Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/execution/AbstractCrossProjectExperiment.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/execution/AbstractCrossProjectExperiment.java (revision 67)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/execution/AbstractCrossProjectExperiment.java (revision 67)
@@ -0,0 +1,329 @@
+// Copyright 2015 Georg-August-Universität Göttingen, Germany
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+// http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+package de.ugoe.cs.cpdp.execution;
+
+import java.io.File;
+import java.util.Collections;
+import java.util.LinkedList;
+import java.util.List;
+import java.util.logging.Level;
+
+import org.apache.commons.collections4.list.SetUniqueList;
+
+import de.ugoe.cs.cpdp.ExperimentConfiguration;
+import de.ugoe.cs.cpdp.dataprocessing.IProcessesingStrategy;
+import de.ugoe.cs.cpdp.dataprocessing.ISetWiseProcessingStrategy;
+import de.ugoe.cs.cpdp.dataselection.IPointWiseDataselectionStrategy;
+import de.ugoe.cs.cpdp.dataselection.ISetWiseDataselectionStrategy;
+import de.ugoe.cs.cpdp.eval.IEvaluationStrategy;
+import de.ugoe.cs.cpdp.loader.IVersionLoader;
+import de.ugoe.cs.cpdp.training.ISetWiseTestdataAwareTrainingStrategy;
+import de.ugoe.cs.cpdp.training.ISetWiseTrainingStrategy;
+import de.ugoe.cs.cpdp.training.ITestAwareTrainingStrategy;
+import de.ugoe.cs.cpdp.training.ITrainer;
+import de.ugoe.cs.cpdp.training.ITrainingStrategy;
+import de.ugoe.cs.cpdp.versions.IVersionFilter;
+import de.ugoe.cs.cpdp.versions.SoftwareVersion;
+import de.ugoe.cs.util.console.Console;
+import weka.core.Instances;
+
+/**
+ * Class responsible for executing an experiment according to an {@link ExperimentConfiguration}.
+ * The steps of an experiment are as follows:
+ *
+ * - load the data from the provided data path
+ * - filter the data sets according to the provided version filters
+ * - execute the following steps for each data sets as test data that is not ignored through the
+ * test version filter:
+ *
+ * - filter the data sets to setup the candidate training data:
+ *
+ * - remove all data sets from the same project
+ * - filter all data sets according to the training data filter
+ *
+ *
+ * - apply the setwise preprocessors
+ * - apply the setwise data selection algorithms
+ * - apply the setwise postprocessors
+ * - train the setwise training classifiers
+ * - unify all remaining training data into one data set
+ * - apply the preprocessors
+ * - apply the pointwise data selection algorithms
+ * - apply the postprocessors
+ * - train the normal classifiers
+ * - evaluate the results for all trained classifiers on the training data
+ *
+ *
+ *
+ *
+ * Note that this class implements {@link Runnable}, i.e., each experiment can be started in its own
+ * thread.
+ *
+ * @author Steffen Herbold
+ */
+public abstract class AbstractCrossProjectExperiment implements IExecutionStrategy {
+
+ /**
+ * configuration of the experiment
+ */
+ protected final ExperimentConfiguration config;
+
+ /**
+ * Constructor. Creates a new experiment based on a configuration.
+ *
+ * @param config
+ * configuration of the experiment
+ */
+ public AbstractCrossProjectExperiment(ExperimentConfiguration config) {
+ this.config = config;
+ }
+
+ /**
+ *
+ * Defines which products are allowed for training.
+ *
+ *
+ * @param trainingVersion
+ * training version
+ * @param testVersion
+ * test candidate
+ * @return true if test candidate can be used for training
+ */
+ protected abstract boolean isTrainingVersion(SoftwareVersion trainingVersion,
+ SoftwareVersion testVersion);
+
+ /**
+ * Helper method that combines a set of Weka {@link Instances} sets into a single
+ * {@link Instances} set.
