Examples with NeuralInputs org.dmg.pmml.neural_network.NeuralInputs used on opensource projects

Example 1 with NeuralInputs

use of org.dmg.pmml.neural_network.NeuralInputs in project jpmml-r by jpmml.

the class ElmNNConverter method encodeModel.

@Override
public NeuralNetwork encodeModel(Schema schema) {
    RGenericVector elmNN = getObject();
    RDoubleVector inpweight = (RDoubleVector) elmNN.getValue("inpweight");
    RDoubleVector biashid = (RDoubleVector) elmNN.getValue("biashid");
    RDoubleVector outweight = (RDoubleVector) elmNN.getValue("outweight");
    RStringVector actfun = (RStringVector) elmNN.getValue("actfun");
    RDoubleVector nhid = (RDoubleVector) elmNN.getValue("nhid");
    Label label = schema.getLabel();
    List<? extends Feature> features = schema.getFeatures();
    switch(actfun.asScalar()) {
        case "purelin":
            break;
        default:
            throw new IllegalArgumentException();
    }
    NeuralInputs neuralInputs = NeuralNetworkUtil.createNeuralInputs(features, DataType.DOUBLE);
    List<? extends Entity> entities = neuralInputs.getNeuralInputs();
    List<NeuralLayer> neuralLayers = new ArrayList<>(2);
    NeuralLayer hiddenNeuralLayer = new NeuralLayer();
    int rows = ValueUtil.asInt(nhid.asScalar());
    int columns = 1 + features.size();
    for (int row = 0; row < rows; row++) {
        List<Double> weights = FortranMatrixUtil.getRow(inpweight.getValues(), rows, columns, row);
        Double bias = biashid.getValue(row);
        bias += weights.remove(0);
        Neuron neuron = NeuralNetworkUtil.createNeuron(entities, weights, bias).setId("hidden/" + String.valueOf(row + 1));
        hiddenNeuralLayer.addNeurons(neuron);
    }
    neuralLayers.add(hiddenNeuralLayer);
    entities = hiddenNeuralLayer.getNeurons();
    NeuralLayer outputNeuralLayer = new NeuralLayer();
    // XXX
    columns = 1;
    for (int column = 0; column < columns; column++) {
        List<Double> weights = FortranMatrixUtil.getColumn(outweight.getValues(), rows, columns, column);
        Double bias = Double.NaN;
        Neuron neuron = NeuralNetworkUtil.createNeuron(entities, weights, bias).setId("output/" + String.valueOf(column + 1));
        outputNeuralLayer.addNeurons(neuron);
    }
    neuralLayers.add(outputNeuralLayer);
    entities = outputNeuralLayer.getNeurons();
    NeuralOutputs neuralOutputs = NeuralNetworkUtil.createRegressionNeuralOutputs(entities, (ContinuousLabel) label);
    NeuralNetwork neuralNetwork = new NeuralNetwork(MiningFunction.REGRESSION, NeuralNetwork.ActivationFunction.IDENTITY, ModelUtil.createMiningSchema(label), neuralInputs, neuralLayers).setNeuralOutputs(neuralOutputs);
    return neuralNetwork;
}

Example 2 with NeuralInputs

use of org.dmg.pmml.neural_network.NeuralInputs in project jpmml-sparkml by jpmml.

the class MultilayerPerceptronClassificationModelConverter method encodeModel.

