/* Copyright (C) 2003 Univ. of Massachusetts Amherst, Computer Science Dept.
This file is part of "MALLET" (MAchine Learning for LanguagE Toolkit).
http://www.cs.umass.edu/~mccallum/mallet
This software is provided under the terms of the Common Public License,
version 1.0, as published by http://www.opensource.org. For further
information, see the file `LICENSE' included with this distribution. */
package cc.mallet.grmm.inference;
import java.util.Iterator;
import cc.mallet.grmm.types.*;
import cc.mallet.types.MatrixOps;
import gnu.trove.THashSet;
/**
* A bunch of static utilities useful for dealing with Inferencers.
* Created: Jun 1, 2005
*
* @author <A HREF="mailto:casutton@cs.umass.edu>casutton@cs.umass.edu</A>
* @version $Id: Utils.java,v 1.1 2007/10/22 21:37:49 mccallum Exp $
*/
public class Utils {
/**
* Returns ths value of -log Z in mdl according to the given inferencer.
* If inf is exact, the answer will be exact; otherwise the answer will be
* approximation
*
* @param mdl
* @param inf An inferencer. <tt>inf.computeMarginals (mdl)</tt> must already have
* been called.
* @return The value of -logZ
*/
public static double lookupMinusLogZ (FactorGraph mdl, Inferencer inf)
{
// heuristically try to find a reasonable assignment (not numerically 0 prob)
int [] vals = new int[mdl.numVariables()];
for (int vi = 0; vi < vals.length; vi++) {
Variable var = mdl.getVariable (vi);
Factor mrg = inf.lookupMarginal (var);
vals[vi] = mrg.argmax();
}
Assignment assn = new Assignment (mdl, vals);
double prob = inf.lookupLogJoint (assn);
double energy = mdl.logValue (assn);
return prob - energy;
}
public static double localMagnetization (Inferencer inferencer, Variable var)
{
if (var.getNumOutcomes () != 2)
throw new IllegalArgumentException ();
Factor marg = inferencer.lookupMarginal (var);
AssignmentIterator it = marg.assignmentIterator ();
double v1 = marg.value (it); it.advance ();
double v2 = marg.value (it);
return v1 - v2;
}
public static double[] allL1MarginalDistance (FactorGraph mdl, Inferencer inf1, Inferencer inf2)
{
double[] dist = new double [mdl.numVariables ()];
int i = 0;
for (Iterator it = mdl.variablesIterator (); it.hasNext();) {
Variable var = (Variable) it.next ();
Factor bel1 = inf1.lookupMarginal (var);
Factor bel2 = inf2.lookupMarginal (var);
dist[i++] = Factors.oneDistance (bel1, bel2);
}
return dist;
}
public static double avgL1MarginalDistance (FactorGraph mdl, Inferencer inf1, Inferencer inf2)
{
double[] dist = allL1MarginalDistance (mdl, inf1, inf2);
return MatrixOps.mean (dist);
}
public static double maxL1MarginalDistance (FactorGraph mdl, Inferencer inf1, Inferencer inf2)
{
double[] dist = allL1MarginalDistance (mdl, inf1, inf2);
return MatrixOps.max (dist);
}
public static int[] toSizesArray (Variable[] vars)
{
int[] szs = new int [vars.length];
for (int i = 0; i < vars.length; i++) {
szs[i] = vars[i].getNumOutcomes ();
}
return szs;
}
public static VarSet defaultIntersection (VarSet v1, VarSet v2)
{// Grossly inefficient implementation
THashSet hset = new THashSet (v1);
hset.retainAll (v2);
Variable[] ret = new Variable [hset.size ()];
int vai = 0;
for (int vi = 0; vi < v1.size(); vi++) {
Variable var = v1.get (vi);
if (hset.contains (var)) { ret[vai++] = var; }
}
return new HashVarSet (ret);
}
}