musad - 経済統計練習帳

package v2;
 
import org.encog.Encog;
import org.encog.engine.network.activation.ActivationSigmoid;
import org.encog.ml.data.MLData;
import org.encog.ml.data.MLDataPair;
import org.encog.ml.data.MLDataSet;
import org.encog.ml.data.basic.BasicMLDataSet;
import org.encog.neural.networks.BasicNetwork;
import org.encog.neural.networks.layers.BasicLayer;
import org.encog.neural.networks.training.propagation.resilient.ResilientPropagation;
 
public class pro{
 
public static void main(String[]args){
pro t=new pro();
}
 
pro(){
 
 
	double XOR_INPUT[][] = { { 0.0, 0.0 }, { 1.0, 0.0 },
			{ 0.0, 1.0 }, { 1.0, 1.0 } };
 
 
	 double XOR_IDEAL[][] = { { 0.0 }, { 1.0 }, { 1.0 }, { 0.0 } };
 
 
		BasicNetwork net = new BasicNetwork();
		net.addLayer(new BasicLayer(null ,false ,2 ));
		net.addLayer(new BasicLayer(new ActivationSigmoid( ),true,3));
                net.addLayer(new BasicLayer(new ActivationSigmoid( ),false,3));
		net.addLayer(new BasicLayer(new ActivationSigmoid( ),false,1));
		net.getStructure().finalizeStructure();
		net.reset();
 
		MLDataSet ts = new BasicMLDataSet(XOR_INPUT, XOR_IDEAL);
 
		final ResilientPropagation train = new ResilientPropagation(net, ts);
 
		int epoch = 1;
 
 
                int mode=0;
		do {
			train.iteration();
			System.out.println("(" + epoch + ") =>" + train.getError());
			epoch= epoch+1;
                        if(train.getError()<0.01)epoch=1000;
		} while( epoch<100);
		train.finishTraining();
 
 
                MLDataSet tsx = new BasicMLDataSet(XOR_INPUT, XOR_IDEAL);
 
		for(MLDataPair pair: tsx ) {
			final MLData output = net.compute(pair.getInput());
			System.out.println(pair.getInput().getData(0) + "," + pair.getInput().getData(1));
		System.out.println(pair.getIdeal().getData(0));
                        System.out.println(output.getData(0));
                }
 
		Encog.getInstance().shutdown();
	}
}