+ *
+ * @param traindataSet
+ * set of {@link Instances} to be combines
+ * @return single {@link Instances} set
+ */
+ public static Instances makeSingleTrainingSet(SetUniqueList traindataSet) {
+ Instances traindataFull = null;
+ for (Instances traindata : traindataSet) {
+ if (traindataFull == null) {
+ traindataFull = new Instances(traindata);
+ }
+ else {
+ for (int i = 0; i < traindata.numInstances(); i++) {
+ traindataFull.add(traindata.instance(i));
+ }
+ }
+ }
+ return traindataFull;
+ }
+
+ /**
+ * Executes the experiment with the steps as described in the class comment.
+ *
+ * @see Runnable#run()
+ */
+ @Override
+ public void run() {
+ final List versions = new LinkedList<>();
+
+ for (IVersionLoader loader : config.getLoaders()) {
+ versions.addAll(loader.load());
+ }
+
+ for (IVersionFilter filter : config.getVersionFilters()) {
+ filter.apply(versions);
+ }
+ boolean writeHeader = true;
+ int versionCount = 1;
+ int testVersionCount = 0;
+
+ for (SoftwareVersion testVersion : versions) {
+ if (isVersion(testVersion, config.getTestVersionFilters())) {
+ testVersionCount++;
+ }
+ }
+
+ // sort versions
+ Collections.sort(versions);
+
+ for (SoftwareVersion testVersion : versions) {
+ if (isVersion(testVersion, config.getTestVersionFilters())) {
+ Console.traceln(Level.INFO,
+ String.format("[%s] [%02d/%02d] %s: starting",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion()));
+
+ // Setup testdata and training data
+ Instances testdata = testVersion.getInstances();
+ SetUniqueList traindataSet =
+ SetUniqueList.setUniqueList(new LinkedList());
+ for (SoftwareVersion trainingVersion : versions) {
+ if (isVersion(trainingVersion, config.getTrainingVersionFilters())) {
+ if (trainingVersion != testVersion) {
+ if (isTrainingVersion(trainingVersion, testVersion)) {
+ traindataSet.add(trainingVersion.getInstances());
+ }
+ }
+ }
+ }
+
+ for (ISetWiseProcessingStrategy processor : config.getSetWisePreprocessors()) {
+ Console.traceln(Level.FINE,
+ String.format(
+ "[%s] [%02d/%02d] %s: applying setwise preprocessor %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ processor.getClass().getName()));
+ processor.apply(testdata, traindataSet);
+ }
+ for (ISetWiseDataselectionStrategy dataselector : config.getSetWiseSelectors()) {
+ Console
+ .traceln(Level.FINE,
+ String.format("[%s] [%02d/%02d] %s: applying setwise selection %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ dataselector.getClass().getName()));
+ dataselector.apply(testdata, traindataSet);
+ }
+ for (ISetWiseProcessingStrategy processor : config.getSetWisePostprocessors()) {
+ Console.traceln(Level.FINE,
+ String.format(
+ "[%s] [%02d/%02d] %s: applying setwise postprocessor %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ processor.getClass().getName()));
+ processor.apply(testdata, traindataSet);
+ }
+ for (ISetWiseTrainingStrategy setwiseTrainer : config.getSetWiseTrainers()) {
+ Console
+ .traceln(Level.FINE,
+ String.format("[%s] [%02d/%02d] %s: applying setwise trainer %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ setwiseTrainer.getName()));
+ setwiseTrainer.apply(traindataSet);
+ }
+ for (ISetWiseTestdataAwareTrainingStrategy setwiseTestdataAwareTrainer : config
