World: The representation of a set of niches with migration for...

In galgo: Genetic Algorithms for Multivariate Statistical Models from Large-Scale Functional Genomics Data

Description

Represents a set of nices for the genetic algorithm. Because the niches are “closed populations”, it is sometimes needed exchange information bewteen niches (or “islands”). The World object implements the exchange of chromosomes between niches, and to be compatible, it also implements the needed methods than an usual niche but considering the immigration property. Thus, the Galgo object can receive a list of Niches, a list of Worlds, or a list of any mixture of them.

Usage

World(id=0,
	niches=list(),
	immigration=0,
	maxFitness=0,
	bestFitness=0,
	maxChromosome=NULL,
	bestChromosome=NULL,
	generation=0,
	...)

Arguments

id	A way to identify the object.
niches	A list of defined niches composing the world. However, it can be a list containing even World objects.
immigration	It can be NULL, a function, or a vector. When it is NULL immigration is disabled. When it is a function it is evaluated using the same World object as parameter, the result should be a numeric value. When the length of immigration is greather than 1 a cycled version is used depending on the generation. If the resulted or selected numeric value is greather than 1 it is interpreted as the number of chromosomes to migrate, otherwise it is assumed to be a probability to migrate one chromosome. The final I best chromosomes to migrate apply to all niches.
bestFitness	The best fitness ever visited.
maxFitness	The maximum fitness from the current chromosomes. It should be 0 initially, but it is included for generalization.
maxChromosome	The chromosome whose fitness is maximum from the current chromosomes. It should be NULL initially, but it is included for generalization.
bestChromosome	The chromosome whose fitness is maximum visited ever. It should be NULL initially, but it is included for generalization.
generation	For internal use only.
...	Other user named values to include in the object (like pMutation, pCrossover or any other).

Class

Package: galgo
Class World

Object
~~|
~~+--World

Directly known subclasses:

public static class World
extends Object

Fields and Methods

Methods:

	best	Returns the best chromosome.
	bestFitness	Returns the fitness of the best chromosome.
	clone	Clones itself and its niches.
	evaluate	Evaluate all niches with a fitness function.
	generateRandom	Generates random values for all niches in the world.
	length	Gets the number of niches defined in the world.
	max	Returns the chromosome whose current fitness is maximum.
	maxFitness	Returns the fitness of the maximum chromosome.
	newCollection	Generates a list cloning an object.
	newRandomCollection	Creates a list of cloned object with its internal values generated by random.
	plot	Plots information about world object.
	print	Prints the representation of a world object.
	progeny	Calls progeny method to all niches in the world object.
	refreshStats	Updates the internal statistics from the current population.
	reInit	Erases all internal values in order to re-use the world object.
	summary	Prints the representation and statistics of the world object.

Methods inherited from Object:
as.list, unObject, $, $<-, [[, [[<-, as.character, attach, clone, detach, equals, extend, finalize, getFields, getInstanciationTime, getStaticInstance, hasField, hashCode, ll, load, objectSize, print, save

Author(s)

Victor Trevino. Francesco Falciani Group. University of Birmingham, U.K. http://www.bip.bham.ac.uk/bioinf

References

Goldberg, David E. 1989 Genetic Algorithms in Search, Optimization and Machine Learning. Addison-Wesley Pub. Co. ISBN: 0201157675

See Also

Gene, Chromosome, Niche, Galgo, BigBang.

Examples

   cr <- Chromosome(genes=newCollection(Gene(shape1=1, shape2=100),5))
   ni <- Niche(chromosomes=newRandomCollection(cr, 10))
   wo <- World(niches=newRandomCollection(ni,2))
   wo

   progeny(wo) # returns the chromosomes indexes that were mutated
 

galgo documentation built on May 2, 2019, 4:20 a.m.