ComputationState (h2o-algos 3.46.0 API)

java.lang.Object
- hex.glm.ComputationState

public final class ComputationState
extends java.lang.Object

Nested Class Summary

Nested Classes
Modifier and Type	Class and Description
`static class`	`ComputationState.GLMSubsetGinfo` This method will grab a subset of the gradient for each multinomial class.
`static class`	`ComputationState.GramXY` Cached state of COD (with covariate updates) solver.

Field Summary

Fields
Modifier and Type Field and Description

DataInfo[] _activeDataMultinomial

boolean _lsNeeded

java.lang.String[] _randCoeffNames

Fields
Modifier and Type	Field and Description
`DataInfo[]`	`_activeDataMultinomial`
`boolean`	`_lsNeeded`
`java.lang.String[]`	`_randCoeffNames`

Constructor Summary

Constructors
Constructor and Description
`ComputationState(water.Job job, GLMModel.GLMParameters parms, DataInfo dinfo, GLM.BetaConstraint bc, GLM.BetaInfo bi)`
`ComputationState(water.Job job, GLMModel.GLMParameters parms, DataInfo dinfo, GLM.BetaConstraint bc, GLM.BetaInfo bi, double[][][] penaltyMat, int[][] gamColInd)`

Method Summary

All Methods Static Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`GLM.BetaConstraint`	`activeBC()`
`DataInfo`	`activeData()`
`DataInfo`	`activeDataMultinomial()`
`DataInfo`	`activeDataMultinomial(int c)`
`void`	`addOffset2Cols(int[] cols)`
`double`	`alpha()`
`protected void`	`applyStrongRules(double lambdaNew, double lambdaOld)` Apply strong rules to filter out expected inactive (with zero coefficient) predictors.
`protected int`	`applyStrongRulesMultinomial_old(double lambdaNew, double lambdaOld)` Apply strong rules to filter out expected inactive (with zero coefficient) predictors.
`protected void`	`applyStrongRulesMultinomial(double lambdaNew, double lambdaOld)` Apply strong rules to filter out expected inactive (with zero coefficient) predictors.
`double[]`	`beta()`
`double[]`	`betaMultinomial()`
`double[]`	`betaMultinomial(int c, double[] beta)`
`double[]`	`betaMultinomialFull(int c, double[] beta)`
`protected boolean`	`checkKKTs()`
`protected boolean`	`checkKKTsMultinomial()`
`ComputationState.GramXY`	`computeGram(double[] beta, GLMModel.GLMParameters.Solver s)`
`ComputationState.GramXY`	`computeGramRCC(double[] beta, GLMModel.GLMParameters.Solver s)` This method is used only for multinomial family.
`protected ComputationState.GramXY`	`computeNewGram(DataInfo activeData, double[] beta, GLMModel.GLMParameters.Solver s)`
`boolean`	`converged()`
`double`	`deviance()`
`boolean`	`dispersionEstimated()`
`void`	`dropActiveData()`
`double[]`	`expandBeta(double[] beta)`
`static double[]`	`expandToFullArray(double[] shortenArr, int[] activeCols, int _totalBetaLength, int nclasses, int betaLengthPerClass)`
