GeneticAlgorithm (evosuite-client 1.0.5 API)

java.lang.Object
- org.evosuite.ga.metaheuristics.GeneticAlgorithm<T>

All Implemented Interfaces:

Serializable, SearchAlgorithm

Direct Known Subclasses:

AbstractMOSA, MonotonicGA, NSGAII, OnePlusOneEA, RandomSearch, SPEA2, StandardGA
```
public abstract class GeneticAlgorithm<T extends Chromosome>
extends Object
implements SearchAlgorithm, Serializable
```
Abstract superclass of genetic algorithms

Author:

Gordon Fraser

See Also:

Serialized Form

Field Summary

Fields
Modifier and Type	Field and Description
`protected Archive<T>`	`archive`
`protected Set<BloatControlFunction>`	`bloatControl` Bloat control, to avoid too long chromosomes
`protected ChromosomeFactory<T>`	`chromosomeFactory` Generator for initial population
`protected CrossOverFunction`	`crossoverFunction` CrossOver function
`protected int`	`currentIteration` Age of the population
`protected List<FitnessFunction<T>>`	`fitnessFunctions` Fitness function to rank individuals
`protected Set<SearchListener>`	`listeners` Listeners
`protected LocalSearchObjective<T>`	`localObjective` Local search might need a different local objective
`protected double`	`localSearchProbability`
`protected List<T>`	`population` Current population
`protected PopulationLimit`	`populationLimit` The population limit decides when an iteration is done
`protected SelectionFunction<T>`	`selectionFunction` Selection function to select parents
`protected Set<StoppingCondition>`	`stoppingConditions` List of conditions on which to end the search

Constructor Summary

Constructors
Constructor and Description

GeneticAlgorithm(ChromosomeFactory<T> factory)
Constructor

Constructors
Constructor and Description
`GeneticAlgorithm(ChromosomeFactory<T> factory)` Constructor

