org.dmg.pmml.tree.TreeModel Java Examples

static
public <C extends ModelConverter<? extends M> & HasTreeOptions, M extends Model<M> & TreeEnsembleModel<T>, T extends Model<T> & DecisionTreeModel> List<TreeModel> encodeDecisionTreeEnsemble(C converter, PredicateManager predicateManager, Schema schema){
	M model = converter.getTransformer();

	Schema segmentSchema = schema.toAnonymousSchema();

	List<TreeModel> treeModels = new ArrayList<>();

	T[] trees = model.trees();
	for(T tree : trees){
		TreeModel treeModel = encodeDecisionTree(converter, tree, predicateManager, segmentSchema);

		treeModels.add(treeModel);
	}

	return treeModels;
}
 

private TreeModel encodeRegression(RGenericVector frame, RIntegerVector rowNames, RIntegerVector var, RIntegerVector n, int[][] splitInfo, RNumberVector<?> splits, RIntegerVector csplit, Schema schema){
	RNumberVector<?> yval = frame.getNumericElement("yval");

	ScoreEncoder scoreEncoder = new ScoreEncoder(){

		@Override
		public Node encode(Node node, int offset){
			Number score = yval.getValue(offset);
			Number recordCount = n.getValue(offset);

			node
				.setScore(score)
				.setRecordCount(recordCount);

			return node;
		}
	};

	Node root = encodeNode(True.INSTANCE, 1, rowNames, var, n, splitInfo, splits, csplit, scoreEncoder, schema);

	TreeModel treeModel = new TreeModel(MiningFunction.REGRESSION, ModelUtil.createMiningSchema(schema.getLabel()), root);

	return configureTreeModel(treeModel);
}
 

@Override
public Model encodeModel(Schema schema){
	RGenericVector boosting = getObject();

	RGenericVector trees = boosting.getGenericElement("trees");
	RDoubleVector weights = boosting.getDoubleElement("weights");

	CategoricalLabel categoricalLabel = (CategoricalLabel)schema.getLabel();

	List<TreeModel> treeModels = encodeTreeModels(trees);

	MiningModel miningModel = new MiningModel(MiningFunction.CLASSIFICATION, ModelUtil.createMiningSchema(categoricalLabel))
		.setSegmentation(MiningModelUtil.createSegmentation(Segmentation.MultipleModelMethod.WEIGHTED_MAJORITY_VOTE, treeModels, weights.getValues()))
		.setOutput(ModelUtil.createProbabilityOutput(DataType.DOUBLE, categoricalLabel));

	return miningModel;
}
 

private TreeModel configureTreeModel(TreeModel treeModel){
	TreeModel.NoTrueChildStrategy noTrueChildStrategy = TreeModel.NoTrueChildStrategy.RETURN_LAST_PREDICTION;
	TreeModel.MissingValueStrategy missingValueStrategy;

	switch(this.useSurrogate){
		case 0:
			missingValueStrategy = TreeModel.MissingValueStrategy.NULL_PREDICTION; // XXX
			break;
		case 1:
			missingValueStrategy = TreeModel.MissingValueStrategy.LAST_PREDICTION;
			break;
		case 2:
			missingValueStrategy = null;
			break;
		default:
			throw new IllegalArgumentException();
	}

	treeModel
		.setNoTrueChildStrategy(noTrueChildStrategy)
		.setMissingValueStrategy(missingValueStrategy);

	return treeModel;
}
 

@Override
public VisitorAction visit(Node node){
	PMMLObject parent = getParent();

	if(parent instanceof TreeModel){
		TreeModel treeModel = (TreeModel)parent;

