com.pervasive.datarush.analytics.regression

Class PMMLRegressionModel

java.lang.Object

com.pervasive.datarush.analytics.pmml.PMMLModel

com.pervasive.datarush.analytics.regression.AbstractPMMLRegressionModel

com.pervasive.datarush.analytics.regression.PMMLRegressionModel

public class PMMLRegressionModel
extends AbstractPMMLRegressionModel

PMML object model for Regression models.

Nested Class Summary

Nested classes/interfaces inherited from class com.pervasive.datarush.analytics.regression.AbstractPMMLRegressionModel

AbstractPMMLRegressionModel.CategoricalPredictor, AbstractPMMLRegressionModel.ModelType, AbstractPMMLRegressionModel.NormalizationMethod, AbstractPMMLRegressionModel.NumericPredictor, AbstractPMMLRegressionModel.RegressionTable

Nested classes/interfaces inherited from class com.pervasive.datarush.analytics.pmml.PMMLModel

PMMLModel.MiningFunction

Constructor Summary

Constructors

Constructor and Description
PMMLRegressionModel(Document document) Construct the PMML model directly from an XML document.
PMMLRegressionModel(PMMLModelSpec objectSpec) Construct the PMML model from a model specification.

Method Summary

Modifier and Type	Method and Description
void	addCategoricalPredictor(AbstractPMMLRegressionModel.CategoricalPredictor categoricalPredictor) Adds a categorical predictor to the PMML model.
void	addNumericPredictor(AbstractPMMLRegressionModel.NumericPredictor numericPredictor) Add a numeric predictor to the PMML model.
protected void	buildModelElement(Element element) Subclasses must implement this method to fill-in the contents of the model element.
List<AbstractPMMLRegressionModel.CategoricalPredictor>	getCategoricalPredictor(String indepVarName) Get the categorical predictor for the given independent variable.
List<AbstractPMMLRegressionModel.CategoricalPredictor>	getCategoricalPredictors()
double	getIntercept() Get the intercept value.
AbstractPMMLRegressionModel.NumericPredictor	getNumericPredictor(String indepVarName) Get the numeric predictor for the given independent variable.
List<AbstractPMMLRegressionModel.NumericPredictor>	getNumericPredictors() Return the list of numeric predictors for this model.
com.pervasive.datarush.analytics.regression.AbstractPMMLRegressionModel.ValuePredictor	getValuePredictor(RecordValued input, DoubleSettable target) Get a value predictor for this model.
protected void	parseModelElement(Element element) Parse the given model element into the respective model object.
void	setIntercept(double value) Set the intercept value.
void	setInterceptDegreesOfFreedom(int value) Set the intercept's degrees of freedom.
void	setInterceptPValue(double value) Set the intercept's p value.
void	setInterceptStandardError(double value) Set the intercept's standard error.
void	setInterceptTTest(double value) Set the intercept's t test.

Methods inherited from class com.pervasive.datarush.analytics.regression.AbstractPMMLRegressionModel

getModelElementName, getModelName, getNormalizationMethod, isScorable, logistic, setModelName, setNormalizationMethod, setScorable

Methods inherited from class com.pervasive.datarush.analytics.pmml.PMMLModel

findModelElement, getAnnotationText, getModelExplanation, getModelSpec, getVersion, parse, setAnnotationText, setModelExplanation, setVersion, toPMML

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

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

getNumericPredictors

public List<AbstractPMMLRegressionModel.NumericPredictor> getNumericPredictors()

Return the list of numeric predictors for this model.

Returns:

numeric predictors

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

getCategoricalPredictors

public List<AbstractPMMLRegressionModel.CategoricalPredictor> getCategoricalPredictors()

Returns:

all the Categorical Predictor elements

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