Module datarush.analytics
Package com.pervasive.datarush.analytics.regression

Class PMMLRegressionModel

    • Constructor Detail

      • PMMLRegressionModel

        public PMMLRegressionModel​(PMMLModelSpec objectSpec)
        Construct the PMML model from a model specification.
        Parameters:
        objectSpec - model specification

      • PMMLRegressionModel

        public PMMLRegressionModel​(Document document)
        Construct the PMML model directly from an XML document.
        Parameters:
        document - XML document containing PMML

    • Method Detail

      • setIntercept

        public void setIntercept​(double value)
        Set the intercept value.
        Parameters:
        value - intercept value

      • setInterceptDegreesOfFreedom

        public void setInterceptDegreesOfFreedom​(int value)
        Set the intercept's degrees of freedom.
        Parameters:
        value - interceptDegreesOfFreedom value

      • setInterceptStandardError

        public void setInterceptStandardError​(double value)
        Set the intercept's standard error.
        Parameters:
        value - interceptStandardError value

      • setInterceptTTest

        public void setInterceptTTest​(double value)
        Set the intercept's t test.
        Parameters:
        value - interceptTTest value

      • setInterceptPValue

        public void setInterceptPValue​(double value)
        Set the intercept's p value.
        Parameters:
        value - interceptPValue value

      • getIntercept

        public double getIntercept()
        Get the intercept value.
        Returns:
        intercept value

      • addNumericPredictor

        public void addNumericPredictor​(AbstractPMMLRegressionModel.NumericPredictor numericPredictor)
        Add a numeric predictor to the PMML model. A numeric predictor contains the model values for an independent variable and can be used to predict an outcome given an instance value of the independent variable.
        Parameters:
        numericPredictor - numeric predictor

      • getNumericPredictor

        public AbstractPMMLRegressionModel.NumericPredictor getNumericPredictor​(String indepVarName)
        Get the numeric predictor for the given independent variable. The given name must match the name of an independent variable in the model.
        Parameters:
        indepVarName - name of the independent variable
        Returns:
        numeric predictor

      • addCategoricalPredictor

        public void addCategoricalPredictor​(AbstractPMMLRegressionModel.CategoricalPredictor categoricalPredictor)
        Adds a categorical predictor to the PMML model. A categorical predictor describes the behavior of the model with regards to a single possible value of a discrete independent variable
        Parameters:
        categoricalPredictor - the Categorical Predictor element

      • getCategoricalPredictor

        public List<AbstractPMMLRegressionModel.CategoricalPredictor> getCategoricalPredictor​(String indepVarName)
        Get the categorical predictor for the given independent variable. The given name must match the name of an independent variable in the model.
        Parameters:
        indepVarName - name of the independent variable
        Returns:
        categorical predictor

      • getValuePredictor

        public com.pervasive.datarush.analytics.regression.AbstractPMMLRegressionModel.ValuePredictor getValuePredictor​(RecordValued input,
                                                                                                                DoubleSettable target)
        Get a value predictor for this model. The input record is provided as input along with a settable target for the predicted value. The value predictor can be used to produce an output prediction for each input record of a dataflow.
        Parameters:
        input - input data source containing independent data values
        target - output field containing result of prediction
        Returns:
        value predictor

      • buildModelElement

        protected void buildModelElement​(Element element)
        Description copied from class: PMMLModel
        Subclasses must implement this method to fill-in the contents of the model element. At the time this method is invoked, the element will have its name and MiningSchema sub-element populated.
        Specified by:
        buildModelElement in class AbstractPMMLRegressionModel
        Parameters:
        element - the model element

      • parseModelElement

        protected void parseModelElement​(Element element)
        Description copied from class: AbstractPMMLRegressionModel
        Parse the given model element into the respective model object. This implementation only reads the normalizationMethod and scorable attributes. Subclasses SHOULD override this and are responsible for Regression Tables and the function name and target field attributes.
        Overrides:
        parseModelElement in class AbstractPMMLRegressionModel
        Parameters:
        element - root element of model object