		MiningFunction miningFunction = treeModel.getMiningFunction();
		switch(miningFunction){
			case CLASSIFICATION:
				break;
			default:
				return VisitorAction.SKIP;
		}
	}

	node.setScore(parseTargetValue(node.getScore()));

	return super.visit(node);
}
 

private <P extends Number> TreeModel encodeTreeModel(MiningFunction miningFunction, ScoreEncoder<P> scoreEncoder, List<? extends Number> leftDaughter, List<? extends Number> rightDaughter, List<P> nodepred, List<? extends Number> bestvar, List<Double> xbestsplit, Schema schema){
	RGenericVector randomForest = getObject();

	Node root = encodeNode(True.INSTANCE, 0, scoreEncoder, leftDaughter, rightDaughter, bestvar, xbestsplit, nodepred, new CategoryManager(), schema);

	TreeModel treeModel = new TreeModel(miningFunction, ModelUtil.createMiningSchema(schema.getLabel()), root)
		.setMissingValueStrategy(TreeModel.MissingValueStrategy.NULL_PREDICTION)
		.setSplitCharacteristic(TreeModel.SplitCharacteristic.BINARY_SPLIT);

	if(this.compact){
		Visitor visitor = new RandomForestCompactor();

		visitor.applyTo(treeModel);
	}

	return treeModel;
}
 

public List<TreeModel> encodeTreeModels(RGenericVector trees){
	List<TreeModel> result = new ArrayList<>();

	if(trees.size() != this.schemas.size()){
		throw new IllegalArgumentException();
	}

	for(int i = 0; i < trees.size(); i++){
		RGenericVector tree = (RGenericVector)trees.getValue(i);
		Schema schema = this.schemas.get(i);

		RPartConverter converter = this.converters.get(tree);
		if(converter == null){
			throw new IllegalArgumentException();
		}

		Schema segmentSchema = schema.toAnonymousSchema();

		TreeModel treeModel = converter.encodeModel(segmentSchema);

		result.add(treeModel);
	}

	return result;
}
 

@Test
public void defaultChildMultiplePenalties() throws Exception {
	Map<FieldName, ?> arguments = createArguments("outlook", null, "temperature", null, "humidity", 70d);

	NodeScoreDistribution<?> targetValue = evaluate(TreeModel.MissingValueStrategy.DEFAULT_CHILD, 0.8d, arguments);

	assertEquals("3", targetValue.getEntityId());

	assertEquals((Double)0.9d, targetValue.getProbability("will play"));
	assertEquals((Double)0.05d, targetValue.getProbability("may play"));
	assertEquals((Double)0.05d, targetValue.getProbability("no play"));

	double missingValuePenalty = (0.8d * 0.8d);

	assertEquals((Double)(0.9d * missingValuePenalty), targetValue.getConfidence("will play"));
	assertEquals((Double)(0.05d * missingValuePenalty), targetValue.getConfidence("may play"));
	assertEquals((Double)(0.05d * missingValuePenalty), targetValue.getConfidence("no play"));
}
 

/**
 * @param parent The parent Node of the Node that evaluated to the missing value.
 * @param node The Node that evaluated to the missing value.
 */
private Trail handleMissingValue(Trail trail, Node parent, Node node, EvaluationContext context){
	TreeModel treeModel = getModel();

	TreeModel.MissingValueStrategy missingValueStrategy = treeModel.getMissingValueStrategy();
	switch(missingValueStrategy){
		case NULL_PREDICTION:
			return trail.selectNull();
		case LAST_PREDICTION:
			return trail.selectLastPrediction();
		case DEFAULT_CHILD:
			return handleDefaultChild(trail, parent, context);
		case NONE:
			return null;
		default:
			throw new UnsupportedAttributeException(treeModel, missingValueStrategy);
	}
}
 

private List<TreeModel> encodeForest(RGenericVector forest, MiningFunction miningFunction, ScoreEncoder scoreEncoder, Schema schema){
	RNumberVector<?> numTrees = forest.getNumericElement("num.trees");
	RGenericVector childNodeIDs = forest.getGenericElement("child.nodeIDs");
	RGenericVector splitVarIDs = forest.getGenericElement("split.varIDs");
	RGenericVector splitValues = forest.getGenericElement("split.values");
	RGenericVector terminalClassCounts = forest.getGenericElement("terminal.class.counts", false);

