void printOverlapReport(const tmodel<COUNT, PROB>& tTable,
sentenceHandler& testHandler, vcbList& trainEList,
vcbList& trainFList, vcbList& testEList, vcbList& testFList)
{
set<pair<WordIndex, WordIndex> > testCoocur ;
sentPair s ;
/* string unseenCoocurFile = Prefix + ".tst.unseen.cooc" ;
ofstream of_unseenCoocur(unseenCoocurFile.c_str());
string seenCoocurFile = Prefix + ".tst.seen.cooc" ;
ofstream of_seenCoocur(seenCoocurFile.c_str());
*/
testHandler.rewind();
int seen_coocur = 0, unseen_coocur = 0, srcUnk = 0, trgUnk = 0 ;
while(testHandler.getNextSentence(s)){
for (WordIndex i = 1 ; i < s.eSent.size() ; i++)
for (WordIndex j = 1 ; j < s.fSent.size() ; j++)
testCoocur.insert(pair<WordIndex, WordIndex> (s.eSent[i], s.fSent[j])) ;
}
set<pair<WordIndex, WordIndex> >::const_iterator i ;
for (i = testCoocur.begin() ; i != testCoocur.end() ; ++i){
if (tTable.getProb((*i).first, (*i).second) > PROB_SMOOTH){
seen_coocur ++ ;
// of_seenCoocur << (*i).first << ' ' << (*i).second << '\n';
}
else {
unseen_coocur++;
// of_unseenCoocur << (*i).first << ' ' << (*i).second << '\n';
}
}
string trgUnkFile = Prefix + ".tst.trg.unk" ;
ofstream of_trgUnk(trgUnkFile.c_str());
for (WordIndex i = 0 ; i < testFList.getVocabList().size() && i < testFList.uniqTokens();i++)
if (testFList.getVocabList()[i].freq > 0 && trainFList.getVocabList()[i].freq <= 0){
of_trgUnk << i << ' ' << testFList.getVocabList()[i].word << ' ' << testFList.getVocabList()[i].freq
<< '\n';
trgUnk++ ;
}
string srcUnkFile = Prefix + ".tst.src.unk" ;
ofstream of_srcUnk(srcUnkFile.c_str());
for (WordIndex j = 0 ; j < testEList.getVocabList().size() && j < testEList.uniqTokens();j++)
if (testEList.getVocabList()[j].freq > 0 && trainEList.getVocabList()[j].freq <= 0){
srcUnk++ ;
of_srcUnk << j << ' ' << testEList.getVocabList()[j].word << ' ' << testEList.getVocabList()[j].freq
<< '\n';
}
string summaryFile = Prefix + ".tst.stats" ;
ofstream of_summary(summaryFile.c_str());
of_summary << "\t\t STATISTICS ABOUT TEST CORPUS\n\n";
of_summary << "source unique tokens: " << testEList.uniqTokens() << '\n';
of_summary << "target unique tokens: " << testFList.uniqTokens() << '\n';
of_summary << "unique unseen source tokens: " << srcUnk << '\n';
of_summary << "unique unseen target tokens: " << trgUnk << '\n';
of_summary << "cooccurrences not found in the final t table: " << unseen_coocur << '\n';
of_summary << "cooccurrences found in the final t table: " << seen_coocur << '\n';
}
void tmodel<COUNT, PROB>::normalizeTable(const vcbList&engl, const vcbList&french, int iter)
// normalize conditional probability P(fj/ei):
// i.e. make sure that Sum over all j of P(fj/e) = 1
// this method reads the counts portion of the table and normalize into
// the probability portion. Then the counts are cleared (i.e. zeroed)
// if the resulting probability of an entry is below a threshold, then
// remove it .
{
if( iter==2 )
{
total2.resize(engl.uniqTokens());for(unsigned int i=0;i<total2.size();i++)total2[i]=0.0;
}
nFrench.resize(engl.uniqTokens());for(unsigned int i=0;i<nFrench.size();i++)nFrench[i]=0;
nEng.resize(french.uniqTokens());for(unsigned int i=0;i<nEng.size();i++)nEng[i]=0;
Vector<double> total(engl.uniqTokens(),0.0);
//Vector<int> nFrench(engl.uniqTokens(), 0);
//Vector<int> nEng(french.uniqTokens(), 0);
typename hash_map<wordPairIds, CPPair, hashpair, equal_to<wordPairIds> >::const_iterator i;
for(i = ef.begin(); i != ef.end(); i++){ // for all possible source words e
if( iter==2 )
total2[((*i).first).first] += (*i).second.count;
total[((*i).first).first] += (*i).second.count;
nFrench[((*i).first).first]++;
nEng[((*i).first).second]++;
}
for(unsigned int k=0;k<engl.uniqTokens();++k)
if( nFrench[k] )
{
double probMass=(french.uniqTokensInCorpus()-nFrench[k])*PROB_SMOOTH;
if( probMass<0.0 )
cout << k << " french.uniqTokensInCorpus(): " << french.uniqTokensInCorpus() << " nFrench[k]:"<< nFrench[k] << '\n';
total[k]+= total[k]*probMass/(1-probMass);
}
typename hash_map<wordPairIds, CPPair, hashpair, equal_to<wordPairIds> >::iterator j, k;
PROB p ;
int nParams=0;
int nor = 0;
for(j = ef.begin(); j != ef.end(); ){
k = j;
k++ ;
if( (total[((*j).first).first])>0.0 )
p = ((((*j).second).count) /(total[((*j).first).first])) ;
else
p= 0.0;
if (p > PROB_CUTOFF)
{
if( iter>0 )
{
if(useWord2Vec && word2vec.Method == 2){
p = word2vec.L * word2vec.getW2VProb(((*j).first).first, ((*j).first).second) + (1. - word2vec.L) * p;
nor = 1;
}
((*j).second).prob = 0 ;
((*j).second).count = p ;
}
else
{
((*j).second).prob = p ;
((*j).second).count = 0 ;
}
nParams++;
}
else {
erase(((*j).first).first, ((*j).first).second);
}
j = k ;
}
if(nor)
cout << "probabilities Normalized in iteration = " << iter << endl;
if( iter>0 )
return normalizeTable(engl, french, iter-1);
else
{
}
}