traindataSet);
+
+ String getName();
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/ITrainer.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/ITrainer.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/ITrainer.java (revision 2)
@@ -0,0 +1,7 @@
+package de.ugoe.cs.cpdp.training;
+
+import de.ugoe.cs.cpdp.IParameterizable;
+
+public interface ITrainer extends IParameterizable {
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/ITrainingStrategy.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/ITrainingStrategy.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/ITrainingStrategy.java (revision 2)
@@ -0,0 +1,8 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.core.Instances;
+
+public interface ITrainingStrategy extends ITrainer {
+
+ void apply(Instances traindata);
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/LogisticRegressionBagging.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/LogisticRegressionBagging.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/LogisticRegressionBagging.java (revision 2)
@@ -0,0 +1,18 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.functions.Logistic;
+
+public class LogisticRegressionBagging extends BaggingTraining {
+
+ @Override
+ public String getName() {
+ return "LogisticRegressionBagging";
+ }
+
+ @Override
+ protected Classifier setupClassifier() {
+ return new Logistic();
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/LogisticRegressionTraining.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/LogisticRegressionTraining.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/LogisticRegressionTraining.java (revision 2)
@@ -0,0 +1,17 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.functions.Logistic;
+
+public class LogisticRegressionTraining extends WekaTraining {
+
+ @Override
+ public String getName() {
+ return "LogisticRegression";
+ }
+
+ @Override
+ protected Classifier setupClassifier() {
+ return new Logistic();
+ }
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NaiveBayesBagging.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NaiveBayesBagging.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NaiveBayesBagging.java (revision 2)
@@ -0,0 +1,18 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.bayes.NaiveBayes;
+
+public class NaiveBayesBagging extends BaggingTraining {
+
+ @Override
+ public String getName() {
+ return "NaiveBayesBagging";
+ }
+
+ @Override
+ protected Classifier setupClassifier() {
+ return new NaiveBayes();
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NaiveBayesTraining.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NaiveBayesTraining.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NaiveBayesTraining.java (revision 2)
@@ -0,0 +1,18 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.bayes.NaiveBayes;
+
+public class NaiveBayesTraining extends WekaTraining {
+
+ @Override
+ public String getName() {
+ return "NaiveBayes";
+ }
+
+ @Override
+ protected Classifier setupClassifier() {
+ return new NaiveBayes();
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NeuralNetworkBagging.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NeuralNetworkBagging.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NeuralNetworkBagging.java (revision 2)
@@ -0,0 +1,18 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.functions.MultilayerPerceptron;
+
+public class NeuralNetworkBagging extends BaggingTraining {
+
+ @Override
+ public String getName() {
+ return "NeuralNetworkBagging";
+ }
+
+ @Override
+ protected Classifier setupClassifier() {
+ return new MultilayerPerceptron();
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NeuralNetworkTraining.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NeuralNetworkTraining.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/NeuralNetworkTraining.java (revision 2)
@@ -0,0 +1,17 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.functions.MultilayerPerceptron;
+
+public class NeuralNetworkTraining extends WekaTraining {
+
+ @Override
+ public String getName() {
+ return "NeuralNetwork";
+ }
+
+ @Override
+ protected Classifier setupClassifier() {
+ return new MultilayerPerceptron();
+ }
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/RandomForestBagging.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/RandomForestBagging.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/RandomForestBagging.java (revision 2)
@@ -0,0 +1,28 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.meta.CVParameterSelection;
+import weka.classifiers.trees.RandomForest;
+
+public class RandomForestBagging extends BaggingTraining {
+
+ @Override
+ protected Classifier setupClassifier() {
+ final CVParameterSelection ps = new CVParameterSelection();
+ ps.setClassifier(new RandomForest());
+ try {
+ // Parameter optimization with 5x5 CV
+ ps.setNumFolds(5);
+ ps.addCVParameter("I 5 25 5");
+ } catch (Exception e) {
+ throw new RuntimeException(e);
+ }
+ return ps;
+ }
+
+ @Override
+ public String getName() {
+ return "RandomForestBagging";
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/RandomForestTraining.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/RandomForestTraining.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/RandomForestTraining.java (revision 2)
@@ -0,0 +1,28 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.meta.CVParameterSelection;
+import weka.classifiers.trees.RandomForest;
+
+public class RandomForestTraining extends WekaTraining {
+
+ @Override
+ protected Classifier setupClassifier() {
+ final CVParameterSelection ps = new CVParameterSelection();
+ ps.setClassifier(new RandomForest());
+ try {
+ // Parameter optimization with 5x5 CV
+ ps.setNumFolds(5);
+ ps.addCVParameter("I 5 25 5");
+ } catch (Exception e) {
+ throw new RuntimeException(e);
+ }
+ return ps;
+ }
+
+ @Override
+ public String getName() {
+ return "RandomForest";
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMOPolyKernelBagging.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMOPolyKernelBagging.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMOPolyKernelBagging.java (revision 2)
@@ -0,0 +1,28 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.functions.SMO;
+
+public class SMOPolyKernelBagging extends BaggingTraining {
+
+ @Override
+ public String getName() {
+ return "SMOPolyBagging";
+ }
+
+ @Override
+ protected Classifier setupClassifier() {
+ return new SMO();
+ /*CVParameterSelection ps = new CVParameterSelection();
+ ps.setClassifier(new SMO());
+ try {
+ // Parameter optimization with 5x5 CV