@Override
public NeuralNetwork encodeModel(Schema schema) {
    MultilayerPerceptronClassificationModel model = getTransformer();
    int[] layers = model.layers();
    Vector weights = model.weights();
    CategoricalLabel categoricalLabel = (CategoricalLabel) schema.getLabel();
    if (categoricalLabel.size() != layers[layers.length - 1]) {
        throw new IllegalArgumentException();
    }
    List<? extends Feature> features = schema.getFeatures();
    if (features.size() != layers[0]) {
        throw new IllegalArgumentException();
    }
    NeuralInputs neuralInputs = NeuralNetworkUtil.createNeuralInputs(features, DataType.DOUBLE);
    List<? extends Entity> entities = neuralInputs.getNeuralInputs();
    List<NeuralLayer> neuralLayers = new ArrayList<>();
    int weightPos = 0;
    for (int layer = 1; layer < layers.length; layer++) {
        NeuralLayer neuralLayer = new NeuralLayer();
        int rows = entities.size();
        int columns = layers[layer];
        List<List<Double>> weightMatrix = new ArrayList<>();
        for (int column = 0; column < columns; column++) {
            List<Double> weightVector = new ArrayList<>();
            for (int row = 0; row < rows; row++) {
                weightVector.add(weights.apply(weightPos + (row * columns) + column));
            }
            weightMatrix.add(weightVector);
        }
        weightPos += (rows * columns);
        for (int column = 0; column < columns; column++) {
            List<Double> weightVector = weightMatrix.get(column);
            Double bias = weights.apply(weightPos);
            Neuron neuron = NeuralNetworkUtil.createNeuron(entities, weightVector, bias).setId(String.valueOf(layer) + "/" + String.valueOf(column + 1));
            neuralLayer.addNeurons(neuron);
            weightPos++;
        }
        if (layer == (layers.length - 1)) {
            neuralLayer.setActivationFunction(NeuralNetwork.ActivationFunction.IDENTITY).setNormalizationMethod(NeuralNetwork.NormalizationMethod.SOFTMAX);
        }
        neuralLayers.add(neuralLayer);
        entities = neuralLayer.getNeurons();
    }
    if (weightPos != weights.size()) {
        throw new IllegalArgumentException();
    }
    NeuralNetwork neuralNetwork = new NeuralNetwork(MiningFunction.CLASSIFICATION, NeuralNetwork.ActivationFunction.LOGISTIC, ModelUtil.createMiningSchema(categoricalLabel), neuralInputs, neuralLayers).setNeuralOutputs(NeuralNetworkUtil.createClassificationNeuralOutputs(entities, categoricalLabel));
    return neuralNetwork;
}

Example 3 with NeuralInputs

use of org.dmg.pmml.neural_network.NeuralInputs in project shifu by ShifuML.

the class NeuralNetworkModelIntegrator method getNeuralInputs.

private NeuralInputs getNeuralInputs(final NeuralNetwork model) {
    NeuralInputs nnInputs = new NeuralInputs();
    // get HashMap for local transform and MiningSchema fields
    HashMap<FieldName, FieldName> reversMiningTransformMap = new HashMap<FieldName, FieldName>();
    HashMap<FieldName, List<FieldName>> treeMapOfTransform = new HashMap<FieldName, List<FieldName>>();
    for (DerivedField dField : model.getLocalTransformations().getDerivedFields()) {
        // Apply z-scale normalization on numerical variables
        FieldName parentField = null;
        if (dField.getExpression() instanceof NormContinuous) {
            parentField = ((NormContinuous) dField.getExpression()).getField();
            reversMiningTransformMap.put(dField.getName(), parentField);
        } else // Apply bin map on categorical variables
        if (dField.getExpression() instanceof MapValues) {
            parentField = ((MapValues) dField.getExpression()).getFieldColumnPairs().get(0).getField();
            reversMiningTransformMap.put(dField.getName(), parentField);
        } else if (dField.getExpression() instanceof Discretize) {
            parentField = ((Discretize) dField.getExpression()).getField();
            reversMiningTransformMap.put(dField.getName(), parentField);
        }
        List<FieldName> fieldNames = treeMapOfTransform.get(parentField);
        if (fieldNames == null) {
            fieldNames = new ArrayList<FieldName>();
        }
        fieldNames.add(dField.getName());
        treeMapOfTransform.put(parentField, fieldNames);
    }
    // comment here
    List<MiningField> miningList = model.getMiningSchema().getMiningFields();
    int index = 0;
    for (DerivedField dField : model.getLocalTransformations().getDerivedFields()) {
        List<FieldName> list = treeMapOfTransform.get(dField.getName());
        boolean isLeaf = (list == null || list.size() == 0);
        FieldName root = getRoot(dField.getName(), reversMiningTransformMap);
        if (isLeaf && isRootInMiningList(root, miningList)) {
            DerivedField field = new DerivedField(OpType.CONTINUOUS, DataType.DOUBLE).setName(dField.getName()).setExpression(new FieldRef(dField.getName()));
            nnInputs.addNeuralInputs(new NeuralInput("0," + (index++), field));
        }
    }
    DerivedField field = new DerivedField(OpType.CONTINUOUS, DataType.DOUBLE).setName(new FieldName(PluginConstants.biasValue)).setExpression(new FieldRef(new FieldName(PluginConstants.biasValue)));
    nnInputs.addNeuralInputs(new NeuralInput(PluginConstants.biasValue, field));
    return nnInputs;
}