+ .getSetWiseTestdataAwareTrainers())
+ {
+ Console.traceln(Level.FINE,
+ String.format(
+ "[%s] [%02d/%02d] %s: applying testdata aware setwise trainer %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ setwiseTestdataAwareTrainer.getName()));
+ setwiseTestdataAwareTrainer.apply(traindataSet, testdata);
+ }
+ Instances traindata = makeSingleTrainingSet(traindataSet);
+ for (IProcessesingStrategy processor : config.getPreProcessors()) {
+ Console.traceln(Level.FINE,
+ String.format("[%s] [%02d/%02d] %s: applying preprocessor %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ processor.getClass().getName()));
+ processor.apply(testdata, traindata);
+ }
+ for (IPointWiseDataselectionStrategy dataselector : config
+ .getPointWiseSelectors())
+ {
+ Console.traceln(Level.FINE,
+ String.format(
+ "[%s] [%02d/%02d] %s: applying pointwise selection %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ dataselector.getClass().getName()));
+ traindata = dataselector.apply(testdata, traindata);
+ }
+ for (IProcessesingStrategy processor : config.getPostProcessors()) {
+ Console.traceln(Level.FINE,
+ String.format(
+ "[%s] [%02d/%02d] %s: applying setwise postprocessor %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ processor.getClass().getName()));
+ processor.apply(testdata, traindata);
+ }
+ for (ITrainingStrategy trainer : config.getTrainers()) {
+ Console.traceln(Level.FINE,
+ String.format("[%s] [%02d/%02d] %s: applying trainer %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ trainer.getName()));
+ trainer.apply(traindata);
+ }
+ for (ITestAwareTrainingStrategy trainer : config.getTestAwareTrainers()) {
+ Console.traceln(Level.FINE,
+ String.format("[%s] [%02d/%02d] %s: applying trainer %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ trainer.getName()));
+ trainer.apply(testdata, traindata);
+ }
+ File resultsDir = new File(config.getResultsPath());
+ if (!resultsDir.exists()) {
+ resultsDir.mkdir();
+ }
+ for (IEvaluationStrategy evaluator : config.getEvaluators()) {
+ Console.traceln(Level.FINE,
+ String.format("[%s] [%02d/%02d] %s: applying evaluator %s",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion(),
+ evaluator.getClass().getName()));
+ List allTrainers = new LinkedList<>();
+ for (ISetWiseTrainingStrategy setwiseTrainer : config.getSetWiseTrainers()) {
+ allTrainers.add(setwiseTrainer);
+ }
+ for (ISetWiseTestdataAwareTrainingStrategy setwiseTestdataAwareTrainer : config
+ .getSetWiseTestdataAwareTrainers())
+ {
+ allTrainers.add(setwiseTestdataAwareTrainer);
+ }
+ for (ITrainingStrategy trainer : config.getTrainers()) {
+ allTrainers.add(trainer);
+ }
+ for (ITestAwareTrainingStrategy trainer : config.getTestAwareTrainers()) {
+ allTrainers.add(trainer);
+ }
+ if (writeHeader) {
+ evaluator.setParameter(config.getResultsPath() + "/" +
+ config.getExperimentName() + ".csv");
+ }
+ evaluator.apply(testdata, traindata, allTrainers, writeHeader);
+ writeHeader = false;
+ }
+ Console.traceln(Level.INFO,
+ String.format("[%s] [%02d/%02d] %s: finished",
+ config.getExperimentName(), versionCount,
+ testVersionCount, testVersion.getVersion()));
+ versionCount++;
+ }
+ }
+ }
+
+ /**
+ * Helper method that checks if a version passes all filters.