+ ps.setNumFolds(5);
+ ps.addCVParameter("C 1 100 4");
+ } catch (Exception e) {
+ throw new RuntimeException();
+ }
+ return ps;*/
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMOPolyKernelTraining.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMOPolyKernelTraining.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMOPolyKernelTraining.java (revision 2)
@@ -0,0 +1,28 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.functions.SMO;
+
+public class SMOPolyKernelTraining extends WekaTraining {
+
+ @Override
+ public String getName() {
+ return "SMOPoly";
+ }
+
+ @Override
+ protected Classifier setupClassifier() {
+ return new SMO();
+ /*CVParameterSelection ps = new CVParameterSelection();
+ ps.setClassifier(new SMO());
+ try {
+ // Parameter optimization with 5x5 CV
+ ps.setNumFolds(5);
+ ps.addCVParameter("C 1 100 4");
+ } catch (Exception e) {
+ throw new RuntimeException();
+ }
+ return ps;*/
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMORBFKernelBagging.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMORBFKernelBagging.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMORBFKernelBagging.java (revision 2)
@@ -0,0 +1,31 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.functions.SMO;
+import weka.classifiers.functions.supportVector.RBFKernel;
+
+public class SMORBFKernelBagging extends BaggingTraining {
+
+ @Override
+ public String getName() {
+ return "SMORBFBagging";
+ }
+
+ @Override
+ protected Classifier setupClassifier() {
+ //CVParameterSelection ps = new CVParameterSelection();
+ final SMO smoRBF = new SMO();
+ smoRBF.setKernel(new RBFKernel());
+ return smoRBF;
+ /*ps.setClassifier(smoRBF);
+ try {
+ // Parameter optimization with 5x5 CV
+ ps.setNumFolds(5);
+ ps.addCVParameter("C 1 100 4");
+ } catch (Exception e) {
+ throw new RuntimeException();
+ }
+ return ps;*/
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMORBFKernelTraining.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMORBFKernelTraining.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/SMORBFKernelTraining.java (revision 2)
@@ -0,0 +1,31 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+import weka.classifiers.functions.SMO;
+import weka.classifiers.functions.supportVector.RBFKernel;
+
+public class SMORBFKernelTraining extends WekaTraining {
+
+ @Override
+ public String getName() {
+ return "SMORBF";
+ }
+
+ @Override
+ protected Classifier setupClassifier() {
+ //CVParameterSelection ps = new CVParameterSelection();
+ final SMO smoRBF = new SMO();
+ smoRBF.setKernel(new RBFKernel());
+ return smoRBF;
+ /*ps.setClassifier(smoRBF);
+ try {
+ // Parameter optimization with 5x5 CV
+ ps.setNumFolds(5);
+ ps.addCVParameter("C 1 100 4");
+ } catch (Exception e) {
+ throw new RuntimeException();
+ }
+ return ps;*/
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaBaggingTraining2.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaBaggingTraining2.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaBaggingTraining2.java (revision 2)
@@ -0,0 +1,126 @@
+package de.ugoe.cs.cpdp.training;
+
+import java.io.PrintStream;
+import java.util.HashSet;
+import java.util.LinkedList;
+import java.util.List;
+import java.util.Set;
+
+import org.apache.commons.collections4.list.SetUniqueList;
+import org.apache.commons.io.output.NullOutputStream;
+
+import weka.classifiers.AbstractClassifier;
+import weka.classifiers.Classifier;
+import weka.core.DenseInstance;
+import weka.core.Instance;
+import weka.core.Instances;
+
+/**
+ * Programmatic WekaBaggingTraining
+ *
+ * first parameter is Trainer Name.
+ * second parameter is class name
+ *
+ * all subsequent parameters are configuration params (for example for trees)
+ *
+ * XML Configurations for Weka Classifiers:
+ *
+ * {@code
+ *
+ *
+ *
+ * }
+ *
+ *
+ */
+public class WekaBaggingTraining2 extends WekaBaseTraining2 implements ISetWiseTrainingStrategy {
+
+ private final TraindatasetBagging classifier = new TraindatasetBagging();
+
+ @Override
+ public Classifier getClassifier() {
+ return classifier;
+ }
+
+ @Override
+ public void apply(SetUniqueList traindataSet) {
+ PrintStream errStr = System.err;
+ System.setErr(new PrintStream(new NullOutputStream()));
+ try {
+ classifier.buildClassifier(traindataSet);
+ } catch (Exception e) {
+ throw new RuntimeException(e);
+ } finally {
+ System.setErr(errStr);
+ }
+ }
+
+ public class TraindatasetBagging extends AbstractClassifier {
+
+ private static final long serialVersionUID = 1L;
+
+ private List trainingData = null;
+
+ private List classifiers = null;
+
+ @Override
+ public double classifyInstance(Instance instance) {
+ if( classifiers==null ) {
+ return 0.0; // TODO check how WEKA expects classifyInstance to behave if no classifier exists yet
+ }
+
+ double classification = 0.0;
+ for( int i=0 ; i attributeNames = new HashSet<>();
+ for( int j=0; j=0.5) ? 1.0 : 0.0;
+ }
+
+ public void buildClassifier(SetUniqueList traindataSet) throws Exception {
+ classifiers = new LinkedList<>();
+ trainingData = new LinkedList<>();
+ for( Instances traindata : traindataSet ) {
+ Classifier classifier = setupClassifier();
+ classifier.buildClassifier(traindata);
+ classifiers.add(classifier);
+ trainingData.add(new Instances(traindata));
+ }
+ }
+
+ @Override
+ public void buildClassifier(Instances traindata) throws Exception {
+ classifiers = new LinkedList<>();
+ trainingData = new LinkedList<>();
+ final Classifier classifier = setupClassifier();
+ classifier.buildClassifier(traindata);
+ classifiers.add(classifier);
+ trainingData.add(new Instances(traindata));
+ }
+ }
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaBaseTraining2.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaBaseTraining2.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaBaseTraining2.java (revision 2)
@@ -0,0 +1,103 @@
+package de.ugoe.cs.cpdp.training;
+
+import java.util.Arrays;
+import java.util.logging.Level;
+
+import de.ugoe.cs.util.console.Console;
+import weka.core.OptionHandler;
+import weka.classifiers.Classifier;
+import weka.classifiers.meta.CVParameterSelection;
+
+public abstract class WekaBaseTraining2 implements WekaCompatibleTrainer {
+
+ protected Classifier classifier = null;
+ protected String classifierClassName;
+ protected String classifierName;
+ protected String[] classifierParams;
+
+ @Override
+ public void setParameter(String parameters) {
+ String[] params = parameters.split(" ");
+
+ // first is classifierName
+ classifierName = params[0];
+
+ // all following parameters can be copied from weka!
