package tr.gov.ulakbim.jDenetX.classifiers;
import tr.gov.ulakbim.jDenetX.classifiers.attributes.AttributeClassObserver;
import weka.core.Instance;
public class RandomHoeffdingOptionTree extends HoeffdingOptionTree {
private static final long serialVersionUID = 1L;
public static class RandomLearningNode extends ActiveLearningNode {
private static final long serialVersionUID = 1L;
protected double weightSeenAtLastSplitEvaluation;
protected int[] listAttributes;
protected int numAttributes;
public RandomLearningNode(double[] initialClassObservations) {
super(initialClassObservations);
}
@Override
public void learnFromInstance(Instance inst, HoeffdingOptionTree ht) {
this.observedClassDistribution.addToValue((int) inst.classValue(),
inst.weight());
if (this.listAttributes == null) {
this.numAttributes = (int) Math.floor(Math.sqrt(inst.numAttributes()));
this.listAttributes = new int[this.numAttributes];
for (int j = 0; j < this.numAttributes; j++) {
boolean isUnique = false;
while (isUnique == false) {
this.listAttributes[j] = ht.classifierRandom.nextInt(inst.numAttributes() - 1);
isUnique = true;
for (int i = 0; i < j; i++) {
if (this.listAttributes[j] == this.listAttributes[i]) {
isUnique = false;
break;
}
}
}
}
}
for (int j = 0; j < this.numAttributes - 1; j++) {
int i = this.listAttributes[j];
int instAttIndex = modelAttIndexToInstanceAttIndex(i, inst);
AttributeClassObserver obs = this.attributeObservers.get(i);
if (obs == null) {
obs = inst.attribute(instAttIndex).isNominal() ? ht
.newNominalClassObserver() : ht
.newNumericClassObserver();
this.attributeObservers.set(i, obs);
}
obs.observeAttributeClass(inst.value(instAttIndex), (int) inst
.classValue(), inst.weight());
}
}
}
public RandomHoeffdingOptionTree() {
this.removePoorAttsOption = null;
}
@Override
protected LearningNode newLearningNode(double[] initialClassObservations) {
return new RandomLearningNode(initialClassObservations);
}
@Override
public boolean isRandomizable() {
return true;
}
}