+ *
+ * @param version
+ * version that is checked
+ * @param filters
+ * list of the filters
+ * @return true, if the version passes all filters, false otherwise
+ */
+ private boolean isVersion(SoftwareVersion version, List filters) {
+ boolean result = true;
+ for (IVersionFilter filter : filters) {
+ result &= !filter.apply(version);
+ }
+ return result;
+ }
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/execution/CrossProjectExperiment.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/execution/CrossProjectExperiment.java (revision 66)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/execution/CrossProjectExperiment.java (revision 67)
@@ -15,69 +15,15 @@
package de.ugoe.cs.cpdp.execution;
-import java.io.File;
-import java.util.Collections;
-import java.util.LinkedList;
-import java.util.List;
-import java.util.logging.Level;
-
-import org.apache.commons.collections4.list.SetUniqueList;
-
-import weka.core.Instances;
import de.ugoe.cs.cpdp.ExperimentConfiguration;
-import de.ugoe.cs.cpdp.dataprocessing.IProcessesingStrategy;
-import de.ugoe.cs.cpdp.dataprocessing.ISetWiseProcessingStrategy;
-import de.ugoe.cs.cpdp.dataselection.IPointWiseDataselectionStrategy;
-import de.ugoe.cs.cpdp.dataselection.ISetWiseDataselectionStrategy;
-import de.ugoe.cs.cpdp.eval.IEvaluationStrategy;
-import de.ugoe.cs.cpdp.loader.IVersionLoader;
-import de.ugoe.cs.cpdp.training.ISetWiseTestdataAwareTrainingStrategy;
-import de.ugoe.cs.cpdp.training.ISetWiseTrainingStrategy;
-import de.ugoe.cs.cpdp.training.ITestAwareTrainingStrategy;
-import de.ugoe.cs.cpdp.training.ITrainer;
-import de.ugoe.cs.cpdp.training.ITrainingStrategy;
-import de.ugoe.cs.cpdp.versions.IVersionFilter;
import de.ugoe.cs.cpdp.versions.SoftwareVersion;
-import de.ugoe.cs.util.console.Console;
/**
- * Class responsible for executing an experiment according to an {@link ExperimentConfiguration}.
- * The steps of an experiment are as follows:
- *
- * - load the data from the provided data path
- * - filter the data sets according to the provided version filters
- * - execute the following steps for each data sets as test data that is not ignored through the
- * test version filter:
- *
- * - filter the data sets to setup the candidate training data:
- *
- * - remove all data sets from the same project
- * - filter all data sets according to the training data filter
- *
- *
- * - apply the setwise preprocessors
- * - apply the setwise data selection algorithms
- * - apply the setwise postprocessors
- * - train the setwise training classifiers
- * - unify all remaining training data into one data set
- * - apply the preprocessors
- * - apply the pointwise data selection algorithms
- * - apply the postprocessors
- * - train the normal classifiers
- * - evaluate the results for all trained classifiers on the training data
- *
- *
- *
- *
- * Note that this class implements {@link Runnable}, i.e., each experiment can be started in its own
- * thread.
+ *
+ * Implements a cross project experiment where all versions not from the same project are used.
+ *
*
* @author Steffen Herbold
*/
-public class CrossProjectExperiment implements IExecutionStrategy {
-
- /**
- * configuration of the experiment
- */
- private final ExperimentConfiguration config;
+public class CrossProjectExperiment extends AbstractCrossProjectExperiment {
/**
@@ -88,206 +34,19 @@
*/
public CrossProjectExperiment(ExperimentConfiguration config) {
- this.config = config;
+ super(config);
}
- /**
- * Executes the experiment with the steps as described in the class comment.
+ /*
+ * (non-Javadoc)
*
- * @see Runnable#run()
+ * @see
+ * de.ugoe.cs.cpdp.execution.AbstractCrossProjectExperiment#isTrainingVersion(de.ugoe.cs.cpdp.
