Changeset 48 for trunk/CrossPare

Timestamp:

12/12/15 15:25:22 (9 years ago)

Author:

atrautsch

Message:

final attribute selection

File:

• trunk/CrossPare/src/de/ugoe/cs/cpdp/training/MetricMatchingTraining.java (modified) (8 diffs)

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

@@ -1 @@
-// Copyright 2015 Georg-August-Universität Göttingen, Germany
+// Copyright 2015 Georg-August-Universit�t G�ttingen, Germany
 //
 //   Licensed under the Apache License, Version 2.0 (the "License");
@@ -85 +85 @@
                 this.traindataSet = traindataSet;
 
-                int rank = 5; // we want at least 5 matching attributes
+                int rank = 0; // we want at least 5 matching attributes
                 int num = 0;
                 int biggest_num = 0;
@@ -93 +93 @@
                         num++;
                         tmp = new MetricMatch(traindata, testdata);
-                        //tmp.kolmogorovSmirnovTest(0.05);
-                        
+
+                        // metric selection may create error, continue to next training set
                         try {
                                 tmp.attributeSelection();
                         }catch(Exception e) {
-                                
+                                e.printStackTrace();
+                                throw new RuntimeException(e);
                         }
                         
@@ -127 +128 @@
                 this.mm = biggest;
                 Instances ilist = this.mm.getMatchedTrain();
-                Console.traceln(Level.INFO, "Chosing the trainingdata set num "+biggest_num +" with " + rank + " matching attributs, " + ilist.size() + " instances out of a possible set of " + traindataSet.size() + " sets");
+                Console.traceln(Level.INFO, "Chosing the trainingdata set num "+biggest_num +" with " + rank + " matching attributes, " + ilist.size() + " instances out of a possible set of " + traindataSet.size() + " sets");
                 
                 // replace traindataSEt
@@ -206 +207 @@
                  
                  public MetricMatch(Instances train, Instances test) {
-                         this.train = train;
-                         this.test = test;
+                         this.train = new Instances(train);  // expensive! but we are dropping the attributes
+                         this.test = new Instances(test);
                          
                          // 1. convert metrics of testdata and traindata to later use in test
@@ -327 +328 @@
                  */
                 public void attributeSelection() throws Exception {
+                        //Console.traceln(Level.INFO, "Attribute Selection on Training Attributes ("+this.train.numAttributes()+")");
                         this.attributeSelection(this.train);
+                        //Console.traceln(Level.INFO, "-----");
+                        //Console.traceln(Level.INFO, "Attribute Selection on Test Attributes ("+this.test.numAttributes()+")");
                         this.attributeSelection(this.test);
+                        //Console.traceln(Level.INFO, "-----");
                 }
                 
@@ -338 +343 @@
                         et.buildEvaluator(which);
                         //double tmp[] = new double[this.train.numAttributes()];
-                        HashMap<Integer,Double> saeval = new HashMap<Integer,Double>();
+                        HashMap<String,Double> saeval = new HashMap<String,Double>();
                         // evaluate all training attributes
                         // select top 15% of metrics
-                        for(int i=0; i < which.numAttributes() - 1; i++) { 
-                                //tmp[i] = et.evaluateAttribute(i);
-                                saeval.put(i, et.evaluateAttribute(i));
+                        for(int i=0; i < which.numAttributes(); i++) { 
+                                if(which.classIndex() != i) {
+                                        saeval.put(which.attribute(i).name(), et.evaluateAttribute(i));
+                                }
                                 //Console.traceln(Level.SEVERE, "Significance Attribute Eval: " + tmp);
                         }
                         
-                        HashMap<Integer, Double> sorted = sortByValues(saeval);
+                        HashMap<String, Double> sorted = sortByValues(saeval);
                         
                         // die letzen 15% wollen wir haben
-                        int last = (saeval.size() / 100) * 15;
-                        int drop_first = saeval.size() - last;
-                        
-                        // drop attributes above last
+                        float last = ((float)saeval.size() / 100) * 15;
+                        int drop_first = saeval.size() - (int)last;
+                        
+                        //Console.traceln(Level.INFO, "Dropping "+ drop_first + " of " + sorted.size() + " attributes (we only keep the best 15% "+last+")");
+                        /*
                         Iterator it = sorted.entrySet().iterator();
                     while (it.hasNext()) {
                         Map.Entry pair = (Map.Entry)it.next();
-                        if(drop_first > 0) {
-                                which.deleteAttributeAt((int)pair.getKey());
-                        }
-                        drop_first--;
-                    }   
-                }
+                        Console.traceln(Level.INFO, "key: "+(int)pair.getKey()+", value: " + (double)pair.getValue() + "");
+                    }*/
+                        
+                        // drop attributes above last
+                        Iterator it = sorted.entrySet().iterator();
+                    while (drop_first > 0) {
+                        Map.Entry pair = (Map.Entry)it.next();
+                        if(which.attribute((String)pair.getKey()).index() != which.classIndex()) {
+                                
+                                which.deleteAttributeAt(which.attribute((String)pair.getKey()).index());
+                                //Console.traceln(Level.INFO, "dropping attribute: "+ (String)pair.getKey());
+                        }
+                        drop_first-=1;
+                 
+                    
+                    }
+//                  //Console.traceln(Level.INFO, "Now we have " + which.numAttributes() + " attributes left (incl. class attribute!)");
+                }
+                
                 
                 private HashMap sortByValues(HashMap map) {
                List list = new LinkedList(map.entrySet());
-               // Defined Custom Comparator here
+
                Collections.sort(list, new Comparator() {
                     public int compare(Object o1, Object o2) {
@@ -374 +394 @@
                });
 
-               // Here I am copying the sorted list in HashMap
-               // using LinkedHashMap to preserve the insertion order
+
                HashMap sortedHashMap = new LinkedHashMap();
                for (Iterator it = list.iterator(); it.hasNext();) {