gmm
Class RandomEstimator

java.lang.Object
  gmm.GMMEstimator
      gmm.RandomEstimator

public class RandomEstimator
extends GMMEstimator

This class implements a random GMM estimator. It generates permutations at random, and from these permutations and their reverses, it picks the one which maximizes the likelihood. Following Cohen, Schapire and Singer, we generate 10n random permutations, where n is the number of items being ranked.

Field Summary

Fields inherited from class gmm.GMMEstimator

GMM, logL, maxIters, n, pi, Q, runningTime, theta, V

Constructor Summary

RandomEstimator(boolean GMM, float[][] Q)

Method Summary

void	run() This method causes the estimator to run, and estimate the GMM parameters.

Methods inherited from class gmm.GMMEstimator

computeLogL, computeThetas, computeVjs, getAvgDk, getLogL, getLogLForQ, getPi, getRunningTime, lnPsiJPrime, minimizeF, minimizeFj, minValueOfFj, printLearnedModel, PsiJ, sumOfLnPsiJPrime

Methods inherited from class java.lang.Object

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

Constructor Detail

RandomEstimator

public RandomEstimator(boolean GMM,
                       float[][] Q)

Method Detail

run

public void run()
         throws java.lang.Exception

Description copied from class: GMMEstimator

This method causes the estimator to run, and estimate the GMM parameters. Any estimator must provide a concrete implementation of this method.

Specified by:

run in class GMMEstimator

Throws:

java.lang.Exception