`static double[]`	`extractSubRange(int N, int c, int[] ids, double[] src)` This method will return a double array that is extracted from src (which includes active and non-active columns) to only include active columns stated in ids.
`static int[]`	`genActiveColsAllClass(int activeColsLen, int numBetaPerClass, int[] activeColsOrig, int nclasses)`
`int[]`	`genActiveColsIndClass(int activeColsLen, int numBetaPerClass, int[] activeColsOrig, int activeClass, int nclasses)`
`double`	`get_correction_HL()`
`double[]`	`get_phi()`
`water.fvec.Frame`	`get_priorw_wpsi()`
`double[]`	`get_psi()`
`double`	`get_tau()`
`boolean`	`getLambdaNull()`
`GLM.GLMGradientInfo`	`ginfo()`
`ComputationState.GLMSubsetGinfo`	`ginfoMultinomial(int c)`
`ComputationState.GLMSubsetGinfo`	`ginfoMultinomialRCC(int c)`
`boolean`	`ginfoNull()`
`GLM.GLMGradientSolver`	`gslvr()`
`OptimizationUtils.GradientSolver`	`gslvrMultinomial(int c)`
`double`	`l1pen()`
`double`	`l2pen()`
`double`	`lambda()`
`double`	`likelihood()`
`double`	`objective()`
`double`	`objective(double[] beta, double likelihood)`
`int[]`	`removeCols(int[] cols)`
`void`	`set_beta_HGLM(double[] beta, int startIdx, int len, boolean interceptFirst)` Copy GLM coefficients stored in beta to _beta of computationState
`void`	`set_likelihoodInfo(double hlik, double pvh, double pbvh, double cAIC)`
`void`	`set_phi(double[] phi)`
`void`	`set_psi(double[] psi)`
`void`	`set_sumEtaSquareConvergence(double[] sumInfo)`
`void`	`set_tau(double tau)`
`void`	`set_ubeta_HGLM(double[] ubeta, int startIdx, int len)`
`void`	`setActiveClass(int activeClass)`
`protected void`	`setActiveDataMultinomialNull()`
`protected void`	`setActiveDataNull()`
`protected void`	`setAllIn(boolean val)`
`void`	`setAlpha(double alpha)`
`void`	`setBC(GLM.BetaConstraint bc)`
`protected void`	`setBeta(double[] beta)`
`protected void`	`setBetaDiff(double betaDiff)`
`void`	`setBetaMultinomial(int c, double[] beta, double[] bc)`
`protected void`	`setGinfo(GLM.GLMGradientInfo ginfo)`
`void`	`setgMax(double gmax)`
`protected void`	`setGradientErr(double gErr)`
`protected void`	`setGslvrNull()`
`protected void`	`setHGLMComputationState(double[] beta, double[] ubeta, double[] psi, double[] phi, double hlcorrection, double tau, water.fvec.Frame wpsi, java.lang.String[] randCoeffNames)`
`protected void`	`setIter(int iteration)`
`void`	`setLambda(double lambda)`
`void`	`setLambdaMax(double lmax)`
`protected void`	`setLambdaSimple(double lambda)`
`protected void`	`setLikelihood(double llk)`
`void`	`setZValues(double[] zValues, boolean dispersionEstimated)`
`double[]`	`shrinkFullArray(double[] fullArray)`
`java.lang.String`	`toString()`
`double[]`	`ubeta()`
`protected double`	`updateState(double[] beta, double likelihood)`
`protected double`	`updateState(double[] beta, GLM.GLMGradientInfo ginfo)`
`double[]`	`zValues()`

