Examples with ComputationGraph org.deeplearning4j.nn.graph.ComputationGraph used on opensource projects

Example 71 with ComputationGraph

use of org.deeplearning4j.nn.graph.ComputationGraph in project deeplearning4j by deeplearning4j.

the class TestCustomLayers method testCustomOutputLayerCG.

@Test
public void testCustomOutputLayerCG() {
    //Create a ComputationGraphConfiguration with custom output layer, and check JSON and YAML config actually works...
    ComputationGraphConfiguration conf = new NeuralNetConfiguration.Builder().seed(12345).learningRate(0.1).graphBuilder().addInputs("in").addLayer("0", new DenseLayer.Builder().nIn(10).nOut(10).build(), "in").addLayer("1", new CustomOutputLayer.Builder(LossFunctions.LossFunction.MCXENT).nIn(10).nOut(10).build(), "0").setOutputs("1").pretrain(false).backprop(true).build();
    String json = conf.toJson();
    String yaml = conf.toYaml();
    System.out.println(json);
    ComputationGraphConfiguration confFromJson = ComputationGraphConfiguration.fromJson(json);
    assertEquals(conf, confFromJson);
    ComputationGraphConfiguration confFromYaml = ComputationGraphConfiguration.fromYaml(yaml);
    assertEquals(conf, confFromYaml);
    //Third: check initialization
    Nd4j.getRandom().setSeed(12345);
    ComputationGraph net = new ComputationGraph(conf);
    net.init();
    assertTrue(net.getLayer(1) instanceof CustomOutputLayerImpl);
    //Fourth: compare to an equivalent standard output layer (should be identical)
    ComputationGraphConfiguration conf2 = new NeuralNetConfiguration.Builder().seed(12345).learningRate(0.1).graphBuilder().addInputs("in").addLayer("0", new DenseLayer.Builder().nIn(10).nOut(10).build(), "in").addLayer("1", new OutputLayer.Builder(LossFunctions.LossFunction.MCXENT).nIn(10).nOut(10).build(), "0").setOutputs("1").pretrain(false).backprop(true).build();
    Nd4j.getRandom().setSeed(12345);
    ComputationGraph net2 = new ComputationGraph(conf2);
    net2.init();
    assertEquals(net2.params(), net.params());
    INDArray testFeatures = Nd4j.rand(1, 10);
    INDArray testLabels = Nd4j.zeros(1, 10);
    testLabels.putScalar(0, 3, 1.0);
    DataSet ds = new DataSet(testFeatures, testLabels);
    assertEquals(net2.output(testFeatures)[0], net.output(testFeatures)[0]);
    assertEquals(net2.score(ds), net.score(ds), 1e-6);
}

Example 72 with ComputationGraph

use of org.deeplearning4j.nn.graph.ComputationGraph in project deeplearning4j by deeplearning4j.

the class CenterLossOutputLayerTest method getCNNMnistConfig.

public ComputationGraph getCNNMnistConfig() {
    // Number of input channels
    int nChannels = 1;
    // The number of possible outcomes
    int outputNum = 10;
    ComputationGraphConfiguration conf = // Training iterations as above
    new NeuralNetConfiguration.Builder().seed(12345).iterations(1).regularization(true).l2(0.0005).learningRate(0.01).weightInit(WeightInit.XAVIER).optimizationAlgo(OptimizationAlgorithm.STOCHASTIC_GRADIENT_DESCENT).updater(Updater.NESTEROVS).momentum(0.9).graphBuilder().addInputs("input").setInputTypes(InputType.convolutionalFlat(28, 28, 1)).addLayer("0", new ConvolutionLayer.Builder(5, 5).nIn(nChannels).stride(1, 1).nOut(20).activation(Activation.IDENTITY).build(), "input").addLayer("1", new SubsamplingLayer.Builder(SubsamplingLayer.PoolingType.MAX).kernelSize(2, 2).stride(2, 2).build(), "0").addLayer("2", new ConvolutionLayer.Builder(5, 5).stride(1, 1).nOut(50).activation(Activation.IDENTITY).build(), "1").addLayer("3", new SubsamplingLayer.Builder(SubsamplingLayer.PoolingType.MAX).kernelSize(2, 2).stride(2, 2).build(), "2").addLayer("4", new DenseLayer.Builder().activation(Activation.RELU).nOut(500).build(), "3").addLayer("output", new org.deeplearning4j.nn.conf.layers.CenterLossOutputLayer.Builder(LossFunction.MCXENT).nOut(outputNum).activation(Activation.SOFTMAX).build(), "4").setOutputs("output").build();
    ComputationGraph graph = new ComputationGraph(conf);
    graph.init();
    return graph;
}

Example 73 with ComputationGraph

use of org.deeplearning4j.nn.graph.ComputationGraph in project deeplearning4j by deeplearning4j.

the class CenterLossOutputLayerTest method testLambdaConf.

@Test
public void testLambdaConf() {
    double[] lambdas = new double[] { 0.1, 0.01 };
    double[] results = new double[2];
    int numClasses = 2;
    INDArray input = Nd4j.rand(150, 4);
    INDArray labels = Nd4j.zeros(150, numClasses);
    Random r = new Random(12345);
    for (int i = 0; i < 150; i++) {
        labels.putScalar(i, r.nextInt(numClasses), 1.0);
    }
    ComputationGraph graph;
    for (int i = 0; i < lambdas.length; i++) {
        graph = getGraph(numClasses, lambdas[i]);
        graph.setInput(0, input);
        graph.setLabel(0, labels);
        graph.computeGradientAndScore();
        results[i] = graph.score();
    }
    assertNotEquals(results[0], results[1]);
}

Example 74 with ComputationGraph

use of org.deeplearning4j.nn.graph.ComputationGraph in project deeplearning4j by deeplearning4j.

the class FrozenLayerTest method testFrozenCompGraph.