Method Summary

All Methods Instance Methods Abstract Methods Concrete Methods
Modifier and Type	Method and Description
`void`	`addBloatControl(BloatControlFunction bloat_control)` Set new bloat control function
`void`	`addFitnessFunction(FitnessFunction<T> function)` Add new fitness function (i.e., for new mutation)
`void`	`addFitnessFunctions(List<FitnessFunction<T>> functions)`
`void`	`addListener(SearchListener listener)` Add a new search listener
`void`	`addStoppingCondition(StoppingCondition condition)` addStoppingCondition
`protected void`	`applyLocalSearch()` Apply local search, starting from the best individual and continue applying it to all individuals until the local search budget is used up.
`protected void`	`calculateFitnessAndSortPopulation()` Calculate fitness for all individuals
`void`	`clearPopulation()` Delete all current individuals
`protected void`	`disableFirstSecondaryCriterion()`
`protected List<T>`	`elitism()` Copy best individuals
`protected void`	`enableFirstSecondaryCriterion()`
`protected abstract void`	`evolve()` Generate one new generation
`protected void`	`generateInitialPopulation(int population_size)` Fills the population at first with recycled chromosomes - for more information see recycleChromosomes() and ChromosomeRecycler - and after that, the population is filled with random chromosomes.
`protected void`	`generateRandomPopulation(int population_size)` Generate random population of given size
`abstract void`	`generateSolution()` Generate solution
`int`	`getAge()` Get number of iterations
`T`	`getBestIndividual()` Return the individual with the highest fitChromosomeess
`List<T>`	`getBestIndividuals()` Return the individual(s) with the highest fitChromosomeess
`String`	`getBudgetString()` getBudgetString
`FitnessFunction<T>`	`getFitnessFunction()` Get currently used fitness function
`List<FitnessFunction<T>>`	`getFitnessFunctions()` Get all used fitness function
`int`	`getNumberOfFitnessFunctions()`
`List<T>`	`getPopulation()` Accessor for population Chromosome *
`int`	`getPopulationSize()` getPopulationSize
`SelectionFunction<T>`	`getSelectionFunction()` Get currently used fitness function
`Set<StoppingCondition>`	`getStoppingConditions()`
`abstract void`	`initializePopulation()` Set up initial population
`protected boolean`	`isBetterOrEqual(Chromosome chromosome1, Chromosome chromosome2)` Returns true if the `chromosome1` is better or equal than `chromosome2` according to the compound fitness function.
`boolean`	`isFinished()` Determine whether any of the stopping conditions hold
`boolean`	`isNextPopulationFull(List<T> nextGeneration)` Determine if the next generation has reached its size limit
`boolean`	`isTooLong(Chromosome chromosome)` Check whether individual is suitable according to bloat control functions
`protected void`	`notifyEvaluation(Chromosome chromosome)` Notify all search listeners of fitness evaluation
`protected void`	`notifyIteration()` Notify all search listeners of iteration
`protected void`	`notifyMutation(Chromosome chromosome)` Notify all search listeners of a mutation
`protected void`	`notifySearchFinished()` Notify all search listeners of search end
`protected void`	`notifySearchStarted()` Notify all search listeners of search start
`void`	`printBudget()` Prints out all information regarding this GAs stopping conditions So far only used for testing purposes in TestSuiteGenerator
`protected List<Chromosome>`	`randomism()` Create random individuals
`void`	`removeListener(SearchListener listener)` Remove a search listener
`void`	`removeStoppingCondition(StoppingCondition condition)` removeStoppingCondition
`void`	`resetStoppingConditions()` resetStoppingConditions
`void`	`setArchive(Archive<T> archive)`
`void`	`setBloatControl(BloatControlFunction bloat_control)` Set new bloat control function
`void`	`setChromosomeFactory(ChromosomeFactory<T> factory)` Set a new factory method
`void`	`setCrossOverFunction(CrossOverFunction crossover)` Set a new xover function
`void`	`setPopulationLimit(PopulationLimit limit)` Set a new population limit function
`void`	`setSelectionFunction(SelectionFunction<T> function)` Set new fitness function (i.e., for new mutation)
`void`	`setStoppingCondition(StoppingCondition condition)` setStoppingCondition
`void`	`setStoppingConditionLimit(int value)` setStoppingConditionLimit
`protected boolean`	`shouldApplyLocalSearch()` Local search is only applied every X generations
`protected void`	`sortPopulation()` Sort the population by fitness WARN: used only with singular objective algorithms, multi-objective algorithms should implement their own 'sort'
`protected void`	`starveByFitness(int limit)` This method can be used to kick out the worst chromosomes in the current population until the given limit is reached again.
`protected void`	`starveRandomly(int limit)` This method can be used to kick out random chromosomes in the current population until the given limit is reached again.
`protected void`	`starveToLimit(int limit)` This method can be used to kick out chromosomes when the population is possibly overcrowded Depending on the Property "starve_by_fitness" chromosome are either kicked out randomly or according to their fitness
`String`	`toString()`
`protected void`	`updateBestIndividualFromArchive()`
`protected void`	`updateFitnessFunctionsAndValues()` update archive fitness functions
`protected void`	`updateProbability(boolean improvement)`
`protected void`	`updateSecondaryCriterion(int starvationCounter)` enable and disable secondary criteria according to the strategy defined in the Properties file.
`void`	`writeIndividuals(List<T> individuals)` Write to a file all fitness values of each individuals.

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

- Field Detail
  - fitnessFunctions
```
protected List<FitnessFunction<T extends Chromosome>> fitnessFunctions
```
    Fitness function to rank individuals
  - selectionFunction
```
protected SelectionFunction<T extends Chromosome> selectionFunction
```
    Selection function to select parents
  - crossoverFunction
```
protected CrossOverFunction crossoverFunction
```
    CrossOver function
  - population
```
protected List<T extends Chromosome> population
```
    Current population
  - chromosomeFactory
```
protected ChromosomeFactory<T extends Chromosome> chromosomeFactory
```
    Generator for initial population
  - listeners
```
protected transient Set<SearchListener> listeners
```
    Listeners
  - stoppingConditions
```
protected transient Set<StoppingCondition> stoppingConditions
```
    List of conditions on which to end the search
  - bloatControl
```
protected Set<BloatControlFunction> bloatControl
```
    Bloat control, to avoid too long chromosomes
  - localObjective
```
protected LocalSearchObjective<T extends Chromosome> localObjective
```
    Local search might need a different local objective
  - populationLimit
```
protected PopulationLimit populationLimit
```
    The population limit decides when an iteration is done
  - currentIteration
```
protected int currentIteration
```
    Age of the population
  - localSearchProbability
```
protected double localSearchProbability
```
  - archive
```
protected transient Archive<T extends Chromosome> archive
```
- Constructor Detail
  - GeneticAlgorithm
```
public GeneticAlgorithm(ChromosomeFactory<T> factory)
```
    Constructor
    