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

_randCoeffNames

public java.lang.String[] _randCoeffNames

_lsNeeded
```
public boolean _lsNeeded
```

_activeDataMultinomial

public DataInfo[] _activeDataMultinomial

Constructor Detail

ComputationState

public ComputationState(water.Job job,
                        GLMModel.GLMParameters parms,
                        DataInfo dinfo,
                        GLM.BetaConstraint bc,
                        GLM.BetaInfo bi)

ComputationState

public ComputationState(water.Job job,
                        GLMModel.GLMParameters parms,
                        DataInfo dinfo,
                        GLM.BetaConstraint bc,
                        GLM.BetaInfo bi,
                        double[][][] penaltyMat,
                        int[][] gamColInd)

Method Detail

set_sumEtaSquareConvergence

public void set_sumEtaSquareConvergence(double[] sumInfo)

set_beta_HGLM
```
public void set_beta_HGLM(double[] beta,
                          int startIdx,
                          int len,
                          boolean interceptFirst)
```
Copy GLM coefficients stored in beta to _beta of computationState

Parameters:

beta: - store coefficients to be copied from

startIdx: - first index of beta to copy from

len: - length of coefficients to copy from beta

interceptFirst: - true if the first index of beta stored the intercept term

set_likelihoodInfo

public void set_likelihoodInfo(double hlik,
                               double pvh,
                               double pbvh,
                               double cAIC)

set_ubeta_HGLM

public void set_ubeta_HGLM(double[] ubeta,
                           int startIdx,
                           int len)

setZValues

public void setZValues(double[] zValues,
                       boolean dispersionEstimated)

get_psi
```
public double[] get_psi()
```

get_correction_HL
```
public double get_correction_HL()
```

get_phi
```
public double[] get_phi()
```

get_priorw_wpsi

public water.fvec.Frame get_priorw_wpsi()

get_tau
```
public double get_tau()
```

getLambdaNull
```
public boolean getLambdaNull()
```

set_tau
```
public void set_tau(double tau)
```

set_psi
```
public void set_psi(double[] psi)
```

set_phi
```
public void set_phi(double[] phi)
```

gslvr
```
public GLM.GLMGradientSolver gslvr()
```

lambda
```
public double lambda()
```

alpha
```
public double alpha()
```

zValues
```
public double[] zValues()
```

dispersionEstimated
```
public boolean dispersionEstimated()
```

setLambdaMax

public void setLambdaMax(double lmax)

setgMax
```
public void setgMax(double gmax)
```

setAlpha
```
public void setAlpha(double alpha)
```

setLambda
```
public void setLambda(double lambda)
```

beta
```
public double[] beta()
```

ubeta
```
public double[] ubeta()
```

ginfo
```
public GLM.GLMGradientInfo ginfo()
```

activeBC
```
public GLM.BetaConstraint activeBC()
```

likelihood
```
public double likelihood()
```

ginfoNull
```
public boolean ginfoNull()
```

activeData
```
public DataInfo activeData()
```

activeDataMultinomial

public DataInfo activeDataMultinomial()

dropActiveData
```
public void dropActiveData()
```

toString
```
public java.lang.String toString()
```
Overrides:

toString in class java.lang.Object

l1pen
```
public double l1pen()
```

l2pen
```
public double l2pen()
```

applyStrongRules

protected void applyStrongRules(double lambdaNew,
                                double lambdaOld)

Apply strong rules to filter out expected inactive (with zero coefficient) predictors.

activeDataMultinomial

public DataInfo activeDataMultinomial(int c)

extractSubRange
```
public static double[] extractSubRange(int N,
                                       int c,
                                       int[] ids,
                                       double[] src)
```
This method will return a double array that is extracted from src (which includes active and non-active columns) to only include active columns stated in ids.

Parameters:

N -

c -

ids -

src -

Returns:

betaMultinomial
```
public double[] betaMultinomial()
```

betaMultinomial

public double[] betaMultinomial(int c,
                                double[] beta)

betaMultinomialFull

public double[] betaMultinomialFull(int c,
                                    double[] beta)

shrinkFullArray

public double[] shrinkFullArray(double[] fullArray)

expandToFullArray

public static double[] expandToFullArray(double[] shortenArr,
                                         int[] activeCols,
                                         int _totalBetaLength,
                                         int nclasses,
                                         int betaLengthPerClass)

genActiveColsAllClass

public static int[] genActiveColsAllClass(int activeColsLen,
                                          int numBetaPerClass,
                                          int[] activeColsOrig,
                                          int nclasses)

genActiveColsIndClass

public int[] genActiveColsIndClass(int activeColsLen,
                                   int numBetaPerClass,
                                   int[] activeColsOrig,
                                   int activeClass,
                                   int nclasses)

ginfoMultinomial

public ComputationState.GLMSubsetGinfo ginfoMultinomial(int c)

ginfoMultinomialRCC

public ComputationState.GLMSubsetGinfo ginfoMultinomialRCC(int c)

setBC

public void setBC(GLM.BetaConstraint bc)

setActiveClass

public void setActiveClass(int activeClass)

deviance
```
public double deviance()
```

gslvrMultinomial

public OptimizationUtils.GradientSolver gslvrMultinomial(int c)

setBetaMultinomial

public void setBetaMultinomial(int c,
                               double[] beta,
                               double[] bc)

applyStrongRulesMultinomial_old
```
protected int applyStrongRulesMultinomial_old(double lambdaNew,
                                              double lambdaOld)
```
Apply strong rules to filter out expected inactive (with zero coefficient) predictors.