+ * versions.SoftwareVersion, de.ugoe.cs.cpdp.versions.SoftwareVersion)
*/
@Override
- public void run() {
- final List versions = new LinkedList<>();
-
- for (IVersionLoader loader : config.getLoaders()) {
- versions.addAll(loader.load());
- }
-
- for (IVersionFilter filter : config.getVersionFilters()) {
- filter.apply(versions);
- }
- boolean writeHeader = true;
- int versionCount = 1;
- int testVersionCount = 0;
-
- for (SoftwareVersion testVersion : versions) {
- if (isVersion(testVersion, config.getTestVersionFilters())) {
- testVersionCount++;
- }
- }
-
- // sort versions
- Collections.sort(versions);
-
- for (SoftwareVersion testVersion : versions) {
- if (isVersion(testVersion, config.getTestVersionFilters())) {
- Console.traceln(Level.INFO, String.format("[%s] [%02d/%02d] %s: starting",
- config.getExperimentName(), versionCount,
- testVersionCount,
- testVersion.getVersion()));
-
- // Setup testdata and training data
- Instances testdata = testVersion.getInstances();
- String testProject = testVersion.getProject();
- SetUniqueList traindataSet =
- SetUniqueList.setUniqueList(new LinkedList());
- for (SoftwareVersion trainingVersion : versions) {
- if (isVersion(trainingVersion, config.getTrainingVersionFilters())) {
- if (trainingVersion != testVersion) {
- if (!trainingVersion.getProject().equals(testProject)) {
- traindataSet.add(trainingVersion.getInstances());
- }
- }
- }
- }
-
- for (ISetWiseProcessingStrategy processor : config.getSetWisePreprocessors()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying setwise preprocessor %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), processor.getClass().getName()));
- processor.apply(testdata, traindataSet);
- }
- for (ISetWiseDataselectionStrategy dataselector : config.getSetWiseSelectors()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying setwise selection %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), dataselector.getClass().getName()));
- dataselector.apply(testdata, traindataSet);
- }
- for (ISetWiseProcessingStrategy processor : config.getSetWisePostprocessors()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying setwise postprocessor %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), processor.getClass().getName()));
- processor.apply(testdata, traindataSet);
- }
- for (ISetWiseTrainingStrategy setwiseTrainer : config.getSetWiseTrainers()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying setwise trainer %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), setwiseTrainer.getName()));
- setwiseTrainer.apply(traindataSet);
- }
- for (ISetWiseTestdataAwareTrainingStrategy setwiseTestdataAwareTrainer : config.getSetWiseTestdataAwareTrainers()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying testdata aware setwise trainer %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), setwiseTestdataAwareTrainer.getName()));
- setwiseTestdataAwareTrainer.apply(traindataSet, testdata);
- }
- Instances traindata = makeSingleTrainingSet(traindataSet);
- for (IProcessesingStrategy processor : config.getPreProcessors()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying preprocessor %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), processor.getClass().getName()));
- processor.apply(testdata, traindata);
- }
- for (IPointWiseDataselectionStrategy dataselector : config.getPointWiseSelectors())
- {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying pointwise selection %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), dataselector.getClass().getName()));
- traindata = dataselector.apply(testdata, traindata);
- }
- for (IProcessesingStrategy processor : config.getPostProcessors()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying setwise postprocessor %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), processor.getClass().getName()));
- processor.apply(testdata, traindata);
- }
- for (ITrainingStrategy trainer : config.getTrainers()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying trainer %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), trainer.getName()));
- trainer.apply(traindata);
- }
- for (ITestAwareTrainingStrategy trainer : config.getTestAwareTrainers()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying trainer %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), trainer.getName()));
- trainer.apply(testdata, traindata);
- }
- File resultsDir = new File(config.getResultsPath());
- if (!resultsDir.exists()) {
- resultsDir.mkdir();
- }
- for (IEvaluationStrategy evaluator : config.getEvaluators()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying evaluator %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), evaluator.getClass().getName()));
- List allTrainers = new LinkedList<>();
- for (ISetWiseTrainingStrategy setwiseTrainer : config.getSetWiseTrainers()) {
- allTrainers.add(setwiseTrainer);
- }
- for (ISetWiseTestdataAwareTrainingStrategy setwiseTestdataAwareTrainer : config.getSetWiseTestdataAwareTrainers()) {
- allTrainers.add(setwiseTestdataAwareTrainer);
- }
- for (ITrainingStrategy trainer : config.getTrainers()) {
- allTrainers.add(trainer);
- }
- for (ITestAwareTrainingStrategy trainer : config.getTestAwareTrainers()) {
- allTrainers.add(trainer);
- }
- if (writeHeader) {
- evaluator.setParameter(config.getResultsPath() + "/" +
- config.getExperimentName() + ".csv");
- }
- evaluator.apply(testdata, traindata, allTrainers, writeHeader);
- writeHeader = false;
- }
- Console.traceln(Level.INFO, String.format("[%s] [%02d/%02d] %s: finished",
- config.getExperimentName(), versionCount,
- testVersionCount,
- testVersion.getVersion()));
- versionCount++;
- }
- }
- }
-
- /**
- * Helper method that checks if a version passes all filters.
