VotedPerceptron Class Reference

VotedPerceptron algorithm (see "Discriminative Training of Markov Logic Networks", Singla and Domingos, 2005). More...

#include <votedperceptron.h>

List of all members.

Public Member Functions
	VotedPerceptron (const Array< Inference * > &inferences, const StringHashArray &nonEvidPredNames, IndexTranslator *const &idxTrans, const bool &lazyInference, const bool &rescaleGradient, const bool &withEM)
	Constructor.
	~VotedPerceptron ()
void	setMeansStdDevs (const int &arrSize, const double *const &priorMeans, const double *const &priorStdDevs)
void	learnWeights (double *const &weights, const int &numWeights, const int &maxIter, const double &learningRate, const double &momentum, bool initWithLogOdds)

---

Detailed Description

VotedPerceptron algorithm (see "Discriminative Training of Markov Logic Networks", Singla and Domingos, 2005).

Definition at line 82 of file votedperceptron.h.

---

Constructor & Destructor Documentation

VotedPerceptron::VotedPerceptron	(	const Array< Inference * > &	inferences,
		const StringHashArray &	nonEvidPredNames,
		IndexTranslator *const &	idxTrans,
		const bool &	lazyInference,
		const bool &	rescaleGradient,
		const bool &	withEM
	)			[inline]

Constructor.

Various variables are initialized, relevant clauses are determined and weights and inference procedures are initialized.

Parameters:

	inferences	Array of inference procedures to be used for inference in each domain.
	nonEvidPredNames	Names of non-evidence predicates. This is used to determine the relevant clauses.
	idxTrans	IndexTranslator needed when multiple dbs are used and they don't line up.
	lazyInference	If true, lazy inference is used.
	rescaleGradient	If true, gradient is rescaled with each iteration.
	withEM	If true, EM is used to fill in missing values.

Definition at line 100 of file votedperceptron.h.

References Array< Type >::append(), Array< Type >::growToSize(), and Array< Type >::size().

    : domainCnt_(inferences.size()), idxTrans_(idxTrans),
      lazyInference_(lazyInference), rescaleGradient_(rescaleGradient),
      withEM_(withEM)
  { 
    cout << endl << "Constructing voted perceptron..." << endl << endl;

    inferences_.append(inferences);
    logOddsPerDomain_.growToSize(domainCnt_);
    clauseCntPerDomain_.growToSize(domainCnt_);
    
    for (int i = 0; i < domainCnt_; i++)
    {
      clauseCntPerDomain_[i] =
        inferences_[i]->getState()->getMLN()->getNumClauses();
      logOddsPerDomain_[i].growToSize(clauseCntPerDomain_[i], 0);
    }

    totalTrueCnts_.growToSize(domainCnt_);
    defaultTrueCnts_.growToSize(domainCnt_);
    relevantClausesPerDomain_.growToSize(domainCnt_);
    //relevantClausesFormulas_ is set in findRelevantClausesFormulas()

    findRelevantClauses(nonEvidPredNames);
    findRelevantClausesFormulas();

      // Initialize the clause wts for lazy version
    if (lazyInference_)
    {
      findCountsInitializeWtsAndSetNonEvidPredsToUnknownInDB(nonEvidPredNames);
    
      for (int i = 0; i < domainCnt_; i++)
      {
        const MLN* mln = inferences_[i]->getState()->getMLN();
        Array<double>& logOdds = logOddsPerDomain_[i];
        assert(mln->getNumClauses() == logOdds.size());
        for (int j = 0; j < mln->getNumClauses(); j++)
          ((Clause*) mln->getClause(j))->setWt(logOdds[j]);
      }
    }
      // Initialize the clause wts for eager version
    else
    {      
      initializeWts();
    }
    
      // Initialize the inference / state
    for (int i = 0; i < inferences_.size(); i++)
      inferences_[i]->init();
  }

---

The documentation for this class was generated from the following file:

• src/learnwts/votedperceptron.h

---