io.improbable.keanu.model.regression

Class RegressionModel<OUTPUT>

java.lang.Object

io.improbable.keanu.model.regression.RegressionModel<OUTPUT>

All Implemented Interfaces:

PredictiveModel<DoubleTensor,OUTPUT>

public class RegressionModel<OUTPUT>
extends java.lang.Object
implements PredictiveModel<DoubleTensor,OUTPUT>

A general linear regression model that can be fitted to input and output training data. You can construct a regression model as follows:

 RegressionModel model = RegressionModel
      .withTrainingData(inputTrainingData, outputTrainingData)
      .withRegularization(RegressionRegularization.RIDGE)
      .build();
 

Method Summary

Modifier and Type	Method and Description
void	fit()
DoubleVertex	getInterceptVertex()
Vertex<OUTPUT>	getOutputVertex()
DoubleVertex	getWeightVertex()
OUTPUT	predict(DoubleTensor tensor)
static RegressionModelBuilder<BooleanTensor>	withTrainingData(DoubleTensor inputTrainingData, BooleanTensor outputTrainingData)
static RegressionModelBuilder<DoubleTensor>	withTrainingData(DoubleTensor inputTrainingData, DoubleTensor outputTrainingData)

Methods inherited from class java.lang.Object

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

Method Detail

withTrainingData

public static RegressionModelBuilder<DoubleTensor> withTrainingData(DoubleTensor inputTrainingData,
                                                                    DoubleTensor outputTrainingData)

withTrainingData

public static RegressionModelBuilder<BooleanTensor> withTrainingData(DoubleTensor inputTrainingData,
                                                                     BooleanTensor outputTrainingData)

getInterceptVertex

public DoubleVertex getInterceptVertex()

getWeightVertex

public DoubleVertex getWeightVertex()

getOutputVertex

public Vertex<OUTPUT> getOutputVertex()

predict

public OUTPUT predict(DoubleTensor tensor)

Specified by:

predict in interface PredictiveModel<DoubleTensor,OUTPUT>

fit

public void fit()