NNRegressionTest.java

package func.test;

import shared.ConvergenceTrainer;
import shared.DataSet;
import shared.Instance;
import shared.SumOfSquaresError;
import func.nn.backprop.*;

/**
 * An XOR test
 * @author Andrew Guillory gtg008g@mail.gatech.edu
 * @version 1.0
 */
public class NNRegressionTest {
	
	/**
	 * Tests out the perceptron with the classic xor test
	 * @param args ignored
	 */
	public static void main(String[] args) {
		BackPropagationNetworkFactory factory = 
			new BackPropagationNetworkFactory();
		double[][][] data = {
			{ { 1, 1 }, { -1 } },
			{ { 1, 0 }, { 1 } },
			{ { 0, 1 }, { 1 } },
			{ { 0, 0 }, { -1 } }
		};
        Instance[] patterns = new Instance[data.length];
        for (int i = 0; i < patterns.length; i++) {
            patterns[i] = new Instance(data[i][0]);
            patterns[i].setLabel(new Instance(data[i][1]));
        }
        BackPropagationNetwork network = factory.createRegressionNetwork(
           new int[] { 2, 2, 1 });
        DataSet set = new DataSet(patterns);
        ConvergenceTrainer trainer = new ConvergenceTrainer(
               new BatchBackPropagationTrainer(set, network,
                   new SumOfSquaresError(), new RPROPUpdateRule()));
        trainer.train();
        System.out.println("Convergence in " 
            + trainer.getIterations() + " iterations");
        for (int i = 0; i < patterns.length; i++) {
            network.setInputValues(patterns[i].getData());
            network.run();
            System.out.println("~~");
            System.out.println(patterns[i].getLabel());
            System.out.println(network.getOutputValues());
        }
	}

}