+
+ // second param is classifierClassName
+ classifierClassName = params[1];
+
+ // rest are params to the specified classifier
+ classifierParams = Arrays.copyOfRange(params, 2, params.length);
+
+ classifier = setupClassifier();
+ }
+
+ @Override
+ public Classifier getClassifier() {
+ return classifier;
+ }
+
+ public Classifier setupClassifier() {
+ Classifier cl = null;
+ try{
+ @SuppressWarnings("rawtypes")
+ Class c = Class.forName(classifierClassName);
+ Classifier obj = (Classifier) c.newInstance();
+
+ // Filter -CVPARAM
+ String[] param = Arrays.copyOf(classifierParams, classifierParams.length);
+ String[] cvparam = {};
+ boolean cv = false;
+ for ( int i=0; i < classifierParams.length; i++ ) {
+ if(classifierParams[i].equals("-CVPARAM")) {
+ // rest of array are cvparam
+ cvparam = Arrays.copyOfRange(classifierParams, i+1, classifierParams.length);
+
+ // before this we have normal params
+ param = Arrays.copyOfRange(classifierParams, 0, i);
+
+ cv = true;
+ break;
+ }
+ }
+
+ // set classifier params
+ ((OptionHandler)obj).setOptions(param);
+ cl = obj;
+
+ // we have cross val params
+ // cant check on cvparam.length may not be initialized
+ if(cv) {
+ final CVParameterSelection ps = new CVParameterSelection();
+ ps.setClassifier(obj);
+ ps.setNumFolds(5);
+ //ps.addCVParameter("I 5 25 5");
+ ps.addCVParameter(Arrays.asList(cvparam).toString().replaceAll(", ", " ").replaceAll("^\\[|\\]$", ""));
+
+ cl = ps;
+ }
+
+ }catch(ClassNotFoundException e) {
+ Console.traceln(Level.WARNING, String.format("class not found: %s", e.toString()));
+ e.printStackTrace();
+ } catch (InstantiationException e) {
+ Console.traceln(Level.WARNING, String.format("Instantiation Exception: %s", e.toString()));
+ e.printStackTrace();
+ } catch (IllegalAccessException e) {
+ Console.traceln(Level.WARNING, String.format("Illegal Access Exception: %s", e.toString()));
+ e.printStackTrace();
+ } catch (Exception e) {
+ Console.traceln(Level.WARNING, String.format("Exception: %s", e.toString()));
+ e.printStackTrace();
+ }
+
+ return cl;
+ }
+
+ @Override
+ public String getName() {
+ return classifierName;
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaClusterTraining2.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaClusterTraining2.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaClusterTraining2.java (revision 2)
@@ -0,0 +1,186 @@
+package de.ugoe.cs.cpdp.training;
+
+import java.io.PrintStream;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.Set;
+import java.util.logging.Level;
+
+import org.apache.commons.io.output.NullOutputStream;
+
+import de.ugoe.cs.util.console.Console;
+import weka.classifiers.AbstractClassifier;
+import weka.classifiers.Classifier;
+import weka.clusterers.EM;
+import weka.core.DenseInstance;
+import weka.core.Instance;
+import weka.core.Instances;
+import weka.filters.Filter;
+import weka.filters.unsupervised.attribute.Remove;
+
+/**
+ * WekaClusterTraining2
+ *
+ * 1. Cluster traindata
+ * 2. for each cluster train a classifier with traindata from cluster
+ * 3. match testdata instance to a cluster, then classify with classifier from the cluster
+ *
+ * XML config:
+ *
+ *
+ *
+ *
+ *
+ *
+ * Questions:
+ * - how do we configure the clustering params?
+ */
+public class WekaClusterTraining2 extends WekaBaseTraining2 implements ITrainingStrategy {
+
+ private final TraindatasetCluster classifier = new TraindatasetCluster();
+
+ @Override
+ public Classifier getClassifier() {
+ return classifier;
+ }
+
+
+ @Override
+ public void apply(Instances traindata) {
+ PrintStream errStr = System.err;
+ System.setErr(new PrintStream(new NullOutputStream()));
+ try {
+ classifier.buildClassifier(traindata);
+ } catch (Exception e) {
+ throw new RuntimeException(e);
+ } finally {
+ System.setErr(errStr);
+ }
+ }
+
+
+ public class TraindatasetCluster extends AbstractClassifier {
+
+ private static final long serialVersionUID = 1L;
+
+ private EM clusterer = null;
+
+ private HashMap cclassifier = new HashMap();
+ private HashMap ctraindata = new HashMap();
+
+
+
+ private Instance createInstance(Instances instances, Instance instance) {
+ // attributes for feeding instance to classifier
+ Set attributeNames = new HashSet<>();
+ for( int j=0; j clusternumber = ctraindata.keySet().iterator();
+ while ( clusternumber.hasNext() ) {
+ cnumber = clusternumber.next();
+ cclassifier.put(cnumber,setupClassifier());
+ cclassifier.get(cnumber).buildClassifier(ctraindata.get(cnumber));
+
+ //Console.traceln(Level.INFO, String.format("classifier in cluster "+cnumber));
+ }
+ }
+ }
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaCompatibleTrainer.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaCompatibleTrainer.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaCompatibleTrainer.java (revision 2)
@@ -0,0 +1,10 @@
+package de.ugoe.cs.cpdp.training;
+
+import weka.classifiers.Classifier;
+
+public interface WekaCompatibleTrainer extends ITrainer {
+
+ Classifier getClassifier();
+
+ String getName();
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaTraining.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaTraining.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaTraining.java (revision 2)
@@ -0,0 +1,38 @@
+package de.ugoe.cs.cpdp.training;
+
+import java.io.PrintStream;
+
+import org.apache.commons.io.output.NullOutputStream;
+
+import weka.classifiers.Classifier;
+import weka.core.Instances;
+
+public abstract class WekaTraining implements ITrainingStrategy, WekaCompatibleTrainer {
+
+ private final Classifier classifier = setupClassifier();
+
+ protected abstract Classifier setupClassifier();
+
+ @Override
+ public Classifier getClassifier() {
+ return classifier;
+ }
+
+ @Override
+ public void apply(Instances traindata) {
+ PrintStream errStr = System.err;
+ System.setErr(new PrintStream(new NullOutputStream()));
+ try {
+ classifier.buildClassifier(traindata);
+ } catch (Exception e) {
+ throw new RuntimeException(e);
+ } finally {
+ System.setErr(errStr);
+ }
+ }
+
+ @Override
+ public void setParameter(String parameters) {
+ // TODO should allow passing of weka parameters to the classifier
+ }
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaTraining2.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaTraining2.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaTraining2.java (revision 2)
@@ -0,0 +1,46 @@
+package de.ugoe.cs.cpdp.training;
+
+import java.io.PrintStream;
+import java.util.logging.Level;
+
+import org.apache.commons.io.output.NullOutputStream;
+
+import de.ugoe.cs.util.console.Console;
+import weka.core.Instances;
+
+/**
+ * Programmatic WekaBaggingTraining
+ *
+ * first parameter is Trainer Name.
