/**
* Bagging method.
*
* @param data the training data to be used for generating the bagged classifier.
* @exception Exception if the classifier could not be built successfully
*/
@Override
public void buildClassifier(Instances data) throws Exception {
super.buildClassifier(data);
if (m_CalcOutOfBag && (m_BagSizePercent != 100)) {
throw new IllegalArgumentException(
"Bag size needs to be 100% if " + "out-of-bag error is to be calculated!");
}
double outOfBagCount = 0.0;
double errorSum = 0.0;
int bagSize = data.numInstances() * m_BagSizePercent / 100;
Random random = new Random(m_Seed);
for (int j = 0; j < m_Classifiers.length; j++) {
Instances bagData = null;
boolean[] inBag = null;
// create the in-bag dataset
if (m_CalcOutOfBag) {
inBag = new boolean[data.numInstances()];
bagData = resampleWithWeights(data, random, inBag);
} else {
bagData = data.resampleWithWeights(random);
if (bagSize < data.numInstances()) {
bagData.randomize(random);
Instances newBagData = new Instances(bagData, 0, bagSize);
bagData = newBagData;
}
}
if (m_Classifier instanceof Randomizable) {
((Randomizable) m_Classifiers[j]).setSeed(random.nextInt());
}
// build the classifier
m_Classifiers[j].buildClassifier(bagData);
if (m_CalcOutOfBag) {
// calculate out of bag error
for (int i = 0; i < inBag.length; i++) {
if (!inBag[i]) {
Instance outOfBagInst = data.instance(i);
outOfBagCount += outOfBagInst.weight();
if (data.classAttribute().isNumeric()) {
errorSum +=
outOfBagInst.weight()
* Math.abs(
m_Classifiers[j].classifyInstance(outOfBagInst)
- outOfBagInst.classValue());
} else {
if (m_Classifiers[j].classifyInstance(outOfBagInst) != outOfBagInst.classValue()) {
errorSum += outOfBagInst.weight();
}
}
}
}
}
}
m_OutOfBagError = errorSum / outOfBagCount;
}
/**
* Parses a given list of options. Valid options are:
*
* <p>-W classname <br>
* Specify the full class name of a weak classifier as the basis for bagging (required).
*
* <p>-I num <br>
* Set the number of bagging iterations (default 10).
*
* <p>-S seed <br>
* Random number seed for resampling (default 1).
*
* <p>-P num <br>
* Size of each bag, as a percentage of the training size (default 100).
*
* <p>-O <br>
* Compute out of bag error.
*
* <p>Options after -- are passed to the designated classifier.
*
* <p>
*
* @param options the list of options as an array of strings
* @exception Exception if an option is not supported
*/
@Override
public void setOptions(String[] options) throws Exception {
String bagSize = Utils.getOption('P', options);
if (bagSize.length() != 0) {
setBagSizePercent(Integer.parseInt(bagSize));
} else {
setBagSizePercent(100);
}
setCalcOutOfBag(Utils.getFlag('O', options));
super.setOptions(options);
}