public static void main(String[] args) throws Exception { //Define a simple ComputationGraph: ComputationGraphConfiguration conf = new NeuralNetConfiguration.Builder() .weightInit(WeightInit.XAVIER) .updater(new Nesterovs(0.01, 0.9)) .graphBuilder() .addInputs("in") .addLayer("layer0", new DenseLayer.Builder().nIn(4).nOut(3).activation(Activation.TANH).build(), "in") .addLayer("layer1", new OutputLayer.Builder(LossFunctions.LossFunction.NEGATIVELOGLIKELIHOOD).activation(Activation.SOFTMAX).nIn(3).nOut(3).build(), "layer0") .setOutputs("layer1") .backprop(true).pretrain(false).build(); ComputationGraph net = new ComputationGraph(conf); net.init(); //Save the model File locationToSave = new File("model/MyComputationGraph.zip"); //Where to save the network. Note: the file is in .zip format - can be opened externally boolean saveUpdater = true; //Updater: i.e., the state for Momentum, RMSProp, Adagrad etc. Save this if you want to train your network more in the future ModelSerializer.writeModel(net, locationToSave, saveUpdater); //Load the model ComputationGraph restored = ModelSerializer.restoreComputationGraph(locationToSave); System.out.println("Saved and loaded parameters are equal: " + net.params().equals(restored.params())); System.out.println("Saved and loaded configurations are equal: " + net.getConfiguration().equals(restored.getConfiguration())); }
.nIn((int) (numHiddenNodes * Math.pow(reduction, 4))).nOut(numOutputsSomaticFrequency).build(), "dense5") .setOutputs(getOutputNames()) .pretrain(false).backprop(true).build();
gb.setInputTypes(InputType.inferInputType(features)); ComputationGraphConfiguration conf = gb.pretrain(false).backprop(true).build(); ComputationGraph model = new ComputationGraph(conf); model.init();
.gradientNormalization(GradientNormalization.RenormalizeL2PerLayer).build(), "embeddings") .setOutputs("lossLayer").backprop(true).pretrain(false) .setInputTypes(InputType.convolutional(inputShape[2], inputShape[1], inputShape[0]));
.nIn(nOut4).nOut(numOutputsIsMutated).build(), "dense5") .setOutputs(getOutputNames()) .pretrain(false).backprop(true).build();
.nIn(nOut4).nOut(numOutputsSomaticFrequency).build(), "dense5") .setOutputs(getOutputNames()) .pretrain(false).backprop(true).build();
.nOut(numClasses).activation(Activation.SOFTMAX).build(), "avgpool") .setOutputs("output").backprop(true).pretrain(false);
.addLayer("output", new OutputLayer.Builder(LossFunctions.LossFunction.NEGATIVELOGLIKELIHOOD) .nIn(1024).nOut(numLabels).activation(Activation.SOFTMAX).build(), "fc1") .setOutputs("output").backprop(true).pretrain(false);
@Override public GraphConfiguration getValue(double[] values) { //Create ComputationGraphConfiguration... NeuralNetConfiguration.Builder builder = randomGlobalConf(values); ComputationGraphConfiguration.GraphBuilder graphBuilder = builder.graphBuilder(); graphBuilder.addInputs(this.networkInputs); graphBuilder.setOutputs(this.networkOutputs); if (inputTypes != null) graphBuilder.setInputTypes(inputTypes.getValue(values)); //Build/add our layers and vertices: for (LayerConf c : layerSpaces) { org.deeplearning4j.nn.conf.layers.Layer l = c.layerSpace.getValue(values); graphBuilder.addLayer(c.getLayerName(), l, c.getInputs()); } for (VertexConf gv : vertices) { graphBuilder.addVertex(gv.getVertexName(), gv.getGraphVertex(), gv.getInputs()); } if (backprop != null) graphBuilder.backprop(backprop.getValue(values)); if (pretrain != null) graphBuilder.pretrain(pretrain.getValue(values)); if (backpropType != null) graphBuilder.backpropType(backpropType.getValue(values)); if (tbpttFwdLength != null) graphBuilder.tBPTTForwardLength(tbpttFwdLength.getValue(values)); if (tbpttBwdLength != null) graphBuilder.tBPTTBackwardLength(tbpttBwdLength.getValue(values)); ComputationGraphConfiguration configuration = graphBuilder.build(); return new GraphConfiguration(configuration, earlyStoppingConfiguration, numEpochs); }
/** * Build the multilayer network defined by the networkconfiguration and the list of layers. */ protected void createModel() throws Exception { final INDArray features = getFirstBatchFeatures(trainData); ComputationGraphConfiguration.GraphBuilder gb = netConfig.builder().seed(getSeed()).graphBuilder(); // Set ouput size final Layer lastLayer = layers[layers.length - 1]; final int nOut = trainData.numClasses(); if (lastLayer instanceof FeedForwardLayer) { ((FeedForwardLayer) lastLayer).setNOut(nOut); } if (getInstanceIterator() instanceof CnnTextEmbeddingInstanceIterator) { makeCnnTextLayerSetup(gb); } else { makeDefaultLayerSetup(gb); } gb.setInputTypes(InputType.inferInputType(features)); ComputationGraphConfiguration conf = gb.pretrain(false).backprop(true).build(); ComputationGraph model = new ComputationGraph(conf); model.init(); this.model = model; }
@Override public ComputationGraph init() { int embeddingSize = 128; ComputationGraphConfiguration.GraphBuilder graph = graphBuilder("input1"); graph.addInputs("input1").setInputTypes(InputType.convolutional(inputShape[2], inputShape[1], inputShape[0])) // Logits .addLayer("bottleneck", new DenseLayer.Builder().nIn(5376).nOut(embeddingSize).build(), "avgpool") // Embeddings .addVertex("embeddings", new L2NormalizeVertex(new int[] {1}, 1e-10), "bottleneck") // Output .addLayer("outputLayer", new CenterLossOutputLayer.Builder() .lossFunction(LossFunctions.LossFunction.NEGATIVELOGLIKELIHOOD) .activation(Activation.SOFTMAX).alpha(0.9).lambda(1e-4) .nIn(embeddingSize).nOut(numClasses).build(), "embeddings") .setOutputs("outputLayer").backprop(true).pretrain(false); ComputationGraphConfiguration conf = graph.build(); ComputationGraph model = new ComputationGraph(conf); model.init(); return model; }