGA::isFinalized, LaralGA::isInitialized, LaralGA::muta, and GeneticAlgo::setGeneration().

bool isElitismEnabled ( )

Return true if elitism is enabled.

Definition at line 174 of file laralga.cpp.

References LaralGA::elitismEnabled.

Referenced by LaralGA::save().

Mutation * mutation ( )

Return the Mutation operator.

Definition at line 142 of file laralga.cpp.

References LaralGA::muta.

Referenced by LaralGA::save().

int numElitism ( )

Return the number of elited Genotypes.

Definition at line 170 of file laralga.cpp.

References LaralGA::nelitism.

Referenced by LaralGA::save().

int numOffspring ( )

Return the number of offspring for each selected Genotype.

Definition at line 166 of file laralga.cpp.

References LaralGA::noffspring.

int numReproducing ( )

Return the number of selected Genotypes.

Definition at line 162 of file laralga.cpp.

References LaralGA::nreproducing.

Referenced by LaralGA::save().

int numThreads ( )

Return the number of thread currently used.

Definition at line 111 of file laralga.cpp.

Referenced by LaralGA::save(), and LaralGA::setNumThreads().

void save	(	ConfigurationParameters &	params,
		QString	prefix
	)		`[virtual]`

Save the actual status of parameters into the ConfigurationParameters object passed.

Parameters:

params	the configuration parameters object on which save actual parameters
prefix	the prefix to use to access the object configuration parameters.

Implements ParameterSettable.

Definition at line 68 of file laralga.cpp.

References ConfigurationParameters::createParameter(), ConfigurationParameters::createSubGroup(), LaralGA::fitnessFunction(), GeneticAlgo::genome(), LaralGA::isElitismEnabled(), LaralGA::mutation(), LaralGA::numElitism(), GeneticAlgo::numGenerations(), LaralGA::numReproducing(), LaralGA::numThreads(), Genome::save(), Mutation::save(), and MultiTrials::save().

void setFitnessFunction ( MultiTrials * fitfunc )

Set the fitness function to use.

Definition at line 115 of file laralga.cpp.

References LaralGA::fitfunc, and Evaluation::setGA().

Referenced by LaralGA::configure().

void setMutation ( Mutation * mutate )

Set the Mutation operator to use.

Definition at line 137 of file laralga.cpp.

References LaralGA::muta, and Mutation::setGA().

Referenced by LaralGA::configure().

void setNumThreads ( int numThreads )

Set the number of thread to use; or in other words the number of parallel evaluation of Genotypes.

Definition at line 97 of file laralga.cpp.

References LaralGA::fitfunc, LaralGA::isFinalized, LaralGA::isInitialized, and LaralGA::numThreads().

Referenced by LaralGA::configure().

void setReproduceParams	(	int	nreproducing,
		bool	useElitism = `false`,
		int	nelitism = `0`
	)

Set Reproducing parameters

Note:: The number of offsprings are automatically calculated dividing nreproducinb by Genome::size()

Parameters:

nreproducing	is the number of Genotypes to select
useElitism	if true use elitism, othersiwe not
nelitism	is the number of Genotypes selected to copy unthouched to next generation

Definition at line 146 of file laralga.cpp.

References LaralGA::elitismEnabled, GeneticAlgo::genomev, LaralGA::nelitism, LaralGA::noffspring, LaralGA::nreproducing, and Genome::size().

Referenced by LaralGA::configure().

void skipEvaluation ( ) [virtual]

Skip the evaluation phase and put the algorithm in the state just after the evaluation.

This can be used after restarting an interrupted evolution to prevent re-evaluation of the genome that was loaded from file.

Implements GeneticAlgo.

Definition at line 267 of file laralga.cpp.

References LaralGA::currPhase, GeneticAlgo::evaluationDone, GeneticAlgo::generation(), and GeneticAlgo::numGens.