+ * second parameter is class name
+ *
+ * all subsequent parameters are configuration params (for example for trees)
+ *
+ * XML Configurations for Weka Classifiers:
+ *
+ * {@code
+ *
+ *
+ *
+ * }
+ *
+ *
+ */
+public class WekaTraining2 extends WekaBaseTraining2 implements ITrainingStrategy {
+
+ @Override
+ public void apply(Instances traindata) {
+ PrintStream errStr = System.err;
+ System.setErr(new PrintStream(new NullOutputStream()));
+ try {
+ if(classifier == null) {
+ Console.traceln(Level.WARNING, String.format("classifier null!"));
+ }
+ classifier.buildClassifier(traindata);
+ } catch (Exception e) {
+ throw new RuntimeException(e);
+ } finally {
+ System.setErr(errStr);
+ }
+ }
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/AbstractVersionFilter.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/AbstractVersionFilter.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/AbstractVersionFilter.java (revision 2)
@@ -0,0 +1,28 @@
+package de.ugoe.cs.cpdp.versions;
+
+import java.util.Iterator;
+import java.util.List;
+
+/**
+ * Implements a skeletal {@link IVersionFilter}.
+ * @author Steffen Herbold
+ */
+public abstract class AbstractVersionFilter implements IVersionFilter {
+
+ /**
+ * @see de.ugoe.cs.cpdp.versions.IVersionFilter#apply(java.util.List)
+ */
+ @Override
+ public int apply(List versions) {
+ int removed = 0;
+ for( final Iterator iter=versions.iterator() ; iter.hasNext() ; ) {
+ SoftwareVersion version = iter.next();
+
+ if( apply(version) ) {
+ iter.remove();
+ removed++;
+ }
+ }
+ return removed;
+ }
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/IVersionFilter.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/IVersionFilter.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/IVersionFilter.java (revision 2)
@@ -0,0 +1,26 @@
+package de.ugoe.cs.cpdp.versions;
+
+import java.util.List;
+
+import de.ugoe.cs.cpdp.IParameterizable;
+
+/**
+ * Implements the interface for a {@link SoftwareVersion} filter.
+ * @author Steffen Herbold
+ */
+public interface IVersionFilter extends IParameterizable {
+
+ /**
+ * Applies the filter to a single version.
+ * @param version the version
+ * @return true if filter applies to version, false otherwise
+ */
+ boolean apply(SoftwareVersion version);
+
+ /**
+ * Applies the filter a a list of versions. Versions were the filter applies are automatically removed from the list.
+ * @param versions list of versions
+ * @return number of removed versions
+ */
+ int apply(List versions);
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/MaxInstanceNumberFilter.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/MaxInstanceNumberFilter.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/MaxInstanceNumberFilter.java (revision 2)
@@ -0,0 +1,31 @@
+package de.ugoe.cs.cpdp.versions;
+
+/**
+ * Applies to large data sets. All data sets that have more than the required maximum number of instances are removed.
+ * @author Steffen Herbold
+ */
+public class MaxInstanceNumberFilter extends AbstractVersionFilter {
+
+ /**
+ * maximum number of instances required
+ */
+ private int maxInstances = 0;
+
+ /**
+ * @see de.ugoe.cs.cpdp.versions.IVersionFilter#apply(de.ugoe.cs.cpdp.versions.SoftwareVersion)
+ */
+ @Override
+ public boolean apply(SoftwareVersion version) {
+ return version.getInstances().numInstances()>maxInstances;
+ }
+
+ /**
+ * Sets the minimal number of instances.
+ * @param parameters number of instances
+ */
+ @Override
+ public void setParameter(String parameters) {
+ maxInstances = Integer.parseInt(parameters);
+ }
+
+}
Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/MinInstanceNumberFilter.java
===================================================================
--- trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/MinInstanceNumberFilter.java (revision 2)
+++ trunk/CrossPare/src/de/ugoe/cs/cpdp/versions/MinInstanceNumberFilter.java (revision 2)
@@ -0,0 +1,31 @@
+package de.ugoe.cs.cpdp.versions;
+
+/**
+ * Applies to small data sets. All data sets that do not have the required minimal number of instances are removed.
