statistics
Class ParallelML

java.lang.Object
  extended by statistics.ParallelML

public class ParallelML
extends java.lang.Object

A parallel implementation of the maximum likelihood estimation. Uses a ChunkedDataSet and high level threads for a convenient and easy parallelization.

Author:
sikoried

Constructor Summary
ParallelML(int featureDimension, ChunkedDataSet data, int numCores, boolean diagonalCovariance)
          Generate a new parallel ML estimator.
 
Method Summary
 Density mlEstimate()
          Return the estimated density.
 
Methods inherited from class java.lang.Object
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

ParallelML

public ParallelML(int featureDimension,
                  ChunkedDataSet data,
                  int numCores,
                  boolean diagonalCovariance)
Generate a new parallel ML estimator.

Parameters:
featureDimension -
data -
numCores -
diagonalCovariance -
Method Detail

mlEstimate

public Density mlEstimate()
                   throws java.io.IOException,
                          java.lang.InterruptedException
Return the estimated density. Caches the result for subsequent calls.

Throws:
java.io.IOException
java.lang.InterruptedException