Returns:

indices of expected active predictors.

applyStrongRulesMultinomial

protected void applyStrongRulesMultinomial(double lambdaNew,
                                           double lambdaOld)

Apply strong rules to filter out expected inactive (with zero coefficient) predictors.

checkKKTsMultinomial

protected boolean checkKKTsMultinomial()

checkKKTs
```
protected boolean checkKKTs()
```

addOffset2Cols

public void addOffset2Cols(int[] cols)

removeCols
```
public int[] removeCols(int[] cols)
```

objective
```
public double objective()
```

objective

public double objective(double[] beta,
                        double likelihood)

updateState

protected double updateState(double[] beta,
                             double likelihood)

converged
```
public boolean converged()
```

updateState

protected double updateState(double[] beta,
                             GLM.GLMGradientInfo ginfo)

setBetaDiff

protected void setBetaDiff(double betaDiff)

setGradientErr

protected void setGradientErr(double gErr)

setGinfo

protected void setGinfo(GLM.GLMGradientInfo ginfo)

setBeta

protected void setBeta(double[] beta)

setIter

protected void setIter(int iteration)

setLikelihood

protected void setLikelihood(double llk)

setAllIn
```
protected void setAllIn(boolean val)
```

setGslvrNull
```
protected void setGslvrNull()
```

setActiveDataMultinomialNull

protected void setActiveDataMultinomialNull()

setActiveDataNull
```
protected void setActiveDataNull()
```

setLambdaSimple

protected void setLambdaSimple(double lambda)

setHGLMComputationState

protected void setHGLMComputationState(double[] beta,
                                       double[] ubeta,
                                       double[] psi,
                                       double[] phi,
                                       double hlcorrection,
                                       double tau,
                                       water.fvec.Frame wpsi,
                                       java.lang.String[] randCoeffNames)

expandBeta

public double[] expandBeta(double[] beta)

computeNewGram

protected ComputationState.GramXY computeNewGram(DataInfo activeData,
                                                 double[] beta,
                                                 GLMModel.GLMParameters.Solver s)

computeGramRCC
```
public ComputationState.GramXY computeGramRCC(double[] beta,
                                              GLMModel.GLMParameters.Solver s)
```
This method is used only for multinomial family. It differs from computeGram because it calls on _activeData which only contains only active columns in its _adaptedFrame. Note activeDataMultinomial(_activeClass) will always contains all predictors in its _adaptedFrame.

Parameters:

beta -

s -

Returns:

computeGram

public ComputationState.GramXY computeGram(double[] beta,
                                           GLMModel.GLMParameters.Solver s)

Class ComputationState

Nested Class Summary

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

_randCoeffNames

_lsNeeded

_activeDataMultinomial

Constructor Detail

ComputationState

ComputationState

Method Detail

set_sumEtaSquareConvergence

set_beta_HGLM

set_likelihoodInfo

set_ubeta_HGLM

setZValues

get_psi

get_correction_HL

get_phi

get_priorw_wpsi

get_tau

getLambdaNull

set_tau

set_psi

set_phi

gslvr

lambda

alpha

zValues

dispersionEstimated

setLambdaMax

setgMax

setAlpha

setLambda

beta

ubeta

ginfo

activeBC

likelihood

ginfoNull

activeData

activeDataMultinomial

dropActiveData

toString

l1pen

l2pen

applyStrongRules

activeDataMultinomial

extractSubRange

betaMultinomial

betaMultinomial

betaMultinomialFull

shrinkFullArray

expandToFullArray

genActiveColsAllClass

genActiveColsIndClass

ginfoMultinomial

ginfoMultinomialRCC

setBC

setActiveClass

deviance

gslvrMultinomial

setBetaMultinomial

applyStrongRulesMultinomial_old

applyStrongRulesMultinomial

checkKKTsMultinomial

checkKKTs

addOffset2Cols

removeCols

objective

objective

updateState

converged

updateState

setBetaDiff

setGradientErr

setGinfo