halfdan/eva2
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Some updates to OptimizerFactory, PostProcess. Minor clean-up in benchmarks.

Browse Source
This commit is contained in:
Marcel Kronfeld
2008-03-19 14:00:16 +00:00
parent f010fc827c
commit 232113e061
39 changed files with 1010 additions and 498 deletions
Download Patch File Download Diff File Expand all files Collapse all files

77
src/javaeva/server/go/operators/postprocess/PostProcess.java
View File

@@ -1,8 +1,6 @@
package javaeva.server.go.operators.postprocess;
import java.io.PrintStream;
import java.util.Collection;
import java.util.List;
import javaeva.OptimizerFactory;
import javaeva.OptimizerRunnable;
@@ -19,7 +17,6 @@ import javaeva.server.go.operators.terminators.EvaluationTerminator;
import javaeva.server.go.populations.Population;
import javaeva.server.go.problems.AbstractOptimizationProblem;
import javaeva.server.go.problems.FM0Problem;
import javaeva.server.go.problems.InterfaceMultimodalProblem;
import javaeva.server.go.problems.InterfaceMultimodalProblemKnown;
import javaeva.server.go.strategies.HillClimbing;
import javaeva.server.stat.InterfaceTextListener;
@@ -36,6 +33,8 @@ public class PostProcess {
private static final boolean TRACE = false;
// the default mutation step size for HC post processing
private static double defaultMutationStepSize = 0.01;
// used for hill climbing post processing and only alive during that period
private static OptimizerRunnable hcRunnable = null;
public static final int BEST_ONLY = 1;
public static final int BEST_RAND = 2;
@@ -108,11 +107,13 @@ public class PostProcess {
for (int i=0; i<optsFound.length; i++) {
if (optsFound[i] != null) result.add(optsFound[i]);
}
result.setPopulationSize(result.size());
return result;
}
/**
* Calls clusterBest with a ClusteringDensitiyBased clustering object with the given sigma.
* Calls clusterBest with a ClusteringDensitiyBased clustering object with the given sigma and a
* minimum group size of 2.
* @see clusterBest(Population pop, InterfaceClustering clustering, double returnQuota, int lonerMode, int takeOverMode)
* @param pop
* @param sigmaCluster
@@ -122,7 +123,7 @@ public class PostProcess {
* @return
*/
public static Population clusterBest(Population pop, double sigmaCluster, double returnQuota, int lonerMode, int takeOverMode) {
return clusterBest(pop, new ClusteringDensityBased(sigmaCluster), returnQuota, lonerMode, takeOverMode);
return clusterBest(pop, new ClusteringDensityBased(sigmaCluster, 2), returnQuota, lonerMode, takeOverMode);
}
/**
@@ -348,10 +349,19 @@ public class PostProcess {
}
hc.setPopulation(pop);
// hc.initByPopulation(pop, false);
OptimizerRunnable runnable = new OptimizerRunnable(OptimizerFactory.makeParams(hc, pop, problem, 0, term), null, true);
runnable.getGOParams().setDoPostProcessing(false);
runnable.run();
hcRunnable = new OptimizerRunnable(OptimizerFactory.makeParams(hc, pop, problem, 0, term), null, true);
hcRunnable.getGOParams().setDoPostProcessing(false);
hcRunnable.run();
hcRunnable = null;
}
/**
* Stop the hill-climbing post processing if its currently running.
*/
public static void stopHC() {
if (hcRunnable != null) synchronized (hcRunnable) {
if (hcRunnable != null) hcRunnable.stopOpt();
}
}
public static void main(String[] args) {
@@ -399,12 +409,12 @@ public class PostProcess {
* @return a pair of the optimized population and the improvement in the mean fitness (not normed) achieved by the HC run
*/
public static Pair<Population, Double> clusterHC(Population pop, AbstractOptimizationProblem problem, double sigmaCluster, int funCalls, double keepClusterRatio, InterfaceMutation mute) {
Population clust = (Population)clusterBest(pop, new ClusteringDensityBased(sigmaCluster), keepClusterRatio, KEEP_LONERS, BEST_RAND).clone();
Population clust = (Population)clusterBest(pop, new ClusteringDensityBased(sigmaCluster, 2), keepClusterRatio, KEEP_LONERS, BEST_RAND).clone();
// System.out.println("keeping " + clust.size() + " for hc....");
double[] meanFit = clust.getMeanFitness();
processWithHC(clust, problem, new EvaluationTerminator(pop.getFunctionCalls()+funCalls), mute);
double improvement = PhenotypeMetric.euclidianDistance(meanFit, clust.getMeanFitness());
System.out.println("improvement by " + improvement);
if (TRACE) System.out.println("improvement by " + improvement);
return new Pair<Population, Double>(clust, improvement);
}
@@ -457,15 +467,18 @@ public class PostProcess {
// }
/**
* Do some data output for multimodal problems with known optima.
* Do some data output for multimodal problems with known optima. The listener may be null, but then the method is
* not really doing much at this state.
*/
public static void procMultiModalKnown(Population solutions, InterfaceMultimodalProblemKnown mmkProb, InterfaceTextListener listener) {
// Population found = getFoundOptima(solutions, mmkProb.getRealOptima(), mmkProb.getEpsilon(), true);
