hex.glrm

Class GLRM

java.lang.Object

water.Iced

hex.ModelBuilder<GLRMModel,GLRMModel.GLRMParameters,GLRMModel.GLRMOutput>

hex.glrm.GLRM

All Implemented Interfaces:

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

public class GLRM
extends hex.ModelBuilder<GLRMModel,GLRMModel.GLRMParameters,GLRMModel.GLRMOutput>

Generalized Low Rank Models This is an algorithm for dimensionality reduction of a dataset. It is a general, parallelized optimization algorithm that applies to a variety of loss and regularization functions. Categorical columns are handled by expansion into 0/1 indicator columns for each level. Generalized Low Rank Models

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
protected static class	GLRM.Archetypes
static class	GLRM.updateXVecs

Nested classes/interfaces inherited from class hex.ModelBuilder

hex.ModelBuilder.BuilderVisibility, hex.ModelBuilder.Driver, hex.ModelBuilder.FilterCols, hex.ModelBuilder.ModelTrainingCoordinator, hex.ModelBuilder.ValidationMessage

Field Summary

Fields inherited from class hex.ModelBuilder

_coordinator, _desc, _eventPublisher, _fold, _input_parms, _job, _messages, _nclass, _offset, _orig_projection_array, _origDomains, _origNames, _origTrain, _parms, _priorClassDist, _removedCols, _response, _result, _startUpOnceModelBuilder, _train, _treatment, _valid, _vresponse, _weights

Constructor Summary

Constructors

Constructor and Description
GLRM(boolean startup_once)
GLRM(GLRMModel.GLRMParameters parms)
GLRM(GLRMModel.GLRMParameters parms, water.Job<GLRMModel> job)

Method Summary

Modifier and Type	Method and Description
hex.ModelCategory[]	can_build()
protected void	checkMemoryFootPrint_impl()
static double[][]	expandCats(double[][] sdata, DataInfo dinfo)
static void	findGoodCidx(water.fvec.Frame xVecs, java.util.ArrayList<java.lang.Integer> currentList, boolean addIndex, int taIndex, int xNChunks, int startTAcidx)
static java.util.ArrayList<java.lang.Integer>	findtAChunkIndices(water.fvec.Frame tAVecs, int xStart, int xEnd, GLRM.Archetypes yt)
static java.util.ArrayList<java.lang.Integer>	findXChunkIndices(water.fvec.Frame xVecs, int taStart, int taEnd, GLRM.Archetypes yt)
static double	frobenius2(double[][] x)
static void	getXChunk(water.fvec.Frame xVecs, int chunkIdx, water.fvec.Chunk[] xChunks)
boolean	hasClosedForm(long na_cnt)
boolean	haveMojo()
boolean	havePojo()
protected static int	idx_xnew(int c, int ncolA, int ncolX)
protected static int	idx_xold(int c, int ncolA)
void	init(boolean expensive) Validate all parameters, and prepare the model for training.
boolean	isSupervised()
void	setWideDataset(boolean isWide)
protected hex.glrm.GLRM.GLRMDriver	trainModelImpl()
static double[][]	transform(double[][] centers, double[] normSub, double[] normMul, int ncats, int nnums)
static water.fvec.Chunk	xFrameVec(water.fvec.Chunk[] chks, int c, int offset)
static double	yArcheTypeVal(GLRM.Archetypes yt, int j, int k)

Methods inherited from class hex.ModelBuilder

algoName, algos, builderVisibility, canLearnFromNAs, checkCustomMetricForEarlyStopping, checkDistributions, checkEarlyStoppingReproducibility, checkMemoryFootPrint, checkResponseVariable, clearInitState, clearValidationErrors, computeCrossValidation, computePriorClassDistribution, cv_buildModels, cv_canBuildMainModelInParallel, cv_computeAndSetOptimalParameters, cv_initStoppingParameters, cv_mainModelScores, cv_makeAggregateModelMetrics, cv_scoreCVModels, cv_updateOptimalParameters, defaultKey, desiredChunks, dest, error_count, error, get, getMessagesByFieldAndSeverity, getName, getSysProperty, getToEigenVec, hasFoldCol, hasOffsetCol, hasTreatmentCol, hasWeightCol, hide, ignoreBadColumns, ignoreConstColumns, ignoreInvalidColumns, ignoreStringColumns, ignoreUuidColumns, info, init_adaptFrameToTrain, init_getNClass, initWorkspace, isClassifier, isResponseOptional, isStopped, javaName, logMe, make, make, make, makeCVMetrics, makeCVModelBuilder, makeParameters, makePojoWriter, message, nclasses, nFoldCV, nFoldWork, nModelsInParallel, nModelsInParallel, nModelsInParallel, numSpecialCols, paramName, raiseReproducibilityWarning, rebalance, remainingTimeSecs, response, schemaDirectory, separateFeatureVecs, setMaxRuntimeSecsForMainModel, setTrain, setValid, shouldReorder, smallDataSize, stop_requested, timeout, train, trainModel, trainModel, trainModelNested, trainModelNested, trainModelOnH2ONode, valid, validateBinaryResponse, validateStoppingMetric, validationErrors, validationWarnings, vresponse, warn

