hex.ensemble

Class StackedEnsembleModel

java.lang.Object

water.Iced<T>

water.Keyed<T>

water.Lockable<M>

hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

hex.ensemble.StackedEnsembleModel

All Implemented Interfaces:

hex.Model.Contributions, java.io.Externalizable, java.io.Serializable, java.lang.Cloneable, water.api.StreamWriter, water.Freezable<StackedEnsembleModel>

public class StackedEnsembleModel
extends hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>
implements hex.Model.Contributions

An ensemble of other models, created by stacking with the SuperLearner algorithm or a variation.

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	StackedEnsembleModel.StackedEnsembleOutput
static class	StackedEnsembleModel.StackedEnsembleParameters
static class	StackedEnsembleModel.StackingStrategy

Nested classes/interfaces inherited from class hex.Model

hex.Model.AdaptFrameParameters, hex.Model.BigScore, hex.Model.BigScoreChunkPredict, hex.Model.BigScorePredict, hex.Model.Contributions, hex.Model.DeepFeatures, hex.Model.ExemplarMembers, hex.Model.FeatureFrequencies, hex.Model.GetMostImportantFeatures, hex.Model.GetNTrees, hex.Model.GLRMArchetypes, hex.Model.GridSortBy, hex.Model.H2OModelDescriptor, hex.Model.InteractionBuilder, hex.Model.InteractionPair, hex.Model.InteractionSpec, hex.Model.JavaModelStreamWriter, hex.Model.JavaScoringOptions, hex.Model.LeafNodeAssignment, hex.Model.Output, hex.Model.Parameters, hex.Model.PredictScoreResult, hex.Model.RowToTreeAssignment, hex.Model.StagedPredictions, hex.Model.UpdateAuxTreeWeights

Nested classes/interfaces inherited from interface hex.Model.Contributions

hex.Model.Contributions.ContributionsOptions, hex.Model.Contributions.ContributionsOutputFormat

Field Summary

Fields

Modifier and Type	Field and Description
hex.ModelCategory	modelCategory
java.lang.String	responseColumn
long	trainingFrameRows

Fields inherited from class hex.Model

_dist, _input_parms, _output, _parms, _toDelete, _warnings, _warningsP, EVAL_AUTO_PARAMS_ENABLED, evalAutoParamsEnabled, scoringInfo

Fields inherited from class water.Lockable

_lockers

Fields inherited from class water.Keyed

_key

Constructor Summary

Constructors

Constructor and Description
StackedEnsembleModel(water.Key selfKey, StackedEnsembleModel.StackedEnsembleParameters parms, StackedEnsembleModel.StackedEnsembleOutput output)

Method Summary

Modifier and Type	Method and Description
void	deleteBaseModelPredictions()
void	deleteCrossValidationFoldAssignment()
void	deleteCrossValidationModels()
void	deleteCrossValidationPreds()
StackedEnsembleMojoWriter	getMojo()
boolean	haveMojo()
void	initActualParamValues()
hex.ModelMetrics.MetricBuilder	makeMetricBuilder(java.lang.String[] domain)
protected hex.Model.PredictScoreResult	predictScoreImpl(water.fvec.Frame fr, water.fvec.Frame adaptFrm, java.lang.String destination_key, water.Job j, boolean computeMetrics, water.udf.CFuncRef customMetricFunc) For StackedEnsemble we call score on all the base_models and then combine the results with the metalearner to create the final predictions frame.
protected water.Keyed	readAll_impl(water.AutoBuffer ab, water.Futures fs) Read in models (base + metalearner)
protected water.Futures	remove_impl(water.Futures fs, boolean cascade)
protected double[]	score0(double[] data, double[] preds) Should never be called: the code paths that normally go here should call predictScoreImpl().
water.fvec.Frame	scoreContributions(water.fvec.Frame frame, water.Key<water.fvec.Frame> destination_key, water.Job<water.fvec.Frame> j, hex.Model.Contributions.ContributionsOptions options, water.fvec.Frame backgroundFrame)
long	scoreContributionsWorkEstimate(water.fvec.Frame frame, water.fvec.Frame backgroundFrame, boolean outputPerReference)
protected water.AutoBuffer	writeAll_impl(water.AutoBuffer ab) Write out models (base + metalearner)

Methods inherited from class hex.Model

adaptFrameForScore, adaptTestForJavaScoring, adaptTestForTrain, adaptTestForTrain, adaptTestForTrain, addMetrics, addModelMetrics, addWarning, aic, auc, AUCPR, checksum_impl, classification_error, compareTo, computeDeviances, containsResponse, data, defaultThreshold, defaultThreshold, deviance, deviance, evaluateAutoModelParameters, exportBinaryModel, exportMojo, fetchAll, fillScoringInfo, getDefaultGridSortBy, getGenModelEncoding, getPojoInterfaces, getToEigenVec, havePojo, importBinaryModel, isDistributionHuber, isFeatureUsedInPredict, isFeatureUsedInPredict, isGeneric, isSupervised, last_scored, lift_top_group, likelihood, logloss, loss, mae, makeAdaptFrameParameters, makeBigScoreTask, makeInteraction, makeInteractions, makeInteractions, makePojoWriter, makeSchema, makeScoringDomains, makeScoringNames, makeScoringNames, mean_per_class_error, modelDescriptor, mse, needsPostProcess, postProcessPredictions, r2, resetThreshold, result, rmsle, score, score, score, score, score, score, score, score, score0, score0, score0, score0PostProcessSupervised, scoreMetrics, scoring_history, scoringDomains, setInputParms, setupBigScorePredict, testJavaScoring, testJavaScoring, testJavaScoring, testJavaScoring, testJavaScoring, toJava, toJava, toJava, toJavaAlgo, toJavaCheckTooBig, toJavaInit, toJavaModelClassName, toJavaPredictBody, toJavaTransform, toJavaUUID, toMojo, toMojo, toString, transform, uploadBinaryModel, writeTo

