hex.glm
Class LSMSolver

java.lang.Object
  water.Iced
      hex.glm.LSMSolver

All Implemented Interfaces:

java.lang.Cloneable, Freezable

Direct Known Subclasses:

LSMSolver.ADMMSolver, LSMSolver.GeneralizedGradientSolver

public abstract class LSMSolver
extends Iced

Distributed least squares solvers

Nested Class Summary

static class	LSMSolver.ADMMSolver
static class	LSMSolver.GeneralizedGradientSolver Generalized gradient solver for solving LSM problem with combination of L1 and L2 penalty.
static class	LSMSolver.LSMSolverException
static class	LSMSolver.LSMSolverType

Field Summary

double	_alpha
protected boolean	_converged
double	_lambda

Constructor Summary

LSMSolver(double lambda, double alpha)

Method Summary

boolean	converged()
abstract java.lang.String	name()
abstract boolean	solve(Gram gram, double[] xy, double yy, double[] newBeta)
abstract com.google.gson.JsonObject	toJson()

Methods inherited from class water.Iced

clone, frozenType, init, newInstance, read, toDocField, write, writeJSON, writeJSONFields

Methods inherited from class java.lang.Object

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

Field Detail

_lambda

public double _lambda

_alpha

public final double _alpha

_converged

protected boolean _converged

Constructor Detail

LSMSolver

public LSMSolver(double lambda,
                 double alpha)

Method Detail

solve

public abstract boolean solve(Gram gram,
                              double[] xy,
                              double yy,
                              double[] newBeta)

Parameters:

xx - - gram matrix. gaussian: X'X, binomial:(1/4)X'X

xy - - guassian: -X'y binomial: -(1/4)X'(XB + (y-p)/(p*1-p))

yy - - /2

beta - - previous vector of coefficients, will be modified/destroyed

newBeta - - resulting vector of coefficients

Returns:

true if converged

toJson

public abstract com.google.gson.JsonObject toJson()

converged

public final boolean converged()

name

public abstract java.lang.String name()