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, toString, wait, wait, wait

Constructor Detail

GLRM

public GLRM(GLRMModel.GLRMParameters parms)

GLRM

public GLRM(GLRMModel.GLRMParameters parms,
            water.Job<GLRMModel> job)

GLRM

public GLRM(boolean startup_once)

Method Detail

trainModelImpl

protected hex.glrm.GLRM.GLRMDriver trainModelImpl()

Specified by:

trainModelImpl in class hex.ModelBuilder<GLRMModel,GLRMModel.GLRMParameters,GLRMModel.GLRMOutput>

can_build

public hex.ModelCategory[] can_build()

Specified by:

can_build in class hex.ModelBuilder<GLRMModel,GLRMModel.GLRMParameters,GLRMModel.GLRMOutput>

isSupervised

public boolean isSupervised()

Specified by:

isSupervised in class hex.ModelBuilder<GLRMModel,GLRMModel.GLRMParameters,GLRMModel.GLRMOutput>

havePojo

public boolean havePojo()

Overrides:

havePojo in class hex.ModelBuilder<GLRMModel,GLRMModel.GLRMParameters,GLRMModel.GLRMOutput>

haveMojo

public boolean haveMojo()

Overrides:

haveMojo in class hex.ModelBuilder<GLRMModel,GLRMModel.GLRMParameters,GLRMModel.GLRMOutput>

checkMemoryFootPrint_impl

protected void checkMemoryFootPrint_impl()

Overrides:

checkMemoryFootPrint_impl in class hex.ModelBuilder<GLRMModel,GLRMModel.GLRMParameters,GLRMModel.GLRMOutput>

setWideDataset

public void setWideDataset(boolean isWide)

init

public void init(boolean expensive)

Validate all parameters, and prepare the model for training.

Overrides:

init in class hex.ModelBuilder<GLRMModel,GLRMModel.GLRMParameters,GLRMModel.GLRMOutput>

frobenius2

public static double frobenius2(double[][] x)

hasClosedForm

public final boolean hasClosedForm(long na_cnt)

transform

public static double[][] transform(double[][] centers,
                                   double[] normSub,
                                   double[] normMul,
                                   int ncats,
                                   int nnums)

expandCats

public static double[][] expandCats(double[][] sdata,
                                    DataInfo dinfo)

idx_xold

protected static int idx_xold(int c,
                              int ncolA)

idx_xnew

protected static int idx_xnew(int c,
                              int ncolA,
                              int ncolX)

yArcheTypeVal

public static double yArcheTypeVal(GLRM.Archetypes yt,
                                   int j,
                                   int k)

xFrameVec

public static water.fvec.Chunk xFrameVec(water.fvec.Chunk[] chks,
                                         int c,
                                         int offset)

getXChunk

public static void getXChunk(water.fvec.Frame xVecs,
                             int chunkIdx,
                             water.fvec.Chunk[] xChunks)

findtAChunkIndices

public static java.util.ArrayList<java.lang.Integer> findtAChunkIndices(water.fvec.Frame tAVecs,
                                                                        int xStart,
                                                                        int xEnd,
                                                                        GLRM.Archetypes yt)

findXChunkIndices

public static java.util.ArrayList<java.lang.Integer> findXChunkIndices(water.fvec.Frame xVecs,
                                                                       int taStart,
                                                                       int taEnd,
                                                                       GLRM.Archetypes yt)

findGoodCidx

public static void findGoodCidx(water.fvec.Frame xVecs,
                                java.util.ArrayList<java.lang.Integer> currentList,
                                boolean addIndex,
                                int taIndex,
                                int xNChunks,
                                int startTAcidx)