dl4jtest1

gistfile1.txt

int rngSeed = 123; // random number seed for reproducibility

            MultiLayerConfiguration conf = new NeuralNetConfiguration.Builder()
                .seed(rngSeed) //include a random seed for reproducibility
                               // use stochastic gradient descent as an optimization algorithm
                .optimizationAlgo(OptimizationAlgorithm.STOCHASTIC_GRADIENT_DESCENT)
                .iterations(1)
                .learningRate(0.006) //specify the learning rate
                .updater(org.deeplearning4j.nn.conf.Updater.NESTEROVS).momentum(0.9) //specify the rate of change of the learning rate.
                .regularization(true).l2(1e-4)
                .list()
                .layer(0, new DenseLayer.Builder() //create the first, input layer with xavier initialization
                        .nIn(1)
                        .nOut(2)
                        .activation(Activation.SIGMOID)
                        .weightInit(WeightInit.XAVIER)
                        .build())
                .layer(1, new OutputLayer.Builder(LossFunction.NEGATIVELOGLIKELIHOOD) //create hidden layer
                        .nIn(2)
                        .nOut(1)
                        .activation(Activation.SIGMOID)
                        .weightInit(WeightInit.XAVIER)
                        //.dist(new UniformDistribution(0, 1))
                        .build())
                .pretrain(false).backprop(true) //use backpropagation to adjust weights
                .build();

            MultiLayerNetwork model = new MultiLayerNetwork(conf);
            model.init();
            //print the score with every 1 iteration
            model.setListeners(new ScoreIterationListener(100));

            double[][] input = new double[2][];
            input[0] = new double[] { 0D };
            input[1] = new double[] { 1D };

            double[][] output = new double[2][];
            output[0] = new double[] { 0D };
            output[1] = new double[] { 1D };

            INDArray ndIn = Nd4j.create(input);
            INDArray ndOut = Nd4j.create(output);

            for (int i=0;i<50;i++)
            {
                model.fit(ndIn, ndOut);
            }

            //int[] res = model.predict(Nd4j.create(input[0]));
            //string msg = string.Format("{0} {1}",res.Length, res[0]);
            //Console.WriteLine(msg);

            INDArray res2 = model.output(Nd4j.create(input[0]));
            string msg = string.Format("{0}", res2.getDouble(0,0));
            Console.WriteLine(msg);