	Schema segmentSchema = schema.toAnonymousSchema();

	List<TreeModel> treeModels = new ArrayList<>();

	for(int i = 0; i < ValueUtil.asInt(numTrees.asScalar()); i++){
		TreeModel treeModel = encodeTreeModel(miningFunction, scoreEncoder, (RGenericVector)childNodeIDs.getValue(i), (RNumberVector<?>)splitVarIDs.getValue(i), (RNumberVector<?>)splitValues.getValue(i), (terminalClassCounts != null ? (RGenericVector)terminalClassCounts.getValue(i) : null), segmentSchema);

		treeModels.add(treeModel);
	}

	return treeModels;
}
 

private Trail handleNoTrueChild(Trail trail){
	TreeModel treeModel = getModel();

	TreeModel.NoTrueChildStrategy noTrueChildStrategy = treeModel.getNoTrueChildStrategy();
	switch(noTrueChildStrategy){
		case RETURN_NULL_PREDICTION:
			return trail.selectNull();
		case RETURN_LAST_PREDICTION:
			Node lastPrediction = trail.getLastPrediction();

			// "Return the parent Node only if it specifies a score attribute"
			if(lastPrediction.hasScore()){
				return trail.selectLastPrediction();
			}
			return trail.selectNull();
		default:
			throw new UnsupportedAttributeException(treeModel, noTrueChildStrategy);
	}
}
 

private MiningModel encodeRegression(RGenericVector ranger, Schema schema){
	RGenericVector forest = ranger.getGenericElement("forest");

	ScoreEncoder scoreEncoder = new ScoreEncoder(){

		@Override
		public Node encode(Node node, Number splitValue, RNumberVector<?> terminalClassCount){
			node.setScore(splitValue);

			return node;
		}
	};

	List<TreeModel> treeModels = encodeForest(forest, MiningFunction.REGRESSION, scoreEncoder, schema);

	MiningModel miningModel = new MiningModel(MiningFunction.REGRESSION, ModelUtil.createMiningSchema(schema.getLabel()))
		.setSegmentation(MiningModelUtil.createSegmentation(Segmentation.MultipleModelMethod.AVERAGE, treeModels));

	return miningModel;
}
 

@Override
public RPartConverter createConverter(RGenericVector rpart){
	return new RPartConverter(rpart){

		@Override
		public boolean hasScoreDistribution(){
			return false;
		}

		@Override
		public TreeModel encodeModel(Schema schema){
			TreeModel treeModel = super.encodeModel(schema)
				.setMiningFunction(MiningFunction.REGRESSION);

			return treeModel;
		}
	};
}
 

static
protected MiningModel createMiningModel(List<RegressionTree> regTrees, float base_score, Schema schema){
    ContinuousLabel continuousLabel = (ContinuousLabel)schema.getLabel();

    Schema segmentSchema = schema.toAnonymousSchema();

    List<TreeModel> treeModels = new ArrayList<>();

    for(RegressionTree regTree : regTrees){
        TreeModel treeModel = regTree.encodeTreeModel(segmentSchema);

        treeModels.add(treeModel);
    }

    MiningModel miningModel = new MiningModel(MiningFunction.REGRESSION, ModelUtil.createMiningSchema(continuousLabel))
            .setMathContext(MathContext.FLOAT)
            .setSegmentation(MiningModelUtil.createSegmentation(Segmentation.MultipleModelMethod.SUM, treeModels))
            .setTargets(ModelUtil.createRescaleTargets(null, ValueUtil.floatToDouble(base_score), continuousLabel));

    return miningModel;
}
 

static
public TreeModel encodeTreeModel(TreePredictor treePredictor, PredicateManager predicateManager, Schema schema){
	int[] leaf = treePredictor.isLeaf();
	int[] leftChildren = treePredictor.getLeft();
	int[] rightChildren = treePredictor.getRight();
	int[] featureIdx = treePredictor.getFeatureIdx();
	double[] thresholds = treePredictor.getThreshold();
	int[] missingGoToLeft = treePredictor.getMissingGoToLeft();
	double[] values = treePredictor.getValues();