+ * @author Steffen Herbold
+ */
+public class MinInstanceNumberFilter extends AbstractVersionFilter {
+
+ /**
+ * minimal number of instances required
+ */
+ private int minInstances = 0;
+
+ /**
+ * @see de.ugoe.cs.cpdp.versions.IVersionFilter#apply(de.ugoe.cs.cpdp.versions.SoftwareVersion)
+ */
+ @Override
+ public boolean apply(SoftwareVersion version) {
+ return version.getInstances().numInstances()= (1-quantil) ||
+ ((double) counts[0])/instances.numInstances() <= (quantil);
+ }
+
+}
Index: trunk/CrossPare/test/de/ugoe/cs/cpdp/ExperimentConfigurationTest.java
===================================================================
--- trunk/CrossPare/test/de/ugoe/cs/cpdp/ExperimentConfigurationTest.java (revision 2)
+++ trunk/CrossPare/test/de/ugoe/cs/cpdp/ExperimentConfigurationTest.java (revision 2)
@@ -0,0 +1,37 @@
+package de.ugoe.cs.cpdp;
+
+import java.io.File;
+
+import org.junit.Test;
+
+import static org.junit.Assert.*;
+
+public class ExperimentConfigurationTest {
+
+ @Test
+ public void validateConfigurations() {
+ /*validateConfigurations("config_shared");
+ validateConfigurations("config_singleclassifier");
+ validateConfigurations("config_bagging");*/
+ validateConfigurations("exp-java/config");
+ validateConfigurations("exp-nasa/config");
+ }
+
+ public void validateConfigurations(String folder) {
+ File configFolder = new File(folder);
+ boolean error = false;
+ for( File configFile : configFolder.listFiles() ) {
+ try {
+ if( configFile.isFile() ) {
+ new ExperimentConfiguration(configFile);
+ }
+ } catch (Exception e) {
+ System.err.println("Failure initializing the experiment configuration for configuration file " + configFile);
+ e.printStackTrace();
+ error = true;
+ }
+ }
+ if(error) fail();
+ }
+
+}
Index: trunk/CrossPare/test/de/ugoe/cs/cpdp/dataselection/PetersFilterTest.java
===================================================================
--- trunk/CrossPare/test/de/ugoe/cs/cpdp/dataselection/PetersFilterTest.java (revision 2)
+++ trunk/CrossPare/test/de/ugoe/cs/cpdp/dataselection/PetersFilterTest.java (revision 2)
@@ -0,0 +1,61 @@
+package de.ugoe.cs.cpdp.dataselection;
+
+import static org.junit.Assert.*;
+
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.Set;
+
+import org.junit.Test;
+
+import weka.core.Attribute;
+import weka.core.DenseInstance;
+import weka.core.Instances;
+
+public class PetersFilterTest {
+
+ @SuppressWarnings("deprecation")
+ @Test
+ public void testSetParameter() {
+ new PetersFilter().setParameter("somestring");
+ }
+
+ @SuppressWarnings("deprecation")
+ @Test
+ public void testApply() {
+ ArrayList attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ attributes.add(new Attribute("class"));
+
+ Instances testdata = new Instances("test", attributes, 0);
+ testdata.setClassIndex(1);
+ testdata.add(new DenseInstance(1.0, new double[]{3.0, 0.0}));
+ testdata.add(new DenseInstance(1.0, new double[]{6.6, 0.0}));
+ testdata.add(new DenseInstance(1.0, new double[]{3.1, 0.0}));
+
+ Instances traindata = new Instances("train", attributes, 0);
+ traindata.setClassIndex(1);
+ traindata.add(new DenseInstance(1.0, new double[]{2.9, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{2.8, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{3.2, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{3.05, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{10.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{9.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{8.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{1.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{5.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{6.6, 0.0}));
+
+ PetersFilter filter = new PetersFilter();
+ Instances selected = filter.apply(testdata, traindata);
+
+ Set selectedSet = new HashSet<>();
+ for( int i=0 ; i attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ attributes.add(new Attribute("class"));
+
+ Instances testdata = new Instances("test", attributes, 0);
+ testdata.setClassIndex(1);
+ testdata.add(new DenseInstance(1.0, new double[]{3.0, 0.0}));
+ testdata.add(new DenseInstance(1.0, new double[]{6.6, 0.0}));
+ testdata.add(new DenseInstance(1.0, new double[]{3.1, 0.0}));
+
+ Instances traindata = new Instances("train", attributes, 0);
+ traindata.setClassIndex(1);
+ traindata.add(new DenseInstance(1.0, new double[]{2.9, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{2.8, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{3.2, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{3.05, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{10.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{9.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{8.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{1.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{5.0, 0.0}));
+
+ SetUniqueList traindataSet = SetUniqueList.setUniqueList(new LinkedList());
+ traindataSet.add(traindata);
+
+ TestAsTraining filter = new TestAsTraining();
+ filter.apply(testdata, traindataSet);
+
+ assertEquals(1, traindataSet.size());
+
+ traindata = traindataSet.get(0);
+ assertNotSame(testdata, traindata);
+ assertEquals(testdata.numInstances(), traindata.numInstances());
+ for( int i=0; i attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ attributes.add(new Attribute("class"));
+
+ Instances testdata = new Instances("test", attributes, 0);
+ testdata.setClassIndex(1);
+ testdata.add(new DenseInstance(1.0, new double[]{3.0, 0.0}));
+ testdata.add(new DenseInstance(1.0, new double[]{6.6, 0.0}));
+ testdata.add(new DenseInstance(1.0, new double[]{3.1, 0.0}));
+
+ Instances traindata = new Instances("train", attributes, 0);
+ traindata.setClassIndex(1);
+ traindata.add(new DenseInstance(1.0, new double[]{2.9, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{2.8, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{3.2, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{3.05, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{10.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{9.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{8.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{1.0, 0.0}));
+ traindata.add(new DenseInstance(1.0, new double[]{5.0, 0.0}));
+
+ TurhanFilter filter = new TurhanFilter();