    Parameters:
    
    factory - a ChromosomeFactory object.
- Method Detail
  - evolve
```
protected abstract void evolve()
```
    Generate one new generation
  - shouldApplyLocalSearch
```
protected boolean shouldApplyLocalSearch()
```
    Local search is only applied every X generations
    
    Returns:
    
    a boolean.
  - disableFirstSecondaryCriterion
```
protected void disableFirstSecondaryCriterion()
```
  - enableFirstSecondaryCriterion
```
protected void enableFirstSecondaryCriterion()
```
  - updateSecondaryCriterion
```
protected void updateSecondaryCriterion(int starvationCounter)
```
    enable and disable secondary criteria according to the strategy defined in the Properties file.
    
    Parameters:
    
    starvationCounter -
  - applyLocalSearch
```
protected void applyLocalSearch()
```
    Apply local search, starting from the best individual and continue applying it to all individuals until the local search budget is used up. The population list is re-ordered if needed.
  - updateProbability
```
protected void updateProbability(boolean improvement)
```
  - initializePopulation
```
public abstract void initializePopulation()
```
    Set up initial population
  - generateSolution
```
public abstract void generateSolution()
```
    Generate solution Generate solution
    
    Specified by:
    
    generateSolution in interface SearchAlgorithm
  - generateInitialPopulation
```
protected void generateInitialPopulation(int population_size)
```
    Fills the population at first with recycled chromosomes - for more information see recycleChromosomes() and ChromosomeRecycler - and after that, the population is filled with random chromosomes. This method guarantees at least a proportion of Properties.initially_enforeced_Randomness % of random chromosomes
    
    Parameters:
    
    population_size - a int.
  - starveToLimit
```
protected void starveToLimit(int limit)
```
    This method can be used to kick out chromosomes when the population is possibly overcrowded Depending on the Property "starve_by_fitness" chromosome are either kicked out randomly or according to their fitness
    
    Parameters:
    
    limit - a int.
  - starveRandomly
```
protected void starveRandomly(int limit)
```
    This method can be used to kick out random chromosomes in the current population until the given limit is reached again.
    
    Parameters:
    
    limit - a int.
  - starveByFitness
```
protected void starveByFitness(int limit)
```
    This method can be used to kick out the worst chromosomes in the current population until the given limit is reached again.
    
    Parameters:
    
    limit - a int.
  - generateRandomPopulation
```
protected void generateRandomPopulation(int population_size)
```
    Generate random population of given size
    
    Parameters:
    
    population_size - a int.
  - clearPopulation
```
public void clearPopulation()
```
    Delete all current individuals
  - addFitnessFunction
```
public void addFitnessFunction(FitnessFunction<T> function)
```
    Add new fitness function (i.e., for new mutation)
    
    Parameters:
    
    function - a FitnessFunction object.
  - addFitnessFunctions
```
public void addFitnessFunctions(List<FitnessFunction<T>> functions)
```
  - getFitnessFunction
```
public FitnessFunction<T> getFitnessFunction()
```
    Get currently used fitness function
    
    Returns:
    
    a FitnessFunction object.
  - getFitnessFunctions
```
public List<FitnessFunction<T>> getFitnessFunctions()
```
    Get all used fitness function
    