	Node root = encodeNode(True.INSTANCE, predicateManager, 0, leaf, leftChildren, rightChildren, featureIdx, thresholds, missingGoToLeft, values, schema);

	TreeModel treeModel = new TreeModel(MiningFunction.REGRESSION, ModelUtil.createMiningSchema(schema.getLabel()), root)
		.setSplitCharacteristic(TreeModel.SplitCharacteristic.BINARY_SPLIT)
		.setMissingValueStrategy(TreeModel.MissingValueStrategy.DEFAULT_CHILD);

	return treeModel;
}
 

static
private TreeModelEvaluator createTreeModelEvaluator(MiningFunction miningFunction, MathContext mathContext, Target target){
	Node root = new LeafNode(null, False.INSTANCE);

	Targets targets = new Targets()
		.addTargets(target);

	TreeModel treeModel = new TreeModel(miningFunction, new MiningSchema(), root)
		.setSplitCharacteristic(TreeModel.SplitCharacteristic.BINARY_SPLIT)
		.setMathContext(mathContext)
		.setTargets(targets);

	PMML pmml = new PMML(Version.PMML_4_3.getVersion(), new Header(), new DataDictionary())
		.addModels(treeModel);

	ModelEvaluatorBuilder modelEvaluatorBuilder = new ModelEvaluatorBuilder(pmml);

	return (TreeModelEvaluator)modelEvaluatorBuilder.build();
}
 

static
protected MiningModel createMiningModel(List<RegressionTree> regTrees, float base_score, Schema schema){
    ContinuousLabel continuousLabel = (ContinuousLabel)schema.getLabel();

    Schema segmentSchema = schema.toAnonymousSchema();

    List<TreeModel> treeModels = new ArrayList<>();

    for(RegressionTree regTree : regTrees){
        TreeModel treeModel = regTree.encodeTreeModel(segmentSchema);

        treeModels.add(treeModel);
    }

    MiningModel miningModel = new MiningModel(MiningFunction.REGRESSION, ModelUtil.createMiningSchema(continuousLabel))
            .setMathContext(MathContext.FLOAT)
            .setSegmentation(MiningModelUtil.createSegmentation(Segmentation.MultipleModelMethod.SUM, treeModels))
            .setTargets(ModelUtil.createRescaleTargets(null, ValueUtil.floatToDouble(base_score), continuousLabel));

    return miningModel;
}
 

static
public <E extends Estimator & HasTree> TreeModel encodeTreeModel(E estimator, PredicateManager predicateManager, ScoreDistributionManager scoreDistributionManager, MiningFunction miningFunction, Schema schema){
	Tree tree = estimator.getTree();

	int[] leftChildren = tree.getChildrenLeft();
	int[] rightChildren = tree.getChildrenRight();
	int[] features = tree.getFeature();
	double[] thresholds = tree.getThreshold();
	double[] values = tree.getValues();

	Node root = encodeNode(True.INSTANCE, predicateManager, scoreDistributionManager, 0, leftChildren, rightChildren, features, thresholds, values, miningFunction, schema);

	TreeModel treeModel = new TreeModel(miningFunction, ModelUtil.createMiningSchema(schema.getLabel()), root)
		.setSplitCharacteristic(TreeModel.SplitCharacteristic.BINARY_SPLIT);

	ClassDictUtil.clearContent(tree);

	return treeModel;
}
 

private MiningModel encodeBinaryClassification(List<TreeModel> treeModels, Double initF, double coefficient, Schema schema){
	Schema segmentSchema = schema.toAnonymousRegressorSchema(DataType.DOUBLE);