- *
- * @param version
- * version that is checked
- * @param filters
- * list of the filters
- * @return true, if the version passes all filters, false otherwise
- */
- private boolean isVersion(SoftwareVersion version, List filters) {
- boolean result = true;
- for (IVersionFilter filter : filters) {
- result &= !filter.apply(version);
- }
- return result;
- }
-
- /**
- * Helper method that combines a set of Weka {@link Instances} sets into a single
- * {@link Instances} set.
- *
- * @param traindataSet
- * set of {@link Instances} to be combines
- * @return single {@link Instances} set
- */
- public static Instances makeSingleTrainingSet(SetUniqueList traindataSet) {
- Instances traindataFull = null;
- for (Instances traindata : traindataSet) {
- if (traindataFull == null) {
- traindataFull = new Instances(traindata);
- }
- else {
- for (int i = 0; i < traindata.numInstances(); i++) {
- traindataFull.add(traindata.instance(i));
- }
- }
- }
- return traindataFull;
+ protected boolean isTrainingVersion(SoftwareVersion trainingVersion,
+ SoftwareVersion testVersion)
+ {
+ return !trainingVersion.getProject().equals(testVersion.getProject());
}
}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/execution/RelaxedCrossProjectExperiment.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/execution/RelaxedCrossProjectExperiment.java (revision 66)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/execution/RelaxedCrossProjectExperiment.java (revision 67)
@@ -15,68 +15,16 @@
package de.ugoe.cs.cpdp.execution;
-import java.io.File;
-import java.util.Collections;
-import java.util.LinkedList;
-import java.util.List;
-import java.util.logging.Level;
-
-import org.apache.commons.collections4.list.SetUniqueList;
-
-import weka.core.Instances;
import de.ugoe.cs.cpdp.ExperimentConfiguration;
-import de.ugoe.cs.cpdp.dataprocessing.IProcessesingStrategy;
-import de.ugoe.cs.cpdp.dataprocessing.ISetWiseProcessingStrategy;
-import de.ugoe.cs.cpdp.dataselection.IPointWiseDataselectionStrategy;
-import de.ugoe.cs.cpdp.dataselection.ISetWiseDataselectionStrategy;
-import de.ugoe.cs.cpdp.eval.IEvaluationStrategy;
-import de.ugoe.cs.cpdp.loader.IVersionLoader;
-import de.ugoe.cs.cpdp.training.ISetWiseTestdataAwareTrainingStrategy;
-import de.ugoe.cs.cpdp.training.ISetWiseTrainingStrategy;
-import de.ugoe.cs.cpdp.training.ITestAwareTrainingStrategy;
-import de.ugoe.cs.cpdp.training.ITrainer;
-import de.ugoe.cs.cpdp.training.ITrainingStrategy;
-import de.ugoe.cs.cpdp.versions.IVersionFilter;
import de.ugoe.cs.cpdp.versions.SoftwareVersion;
-import de.ugoe.cs.util.console.Console;
/**
- * Class responsible for executing an experiment according to an {@link ExperimentConfiguration}.
- * The steps of an experiment are as follows:
- *
- * - load the data from the provided data path
- * - filter the data sets according to the provided version filters
- * - execute the following steps for each data sets as test data that is not ignored through the
- * test version filter:
- *
- * - filter the data sets to setup the candidate training data:
- *
- * - filter all data sets according to the training data filter
- *
- *
- * - apply the setwise preprocessors
- * - apply the setwise data selection algorithms
- * - apply the setwise postprocessors
- * - train the setwise training classifiers
- * - unify all remaining training data into one data set
- * - apply the preprocessors
- * - apply the pointwise data selection algorithms
- * - apply the postprocessors
- * - train the normal classifiers
- * - evaluate the results for all trained classifiers on the training data
- *
- *
- *
- *
- * Note that this class implements {@link Runnable}, i.e., each experiment can be started in its own
- * thread.