    Returns:
    
    a FitnessFunction object.
  - getNumberOfFitnessFunctions
```
public int getNumberOfFitnessFunctions()
```
  - toString
```
public String toString()
```
    Overrides:
    
    toString in class Object
    
    Returns:
  - setArchive
```
public void setArchive(Archive<T> archive)
```
  - setSelectionFunction
```
public void setSelectionFunction(SelectionFunction<T> function)
```
    Set new fitness function (i.e., for new mutation)
    
    Parameters:
    
    function - a SelectionFunction object.
  - getSelectionFunction
```
public SelectionFunction<T> getSelectionFunction()
```
    Get currently used fitness function
    
    Returns:
    
    a SelectionFunction object.
  - setBloatControl
```
public void setBloatControl(BloatControlFunction bloat_control)
```
    Set new bloat control function
    
    Parameters:
    
    bloat_control - a BloatControlFunction object.
  - addBloatControl
```
public void addBloatControl(BloatControlFunction bloat_control)
```
    Set new bloat control function
    
    Parameters:
    
    bloat_control - a BloatControlFunction object.
  - isTooLong
```
public boolean isTooLong(Chromosome chromosome)
```
    Check whether individual is suitable according to bloat control functions
    
    Parameters:
    
    chromosome - a Chromosome object.
    
    Returns:
    
    a boolean.
  - getAge
```
public int getAge()
```
    Get number of iterations
    
    Returns:
    
    Number of iterations
  - calculateFitnessAndSortPopulation
```
protected void calculateFitnessAndSortPopulation()
```
    Calculate fitness for all individuals
  - getPopulationSize
```
public int getPopulationSize()
```
    getPopulationSize
    
    Returns:
    
    a int.
  - elitism
```
protected List<T> elitism()
```
    Copy best individuals
    
    Returns:
    
    a List object.
  - randomism
```
protected List<Chromosome> randomism()
```
    Create random individuals
    
    Returns:
    
    a List object.
  - updateFitnessFunctionsAndValues
```
protected void updateFitnessFunctionsAndValues()
```
    update archive fitness functions
  - getBestIndividual
```
public T getBestIndividual()
```
    Return the individual with the highest fitChromosomeess
    
    Returns:
    
    a Chromosome object.
  - getBestIndividuals
```
public List<T> getBestIndividuals()
```
    Return the individual(s) with the highest fitChromosomeess
    
    Returns:
    
    a list of Chromosome object(s).
  - writeIndividuals
```
public void writeIndividuals(List<T> individuals)
```
    Write to a file all fitness values of each individuals.
    
    Parameters:
    
    solutions - a list of Chromosome object(s).
  - setChromosomeFactory
```
public void setChromosomeFactory(ChromosomeFactory<T> factory)
```
    Set a new factory method
    
    Parameters:
    
    factory - a ChromosomeFactory object.
  - setCrossOverFunction
```
public void setCrossOverFunction(CrossOverFunction crossover)
```
    Set a new xover function
    
    Parameters:
    
    crossover - a CrossOverFunction object.
  - addListener
```
public void addListener(SearchListener listener)
```
    Add a new search listener
    
    Parameters:
    
    listener - a SearchListener object.
  - removeListener
```
public void removeListener(SearchListener listener)
```
    Remove a search listener
    
    Parameters:
    
    listener - a SearchListener object.
  - notifySearchStarted
```
protected void notifySearchStarted()
```
    Notify all search listeners of search start
  - notifySearchFinished
```
protected void notifySearchFinished()
```
    Notify all search listeners of search end
  - notifyIteration
```
protected void notifyIteration()
```
    Notify all search listeners of iteration
  - notifyEvaluation
```
protected void notifyEvaluation(Chromosome chromosome)
```
    Notify all search listeners of fitness evaluation
    
    Parameters:
    
    chromosome - a Chromosome object.
  - notifyMutation
```
protected void notifyMutation(Chromosome chromosome)
```
    Notify all search listeners of a mutation
    
    Parameters:
    
    chromosome - a Chromosome object.
  - sortPopulation
```
protected void sortPopulation()
```
    Sort the population by fitness WARN: used only with singular objective algorithms, multi-objective algorithms should implement their own 'sort'
  - getPopulation
```
public List<T> getPopulation()
```
    Accessor for population Chromosome *
    
    Returns:
    
    a List object.
  - isNextPopulationFull
```
public boolean isNextPopulationFull(List<T> nextGeneration)
```
    Determine if the next generation has reached its size limit
    
    Parameters:
    
    nextGeneration - a List object.
    