+ filter.setParameter("2");
+ Instances selected = filter.apply(testdata, traindata);
+
+ Set selectedSet = new HashSet<>();
+ for( int i=0 ; i versions = loader.load();
+
+ assertEquals(65, versions.size());
+ }
+
+}
Index: trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/AttributeRemoverTest.java
===================================================================
--- trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/AttributeRemoverTest.java (revision 2)
+++ trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/AttributeRemoverTest.java (revision 2)
@@ -0,0 +1,81 @@
+package de.ugoe.cs.cpdp.preprocessing;
+
+import static org.junit.Assert.*;
+
+import java.util.ArrayList;
+import java.util.LinkedList;
+
+import org.apache.commons.collections4.list.SetUniqueList;
+import org.junit.Test;
+
+import de.ugoe.cs.cpdp.dataprocessing.AttributeRemoval;
+
+import weka.core.Attribute;
+import weka.core.DenseInstance;
+import weka.core.Instances;
+
+public class AttributeRemoverTest {
+
+ @Test
+ public void testApply_1() {
+ ArrayList attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ attributes.add(new Attribute("class"));
+ attributes.add(new Attribute("attr2"));
+ Instances instances = new Instances("test", attributes, 0);
+ instances.setClassIndex(1);
+
+ double[] value1 = new double[]{1.5, 0.0, 3.0};
+ double[] value2 = new double[]{1.4, 1.0, 6.0};
+ double[] value3 = new double[]{1.6, 0.0, 15.0};
+
+ instances.add(new DenseInstance(1.0, value1));
+ instances.add(new DenseInstance(1.0, value2));
+ instances.add(new DenseInstance(1.0, value3));
+
+ AttributeRemoval processor = new AttributeRemoval();
+ processor.setParameter("attr2");
+ processor.apply(instances, SetUniqueList.setUniqueList(new LinkedList()) );
+
+ double[] expected1 = new double[]{1.5, 0.0};
+ double[] expected2 = new double[]{1.4, 1.0};
+ double[] expected3 = new double[]{1.6, 0.0};
+
+ assertEquals(2, instances.numAttributes());
+ assertArrayEquals(expected1, instances.instance(0).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected2, instances.instance(1).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected3, instances.instance(2).toDoubleArray(), 0.0001);
+ }
+
+ @Test
+ public void testApply_2() {
+ ArrayList attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ attributes.add(new Attribute("class"));
+ attributes.add(new Attribute("attr2"));
+ Instances instances = new Instances("test", attributes, 0);
+ instances.setClassIndex(1);
+
+ double[] value1 = new double[]{1.5, 0.0, 3.0};
+ double[] value2 = new double[]{1.4, 1.0, 6.0};
+ double[] value3 = new double[]{1.6, 0.0, 15.0};
+
+ instances.add(new DenseInstance(1.0, value1));
+ instances.add(new DenseInstance(1.0, value2));
+ instances.add(new DenseInstance(1.0, value3));
+
+ AttributeRemoval processor = new AttributeRemoval();
+ processor.setParameter("attr2 attr1");
+ processor.apply(instances, SetUniqueList.setUniqueList(new LinkedList()) );
+
+ double[] expected1 = new double[]{0.0};
+ double[] expected2 = new double[]{1.0};
+ double[] expected3 = new double[]{0.0};
+
+ assertEquals(1, instances.numAttributes());
+ assertArrayEquals(expected1, instances.instance(0).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected2, instances.instance(1).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected3, instances.instance(2).toDoubleArray(), 0.0001);
+ }
+
+}
Index: trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/AverageStandardizationTest.java
===================================================================
--- trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/AverageStandardizationTest.java (revision 2)
+++ trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/AverageStandardizationTest.java (revision 2)
@@ -0,0 +1,65 @@
+package de.ugoe.cs.cpdp.preprocessing;
+
+import static org.junit.Assert.*;
+
+import java.util.ArrayList;
+import java.util.LinkedList;
+
+import org.apache.commons.collections4.list.SetUniqueList;
+import org.junit.Test;
+
+import de.ugoe.cs.cpdp.dataprocessing.AverageStandardization;
+
+import weka.core.Attribute;
+import weka.core.DenseInstance;
+import weka.core.Instances;
+
+public class AverageStandardizationTest {
+
+ @Test
+ public void testSetParameter() {
+ new AverageStandardization().setParameter("somestring");
+ }
+
+ @Test
+ public void testApply() {
+ ArrayList attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ attributes.add(new Attribute("class"));
+ attributes.add(new Attribute("attr2"));
+
+ double[] value1 = new double[]{1.5, 0.0, 3.0};
+ double[] value2 = new double[]{1.4, 1.0, 6.0};
+ double[] value3 = new double[]{1.6, 0.0, 15.0};
+ double[] value4 = new double[]{ 3.0, 0.0, 1.5};
+ double[] value5 = new double[]{ 6.0, 1.0, 1.4 };
+ double[] value6 = new double[]{15.0, 0.0, 1.6};
+
+ Instances instances = new Instances("test", attributes, 0);
+ instances.setClassIndex(1);
+
+ instances.add(new DenseInstance(1.0, value1));
+ instances.add(new DenseInstance(1.0, value2));
+ instances.add(new DenseInstance(1.0, value3));
+
+ Instances instTest = new Instances("foo", attributes, 0);
+ instTest.add(new DenseInstance(1.0, value4));
+ instTest.add(new DenseInstance(1.0, value5));
+ instTest.add(new DenseInstance(1.0, value6));
+
+ SetUniqueList instSet = SetUniqueList.setUniqueList(new LinkedList());
+ instSet.add(instTest);
+
+ AverageStandardization processor = new AverageStandardization();
+ processor.apply(instances, instSet);
+
+ double[] expected1 = new double[]{0.5625, 0.0, 8.0};
+ double[] expected2 = new double[]{1.125 , 1.0, 7.466666666666};
+ double[] expected3 = new double[]{2.8125, 0.0, 8.533333333333};
+
+ assertArrayEquals(expected1, instSet.get(0).instance(0).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected2, instSet.get(0).instance(1).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected3, instSet.get(0).instance(2).toDoubleArray(), 0.0001);
+ }
+
+}
Index: trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/LogarithmTranformTest.java
===================================================================
--- trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/LogarithmTranformTest.java (revision 2)
+++ trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/LogarithmTranformTest.java (revision 2)
@@ -0,0 +1,53 @@
