Classifier Training

To train a classifier, you first need a list of training samples. In order to be able to work on any type of input space, the datatype of the samples are encapsulated into a generic class called TrainingSample. This class contains the sample of generic type T and the associated label (if the sample is labeled):

public class TrainingSample<T> implements Serializable {
    T sample;
    int label;
}

Training an SVM classifier consists in calling the method train with a list of TrainingSample as parameters. For example, A SVM with GaussianKernel on vectors of double would be done as follows:

List<TrainingSample<double[]>> train = new ArrayList<TrainingSample<double[]>>();

/* here goes the code where you feed the list with labeled samples */

DoubleGaussL2 k = new DoubleGaussL2();
LaSVM<double[]> lasvm = new LaSVM<double[]>(k);
lasvm.train(train);

In this example, the evaluation of a new sample is as simple as:

double[] sample;
// filling sample

double value = lasvm.valueOf(sample);