@Test
public void testFrozenCompGraph() {
    DataSet randomData = new DataSet(Nd4j.rand(10, 4), Nd4j.rand(10, 3));
    NeuralNetConfiguration.Builder overallConf = new NeuralNetConfiguration.Builder().learningRate(0.1).optimizationAlgo(OptimizationAlgorithm.STOCHASTIC_GRADIENT_DESCENT).updater(Updater.SGD).activation(Activation.IDENTITY);
    ComputationGraph modelToFineTune = new ComputationGraph(overallConf.graphBuilder().addInputs("layer0In").addLayer("layer0", new DenseLayer.Builder().nIn(4).nOut(3).build(), "layer0In").addLayer("layer1", new DenseLayer.Builder().nIn(3).nOut(2).build(), "layer0").addLayer("layer2", new DenseLayer.Builder().nIn(2).nOut(3).build(), "layer1").addLayer("layer3", new org.deeplearning4j.nn.conf.layers.OutputLayer.Builder(LossFunctions.LossFunction.MCXENT).activation(Activation.SOFTMAX).nIn(3).nOut(3).build(), "layer2").setOutputs("layer3").build());
    modelToFineTune.init();
    INDArray asFrozenFeatures = modelToFineTune.feedForward(randomData.getFeatures(), false).get("layer1");
    ComputationGraph modelNow = new TransferLearning.GraphBuilder(modelToFineTune).setFeatureExtractor("layer1").build();
    ComputationGraph notFrozen = new ComputationGraph(overallConf.graphBuilder().addInputs("layer0In").addLayer("layer0", new DenseLayer.Builder().nIn(2).nOut(3).build(), "layer0In").addLayer("layer1", new org.deeplearning4j.nn.conf.layers.OutputLayer.Builder(LossFunctions.LossFunction.MCXENT).activation(Activation.SOFTMAX).nIn(3).nOut(3).build(), "layer0").setOutputs("layer1").build());
    notFrozen.init();
    notFrozen.setParams(Nd4j.hstack(modelToFineTune.getLayer("layer2").params(), modelToFineTune.getLayer("layer3").params()));
    int i = 0;
    while (i < 5) {
        notFrozen.fit(new DataSet(asFrozenFeatures, randomData.getLabels()));
        modelNow.fit(randomData);
        i++;
    }
    assertEquals(Nd4j.hstack(modelToFineTune.getLayer("layer0").params(), modelToFineTune.getLayer("layer1").params(), notFrozen.params()), modelNow.params());
}

Example 75 with ComputationGraph

use of org.deeplearning4j.nn.graph.ComputationGraph in project deeplearning4j by deeplearning4j.

the class TestGraphNodes method testLastTimeStepVertex.

@Test
public void testLastTimeStepVertex() {
    ComputationGraphConfiguration conf = new NeuralNetConfiguration.Builder().graphBuilder().addInputs("in").addVertex("lastTS", new LastTimeStepVertex("in"), "in").addLayer("out", new OutputLayer.Builder().nIn(1).nOut(1).build(), "lastTS").setOutputs("out").build();
    ComputationGraph graph = new ComputationGraph(conf);
    graph.init();
    //First: test without input mask array
    Nd4j.getRandom().setSeed(12345);
    INDArray in = Nd4j.rand(new int[] { 3, 5, 6 });
    INDArray expOut = in.get(NDArrayIndex.all(), NDArrayIndex.all(), NDArrayIndex.point(5));
    GraphVertex gv = graph.getVertex("lastTS");
    gv.setInputs(in);
    //Forward pass:
    INDArray outFwd = gv.doForward(true);
    assertEquals(expOut, outFwd);
    //Backward pass:
    gv.setEpsilon(expOut);
    Pair<Gradient, INDArray[]> pair = gv.doBackward(false);
    INDArray eps = pair.getSecond()[0];
    assertArrayEquals(in.shape(), eps.shape());
    assertEquals(Nd4j.zeros(3, 5, 5), eps.get(NDArrayIndex.all(), NDArrayIndex.all(), NDArrayIndex.interval(0, 4, true)));
    assertEquals(expOut, eps.get(NDArrayIndex.all(), NDArrayIndex.all(), NDArrayIndex.point(5)));
    //Second: test with input mask array
    INDArray inMask = Nd4j.zeros(3, 6);
    inMask.putRow(0, Nd4j.create(new double[] { 1, 1, 1, 0, 0, 0 }));
    inMask.putRow(1, Nd4j.create(new double[] { 1, 1, 1, 1, 0, 0 }));
    inMask.putRow(2, Nd4j.create(new double[] { 1, 1, 1, 1, 1, 0 }));
    graph.setLayerMaskArrays(new INDArray[] { inMask }, null);
    expOut = Nd4j.zeros(3, 5);
    expOut.putRow(0, in.get(NDArrayIndex.point(0), NDArrayIndex.all(), NDArrayIndex.point(2)));
    expOut.putRow(1, in.get(NDArrayIndex.point(1), NDArrayIndex.all(), NDArrayIndex.point(3)));
    expOut.putRow(2, in.get(NDArrayIndex.point(2), NDArrayIndex.all(), NDArrayIndex.point(4)));
    gv.setInputs(in);
    outFwd = gv.doForward(true);
    assertEquals(expOut, outFwd);
    String json = conf.toJson();
    ComputationGraphConfiguration conf2 = ComputationGraphConfiguration.fromJson(json);
    assertEquals(conf, conf2);
}