Changeset 19 for trunk/CrossPare/src/de

Timestamp:

10/16/14 10:37:01 (10 years ago)

Author:

atrautsch

Message:

Update wegen Speicherverbrauch, Debug Output auskommentiert.

Location:

trunk/CrossPare/src/de/ugoe/cs/cpdp/training

Files:

• WekaClusterTraining2.java (modified) (5 diffs)
• WekaLocalTraining2.java (modified) (11 diffs)

trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaClusterTraining2.java

@@ -67 +67 @@
                 private EM clusterer = null;
 
-                private HashMap<Integer, Classifier> cclassifier = new HashMap<Integer, Classifier>();
-                private HashMap<Integer, Instances> ctraindata = new HashMap<Integer, Instances>(); 
+                private HashMap<Integer, Classifier> cclassifier;
+                private HashMap<Integer, Instances> ctraindata; 
                 
                 
@@ -115 +115 @@
                                 int cnum = clusterer.clusterInstance(clusterInstance);
                                 
+                                //Console.traceln(Level.INFO, String.format("instance is in cluster: " + cnum));
+                                                
                                 // 2. classify testata instance to the classifier
                                 ret = cclassifier.get(cnum).classifyInstance(classInstance);
@@ -141 +143 @@
                         // 3. cluster data
                         //Console.traceln(Level.INFO, String.format("starting clustering"));
+                        
+                        cclassifier = new HashMap<Integer, Classifier>();
+                        ctraindata = new HashMap<Integer, Instances>();
                         
                         // use standard params for now
@@ -173 +178 @@
                         }
                         
+                        // Debug output
+                        //Console.traceln(Level.INFO, String.format("number of clusters: " + clusterer.numberOfClusters()));
+                        
                         // train one classifier per cluster, we get the clusternumber from the traindata
                         Iterator<Integer> clusternumber = ctraindata.keySet().iterator();
@@ -180 +188 @@
                                 cclassifier.get(cnumber).buildClassifier(ctraindata.get(cnumber));
                                 
-                                //Console.traceln(Level.INFO, String.format("classifier in cluster "+cnumber));
+                                //Console.traceln(Level.INFO, String.format("building classifier in cluster "+cnumber + " with " + ctraindata.get(cnumber).size() + " traindata instances"));
                         }
                 }

trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaLocalTraining2.java

@@ -70 +70 @@
                 
                 /* classifier per cluster */
-                private HashMap<Integer, Classifier> cclassifier = new HashMap<Integer, Classifier>();
+                private HashMap<Integer, Classifier> cclassifier;
                 
                 /* instances per cluster */
-                private HashMap<Integer, Instances> ctraindata = new HashMap<Integer, Instances>(); 
+                private HashMap<Integer, Instances> ctraindata; 
                 
                 /* holds the instances and indices of the pivot objects of the Fastmap calculation in buildClassifier*/
-                private HashMap<Integer, Instance> cpivots = new HashMap<Integer, Instance>();
+                private HashMap<Integer, Instance> cpivots;
                 
                 /* holds the indices of the pivot objects for x,y and the dimension [x,y][dimension]*/
-                private int[][] cpivotindices = new int[2][2];
+                private int[][] cpivotindices;
 
                 /* holds the sizes of the cluster multiple "boxes" per cluster */
@@ -144 +144 @@
                                 Fastmap FMAP = new Fastmap(2);
                                 EuclideanDistance dist = new EuclideanDistance(traindata);
-                                
                                 
                                 // we set our pivot indices [x=0,y=1][dimension]
@@ -234 +233 @@
                             */
                                 
-                                // TODO: can we be in more cluster than one?
                                 // now we iterate over all clusters (well, boxes of sizes per cluster really) and save the number of the 
                                 // cluster in which we are
@@ -244 +242 @@
                                         
                                         // now iterate over the boxes of the cluster and hope we find one (cluster could have been removed)
-                                        // or we are too far away from any cluster
+                                        // or we are too far away from any cluster because of the fastmap calculation with the initial pivot objects
                                         for ( int box=0; box < this.csize.get(cnumber).size(); box++ ) { 
                                                 Double[][] current = this.csize.get(cnumber).get(box);
@@ -256 +254 @@
                                 
                                 // we want to count how often we are really inside a cluster
-                                if ( found_cnumber == -1 ) {
-                                        CNOTFOUND += 1;
-                                }else {
-                                        CFOUND += 1;
-                                }
+                                //if ( found_cnumber == -1 ) {
+                                //      CNOTFOUND += 1;
+                                //}else {
+                                //      CFOUND += 1;
+                                //}
 
                                 // now it can happen that we dont find a cluster because we deleted it previously (too few instances)
@@ -306 +304 @@
                         this.show_biggest = true;
                         
+                        cclassifier = new HashMap<Integer, Classifier>();
+                        ctraindata = new HashMap<Integer, Instances>();
+                        cpivots = new HashMap<Integer, Instance>();
+                        cpivotindices = new int[2][2];
                         
                         // 1. copy traindata
@@ -339 +341 @@
                         
                         double[][] X = FMAP.getX();
+                        distmat = new double[0][0];
+                        System.gc();
                         
                         // quadtree payload generation
                         ArrayList<QuadTreePayload<Instance>> qtp = new ArrayList<QuadTreePayload<Instance>>();
                     
-                        // die max und min brauchen wir für die größenangaben der sektoren
+                        // we need these for the sizes of the quadrants
                         double[] big = {0,0};
                         double[] small = {9999999,99999999};
@@ -365 +369 @@
                     }
                     
-                    Console.traceln(Level.INFO, String.format("size for cluster ("+small[0]+","+small[1]+") - ("+big[0]+","+big[1]+")"));
+                    //Console.traceln(Level.INFO, String.format("size for cluster ("+small[0]+","+small[1]+") - ("+big[0]+","+big[1]+")"));
                     
                     // 5. generate quadtree
@@ -410 +414 @@
                         
                         // here we keep things we need later on
-                        // QuadTree sizes for later use
+                        // QuadTree sizes for later use (matching new instances)
                         this.csize = new HashMap<Integer, ArrayList<Double[][]>>(QuadTree.csize);
                 
@@ -436 +440 @@
                         Iterator<Integer> clusternumber = ctraindata.keySet().iterator();
                         //cclassifier.clear();
+                        
+                        //int traindata_count = 0;
                         while ( clusternumber.hasNext() ) {
                                 cnumber = clusternumber.next();
@@ -441 +447 @@
                                 cclassifier.get(cnumber).buildClassifier(ctraindata.get(cnumber));
                                 //Console.traceln(Level.INFO, String.format("classifier in cluster "+cnumber));
-                                //Console.traceln(Level.INFO, String.format("" + ctraindata.get(cnumber).size() + " instances in cluster "+cnumber));
-                        }
-                        
-                        //Console.traceln(Level.INFO, String.format("num clusters: "+cclassifier.size()));
+                                //traindata_count += ctraindata.get(cnumber).size();
+                                //Console.traceln(Level.INFO, String.format("building classifier in cluster "+cnumber +"  with "+ ctraindata.get(cnumber).size() +" traindata instances"));
+                        }
+                        
+                        // add all traindata
+                        //Console.traceln(Level.INFO, String.format("traindata in all clusters: " + traindata_count));
                 }
         }