grail::evolution::Crossover Struct Reference

The Crossover structure that encapsulates the crossover operation and its configuration in evolutionary algorithm. More...

#include <Crossover.hh>

Public Member Functions
	Crossover (double crossoverIndividualRate=0.45, CrossoverSelectionType selectionType=CrossoverSelectionType::PSEUDO_ROULETTE, double pseudoRouletteRandomPortion=0.3, double crossoverPhaseRate=1.0, std::mt19937_64::result_type seed=std::random_device{}())
	Crossover - Constructor. More...
void	Perform (const std::vector< std::unique_ptr< Individual >> &inputPopulation)
	Perform - the function that performs the crossover phase. More...

Public Attributes
std::vector< std::unique_ptr< Individual > >	populationAfter {}
	The population of children, i.e., after the crossover operation has been performed. More...
double	crossoverIndividualRate = 0.45
	The average rate of crossover. The expected number of children is: populationSize*crossoverIndividualRate.
CrossoverSelectionType	selectionType = CrossoverSelectionType::PSEUDO_ROULETTE
	The method that chooses parents to be recombined.
double	pseudoRouletteRandomPortion = 0.3
	Used only with pseudo-roulette. This method selects pseudoRouletteRandomPortion*100% of individuals at random and (1-pseudoRouletteRandomPortion)*100% according to roulette.
double	crossoverPhaseRate = 1.0
	The probability of a crossover phase to happen (globally). If a crossover is included in the algorithm, this value should most likely be equal to 1.0.

Protected Member Functions
virtual void	CrossoverFunctionOnePoint (Individual &individual1, Individual &individual2)
	Implementation of the standard one-point crossover operation. More...
virtual void	CrossoverFunctionMultiPoint (Individual &individual1, Individual &individual2, size_t length)
	Implementation of the standard k-point (multi-point) crossover operation. More...
virtual void	CrossoverIndividuals (Individual &individual1, Individual &individual2)
	The function that performs a single crossover operation between two individuals. More...

Protected Attributes
std::mt19937_64	randGen

Detailed Description

The Crossover structure that encapsulates the crossover operation and its configuration in evolutionary algorithm.

Constructor & Destructor Documentation

Crossover()

grail::evolution::Crossover::Crossover	(	double	crossoverIndividualRate = 0.45,
		CrossoverSelectionType	selectionType = CrossoverSelectionType::PSEUDO_ROULETTE,
		double	pseudoRouletteRandomPortion = 0.3,
		double	crossoverPhaseRate = 1.0,
		std::mt19937_64::result_type	seed = std::random_device{}()
	)

Crossover - Constructor.

Parameters

crossoverIndividualRate	- the average rate of crossover. The expected number of children is: populationSize*crossoverIndividualRate.
selectionType	- the method that chooses parents to be recombined. See the CrossoverSelectionType enum.
pseudoRouletteRandomPortion	- used only with pseudo-roulette. This method selects pseudoRouletteRandomPortion*100% of individuals at random and (1-pseudoRouletteRandomPortion)*100% according to roulette.
crossoverPhaseRate	- the probability of a crossover phase to happen (globally). If a crossover is included in the algorithm, this value should most likely be equal to 1.0.
seed	- the initial seed for random numbers generator.

Member Function Documentation

CrossoverFunctionMultiPoint()

void grail::evolution::Crossover::CrossoverFunctionMultiPoint ( Individual &  individual1, Individual &  individual2, size_t  length  )

protectedvirtual

Implementation of the standard k-point (multi-point) crossover operation.

Parameters

individual1	- the first parent to produce offspring.
individual2	- the second parent to produce offspring.

CrossoverFunctionOnePoint()

void grail::evolution::Crossover::CrossoverFunctionOnePoint ( Individual &  individual1, Individual &  individual2  )

protectedvirtual

Implementation of the standard one-point crossover operation.

Parameters

individual1	- the first parent to produce offspring.
individual2	- the second parent to produce offspring.

CrossoverIndividuals()

void grail::evolution::Crossover::CrossoverIndividuals ( Individual &  individual1, Individual &  individual2  )

protectedvirtual

The function that performs a single crossover operation between two individuals.

Parameters

individual1	- the first parent to produce offspring.
individual2	- the second parent to produce offspring.

Perform()

void grail::evolution::Crossover::Perform

(

const std::vector< std::unique_ptr< Individual >> &

inputPopulation

)

Perform - the function that performs the crossover phase.

Parameters

inputPopulation

- the population containing individuals that are candidates for the crossover. Please note that the crossover might not choose all individuals depending on its configuration.

Member Data Documentation

populationAfter

std::vector<std::unique_ptr<Individual> > grail::evolution::Crossover::populationAfter {}

The population of children, i.e., after the crossover operation has been performed.

---

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

• GrailEvolution/Crossover.hh
• GrailEvolution/Crossover.cpp