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

Class AbstractPMMLRegressionModel

java.lang.Object
com.pervasive.datarush.analytics.pmml.PMMLModel
com.pervasive.datarush.analytics.regression.AbstractPMMLRegressionModel
Direct Known Subclasses:
PMMLRegressionClassificationModel, PMMLRegressionModel
public abstract class AbstractPMMLRegressionModel extends PMMLModel
A representation of features found in PMML regressionModel elements that do not depend on its attributes.

  • Constructor Details

    • AbstractPMMLRegressionModel

      public AbstractPMMLRegressionModel(PMMLModelSpec objectSpec)

    • AbstractPMMLRegressionModel

      public AbstractPMMLRegressionModel()

  • Method Details

    • logistic

      public static double logistic(double value)
      An implementation of the logit model's normalization function.
      Parameters:
      value - the real number to normalize.
      Returns:
      the normalized value, between 0 and 1, or NaN if the input value is not-a-number.

    • setScorable

      public void setScorable(boolean b)
      Whether a consumer can use the model to make predictions (set to false to indicate the model is for information purposes only.)
      Parameters:
      b -

    • isScorable

      public boolean isScorable()
      Returns:
      whether the model should be used for predictions.

    • setNormalizationMethod

      public void setNormalizationMethod(AbstractPMMLRegressionModel.NormalizationMethod normalizationMethod)
      Parameters:
      normalizationMethod - the normalization method to use to transform the linear result to a prediction

    • getNormalizationMethod

      public AbstractPMMLRegressionModel.NormalizationMethod getNormalizationMethod()
      Returns:
      the normalization method used to transform the linear result to a prediction

    • setModelName

      public void setModelName(String modelName)
      Parameters:
      modelName - the desired "modelName" PMML attribute

    • getModelName

      public String getModelName()
      Returns:
      the current value of the "modelName" PMML attribute. The modelName is a brief natural-language description of the model.

    • getModelElementName

      protected String getModelElementName()
      Description copied from class: PMMLModel
      Return the element name of the primary model element associated with this PMML
      Specified by:
      getModelElementName in class PMMLModel
      Returns:
      the element name of the primary model element

    • buildModelElement

      protected abstract 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 PMMLModel
      Parameters:
      element - the model element

    • parseModelElement

      protected void parseModelElement(Element element)
      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.
      Specified by:
      parseModelElement in class PMMLModel
      Parameters:
      element - root element of model object
      Throws:
      UnsupportedPPMMLException - if the normalization method is not represented in the NormalizationMethod enum.