	MiningModel miningModel = createMiningModel(treeModels, initF, segmentSchema)
		.setOutput(ModelUtil.createPredictedOutput(FieldName.create("gbmValue"), OpType.CONTINUOUS, DataType.DOUBLE));

	return MiningModelUtil.createBinaryLogisticClassification(miningModel, -coefficient, 0d, RegressionModel.NormalizationMethod.LOGIT, true, schema);
}
 

private TreeModel encodeTreeModel(RGenericVector tree, Schema schema){
	Node root = encodeNode(True.INSTANCE, tree, schema);

	TreeModel treeModel = new TreeModel(this.miningFunction, ModelUtil.createMiningSchema(schema.getLabel()), root)
		.setSplitCharacteristic(TreeModel.SplitCharacteristic.BINARY_SPLIT);

	return treeModel;
}
 

public RegressorManager(PMML pmml, TreeModel treeModel){
	super(pmml, treeModel);

	MiningFunction miningFunction = treeModel.getMiningFunction();
	switch(miningFunction){
		case REGRESSION:
			break;
		default:
			throw new UnsupportedAttributeException(treeModel, miningFunction);
	}
}
 

@Override
public BiMap<String, Node> load(TreeModel treeModel){
	ImmutableBiMap.Builder<String, Node> builder = new ImmutableBiMap.Builder<>();

	builder = collectNodes(treeModel.getNode(), new AtomicInteger(1), builder);

	return builder.build();
}
 

static
private MiningModel createMiningModel(List<TreeModel> treeModels, Double initF, Schema schema){
	ContinuousLabel continuousLabel = (ContinuousLabel)schema.getLabel();

	MiningModel miningModel = new MiningModel(MiningFunction.REGRESSION, ModelUtil.createMiningSchema(continuousLabel))
		.setSegmentation(MiningModelUtil.createSegmentation(Segmentation.MultipleModelMethod.SUM, treeModels))
		.setTargets(ModelUtil.createRescaleTargets(null, initF, continuousLabel));

	return miningModel;
}
 

private TreeModel encodeTreeModel(RGenericVector trees, int index, Schema schema){
	RIntegerVector nrnodes = trees.getIntegerElement("nrnodes");
	RIntegerVector ntree = trees.getIntegerElement("ntree");
	RIntegerVector nodeStatus = trees.getIntegerElement("nodeStatus");
	RIntegerVector leftDaughter = trees.getIntegerElement("lDaughter");
	RIntegerVector rightDaughter = trees.getIntegerElement("rDaughter");
	RIntegerVector splitAtt = trees.getIntegerElement("splitAtt");
	RDoubleVector splitPoint = trees.getDoubleElement("splitPoint");
	RIntegerVector nSam = trees.getIntegerElement("nSam");

	int rows = nrnodes.asScalar();
	int columns = ntree.asScalar();

	Node root = encodeNode(
		True.INSTANCE,
		0,
		0,
		FortranMatrixUtil.getColumn(nodeStatus.getValues(), rows, columns, index),
		FortranMatrixUtil.getColumn(nSam.getValues(), rows, columns, index),
		FortranMatrixUtil.getColumn(leftDaughter.getValues(), rows, columns, index),
		FortranMatrixUtil.getColumn(rightDaughter.getValues(), rows, columns, index),
		FortranMatrixUtil.getColumn(splitAtt.getValues(), rows, columns, index),
		FortranMatrixUtil.getColumn(splitPoint.getValues(), rows, columns, index),
		schema
	);

	TreeModel treeModel = new TreeModel(MiningFunction.REGRESSION, ModelUtil.createMiningSchema(schema.getLabel()), root)
		.setSplitCharacteristic(TreeModel.SplitCharacteristic.BINARY_SPLIT);

	return treeModel;
}
 

private TreeModel encodeTreeModel(MiningFunction miningFunction, ScoreEncoder scoreEncoder, RGenericVector childNodeIDs, RNumberVector<?> splitVarIDs, RNumberVector<?> splitValues, RGenericVector terminalClassCounts, Schema schema){
	RNumberVector<?> leftChildIDs = (RNumberVector<?>)childNodeIDs.getValue(0);
	RNumberVector<?> rightChildIDs = (RNumberVector<?>)childNodeIDs.getValue(1);