+package de.ugoe.cs.cpdp.preprocessing;
+
+import static org.junit.Assert.*;
+
+import java.util.ArrayList;
+import java.util.LinkedList;
+
+import org.apache.commons.collections4.list.SetUniqueList;
+import org.junit.Test;
+
+import de.ugoe.cs.cpdp.dataprocessing.LogarithmTransform;
+
+import weka.core.Attribute;
+import weka.core.DenseInstance;
+import weka.core.Instances;
+
+public class LogarithmTranformTest {
+
+ @Test
+ public void testSetParameter() {
+ new LogarithmTransform().setParameter("somestring");
+ }
+
+ @Test
+ public void testApply() {
+ ArrayList attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ attributes.add(new Attribute("class"));
+ attributes.add(new Attribute("attr2"));
+ Instances instances = new Instances("test", attributes, 0);
+ instances.setClassIndex(1);
+
+ double[] value1 = new double[]{1.5, 0.0, 3.0};
+ double[] value2 = new double[]{1.4, 1.0, 6.0};
+ double[] value3 = new double[]{1.6, 0.0, 15.0};
+
+ instances.add(new DenseInstance(1.0, value1));
+ instances.add(new DenseInstance(1.0, value2));
+ instances.add(new DenseInstance(1.0, value3));
+
+ LogarithmTransform processor = new LogarithmTransform();
+ processor.apply(instances, SetUniqueList.setUniqueList(new LinkedList()) );
+
+ double[] expected1 = new double[]{Math.log(1.5+1), 0.0, Math.log(3.0+1)};
+ double[] expected2 = new double[]{Math.log(1.4+1), 1.0, Math.log(6.0+1)};
+ double[] expected3 = new double[]{Math.log(1.6+1), 0.0, Math.log(15.0+1)};
+
+ assertArrayEquals(expected1, instances.instance(0).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected2, instances.instance(1).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected3, instances.instance(2).toDoubleArray(), 0.0001);
+ }
+
+}
Index: trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/MedianAsReferenceTest.java
===================================================================
--- trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/MedianAsReferenceTest.java (revision 2)
+++ trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/MedianAsReferenceTest.java (revision 2)
@@ -0,0 +1,54 @@
+package de.ugoe.cs.cpdp.preprocessing;
+
+import static org.junit.Assert.*;
+
+import java.util.ArrayList;
+import java.util.LinkedList;
+
+import org.apache.commons.collections4.list.SetUniqueList;
+import org.junit.Test;
+
+import de.ugoe.cs.cpdp.dataprocessing.MedianAsReference;
+
+import weka.core.Attribute;
+import weka.core.DenseInstance;
+import weka.core.Instances;
+
+public class MedianAsReferenceTest {
+
+ @Test
+ public void testSetParameter() {
+ // simple crashstest
+ new MedianAsReference().setParameter("somestring");
+ }
+
+ @Test
+ public void testApply() {
+ ArrayList attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ attributes.add(new Attribute("class"));
+ attributes.add(new Attribute("attr2"));
+ Instances instances = new Instances("test", attributes, 0);
+ instances.setClassIndex(1);
+
+ double[] value1 = new double[]{1.5, 0.0, 3.0};
+ double[] value2 = new double[]{1.4, 1.0, 6.0};
+ double[] value3 = new double[]{1.6, 0.0, 15.0};
+
+ instances.add(new DenseInstance(1.0, value1));
+ instances.add(new DenseInstance(1.0, value2));
+ instances.add(new DenseInstance(1.0, value3));
+
+ MedianAsReference processor = new MedianAsReference();
+ processor.apply(instances, SetUniqueList.setUniqueList(new LinkedList()) );
+
+ double[] expected1 = new double[]{0.0, 0.0, -3.0};
+ double[] expected2 = new double[]{-0.1,1.0, 0.0};
+ double[] expected3 = new double[]{0.1, 0.0, 9.0};
+
+ assertArrayEquals(expected1, instances.instance(0).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected2, instances.instance(1).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected3, instances.instance(2).toDoubleArray(), 0.0001);
+ }
+
+}
Index: trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/NormalizationPreprocessorTest.java
===================================================================
--- trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/NormalizationPreprocessorTest.java (revision 2)
+++ trunk/CrossPare/test/de/ugoe/cs/cpdp/preprocessing/NormalizationPreprocessorTest.java (revision 2)
@@ -0,0 +1,54 @@
+package de.ugoe.cs.cpdp.preprocessing;
+
+import static org.junit.Assert.*;
+
+import java.util.ArrayList;
+import java.util.LinkedList;
+
+import org.apache.commons.collections4.list.SetUniqueList;
+import org.junit.Test;
+
+import de.ugoe.cs.cpdp.dataprocessing.Normalization;
+
+import weka.core.Attribute;
+import weka.core.DenseInstance;
+import weka.core.Instances;
+
+public class NormalizationPreprocessorTest {
+
+ @Test
+ public void testApply() {
+ ArrayList attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ attributes.add(new Attribute("class"));
+ attributes.add(new Attribute("attr2"));
+ Instances instances = new Instances("test", attributes, 0);
+ instances.setClassIndex(1);
+
+ double[] value1 = new double[]{1.5, 0.0, 3.0};
+ double[] value2 = new double[]{1.4, 1.0, 6.0};
+ double[] value3 = new double[]{1.6, 0.0, 15.0};
+
+ instances.add(new DenseInstance(1.0, value1));
+ instances.add(new DenseInstance(1.0, value2));
+ instances.add(new DenseInstance(1.0, value3));
+
+ Normalization processor = new Normalization();
+ processor.apply(instances, SetUniqueList.setUniqueList(new LinkedList()) );
+
+ double[] expected1 = new double[]{0.5, 0.0, 0.0};
+ double[] expected2 = new double[]{0.0, 1.0, 0.25};
+ double[] expected3 = new double[]{1.0, 0.0, 1.0};
+
+ assertArrayEquals(expected1, instances.instance(0).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected2, instances.instance(1).toDoubleArray(), 0.0001);
+ assertArrayEquals(expected3, instances.instance(2).toDoubleArray(), 0.0001);
+ }
+
+ @Test
+ public void testSetParameter() {
+ // just a simple crash test
+ new Normalization().setParameter("somestring");
+ }
+
+}
Index: trunk/CrossPare/test/de/ugoe/cs/cpdp/versions/InstanceNumberFilterTest.java
===================================================================
--- trunk/CrossPare/test/de/ugoe/cs/cpdp/versions/InstanceNumberFilterTest.java (revision 2)
+++ trunk/CrossPare/test/de/ugoe/cs/cpdp/versions/InstanceNumberFilterTest.java (revision 2)
@@ -0,0 +1,44 @@
+package de.ugoe.cs.cpdp.versions;
+
+import static org.junit.Assert.*;
+
