org.apache.mahout.math.hadoop.solver

Class DistributedConjugateGradientSolver

java.lang.Object

org.apache.mahout.math.solver.ConjugateGradientSolver

org.apache.mahout.math.hadoop.solver.DistributedConjugateGradientSolver

All Implemented Interfaces:

org.apache.hadoop.conf.Configurable, org.apache.hadoop.util.Tool

public class DistributedConjugateGradientSolver
extends ConjugateGradientSolver
implements org.apache.hadoop.util.Tool

Distributed implementation of the conjugate gradient solver. More or less, this is just the standard solver but wrapped with some methods that make it easy to run it on a DistributedRowMatrix.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
class	DistributedConjugateGradientSolver.DistributedConjugateGradientSolverJob

Field Summary

Fields inherited from class org.apache.mahout.math.solver.ConjugateGradientSolver

DEFAULT_MAX_ERROR

Constructor Summary

Constructors

Constructor and Description
DistributedConjugateGradientSolver()

Method Summary

Modifier and Type	Method and Description
org.apache.hadoop.conf.Configuration	getConf()
DistributedConjugateGradientSolver.DistributedConjugateGradientSolverJob	job()
static void	main(String[] args)
int	run(String[] strings)
Vector	runJob(org.apache.hadoop.fs.Path inputPath, org.apache.hadoop.fs.Path tempPath, int numRows, int numCols, Vector b, Preconditioner preconditioner, int maxIterations, double maxError) Runs the distributed conjugate gradient solver programmatically to solve the system (A + lambda*I)x = b.
void	setConf(org.apache.hadoop.conf.Configuration conf)

Methods inherited from class org.apache.mahout.math.solver.ConjugateGradientSolver

getIterations, getResidualNorm, solve, solve, solve

Methods inherited from class java.lang.Object

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

Constructor Detail

DistributedConjugateGradientSolver

public DistributedConjugateGradientSolver()

Method Detail

runJob

public Vector runJob(org.apache.hadoop.fs.Path inputPath,
                     org.apache.hadoop.fs.Path tempPath,
                     int numRows,
                     int numCols,
                     Vector b,
                     Preconditioner preconditioner,
                     int maxIterations,
                     double maxError)

Runs the distributed conjugate gradient solver programmatically to solve the system (A + lambda*I)x = b.

Parameters:

inputPath - Path to the matrix A

tempPath - Path to scratch output path, deleted after the solver completes

numRows - Number of rows in A

numCols - Number of columns in A

b - Vector b

preconditioner - Optional preconditioner for the system

maxIterations - Maximum number of iterations to run, defaults to numCols

maxError - Maximum error tolerated in the result. If the norm of the residual falls below this, then the algorithm stops and returns.

Returns:

The vector that solves the system.

getConf

public org.apache.hadoop.conf.Configuration getConf()

Specified by:

getConf in interface org.apache.hadoop.conf.Configurable

setConf

public void setConf(org.apache.hadoop.conf.Configuration conf)

Specified by:

setConf in interface org.apache.hadoop.conf.Configurable

run

public int run(String[] strings)
        throws Exception

Specified by:

run in interface org.apache.hadoop.util.Tool

Throws:

Exception

job

public DistributedConjugateGradientSolver.DistributedConjugateGradientSolverJob job()

main

public static void main(String[] args)
                 throws Exception

Throws:

Exception