listener.println("default epsilon is " + mmkProb.getEpsilon());
listener.println("max peak ratio is " + mmkProb.getMaximumPeakRatio(getFoundOptima(solutions, mmkProb.getRealOptima(), mmkProb.getEpsilon(), true)));
if (listener != null) {
listener.println("default epsilon is " + mmkProb.getEpsilon());
listener.println("max peak ratio is " + mmkProb.getMaximumPeakRatio(getFoundOptima(solutions, mmkProb.getRealOptima(), mmkProb.getEpsilon(), true)));
}
for (double epsilon=0.1; epsilon > 0.00000001; epsilon/=10.) {
// out.println("no optima found: " + ((InterfaceMultimodalProblemKnown)mmProb).getNumberOfFoundOptima(pop));
listener.println("found " + getFoundOptima(solutions, mmkProb.getRealOptima(), epsilon, true).size() + " for epsilon = " + epsilon);
if (listener != null) listener.println("found " + getFoundOptima(solutions, mmkProb.getRealOptima(), epsilon, true).size() + " for epsilon = " + epsilon);
}
}
@@ -478,38 +491,50 @@ public class PostProcess {
* @param solutions
* @param problem
* @param listener
* @return the clustered, post-processed population
*/
public static void postProcess(InterfacePostProcessParams params, Population solutions, AbstractOptimizationProblem problem, InterfaceTextListener listener) {
public static Population postProcess(InterfacePostProcessParams params, Population solutions, AbstractOptimizationProblem problem, InterfaceTextListener listener) {
if (params.isDoPostProcessing()) {
Population outputPop;
if (params.getPostProcessClusterSigma() > 0) {
outputPop = (Population)PostProcess.clusterBest(solutions, params.getPostProcessClusterSigma(), 0, PostProcess.KEEP_LONERS, PostProcess.BEST_ONLY).clone();
if (outputPop.size() < solutions.size()) {
listener.println("Clustering reduced population size from " + solutions.size() + " to " + outputPop.size());
} else listener.println("Clustering yielded no size reduction.");
if (listener != null) listener.println("Initial clustering reduced population size from " + solutions.size() + " to " + outputPop.size());
} else if (listener != null) listener.println("Initial clustering yielded no size reduction.");
} else outputPop = solutions;
if (params.getPostProcessSteps() > 0) {
processWithHC(outputPop, problem, params.getPostProcessSteps());
listener.println("HC post processing done.");
if (listener != null) listener.println("HC post processing done.");
// some individuals may have now converged again
if (params.getPostProcessClusterSigma() > 0) {
// so if wished, cluster again.
outputPop = (Population)PostProcess.clusterBest(outputPop, params.getPostProcessClusterSigma(), 0, PostProcess.KEEP_LONERS, PostProcess.BEST_ONLY).clone();
if (outputPop.size() < solutions.size()) {
if (listener != null) listener.println("Second clustering reduced population size from " + solutions.size() + " to " + outputPop.size());
} else if (listener != null) listener.println("Second clustering yielded no size reduction.");
}
}
double upBnd = PhenotypeMetric.norm(outputPop.getWorstEAIndividual().getFitness())*1.1;
upBnd = Math.pow(10,Math.floor(Math.log10(upBnd)+1));
double lowBnd = 0;
int[] sols = PostProcess.createFitNormHistogram(outputPop, lowBnd, upBnd, 20);
// PostProcessInterim.outputResult((AbstractOptimizationProblem)goParams.getProblem(), outputPop, 0.01, System.out, 0, 2000, 20, goParams.getPostProcessSteps());
listener.println("measures: " + BeanInspector.toString(outputPop.getPopulationMeasures()));
listener.println("solution histogram in [" + lowBnd + "," + upBnd + "]: " + BeanInspector.toString(sols));
if (listener != null) listener.println("measures: " + BeanInspector.toString(outputPop.getPopulationMeasures()));
if (listener != null) listener.println("solution histogram in [" + lowBnd + "," + upBnd + "]: " + BeanInspector.toString(sols));
//////////// multimodal data output?
if (problem instanceof InterfaceMultimodalProblemKnown) procMultiModalKnown(outputPop, (InterfaceMultimodalProblemKnown)problem, listener);
Population resPop = outputPop.getBestNIndividuals(params.getPrintNBest()); // n individuals are returned and sorted, all of them if n<=0
if (listener != null) listener.println("Best after post process:" + ((outputPop.size()>resPop.size()) ? ( "(first " + resPop.size() + " of " + outputPop.size() + ")") : ""));
//////////// output some individual data
List<AbstractEAIndividual> bestList = outputPop.getBestNIndividuals(params.getPrintNBest()); // n individuals are returned and sorted, all of them if n<=0
listener.println("Best after post process:" + ((outputPop.size()>bestList.size()) ? ( "(first " + bestList.size() + " of " + outputPop.size() + ")") : ""));
for (AbstractEAIndividual indy : bestList) {
listener.println(AbstractEAIndividual.getDefaultStringRepresentation(indy));
if (listener != null) for (int i=0; i<resPop.size(); i++) {
listener.println(AbstractEAIndividual.getDefaultStringRepresentation(resPop.getEAIndividual(i)));
}
}
return resPop;
} else return solutions;
}
}