Add files via upload

RP_DBSCAN/src/dm/kaist/algorithm/Conf.java

@@ -57,6 +57,10 @@ public static void setInputParameters(String[] args)
 			System.out.println(" -bs : the block size for virtually combining two-level cell dictionary (default : 1).");
 			System.out.println(" -l : the hdfs path to write labeled points, <point id, cluster label> (default : no output).");
 			System.exit(1);
+		}else if(rho < 0.009999)
+		{
+			System.out.println("This version deos not support rho < 0.01!. We recommend you to set rho = 0.02 or 0.01 for accuracy.");
+			System.exit(1);
 		}
 	}
 

RP_DBSCAN/src/dm/kaist/algorithm/Methods.java

@@ -182,13 +182,15 @@ public static class PseudoRandomPartition implements PairFunction<Tuple2<List<In
 		private int dim = 0;
 		private float levelpSideLen = 0;
 		private int metaBlockWindow = 0;
+		private String pairOutputPath = null;
 
-		public PseudoRandomPartition(int dim, float epsilon, float p, int metaBlockWindow) {
+		public PseudoRandomPartition(int dim, float epsilon, float p, int metaBlockWindow, String pairOutputPath) {
 			// TODO Auto-generated constructor stub
 			this.dim = dim;
 		    int LOWEST_LEVEL = (int)(Math.ceil(1 - Math.log10(p)/Math.log10(2)));
 		    this.levelpSideLen = epsilon / ((float)Math.sqrt(dim) * (1 << LOWEST_LEVEL-1));
 		    this.metaBlockWindow = metaBlockWindow;
+		    this.pairOutputPath = pairOutputPath;
 		}
 
 		@Override
@@ -206,10 +208,14 @@ public Tuple2<List<Integer>, ApproximatedCell> call(Tuple2<List<Integer>, Iterab
 				{
 					ApproximatedPoint apprPt = new ApproximatedPoint(pt.id, pt.coords);
 					map.put(lvH, apprPt);
-					apprPt.ptsIds = new ArrayList<Long>();
+
+					if(pairOutputPath != null)
+						apprPt.ptsIds = new ArrayList<Long>();
 				}
 				map.get(lvH).count++;
-				map.get(lvH).ptsIds.add(pt.id);
+
+				if(pairOutputPath != null)
+					map.get(lvH).ptsIds.add(pt.id);
 			}
 
 			ApproximatedCell cell = new ApproximatedCell(pts._1);
@@ -745,7 +751,8 @@ public void findDDRWithSpecificMeta(int readOrder, HashSet<Edge> edges, List<App
 							closest = neighbor.lv_p_kdtree.cloestNode(node.coords,sqr_r);
 
 						if(node == null || closest == null || Norm.sqr_L2_norm(node.coords, closest.coords) <= sqr_r)
-							edges.add(edge);
+								edges.add(edge);
+
 					}
 				}
 				meta = null;

RP_DBSCAN/src/dm/kaist/algorithm/RP_DBSCAN.java

@@ -86,7 +86,7 @@ public void phaseI()
 			.combineByKey(new Methods.CreateLocalApproximatedPoint(Conf.dim, Conf.epsilon, Conf.rho), new Methods.LocalApproximation(Conf.dim, Conf.epsilon, Conf.rho), new Methods.GlobalApproximation(Conf.dim))
 			.mapToPair(new Methods.PseudoRandomPartition2(Conf.metaBlockWindow)).persist(StorageLevel.MEMORY_AND_DISK_SER());
 		}else
-			dataMap = lines.mapToPair(new Methods.PointToCell(Conf.dim, Conf.epsilon)).groupByKey().mapToPair(new Methods.PseudoRandomPartition(Conf.dim, Conf.epsilon, Conf.rho, Conf.metaBlockWindow)).persist(StorageLevel.MEMORY_AND_DISK_SER());
+			dataMap = lines.mapToPair(new Methods.PointToCell(Conf.dim, Conf.epsilon)).groupByKey().mapToPair(new Methods.PseudoRandomPartition(Conf.dim, Conf.epsilon, Conf.rho, Conf.metaBlockWindow, Conf.pairOutputPath)).persist(StorageLevel.MEMORY_AND_DISK_SER());
 
 		numOfCells = dataMap.count();
 

RP_DBSCAN/src/dm/kaist/io/FileIO.java

@@ -66,14 +66,19 @@ public static List<String> broadCastData(JavaSparkContext sc, Configuration conf
 		FileStatus[] status = fs.listStatus(new Path(dirPath));
 		List<String> metaPaths = new ArrayList<String>();
 
+		long size = 0;
+
 		for(int i=0; i<status.length; i++)
 		{
 			String path = status[i].getPath().toString();
 			String fileName = status[i].getPath().getName();
 			sc.addFile(path);
 			metaPaths.add(fileName);
+
+			size += status[i].getLen();
 		}
 
+		System.out.println("size : " + size);
 		return metaPaths;
 	}
 }