hex.genmodel.easy

Class EasyPredictModelWrapper.Config

java.lang.Object

hex.genmodel.easy.EasyPredictModelWrapper.Config

Enclosing class:

EasyPredictModelWrapper

public static class EasyPredictModelWrapper.Config
extends java.lang.Object

Configuration builder for instantiating a Wrapper.

Constructor Summary

Constructors

Constructor and Description
Config()

Method Summary

Modifier and Type	Method and Description
boolean	getConvertInvalidNumbersToNa()
boolean	getConvertUnknownCategoricalLevelsToNa()
boolean	getEnableGLRMReconstrut()
boolean	getEnableLeafAssignment()
boolean	getEnableStagedProbabilities()
EasyPredictModelWrapper.ErrorConsumer	getErrorConsumer()
GenModel	getModel()
boolean	getUseExtendedOutput()
EasyPredictModelWrapper.Config	setConvertInvalidNumbersToNa(boolean value) Specify the default action when a string value cannot be converted to a number.
EasyPredictModelWrapper.Config	setConvertUnknownCategoricalLevelsToNa(boolean value) Specify how to handle unknown categorical levels.
EasyPredictModelWrapper.Config	setEnableGLRMReconstrut(boolean value)
EasyPredictModelWrapper.Config	setEnableLeafAssignment(boolean val)
EasyPredictModelWrapper.Config	setEnableStagedProbabilities(boolean val)
EasyPredictModelWrapper.Config	setErrorConsumer(EasyPredictModelWrapper.ErrorConsumer errorConsumer) Specify an instance of EasyPredictModelWrapper.ErrorConsumer the EasyPredictModelWrapper is going to call whenever an error defined by the EasyPredictModelWrapper.ErrorConsumer instance occurs.
EasyPredictModelWrapper.Config	setModel(GenModel value) Specify model object to wrap.
EasyPredictModelWrapper.Config	setUseExtendedOutput(boolean value) Specify whether to include additional metadata in the prediction output.

Methods inherited from class java.lang.Object

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

Constructor Detail

Config

public Config()

Method Detail

setModel

public EasyPredictModelWrapper.Config setModel(GenModel value)

Specify model object to wrap.

Parameters:

value - model

Returns:

this config object

getModel

public GenModel getModel()

Returns:

model object being wrapped

setConvertUnknownCategoricalLevelsToNa

public EasyPredictModelWrapper.Config setConvertUnknownCategoricalLevelsToNa(boolean value)

Specify how to handle unknown categorical levels.

Parameters:

value - false: throw exception; true: convert to N/A

Returns:

this config object

setEnableLeafAssignment

public EasyPredictModelWrapper.Config setEnableLeafAssignment(boolean val)
                                                       throws java.io.IOException

Throws:

java.io.IOException

setEnableGLRMReconstrut

public EasyPredictModelWrapper.Config setEnableGLRMReconstrut(boolean value)
                                                       throws java.io.IOException

Throws:

java.io.IOException

setEnableStagedProbabilities

public EasyPredictModelWrapper.Config setEnableStagedProbabilities(boolean val)
                                                            throws java.io.IOException

Throws:

java.io.IOException

getEnableGLRMReconstrut

public boolean getEnableGLRMReconstrut()

getConvertUnknownCategoricalLevelsToNa

public boolean getConvertUnknownCategoricalLevelsToNa()

Returns:

Setting for unknown categorical levels handling

setConvertInvalidNumbersToNa

public EasyPredictModelWrapper.Config setConvertInvalidNumbersToNa(boolean value)

Specify the default action when a string value cannot be converted to a number.

Parameters:

value - if true, then an N/A value will be produced, if false an exception will be thrown.

getConvertInvalidNumbersToNa

public boolean getConvertInvalidNumbersToNa()

setUseExtendedOutput

public EasyPredictModelWrapper.Config setUseExtendedOutput(boolean value)

Specify whether to include additional metadata in the prediction output. This feature needs to be supported by a particular model and type of metadata is model specific.

Parameters:

value - if true, then the Prediction result will contain extended information about the prediction (this will be specific to a particular model).

Returns:

this config object

getUseExtendedOutput

public boolean getUseExtendedOutput()

getEnableLeafAssignment

public boolean getEnableLeafAssignment()

getEnableStagedProbabilities

public boolean getEnableStagedProbabilities()

getErrorConsumer

public EasyPredictModelWrapper.ErrorConsumer getErrorConsumer()

Returns:

An instance of ErrorConsumer used to build the EasyPredictModelWrapper. Null if there is no instance.

setErrorConsumer

public EasyPredictModelWrapper.Config setErrorConsumer(EasyPredictModelWrapper.ErrorConsumer errorConsumer)

Specify an instance of EasyPredictModelWrapper.ErrorConsumer the EasyPredictModelWrapper is going to call whenever an error defined by the EasyPredictModelWrapper.ErrorConsumer instance occurs.

Parameters:

errorConsumer - An instance of EasyPredictModelWrapper.ErrorConsumer

Returns:

This EasyPredictModelWrapper.Config object