public class GenerationSwitchPipeline extends BreedingPipeline
Important Note: Because GenerationSwitchPipeline gets the generation number from the EvolutionState, and this number is not protected by a mutex, if you create an EvolutionState or Breeder which uses multiple threads that can update the generation number as they like, you could cause a race condition. This doesn't occur with the present EvolutionState objects provided with ECJ, but you should be aware of the possibility.
Typical Number of Individuals Produced Per produce(...) call
Defined as the max of its children's responses.
Number of Sources
2
Parameters
| base.generate-max bool = true (default) or false |
(Each time produce(...) is called, should the GenerationSwitchPipeline force all its sources to produce exactly the same number of individuals as the largest typical number of individuals produced by any source in the group?) |
| base.switch-at int >= 0 |
(The generation we will switch at) |
Default Base
breed.generation-switch
| Modifier and Type | Field and Description |
|---|---|
boolean |
generateMax |
int |
generationSwitch |
int |
maxGeneratable |
static int |
NUM_SOURCES |
static String |
P_GEN_MAX |
static String |
P_MULTIBREED |
static String |
P_SWITCHAT |
DYNAMIC_SOURCES, likelihood, mybase, P_LIKELIHOOD, P_NUMSOURCES, P_SOURCE, sources, V_SAMENO_PROBABILITY, P_PROB, probability| Constructor and Description |
|---|
GenerationSwitchPipeline() |
| Modifier and Type | Method and Description |
|---|---|
Parameter |
defaultBase()
Returns the default base for this prototype.
|
int |
numSources()
Returns the number of sources to this pipeline.
|
int |
produce(int min,
int max,
int start,
int subpopulation,
Individual[] inds,
EvolutionState state,
int thread)
Produces n individuals from the given subpopulation
and puts them into inds[start...start+n-1],
where n = Min(Max(q,min),max), where q is the "typical" number of
individuals the BreedingSource produces in one shot, and returns
n.
|
void |
setup(EvolutionState state,
Parameter base)
Sets up the BreedingPipeline.
|
int |
typicalIndsProduced()
Returns the max of typicalIndsProduced() of all its children
|
clone, finishProducing, individualReplaced, maxChildProduction, minChildProduction, preparePipeline, prepareToProduce, produces, reproduce, sourcesAreProperFormgetProbability, pickRandom, setProbability, setupProbabilitiespublic static final String P_SWITCHAT
public static final String P_MULTIBREED
public static final String P_GEN_MAX
public static final int NUM_SOURCES
public int maxGeneratable
public boolean generateMax
public int generationSwitch
public Parameter defaultBase()
Prototypepublic int numSources()
BreedingPipelinenumSources in class BreedingPipelinepublic void setup(EvolutionState state, Parameter base)
BreedingSourceThe most common modification is to normalize it with some other set of probabilities, then set all of them up in increasing summation; this allows the use of the fast static BreedingSource-picking utility method, BreedingSource.pickRandom(...). In order to use this method, for example, if four breeding source probabilities are {0.3, 0.2, 0.1, 0.4}, then they should get normalized and summed by the outside owners as: {0.3, 0.5, 0.6, 1.0}.
setup in interface Prototypesetup in interface Setupsetup in class BreedingPipelinePrototype.setup(EvolutionState,Parameter)public int typicalIndsProduced()
typicalIndsProduced in class BreedingPipelinepublic int produce(int min,
int max,
int start,
int subpopulation,
Individual[] inds,
EvolutionState state,
int thread)
BreedingSourceproduce in class BreedingSourceCopyright © 2014 Evolutionary Computation Laboratory at George Mason University. All rights reserved.