+import java.util.List;
+
+import org.junit.Test;
+
+import de.ugoe.cs.cpdp.loader.CSVFolderLoader;
+
+public class InstanceNumberFilterTest {
+
+ @Test
+ public void testApply_1() {
+ CSVFolderLoader loader = new CSVFolderLoader();
+ loader.setLocation("data");
+ List versions = loader.load();
+
+ int sizeVersions = versions.size();
+
+ MinInstanceNumberFilter filter = new MinInstanceNumberFilter();
+ filter.apply(versions);
+
+ assertEquals(sizeVersions, versions.size());
+ }
+
+ @Test
+ public void testApply_2() {
+ CSVFolderLoader loader = new CSVFolderLoader();
+ loader.setLocation("data");
+ List versions = loader.load();
+
+ int expected = 0;
+ for( SoftwareVersion version : versions ) {
+ if(version.getInstances().numInstances() >= 100) expected++;
+ }
+
+ MinInstanceNumberFilter filter = new MinInstanceNumberFilter();
+ filter.setParameter("100");
+ filter.apply(versions);
+
+ assertEquals(expected, versions.size());
+ }
+}
Index: trunk/CrossPare/test/de/ugoe/cs/cpdp/versions/UnbalancedFilterTest.java
===================================================================
--- trunk/CrossPare/test/de/ugoe/cs/cpdp/versions/UnbalancedFilterTest.java (revision 2)
+++ trunk/CrossPare/test/de/ugoe/cs/cpdp/versions/UnbalancedFilterTest.java (revision 2)
@@ -0,0 +1,25 @@
+package de.ugoe.cs.cpdp.versions;
+
+import static org.junit.Assert.*;
+
+import java.util.List;
+
+import org.junit.Test;
+
+import de.ugoe.cs.cpdp.loader.CSVFolderLoader;
+
+public class UnbalancedFilterTest {
+
+ @Test
+ public void testApply() {
+ CSVFolderLoader loader = new CSVFolderLoader();
+ loader.setLocation("data");
+ List versions = loader.load();
+
+ UnbalancedFilter filter = new UnbalancedFilter();
+ filter.apply(versions);
+
+ assertEquals(57, versions.size());
+ }
+
+}
Index: trunk/CrossPare/test/de/ugoe/cs/cpdp/weighting/BiasedWeightsTest.java
===================================================================
--- trunk/CrossPare/test/de/ugoe/cs/cpdp/weighting/BiasedWeightsTest.java (revision 2)
+++ trunk/CrossPare/test/de/ugoe/cs/cpdp/weighting/BiasedWeightsTest.java (revision 2)
@@ -0,0 +1,110 @@
+package de.ugoe.cs.cpdp.weighting;
+
+import static org.junit.Assert.*;
+
+import java.util.ArrayList;
+import java.util.List;
+
+import org.junit.Test;
+
+import de.ugoe.cs.cpdp.dataprocessing.BiasedWeights;
+
+import weka.core.Attribute;
+import weka.core.DenseInstance;
+import weka.core.Instances;
+
+public class BiasedWeightsTest {
+
+ @Test
+ public void testApply_1() {
+ ArrayList attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ List classAttVals = new ArrayList();
+ classAttVals.add("0");
+ classAttVals.add("1");
+ attributes.add(new Attribute("bug", classAttVals));
+ attributes.add(new Attribute("attr2"));
+
+ double[] value1 = new double[]{1.5, 0.0, 3.0};
+ double[] value2 = new double[]{1.4, 1.0, 6.0};
+ double[] value3 = new double[]{1.6, 0.0, 15.0};
+ double[] value4 = new double[]{ 3.0, 0.0, 1.5};
+ double[] value5 = new double[]{ 6.0, 1.0, 1.4 };
+ double[] value6 = new double[]{15.0, 0.0, 1.6};
+ double[] value7 = new double[]{ 6.0, 0.0, 1.4 };
+ double[] value8 = new double[]{15.0, 0.0, 1.6};
+
+ Instances instances = new Instances("test", attributes, 0);
+ instances.setClassIndex(1);
+
+ instances.add(new DenseInstance(1.0, value1));
+ instances.add(new DenseInstance(1.0, value2));
+ instances.add(new DenseInstance(1.0, value3));
+ instances.add(new DenseInstance(1.0, value4));
+ instances.add(new DenseInstance(1.0, value5));
+ instances.add(new DenseInstance(1.0, value6));
+ instances.add(new DenseInstance(1.0, value7));
+ instances.add(new DenseInstance(1.0, value8));
+
+ BiasedWeights processor = new BiasedWeights();
+ processor.apply(new Instances(instances), instances);
+
+ assertEquals(0.6666666d, instances.instance(0).weight(), 0.00001);
+ assertEquals(2.0d, instances.instance(1).weight(), 0.00001);
+ assertEquals(0.6666666d, instances.instance(2).weight(), 0.00001);
+ assertEquals(0.6666666d, instances.instance(3).weight(), 0.00001);
+ assertEquals(2.0d, instances.instance(4).weight(), 0.00001);
+ assertEquals(0.6666666d, instances.instance(5).weight(), 0.00001);
+ assertEquals(0.6666666d, instances.instance(6).weight(), 0.00001);
+ assertEquals(0.6666666d, instances.instance(7).weight(), 0.00001);
+ assertEquals(instances.numInstances(), instances.sumOfWeights(), 0.0001);
+ }
+
+ @Test
+ public void testApply_2() {
+ ArrayList attributes = new ArrayList();
+ attributes.add(new Attribute("attr1"));
+ List classAttVals = new ArrayList();
+ classAttVals.add("0");
+ classAttVals.add("1");
+ attributes.add(new Attribute("bug", classAttVals));
+ attributes.add(new Attribute("attr2"));
+
+ double[] value1 = new double[]{1.5, 0.0, 3.0};
+ double[] value2 = new double[]{1.4, 1.0, 6.0};
+ double[] value3 = new double[]{1.6, 0.0, 15.0};
+ double[] value4 = new double[]{ 3.0, 0.0, 1.5};
+ double[] value5 = new double[]{ 6.0, 1.0, 1.4 };
+ double[] value6 = new double[]{15.0, 0.0, 1.6};
+ double[] value7 = new double[]{ 6.0, 0.0, 1.4 };
+ double[] value8 = new double[]{15.0, 0.0, 1.6};
+
+ Instances instances = new Instances("test", attributes, 0);
+ instances.setClassIndex(1);
+
+ instances.add(new DenseInstance(1.0, value1));
+ instances.add(new DenseInstance(1.0, value2));
+ instances.add(new DenseInstance(1.0, value3));
+ instances.add(new DenseInstance(1.0, value4));
+ instances.add(new DenseInstance(1.0, value5));
+ instances.add(new DenseInstance(1.0, value6));
+ instances.add(new DenseInstance(1.0, value7));
+ instances.add(new DenseInstance(1.0, value8));
+
+
+ BiasedWeights processor = new BiasedWeights();
+ processor.setParameter("0.7");
+ processor.apply(new Instances(instances), instances);
+
+ assertEquals(0.4, instances.instance(0).weight(), 0.00001);
+ assertEquals(2.8, instances.instance(1).weight(), 0.00001);
+ assertEquals(0.4, instances.instance(2).weight(), 0.00001);
+ assertEquals(0.4, instances.instance(3).weight(), 0.00001);
+ assertEquals(2.8, instances.instance(4).weight(), 0.00001);
+ assertEquals(0.4, instances.instance(5).weight(), 0.00001);
+ assertEquals(0.4, instances.instance(6).weight(), 0.00001);
+ assertEquals(0.4, instances.instance(7).weight(), 0.00001);
+ assertEquals(instances.numInstances(), instances.sumOfWeights(), 0.0001);
+ }
+
+}