+ *
+ * Implements a cross project experiment where all versions not from the same project are used
+ * together with all older version from the test product.
+ *
*
* @author Steffen Herbold
*/
-public class RelaxedCrossProjectExperiment implements IExecutionStrategy {
-
- /**
- * configuration of the experiment
- */
- private final ExperimentConfiguration config;
+public class RelaxedCrossProjectExperiment extends AbstractCrossProjectExperiment {
/**
@@ -87,212 +35,28 @@
*/
public RelaxedCrossProjectExperiment(ExperimentConfiguration config) {
- this.config = config;
+ super(config);
}
- /**
- * Executes the experiment with the steps as described in the class comment.
+ /*
+ * (non-Javadoc)
*
- * @see Runnable#run()
+ * @see
+ * de.ugoe.cs.cpdp.execution.AbstractCrossProjectExperiment#isTrainingVersion(de.ugoe.cs.cpdp.
+ * versions.SoftwareVersion, de.ugoe.cs.cpdp.versions.SoftwareVersion)
*/
@Override
- public void run() {
- final List versions = new LinkedList<>();
-
- for (IVersionLoader loader : config.getLoaders()) {
- versions.addAll(loader.load());
- }
-
- for (IVersionFilter filter : config.getVersionFilters()) {
- filter.apply(versions);
- }
- boolean writeHeader = true;
- int versionCount = 1;
- int testVersionCount = 0;
-
- for (SoftwareVersion testVersion : versions) {
- if (isVersion(testVersion, config.getTestVersionFilters())) {
- testVersionCount++;
+ protected boolean isTrainingVersion(SoftwareVersion trainingVersion,
+ SoftwareVersion testVersion)
+ {
+ if (trainingVersion.getProject().equals(testVersion.getProject())) {
+ if (trainingVersion.compareTo(testVersion) < 0) {
+ // only add if older
+ return true;
}
}
-
- // sort versions
- Collections.sort(versions);
-
- for (SoftwareVersion testVersion : versions) {
- if (isVersion(testVersion, config.getTestVersionFilters())) {
- Console.traceln(Level.INFO, String.format("[%s] [%02d/%02d] %s: starting",
- config.getExperimentName(), versionCount,
- testVersionCount,
- testVersion.getVersion()));
-
- // Setup testdata and training data
- Instances testdata = testVersion.getInstances();
- String testProject = testVersion.getProject();
- SetUniqueList traindataSet =
- SetUniqueList.setUniqueList(new LinkedList());
- for (SoftwareVersion trainingVersion : versions) {
- if (isVersion(trainingVersion, config.getTrainingVersionFilters())) {
- if (trainingVersion != testVersion) {
- if (trainingVersion.getProject().equals(testProject)) {
- if (trainingVersion.compareTo(testVersion) < 0) {
- // only add if older
- traindataSet.add(trainingVersion.getInstances());
- }
- }
- else {
- traindataSet.add(trainingVersion.getInstances());
- }
- }
- }
- }
-
- for (ISetWiseProcessingStrategy processor : config.getSetWisePreprocessors()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying setwise preprocessor %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), processor.getClass().getName()));
- processor.apply(testdata, traindataSet);
- }
- for (ISetWiseDataselectionStrategy dataselector : config.getSetWiseSelectors()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying setwise selection %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), dataselector.getClass().getName()));
- dataselector.apply(testdata, traindataSet);
- }
- for (ISetWiseProcessingStrategy processor : config.getSetWisePostprocessors()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying setwise postprocessor %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), processor.getClass().getName()));
- processor.apply(testdata, traindataSet);
- }
- for (ISetWiseTrainingStrategy setwiseTrainer : config.getSetWiseTrainers()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying setwise trainer %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), setwiseTrainer.getName()));
- setwiseTrainer.apply(traindataSet);
- }
- for (ISetWiseTestdataAwareTrainingStrategy setwiseTestdataAwareTrainer : config.getSetWiseTestdataAwareTrainers()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying testdata aware setwise trainer %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), setwiseTestdataAwareTrainer.getName()));