Methods inherited from class water.Lockable

delete_and_lock, delete_and_lock, delete_and_lock, delete_and_lock, delete, delete, delete, delete, read_lock, read_lock, read_lock, unlock_all, unlock, unlock, unlock, unlock, update, update, update, write_lock_to_read_lock, write_lock, write_lock, write_lock

Methods inherited from class water.Keyed

checksum_impl, checksum, checksum, getKey, readAll, remove_impl, remove_self_key_impl, remove, remove, remove, remove, remove, remove, removeQuietly, writeAll

Methods inherited from class water.Iced

asBytes, clone, copyOver, frozenType, read, readExternal, readJSON, reloadFromBytes, toJsonBytes, toJsonString, write, writeExternal, writeJSON

Methods inherited from class java.lang.Object

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

Methods inherited from interface hex.Model.Contributions

composeScoreContributionTaskMetadata, scoreContributions, scoreContributions, scoreContributions

Field Detail

modelCategory

public hex.ModelCategory modelCategory

trainingFrameRows

public long trainingFrameRows

responseColumn

public java.lang.String responseColumn

Constructor Detail

StackedEnsembleModel

public StackedEnsembleModel(water.Key selfKey,
                            StackedEnsembleModel.StackedEnsembleParameters parms,
                            StackedEnsembleModel.StackedEnsembleOutput output)

Method Detail

scoreContributionsWorkEstimate

public long scoreContributionsWorkEstimate(water.fvec.Frame frame,
                                           water.fvec.Frame backgroundFrame,
                                           boolean outputPerReference)

Specified by:

scoreContributionsWorkEstimate in interface hex.Model.Contributions

scoreContributions

public water.fvec.Frame scoreContributions(water.fvec.Frame frame,
                                           water.Key<water.fvec.Frame> destination_key,
                                           water.Job<water.fvec.Frame> j,
                                           hex.Model.Contributions.ContributionsOptions options,
                                           water.fvec.Frame backgroundFrame)

Specified by:

scoreContributions in interface hex.Model.Contributions

initActualParamValues

public void initActualParamValues()

Overrides:

initActualParamValues in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

haveMojo

public boolean haveMojo()

Overrides:

haveMojo in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

predictScoreImpl

protected hex.Model.PredictScoreResult predictScoreImpl(water.fvec.Frame fr,
                                                        water.fvec.Frame adaptFrm,
                                                        java.lang.String destination_key,
                                                        water.Job j,
                                                        boolean computeMetrics,
                                                        water.udf.CFuncRef customMetricFunc)

For StackedEnsemble we call score on all the base_models and then combine the results with the metalearner to create the final predictions frame.

Overrides:

predictScoreImpl in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

Parameters:

adaptFrm - Already adapted frame

computeMetrics -

Returns:

A Frame containing the prediction column, and class distribution

See Also:

Model.predictScoreImpl(Frame, Frame, String, Job, boolean, CFuncRef)

score0

protected double[] score0(double[] data,
                          double[] preds)

Should never be called: the code paths that normally go here should call predictScoreImpl().

Specified by:

score0 in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

See Also:

Model.score0(double[], double[])

makeMetricBuilder

public hex.ModelMetrics.MetricBuilder makeMetricBuilder(java.lang.String[] domain)

Specified by:

makeMetricBuilder in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

deleteBaseModelPredictions

public void deleteBaseModelPredictions()

remove_impl

protected water.Futures remove_impl(water.Futures fs,
                                    boolean cascade)

Overrides:

remove_impl in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

writeAll_impl

protected water.AutoBuffer writeAll_impl(water.AutoBuffer ab)

Write out models (base + metalearner)

Overrides:

writeAll_impl in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

readAll_impl

protected water.Keyed readAll_impl(water.AutoBuffer ab,
                                   water.Futures fs)

Read in models (base + metalearner)

Overrides:

readAll_impl in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

getMojo

public StackedEnsembleMojoWriter getMojo()

Overrides:

getMojo in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

deleteCrossValidationModels

public void deleteCrossValidationModels()

Overrides:

deleteCrossValidationModels in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

deleteCrossValidationPreds

public void deleteCrossValidationPreds()

Overrides:

deleteCrossValidationPreds in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>

deleteCrossValidationFoldAssignment

public void deleteCrossValidationFoldAssignment()

Overrides:

deleteCrossValidationFoldAssignment in class hex.Model<StackedEnsembleModel,StackedEnsembleModel.StackedEnsembleParameters,StackedEnsembleModel.StackedEnsembleOutput>