	Node root = encodeNode(True.INSTANCE, 0, scoreEncoder, leftChildIDs, rightChildIDs, splitVarIDs, splitValues, terminalClassCounts, new CategoryManager(), schema);

	TreeModel treeModel = new TreeModel(miningFunction, ModelUtil.createMiningSchema(schema.getLabel()), root)
		.setSplitCharacteristic(TreeModel.SplitCharacteristic.BINARY_SPLIT);

	return treeModel;
}
 

private MiningModel encodeMiningModel(RStringVector distributionName, List<TreeModel> treeModels, Double initF, Schema schema){

		switch(distributionName.asScalar()){
			case "gaussian":
				return encodeRegression(treeModels, initF, schema);
			case "adaboost":
				return encodeBinaryClassification(treeModels, initF, -2d, schema);
			case "bernoulli":
				return encodeBinaryClassification(treeModels, initF, -1d, schema);
			case "multinomial":
				return encodeMultinomialClassification(treeModels, initF, schema);
			default:
				throw new IllegalArgumentException();
		}
	}
 

@Override
public void enterTreeModel(TreeModel treeModel){
	TreeModel.MissingValueStrategy missingValueStrategy = treeModel.getMissingValueStrategy();
	TreeModel.NoTrueChildStrategy noTrueChildStrategy = treeModel.getNoTrueChildStrategy();
	TreeModel.SplitCharacteristic splitCharacteristic = treeModel.getSplitCharacteristic();

	if(!(TreeModel.MissingValueStrategy.NONE).equals(missingValueStrategy) || !(TreeModel.NoTrueChildStrategy.RETURN_NULL_PREDICTION).equals(noTrueChildStrategy) || !(TreeModel.SplitCharacteristic.BINARY_SPLIT).equals(splitCharacteristic)){
		throw new IllegalArgumentException();
	}

	this.miningFunction = treeModel.getMiningFunction();
}
 

public TreeModel encodeTreeModel(Schema schema){
    org.dmg.pmml.tree.Node root = new org.dmg.pmml.tree.Node()
            .setPredicate(new True());

    encodeNode(root, 0, schema);

    TreeModel treeModel = new TreeModel(MiningFunction.REGRESSION, ModelUtil.createMiningSchema(schema.getLabel()), root)
            .setSplitCharacteristic(TreeModel.SplitCharacteristic.BINARY_SPLIT)
            .setMissingValueStrategy(TreeModel.MissingValueStrategy.NONE)
            .setMathContext(MathContext.FLOAT);

    return treeModel;
}
 

static
public <E extends Estimator & HasEstimatorEnsemble<T>, T extends Estimator & HasTree> List<TreeModel> encodeTreeModelEnsemble(E estimator, MiningFunction miningFunction, Schema schema){
	PredicateManager predicateManager = new PredicateManager();
	ScoreDistributionManager scoreDistributionManager = new ScoreDistributionManager();

	return encodeTreeModelEnsemble(estimator, predicateManager, scoreDistributionManager, miningFunction, schema);
}
 

private TreeModel encodeTreeModel(MiningFunction miningFunction, RGenericVector tree, RGenericVector c_splits, Schema schema){
	Node root = encodeNode(True.INSTANCE, 0, tree, c_splits, new FlagManager(), new CategoryManager(), schema);

	TreeModel treeModel = new TreeModel(miningFunction, ModelUtil.createMiningSchema(schema.getLabel()), root)
		.setSplitCharacteristic(TreeModel.SplitCharacteristic.MULTI_SPLIT);

	return treeModel;
}