com.imsl.datamining

Class GradientBoostingModelObject

java.lang.Object

com.imsl.datamining.GradientBoostingModelObject

All Implemented Interfaces:

Serializable

public class GradientBoostingModelObject
extends Object
implements Serializable

Predicts a data set using a trained gradient boosting model.

This class extracts the decision trees and other necessary parameters from a trained gradient boosting model for the purpose of predicting a new data set separately from the original training data.

See Also:

Example 1, Example 2, Serialized Form

Constructor Summary

Constructors

Constructor and Description
GradientBoostingModelObject(GradientBoosting gbm) Constructs the GradientBoostingModelObject.

Method Summary

Modifier and Type	Method and Description
double[][]	getClassPredictorValues() Returns the predictor function values \(\{f(x_i)\}\) on the test data for a categorical (binomial or multinomial) response variable.
double[][]	getClassProbabilities() Returns the predicted probabilities on the test data for a categorical response variable.
double[]	getContinuousPredictorValues() Returns the predictor function values \(\{f(x_i)\}\) for a continuous response variable on the test data.
int	getPrintLevel() Returns the current print level.
double	getTestLossValue() Returns the value of the loss function on the test data.
double[]	predict(double[][] testData) Returns the predictions on the test data.
double[]	predict(double[][] testData, double[] testDataWeights) Returns the predictions on the test data using data weights.
void	setMissingTestYFlag(boolean missingTestY) Sets the flag for whether or not the test data has missing response values.
void	setPrintLevel(int printLevel) Sets the print level.

Methods inherited from class java.lang.Object

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

Constructor Detail

GradientBoostingModelObject

public GradientBoostingModelObject(GradientBoosting gbm)

Constructs the GradientBoostingModelObject.

Parameters:

gbm - a previously trained GradientBoosting model

Method Detail

predict

public double[] predict(double[][] testData)
                 throws PredictiveModel.PredictiveModelException

Returns the predictions on the test data.

Parameters:

testData - a double matrix containing the data. It must be compatible with the original training data (it must have the same variable types and column orders).

Returns:

a double array containing the predictions

Throws:

PredictiveModel.PredictiveModelException

predict

public double[] predict(double[][] testData,
                        double[] testDataWeights)
                 throws PredictiveModel.PredictiveModelException

Returns the predictions on the test data using data weights.

Parameters:

testData - a double matrix containing the data. It must be compatible with the original training data (it must have the same variable types and column orders).

testDataWeights - a double array containing weights for each row of testData. The weights must be non-negative.

Returns:

a double array containing the predictions

Throws:

PredictiveModel.PredictiveModelException

getTestLossValue

public double getTestLossValue()

Returns the value of the loss function on the test data.

The result will be NaN if the test data response variable is missing (the default).

Returns:

a double, the value of the loss function.

getClassPredictorValues

public double[][] getClassPredictorValues()

Returns the predictor function values \(\{f(x_i)\}\) on the test data for a categorical (binomial or multinomial) response variable.

Returns:

a double matrix containing the predictor function values

getContinuousPredictorValues

public double[] getContinuousPredictorValues()

Returns the predictor function values \(\{f(x_i)\}\) for a continuous response variable on the test data.

Returns:

a double array containing the predictor function values

getClassProbabilities

public double[][] getClassProbabilities()

Returns the predicted probabilities on the test data for a categorical response variable.

Returns:

a double matrix containing the class probabilities on the test data. The i,k-th element of the matrix is the estimated probability that the observation at row index i belongs to the k+1-st class, where k=0,..., nClasses-1.

setMissingTestYFlag

public void setMissingTestYFlag(boolean missingTestY)

Sets the flag for whether or not the test data has missing response values. The default is true (the methods will assume the response values are not present). If the response variable is not missing, the value of the loss function on the test data can be calculated.

Parameters:

missingTestY - a boolean indicating whether or not the test data response variable is missing

getPrintLevel

public int getPrintLevel()

Returns the current print level.

Returns:

an int, the current print level

setPrintLevel

public void setPrintLevel(int printLevel)

Sets the print level.

When printLevel>0 and the response values are non-missing, the loss function values for each iteration are printed. The default is printLevel=0.

Parameters:

printLevel - an int specifying the print level