Update log messages

Author: oshando

swan-pipeline/src/main/java/de/fraunhofer/iem/swan/features/WekaFeatureSet.java

@@ -6,6 +6,11 @@
 import de.fraunhofer.iem.swan.io.dataset.Dataset;
 import de.fraunhofer.iem.swan.model.ModelEvaluator;
 import de.fraunhofer.iem.swan.util.Util;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+import weka.attributeSelection.AttributeSelection;
+import weka.attributeSelection.InfoGainAttributeEval;
+import weka.attributeSelection.Ranker;
 import weka.core.Attribute;
 import weka.core.Instances;
 import weka.core.converters.ArffLoader;
@@ -20,6 +25,7 @@
 public class WekaFeatureSet extends FeatureSet implements IFeatureSet {
 
     HashMap<String, Instances> structures;
+    private static final Logger logger = LoggerFactory.getLogger(WekaFeatureSet.class);
 
     public WekaFeatureSet(Dataset dataset, SwanOptions options) {
         super(dataset, options, ModelEvaluator.Toolkit.WEKA);
@@ -57,6 +63,9 @@ public void createFeatures() {
 
             ArffLoader loader = new ArffLoader();
 
+            logger.info("Using default {} TRAIN dataset(s) file(s) in {}",
+                    options.getAllClasses(), options.getArffInstancesFiles());
+
             for (Category category : options.getAllClasses().stream().map(Category::fromText).collect(Collectors.toList())) {
 
                 List<String> instancesFile = options.getArffInstancesFiles().stream().filter(c -> c.contains(category.getId().toLowerCase())).collect(Collectors.toList());
@@ -74,8 +83,8 @@ public void createFeatures() {
                             ArffLoader arffLoader = new ArffLoader();
                             arffLoader.setSource(new File(instancesFile.get(x)));
 
-                            trainInstances = mergeInstances(trainInstances, arffLoader.getDataSet());
-                            structure = mergeInstances(structure, arffLoader.getStructure());
+                            trainInstances = joinInstances(trainInstances, arffLoader.getDataSet());
+                            structure = joinInstances(structure, arffLoader.getStructure());
                         }
                     }
 

swan-pipeline/src/main/java/de/fraunhofer/iem/swan/model/MonteCarloValidator.java

@@ -6,6 +6,7 @@
 import weka.classifiers.evaluation.output.prediction.CSV;
 import weka.core.Instances;
 
+import java.text.DecimalFormat;
 import java.util.ArrayList;
 import java.util.Collections;
 import java.util.HashMap;
@@ -18,6 +19,7 @@ public class MonteCarloValidator {
 
     private ArrayList<String> predictions;
     private HashMap<String, ArrayList<Double>> fMeasure;
+    private DecimalFormat df = new DecimalFormat("####0.00");
 
     public MonteCarloValidator() {
         predictions = new ArrayList<>();
@@ -98,9 +100,9 @@ public void updateResultSet(Instances instances, Evaluation eval) {
 
             if (!currentClass.contentEquals("0")) {
                 if (!fMeasure.containsKey(currentClass))
-                    fMeasure.put(currentClass, new ArrayList<>(Collections.singletonList(Double.isNaN(eval.fMeasure(c)) ? 0 : eval.fMeasure(c))));
+                    fMeasure.put(currentClass, new ArrayList<>(Collections.singletonList(Double.isNaN(eval.fMeasure(c)) ? 0 : Double.parseDouble(df.format(eval.fMeasure(c))))));
                 else {
-                    fMeasure.get(currentClass).add(Double.isNaN(eval.fMeasure(c)) ? 0 : eval.fMeasure(c));
+                    fMeasure.get(currentClass).add(Double.isNaN(eval.fMeasure(c)) ? 0 : Double.parseDouble(df.format(eval.fMeasure(c))));
                 }
             }
         }

swan-pipeline/src/main/java/de/fraunhofer/iem/swan/model/toolkit/Weka.java

@@ -51,11 +51,13 @@ public HashMap<String, HashMap<String, String>> trainModel() {
             case VALIDATE:
 
                 //Phase 1: classify SRM classes
+                logger.info("Performing {}-fold cross-validation for {} using WEKA", options.getIterations(), options.getSrmClasses());
                 for (String srm : options.getSrmClasses())
                     runManualEvaluation(features.getTrainInstances().get(srm));
 
                 //Filter methods from CWE instances that were not classified into one of the SRM classes
                 //Phase 2: classify CWE classes
+                logger.info("Performing {}-fold cross-validation for {} using WEKA", options.getIterations(), options.getCweClasses());
                 for (String cwe : options.getCweClasses())
                     runManualEvaluation(features.getTrainInstances().get(cwe));