Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaClusterTraining2.java =================================================================== --- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaClusterTraining2.java (revision 18) +++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaClusterTraining2.java (revision 19) @@ -67,6 +67,6 @@ private EM clusterer = null; - private HashMap cclassifier = new HashMap(); - private HashMap ctraindata = new HashMap(); + private HashMap cclassifier; + private HashMap ctraindata; @@ -115,4 +115,6 @@ 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); @@ -141,4 +143,7 @@ // 3. cluster data //Console.traceln(Level.INFO, String.format("starting clustering")); + + cclassifier = new HashMap(); + ctraindata = new HashMap(); // use standard params for now @@ -173,4 +178,7 @@ } + // 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 clusternumber = ctraindata.keySet().iterator(); @@ -180,5 +188,5 @@ 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")); } } Index: trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaLocalTraining2.java =================================================================== --- trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaLocalTraining2.java (revision 18) +++ trunk/CrossPare/src/de/ugoe/cs/cpdp/training/WekaLocalTraining2.java (revision 19) @@ -70,14 +70,14 @@ /* classifier per cluster */ - private HashMap cclassifier = new HashMap(); + private HashMap cclassifier; /* instances per cluster */ - private HashMap ctraindata = new HashMap(); + private HashMap ctraindata; /* holds the instances and indices of the pivot objects of the Fastmap calculation in buildClassifier*/ - private HashMap cpivots = new HashMap(); + private HashMap 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 */ @@ -144,5 +144,4 @@ Fastmap FMAP = new Fastmap(2); EuclideanDistance dist = new EuclideanDistance(traindata); - // we set our pivot indices [x=0,y=1][dimension] @@ -234,5 +233,4 @@ */ - // 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 @@ -244,5 +242,5 @@ // 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); @@ -256,9 +254,9 @@ // 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) @@ -306,4 +304,8 @@ this.show_biggest = true; + cclassifier = new HashMap(); + ctraindata = new HashMap(); + cpivots = new HashMap(); + cpivotindices = new int[2][2]; // 1. copy traindata @@ -339,9 +341,11 @@ double[][] X = FMAP.getX(); + distmat = new double[0][0]; + System.gc(); // quadtree payload generation ArrayList> qtp = new ArrayList>(); - // 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}; @@ -365,5 +369,5 @@ } - 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 @@ -410,5 +414,5 @@ // 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>(QuadTree.csize); @@ -436,4 +440,6 @@ Iterator clusternumber = ctraindata.keySet().iterator(); //cclassifier.clear(); + + //int traindata_count = 0; while ( clusternumber.hasNext() ) { cnumber = clusternumber.next(); @@ -441,8 +447,10 @@ 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)); } }