1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
|
package hu.bme.mit.inf.dslreasoner.domains.cps.mdeo
import java.util.Properties
import org.moeaframework.Executor
import org.moeaframework.Instrumenter
import org.moeaframework.algorithm.PeriodicAction
import org.moeaframework.core.TerminationCondition
import org.moeaframework.core.spi.AlgorithmFactory
import uk.ac.kcl.inf.mdeoptimiser.interfaces.cli.Run
import uk.ac.kcl.inf.mdeoptimiser.languages.mopt.Optimisation
import uk.ac.kcl.inf.mdeoptimiser.languages.mopt.SolverSpec
import uk.ac.kcl.inf.mdeoptimiser.libraries.core.optimisation.OptimisationInterpreter
import uk.ac.kcl.inf.mdeoptimiser.libraries.core.optimisation.executor.SolutionGenerator
import uk.ac.kcl.inf.mdeoptimiser.libraries.core.optimisation.moea.MoeaOptimisation
import uk.ac.kcl.inf.mdeoptimiser.libraries.core.optimisation.moea.algorithms.MoeaOptimisationAlgorithmProvider
import uk.ac.kcl.inf.mdeoptimiser.libraries.core.optimisation.moea.instrumentation.PopulationCollector
import uk.ac.kcl.inf.mdeoptimiser.libraries.core.optimisation.moea.problem.MoeaOptimisationProblem
import uk.ac.kcl.inf.mdeoptimiser.libraries.core.optimisation.operators.adaptation.MutationStepSizeStrategyFactory
import uk.ac.kcl.inf.mdeoptimiser.libraries.core.optimisation.output.MDEOBatch
class ExcludedMoeaOptimisation extends MoeaOptimisation {
SolutionGenerator solutionGenerator
Instrumenter algorithmStepSizeInstrumenter
override execute(SolverSpec solverSpec, SolutionGenerator solutionGenerator) {
this.solutionGenerator = solutionGenerator
super.execute(solverSpec, solutionGenerator)
}
override Instrumenter runOptimisation(SolverSpec solverSpec, Properties optimisationProperties) {
val algorithmFactory = new AlgorithmFactory
algorithmFactory.addProvider(new MoeaOptimisationAlgorithmProvider)
algorithmStepSizeInstrumenter = new Instrumenter().addExcludedPackage("org.eclipse").withProblemClass(
MoeaOptimisationProblem, solutionGenerator).attachApproximationSetCollector().attachElapsedTimeCollector().
attachPopulationSizeCollector.attach(new PopulationCollector()).withFrequency(1).withFrequencyType(
PeriodicAction.FrequencyType.STEPS)
var stepSizeStrategy = new MutationStepSizeStrategyFactory(solverSpec.algorithm, algorithmStepSizeInstrumenter).
strategy
solutionGenerator.setMutationStepSizeStrategy(stepSizeStrategy)
// TODO: Place this in a better location.
// Exclude JDK packages from Instrumenter
this.algorithmStepSizeInstrumenter.addExcludedPackage("jdk")
new Executor().usingAlgorithmFactory(algorithmFactory).withAlgorithm(solverSpec.algorithm.name) // Initialize problem with our solution generator
.withProblemClass(MoeaOptimisationProblem, solutionGenerator).withProperties(optimisationProperties).
withInstrumenter(algorithmStepSizeInstrumenter).withTerminationCondition(
optimisationProperties.get("terminationCondition") as TerminationCondition).run()
return algorithmStepSizeInstrumenter
}
}
class ExcludedOptimisationInterpreter extends OptimisationInterpreter {
val Optimisation model
new(String projectPath, Optimisation model) {
super(projectPath, model)
this.model = model
}
override start() {
// This model provider loads the model given by the user in the DSL
var solutionGenerator = new SolutionGenerator(model, getBreedingOperators, getMutationOperators,
getModelProvider, getMetamodel);
return new ExcludedMoeaOptimisation().execute(model.solver, solutionGenerator)
}
}
class ExcludedRun extends Run {
override runBatch(String moptProjectPath, Optimisation optimisationModel, Integer batch, boolean singleBatch) {
val optimisationInterpreter = new ExcludedOptimisationInterpreter(moptProjectPath, optimisationModel);
val startTime = System.nanoTime();
val optimisationOutcome = optimisationInterpreter.start();
val endTime = System.nanoTime();
val experimentDuration = ((endTime - startTime) / 1000000);
val generatedRules = optimisationInterpreter.getRulegenOperators();
return new MDEOBatch(batch, experimentDuration, optimisationOutcome, generatedRules, singleBatch);
}
}
|