    Returns:
    
    a boolean.
  - setPopulationLimit
```
public void setPopulationLimit(PopulationLimit limit)
```
    Set a new population limit function
    
    Parameters:
    
    limit - a PopulationLimit object.
  - isFinished
```
public boolean isFinished()
```
    Determine whether any of the stopping conditions hold
    
    Returns:
    
    a boolean.
  - addStoppingCondition
```
public void addStoppingCondition(StoppingCondition condition)
```
    addStoppingCondition
    
    Parameters:
    
    condition - a StoppingCondition object.
  - getStoppingConditions
```
public Set<StoppingCondition> getStoppingConditions()
```
  - setStoppingCondition
```
public void setStoppingCondition(StoppingCondition condition)
```
    setStoppingCondition
    
    Parameters:
    
    condition - a StoppingCondition object.
  - removeStoppingCondition
```
public void removeStoppingCondition(StoppingCondition condition)
```
    removeStoppingCondition
    
    Parameters:
    
    condition - a StoppingCondition object.
  - resetStoppingConditions
```
public void resetStoppingConditions()
```
    resetStoppingConditions
  - setStoppingConditionLimit
```
public void setStoppingConditionLimit(int value)
```
    setStoppingConditionLimit
    
    Parameters:
    
    value - a int.
  - updateBestIndividualFromArchive
```
protected void updateBestIndividualFromArchive()
```
  - isBetterOrEqual
```
protected boolean isBetterOrEqual(Chromosome chromosome1,
                                  Chromosome chromosome2)
```
    Returns true if the chromosome1 is better or equal than chromosome2 according to the compound fitness function.
    
    Parameters:
    
    chromosome1 - a Chromosome object.
    
    chromosome2 - a Chromosome object.
    
    Returns:
    
    a boolean.
  - printBudget
```
public void printBudget()
```
    Prints out all information regarding this GAs stopping conditions So far only used for testing purposes in TestSuiteGenerator
  - getBudgetString
```
public String getBudgetString()
```
    getBudgetString
    
    Returns:
    
    a String object.

Class GeneticAlgorithm<T extends Chromosome>

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

fitnessFunctions

selectionFunction

crossoverFunction

population

chromosomeFactory

listeners

stoppingConditions

bloatControl

localObjective

populationLimit

currentIteration

localSearchProbability

archive

Constructor Detail

GeneticAlgorithm

Method Detail

evolve

shouldApplyLocalSearch

disableFirstSecondaryCriterion

enableFirstSecondaryCriterion

updateSecondaryCriterion

applyLocalSearch

updateProbability

initializePopulation

generateSolution

generateInitialPopulation

starveToLimit

starveRandomly

starveByFitness

generateRandomPopulation

clearPopulation

addFitnessFunction

addFitnessFunctions

getFitnessFunction

getFitnessFunctions

getNumberOfFitnessFunctions

toString

setArchive

setSelectionFunction

getSelectionFunction

setBloatControl

addBloatControl

isTooLong

getAge

calculateFitnessAndSortPopulation

getPopulationSize

elitism

randomism

updateFitnessFunctionsAndValues

getBestIndividual

getBestIndividuals

writeIndividuals

setChromosomeFactory

setCrossOverFunction

addListener

removeListener

notifySearchStarted

notifySearchFinished

notifyIteration

notifyEvaluation

notifyMutation

sortPopulation

getPopulation

isNextPopulationFull

setPopulationLimit

isFinished

addStoppingCondition

getStoppingConditions

setStoppingCondition

removeStoppingCondition

resetStoppingConditions

setStoppingConditionLimit

updateBestIndividualFromArchive

isBetterOrEqual