hex.deeplearning

Class DeepLearning

java.lang.Object

water.Iced

hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

hex.deeplearning.DeepLearning

All Implemented Interfaces:

java.io.Externalizable, java.io.Serializable, java.lang.Cloneable, water.Freezable

public class DeepLearning
extends hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

Deep Learning Neural Net implementation based on MRTask

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
class	DeepLearning.DeepLearningDriver

Nested classes/interfaces inherited from class hex.ModelBuilder

hex.ModelBuilder.BuilderVisibility, hex.ModelBuilder.Driver, hex.ModelBuilder.ValidationMessage

Field Summary

Fields inherited from class hex.ModelBuilder

_fold, _job, _messages, _nclass, _offset, _origDomains, _origNames, _parms, _priorClassDist, _removedCols, _response, _result, _train, _valid, _vresponse, _weights

Constructor Summary

Constructors

Constructor and Description
DeepLearning(boolean startup_once)
DeepLearning(DeepLearningModel.DeepLearningParameters parms) Main constructor from Deep Learning parameters
DeepLearning(DeepLearningModel.DeepLearningParameters parms, water.Key<DeepLearningModel> key)

Method Summary

Methods

Modifier and Type	Method and Description
hex.ModelCategory[]	can_build() Types of models we can build with DeepLearning
protected void	checkMemoryFootPrint()
protected int	desiredChunks(water.fvec.Frame original_fr, boolean local)
hex.ToEigenVec	getToEigenVec()
void	init(boolean expensive) Initialize the ModelBuilder, validating all arguments and preparing the training frame.
boolean	isSupervised()
void	modifyParmsForCrossValidationMainModel(hex.ModelBuilder[] cvModelBuilders)
protected int	nModelsInParallel()
protected water.fvec.Frame	rebalance(water.fvec.Frame original_fr, boolean local, java.lang.String name)
protected DeepLearning.DeepLearningDriver	trainModelImpl()

Methods inherited from class hex.ModelBuilder

algoName, algos, builderVisibility, checkDistributions, clearInitState, clearValidationErrors, computeCrossValidation, computePriorClassDistribution, cv_AssignFold, cv_buildModels, cv_mainModelScores, cv_makeFramesAndBuilders, cv_makeWeights, cv_scoreCVModels, defaultKey, dest, error_count, error, get, hasFoldCol, hasOffsetCol, hasWeightCol, hide, ignoreBadColumns, ignoreConstColumns, ignoreStringColumns, info, isClassifier, isStopped, javaName, logMe, make, message, nclasses, nFoldCV, nFoldWork, numSpecialCols, paramName, response, schemaDirectory, separateFeatureVecs, specialColNames, stop_requested, timeout, train, trainModel, trainModelNested, valid, validationErrors, vresponse, warn

Methods inherited from class water.Iced

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

Methods inherited from class java.lang.Object

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

Constructor Detail

DeepLearning

public DeepLearning(DeepLearningModel.DeepLearningParameters parms)

Main constructor from Deep Learning parameters

DeepLearning

public DeepLearning(DeepLearningModel.DeepLearningParameters parms,
            water.Key<DeepLearningModel> key)

DeepLearning

public DeepLearning(boolean startup_once)

Method Detail

can_build

public hex.ModelCategory[] can_build()

Types of models we can build with DeepLearning

Specified by:

can_build in class hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

getToEigenVec

public hex.ToEigenVec getToEigenVec()

Overrides:

getToEigenVec in class hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

isSupervised

public boolean isSupervised()

Overrides:

isSupervised in class hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

nModelsInParallel

protected int nModelsInParallel()

Overrides:

nModelsInParallel in class hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

trainModelImpl

protected DeepLearning.DeepLearningDriver trainModelImpl()

Specified by:

trainModelImpl in class hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

init

public void init(boolean expensive)

Initialize the ModelBuilder, validating all arguments and preparing the training frame. This call is expected to be overridden in the subclasses and each subclass will start with "super.init();". This call is made by the front-end whenever the GUI is clicked, and needs to be fast; heavy-weight prep needs to wait for the trainModel() call. Validate the very large number of arguments in the DL Parameter directly.

Overrides:

init in class hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

checkMemoryFootPrint

protected void checkMemoryFootPrint()

Overrides:

checkMemoryFootPrint in class hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

modifyParmsForCrossValidationMainModel

public void modifyParmsForCrossValidationMainModel(hex.ModelBuilder[] cvModelBuilders)

Overrides:

modifyParmsForCrossValidationMainModel in class hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

rebalance

protected water.fvec.Frame rebalance(water.fvec.Frame original_fr,
                         boolean local,
                         java.lang.String name)

Overrides:

rebalance in class hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>

desiredChunks

protected int desiredChunks(water.fvec.Frame original_fr,
                boolean local)

Overrides:

desiredChunks in class hex.ModelBuilder<DeepLearningModel,DeepLearningModel.DeepLearningParameters,DeepLearningModel.DeepLearningModelOutput>