- setwiseTestdataAwareTrainer.apply(traindataSet, testdata);
- }
- Instances traindata = makeSingleTrainingSet(traindataSet);
- for (IProcessesingStrategy processor : config.getPreProcessors()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying preprocessor %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), processor.getClass().getName()));
- processor.apply(testdata, traindata);
- }
- for (IPointWiseDataselectionStrategy dataselector : config.getPointWiseSelectors())
- {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying pointwise selection %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), dataselector.getClass().getName()));
- traindata = dataselector.apply(testdata, traindata);
- }
- for (IProcessesingStrategy processor : config.getPostProcessors()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying setwise postprocessor %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), processor.getClass().getName()));
- processor.apply(testdata, traindata);
- }
- for (ITrainingStrategy trainer : config.getTrainers()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying trainer %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), trainer.getName()));
- trainer.apply(traindata);
- }
- for (ITestAwareTrainingStrategy trainer : config.getTestAwareTrainers()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying trainer %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), trainer.getName()));
- trainer.apply(testdata, traindata);
- }
- File resultsDir = new File(config.getResultsPath());
- if (!resultsDir.exists()) {
- resultsDir.mkdir();
- }
- for (IEvaluationStrategy evaluator : config.getEvaluators()) {
- Console.traceln(Level.FINE, String
- .format("[%s] [%02d/%02d] %s: applying evaluator %s",
- config.getExperimentName(), versionCount, testVersionCount,
- testVersion.getVersion(), evaluator.getClass().getName()));
- List allTrainers = new LinkedList<>();
- for (ISetWiseTrainingStrategy setwiseTrainer : config.getSetWiseTrainers()) {
- allTrainers.add(setwiseTrainer);
- }
- for (ISetWiseTestdataAwareTrainingStrategy setwiseTrainer : config.getSetWiseTestdataAwareTrainers()) {
- allTrainers.add(setwiseTrainer);
- }
- for (ITrainingStrategy trainer : config.getTrainers()) {
- allTrainers.add(trainer);
- }
- for (ITestAwareTrainingStrategy trainer : config.getTestAwareTrainers()) {
- allTrainers.add(trainer);
- }
- if (writeHeader) {
- evaluator.setParameter(config.getResultsPath() + "/" +
- config.getExperimentName() + ".csv");
- }
- evaluator.apply(testdata, traindata, allTrainers, writeHeader);
- writeHeader = false;
- }
- Console.traceln(Level.INFO, String.format("[%s] [%02d/%02d] %s: finished",
- config.getExperimentName(), versionCount,
- testVersionCount,
- testVersion.getVersion()));
- versionCount++;
- }
+ else {
+ return true;
}
- }
-
- /**
- * Helper method that checks if a version passes all filters.
- *
- * @param version
- * version that is checked
- * @param filters
- * list of the filters
- * @return true, if the version passes all filters, false otherwise
- */
- private boolean isVersion(SoftwareVersion version, List filters) {
- boolean result = true;
- for (IVersionFilter filter : filters) {
- result &= !filter.apply(version);
- }
- return result;
- }
-
- /**
- * Helper method that combines a set of Weka {@link Instances} sets into a single
- * {@link Instances} set.
- *
- * @param traindataSet
- * set of {@link Instances} to be combines
- * @return single {@link Instances} set
- */
- public static Instances makeSingleTrainingSet(SetUniqueList traindataSet) {
- Instances traindataFull = null;
- for (Instances traindata : traindataSet) {
- if (traindataFull == null) {
- traindataFull = new Instances(traindata);
- }
- else {
- for (int i = 0; i < traindata.numInstances(); i++) {
- traindataFull.add(traindata.instance(i));
- }
- }
- }
- return traindataFull;
+ return false;
}
}