label less fpof & pbad

pom.xml

@@ -35,13 +35,6 @@
 				<groupId>org.springframework.boot</groupId>
 				<artifactId>spring-boot-maven-plugin</artifactId>
 			</plugin>
-			<!--
-			<plugin>
- 				<groupId>org.eclipse.jetty</groupId>
-  				<artifactId>jetty-maven-plugin</artifactId>
-  				<version>9.4.26-SNAPSHOT</version>
-			</plugin>
-			-->
 		</plugins>
 		<resources>
 			<resource>
@@ -52,14 +45,6 @@
 	</build>
 
 	<dependencies>
-		<!-- internal -->
-		<!--
-		<dependency>
-			<groupId>be.uantwerpen</groupId>
-			<artifactId>QCSP-pub</artifactId>
-			<version>1.0.0</version>
-		</dependency>
-		 -->
 		<dependency>
 			<groupId>com.h2database</groupId>
 			<artifactId>h2</artifactId>

src/main/java/be/uantwerpen/datamining/pattern_mining/RunPBADEmbeddingOnly.java

@@ -82,8 +82,8 @@ public class RunPBADEmbeddingOnly {
 				.filter((p) -> p.getType().equals("sequential patterns"))
 				.map((p) -> new File(Settings.FILE_FOLDER + p.getFilename()).getAbsolutePath())
 				.collect(Collectors.toList());
-		boolean onlyItemsets = item.getPatterns().size() > 0 && itemsetFilenames.size() == item.getPatterns().size();
-		boolean onlySequentialPatterns = item.getPatterns().size() > 0  && sequentialPatternFilenames.size() == item.getPatterns().size();
+		boolean onlyItemsets = !item.getPatterns().isEmpty() && itemsetFilenames.size() == item.getPatterns().size();
+		boolean onlySequentialPatterns = !item.getPatterns().isEmpty()  && sequentialPatternFilenames.size() == item.getPatterns().size();
 		if(onlyItemsets) {
 			type = "itemset";
 		}

src/main/java/be/uantwerpen/mime_webapp/controller/PatternMiningController.java

@@ -574,7 +574,11 @@ public class PatternMiningController {
 		long elapsed = timer.elapsed();
 		//3.2 get evaluation metrics
 		Map<String,Double> results = computeAnomalyEvaluationMetrics(scoreFile, current_item.getFile(), "FPOF");
-		label = label.substring(0,label.length()-1) + String.format(Locale.ENGLISH, "; AUC=%.3f, AP=%.3f)", results.get("AUC"), results.get("AP"));
+
+		if(results.containsKey("AUC") && results.containsKey("AP"))
+			label = label.substring(0,label.length()-1) + String.format(Locale.ENGLISH, "; AUC=%.3f, AP=%.3f", results.get("AUC"), results.get("AP"));
+		else label = label.substring(0,label.length()-1) + "; Unkown precision missing labels";
+
 		//3.3  save project
 		AnomalyScores score = new AnomalyScores();
 		score.setAlgorithm("FPOF");
@@ -585,7 +589,9 @@ public class PatternMiningController {
 		current_item.getScores().add(score);
 		current_item.getStackOperations().add("Anomaly detection " + label);
 		repository.save();
+		if(results.containsKey("AUC") && results.containsKey("AP"))
 			return String.format(Locale.ENGLISH, "Finished FPOF. Took %s. AUC=%.3f, AP=%.3f", Utils.milisToStringReadable(elapsed), results.get("AUC"), results.get("AP"));
+		else return String.format(Locale.ENGLISH, "Finished FPOF. Took %s. Unkown precision missing labels", Utils.milisToStringReadable(elapsed));
 	}
 
 	@RequestMapping(value="/rest/mining/anomaly-detection-pbad", method=RequestMethod.POST)
@@ -624,7 +630,11 @@ public class PatternMiningController {
 		current_item.getScores().add(score);
 		current_item.getStackOperations().add("Anomaly detection " + label);
 		repository.save();
+		if(results.containsKey("AUC") && results.containsKey("AP")) {
 			return String.format("Finished PBAD. Took %s. AUC=%.3f, AP=%.3f", Utils.milisToStringReadable(elapsed), results.get("AUC"), results.get("AP"));
+		} else {
+			return String.format("Finished PBAD. Took %s. Unevluated missing labels", Utils.milisToStringReadable(elapsed));
+		}
 	}
 
 	private Map<String,Double> computeAnomalyEvaluationMetrics(File scoreFile, File itemFile, String method) throws IOException, InterruptedException {
@@ -677,17 +687,16 @@ public class PatternMiningController {
 			//AP: 0.259
 			//<Finished took  1.04 seconds
 
-			Map<String,Double> evalMap = new HashMap<>();
-			try {
+			if(logStr.contains("AUC") && logStr.contains("AP")) {
+				Map<String,Double> evalMap = new HashMap<>(2);
+				//last line
 				Double auc = Double.valueOf(logStr.get(logStr.size()-3).split(":")[1]);
 				Double ap = Double.valueOf(logStr.get(logStr.size()-2).split(":")[1]);
-				//last line
 				evalMap.put("AUC", auc);
 				evalMap.put("AP",ap);
 				return evalMap;
-			}catch(Exception e) {
-				throw new RuntimeException(e);
 			}
+			return new HashMap<>();
 		}
 	}
 

src/main/python/compute_anomaly_evaluation.py

@@ -49,9 +49,9 @@ if __name__ == '__main__':
     window_labels = df['Label'].values
     ixl = np.where(window_labels != 0)[0]
 
-
     # the roc score is a methode of evaluating reliability but it doesn't work if we didn't properly fill in the labels
-    auc = roc_auc_score(y_true=window_labels[ixl], y_score=scores[ixl]) if has_multiple_values(window_labels) else 0
+    if has_multiple_values(window_labels):
+        auc = roc_auc_score(y_true=window_labels[ixl], y_score=scores[ixl])
         ap = average_precision_score(y_true=window_labels[ixl], y_score=scores[ixl])
         print("AUC: {:.3f}".format(auc))
         print("AP: {:.3f}".format(ap))