version 0.1

pom.xml

@@ -4,7 +4,7 @@
 	<groupId>de.charite.compbio</groupId>
 	<artifactId>weka-GWAVA</artifactId>
 	<packaging>jar</packaging>
-	<version>0.1-SNAPSHOT</version>
+	<version>0.1</version>
 	<name>weka-GWAVA</name>
 	<url>https://charite.github.io/weka-GWAVA/</url>
 	<licenses>

src/main/java/weka/classifiers/meta/GWAVABagging.java

@@ -5,25 +5,31 @@
 import weka.filters.Filter;
 import weka.filters.supervised.instance.SpreadSubsample;
 
+/**
+ * Extended class for bagging. Important that for every training set the majority class must be subsampled to the
+ * minority class.
+ * 
+ * @author <a href="mailto:[email protected]">Max Schubach</a>
+ *
+ */
 public class GWAVABagging extends Bagging {
 
 	/**
 	 * for serialization
 	 */
 	private static final long serialVersionUID = -3101726478201686871L;
-
-
+
 	@Override
 	protected synchronized Instances getTrainingSet(int iteration) throws Exception {
-		Instances bagData =  super.getTrainingSet(iteration);
-		
+		Instances bagData = super.getTrainingSet(iteration);
+
 		SpreadSubsample subsample = new SpreadSubsample();
-		
+
 		subsample.setRandomSeed(m_Seed);
 		subsample.setDistributionSpread(1);
 		subsample.setInputFormat(bagData);
 		return Filter.useFilter(bagData, subsample);
-		
+
 	}
 
 }

src/main/java/weka/classifiers/trees/GWAVARandomForest.java

@@ -1,10 +1,16 @@
 package weka.classifiers.trees;
 
-import weka.classifiers.trees.RandomForest;
 import weka.core.Instances;
 import weka.filters.Filter;
 import weka.filters.supervised.instance.SpreadSubsample;
 
+/**
+ * The GWAVA random forest. Important that for every training set the majority class must be subsampled to the minority
+ * class.
+ * 
+ * @author <a href="mailto:[email protected]">Max Schubach</a>
+ *
+ */
 public class GWAVARandomForest extends RandomForest {
 
 	/**

src/test/java/weka/classifiers/meta/GWAVABaggingTest.java

@@ -17,6 +17,13 @@
 import weka.classifiers.trees.J48;
 import weka.core.Instances;
 
+/**
+ * 
+ * test class for {@link GWAVABagging}
+ * 
+ * @author <a href="mailto:[email protected]">Max Schubach</a>
+ *
+ */
 public class GWAVABaggingTest {
 
 	private static Instances data;
@@ -25,6 +32,11 @@ public class GWAVABaggingTest {
 	private static int seed = 42;
 	private int folds = 10;
 
+	/**
+	 * set up
+	 * 
+	 * @throws Exception
+	 */
 	@Before
 	public void setUp() throws Exception {
 		File file = new File(Resources.getResource(diabetesFile).getPath());
@@ -38,6 +50,11 @@ public void setUp() throws Exception {
 		randData.randomize(rand);
 	}
 
+	/**
+	 * test the tree
+	 * 
+	 * @throws Exception
+	 */
 	@Test
 	public void classifyJ48Test() throws Exception {
 

src/test/java/weka/classifiers/trees/GWAVARandomForestTest.java

@@ -19,6 +19,13 @@
 import weka.classifiers.Evaluation;
 import weka.core.Instances;
 
+/**
+ * 
+ * test class for {@link GWAVARandomForest}
+ * 
+ * @author <a href="mailto:[email protected]">Max Schubach</a>
+ *
+ */
 public class GWAVARandomForestTest {
 
 	private Instances randDiabetesData;
@@ -30,6 +37,12 @@ public class GWAVARandomForestTest {
 	private int seed = 42;
 	private int folds = 10;
 
+	/**
+	 * 
+	 * setup (load data etc.)
+	 * 
+	 * @throws Exception
+	 */
 	@Before
 	public void setUp() throws Exception {
 		File file = new File(Resources.getResource(diabetesFile).getPath());
@@ -60,6 +73,11 @@ public void setUp() throws Exception {
 		randGeneratedImbalancedData.randomize(rand);
 	}
 
+	/**
+	 * test GWAVA vs tree
+	 * 
+	 * @throws Exception
+	 */
 	@Test
 	public void classifyJ48Test() throws Exception {
 
@@ -69,7 +87,7 @@ public void classifyJ48Test() throws Exception {
 		eval.crossValidateModel(gwava, randDiabetesData, folds, new Random(seed));
 
 		double prcGWAVA = eval.areaUnderPRC(1);
-		double rocGWAVA= eval.areaUnderROC(1);
+		double rocGWAVA = eval.areaUnderROC(1);
 
 		eval = new Evaluation(randDiabetesData);
 		eval.crossValidateModel(new J48(), randDiabetesData, folds, new Random(seed));
@@ -80,6 +98,11 @@ public void classifyJ48Test() throws Exception {
 		assertThat(rocGWAVA, Matchers.greaterThan(rocJ48));
 	}
 
+	/**
+	 * test gwava vs tree bin data
+	 * 
+	 * @throws Exception
+	 */
 	@Test
 	public void classifyRFRandomBinDataTest() throws Exception {
 
@@ -89,17 +112,22 @@ public void classifyRFRandomBinDataTest() throws Exception {
 
 		Evaluation eval = new Evaluation(randGeneratedImbalancedBinData);
 		eval.crossValidateModel(gwava, randGeneratedImbalancedBinData, folds, new Random(seed));
-		double prcHyperSMURF = eval.areaUnderPRC(1);
-		double rocHyperSMURF = eval.areaUnderROC(1);
+		double prcGWAVA = eval.areaUnderPRC(1);
+		double rocGWAVA = eval.areaUnderROC(1);
 		eval = new Evaluation(randGeneratedImbalancedBinData);
 		eval.crossValidateModel(new J48(), randGeneratedImbalancedBinData, folds, new Random(seed));
 		double prcJ48 = eval.areaUnderPRC(1);
 		double rocJ48 = eval.areaUnderROC(1);
 
-		assertThat(prcHyperSMURF, Matchers.greaterThan(prcJ48));
-		assertThat(rocHyperSMURF, Matchers.greaterThan(rocJ48));
+		assertThat(prcGWAVA, Matchers.greaterThan(prcJ48));
+		assertThat(rocGWAVA, Matchers.greaterThan(rocJ48));
 	}
 
+	/**
+	 * test gwava vst tree using random data
+	 * 
+	 * @throws Exception
+	 */
 	@Test
 	public void classifyRFRandomDataTest() throws Exception {