std::vector<TargetPhrase*> PhraseDictionaryTransliteration::CreateTargetPhrases(const Phrase &sourcePhrase, const string &outDir) const
{
std::vector<TargetPhrase*> ret;
string outPath = outDir + "/out.txt";
ifstream outStream(outPath.c_str());
string line;
while (getline(outStream, line)) {
vector<string> toks;
Tokenize(toks, line, "\t");
UTIL_THROW_IF2(toks.size() != 2, "Error in transliteration output file. Expecting word\tscore");
TargetPhrase *tp = new TargetPhrase();
Word &word = tp->AddWord();
word.CreateFromString(Output, m_output, toks[0], false);
float score = Scan<float>(toks[1]);
tp->GetScoreBreakdown().PlusEquals(this, score);
// score of all other ff when this rule is being loaded
tp->Evaluate(sourcePhrase, GetFeaturesToApply());
ret.push_back(tp);
}
outStream.close();
return ret;
}
void PhraseDictionaryMultiModelCounts::CollectSufficientStatistics(const Phrase& src, vector<float> &fs, map<string,multiModelCountsStatistics*>* allStats) const
//fill fs and allStats with statistics from models
{
for(size_t i = 0; i < m_numModels; ++i) {
const PhraseDictionary &pd = *m_pd[i];
TargetPhraseCollection *ret_raw = (TargetPhraseCollection*) pd.GetTargetPhraseCollection( src);
if (ret_raw != NULL) {
TargetPhraseCollection::iterator iterTargetPhrase;
for (iterTargetPhrase = ret_raw->begin(); iterTargetPhrase != ret_raw->end(); ++iterTargetPhrase) {
TargetPhrase * targetPhrase = *iterTargetPhrase;
vector<float> raw_scores = targetPhrase->GetScoreBreakdown().GetScoresForProducer(&pd);
string targetString = targetPhrase->GetStringRep(m_output);
if (allStats->find(targetString) == allStats->end()) {
multiModelCountsStatistics * statistics = new multiModelCountsStatistics;
statistics->targetPhrase = new TargetPhrase(*targetPhrase); //make a copy so that we don't overwrite the original phrase table info
// zero out scores from original phrase table
statistics->targetPhrase->GetScoreBreakdown().ZeroDenseFeatures(&pd);
statistics->fst.resize(m_numModels);
statistics->ft.resize(m_numModels);
Scores scoreVector(5);
scoreVector[0] = -raw_scores[0];
scoreVector[1] = -raw_scores[1];
scoreVector[2] = -raw_scores[2];
statistics->targetPhrase->GetScoreBreakdown().Assign(this, scoreVector); // set scores to 0
statistics->targetPhrase->Evaluate(src, GetFeaturesToApply());
(*allStats)[targetString] = statistics;
}
multiModelCountsStatistics * statistics = (*allStats)[targetString];
statistics->fst[i] = UntransformScore(raw_scores[0]);
statistics->ft[i] = UntransformScore(raw_scores[1]);
fs[i] = UntransformScore(raw_scores[2]);
(*allStats)[targetString] = statistics;
}
}
}
// get target phrase frequency for models which have not seen the phrase pair
for ( map< string, multiModelCountsStatistics*>::const_iterator iter = allStats->begin(); iter != allStats->end(); ++iter ) {
multiModelCountsStatistics * statistics = iter->second;
for (size_t i = 0; i < m_numModels; ++i) {
if (!statistics->ft[i]) {
statistics->ft[i] = GetTargetCount(static_cast<const Phrase&>(*statistics->targetPhrase), i);
}
}
}
}
TargetPhrase *SkeletonPT::CreateTargetPhrase(const Phrase &sourcePhrase) const
{
// create a target phrase from the 1st word of the source, prefix with 'SkeletonPT:'
CHECK(sourcePhrase.GetSize());
CHECK(m_output.size() == 1);
string str = sourcePhrase.GetWord(0).GetFactor(0)->GetString().as_string();
str = "SkeletonPT:" + str;
TargetPhrase *tp = new TargetPhrase();
Word &word = tp->AddWord();
word.CreateFromString(Output, m_output, str, false);
// score for this phrase table
vector<float> scores(m_numScoreComponents, 1.3);
tp->GetScoreBreakdown().PlusEquals(this, scores);
// score of all other ff when this rule is being loaded
tp->Evaluate(sourcePhrase, GetFeaturesToApply());
return tp;
}
TargetPhraseCollection* PhraseDictionaryMultiModelCounts::CreateTargetPhraseCollectionCounts(const Phrase &src, vector<float> &fs, map<string,multiModelCountsStatistics*>* allStats, vector<vector<float> > &multimodelweights) const
{
TargetPhraseCollection *ret = new TargetPhraseCollection();
for ( map< string, multiModelCountsStatistics*>::const_iterator iter = allStats->begin(); iter != allStats->end(); ++iter ) {
multiModelCountsStatistics * statistics = iter->second;
if (statistics->targetPhrase->GetAlignTerm().GetSize() == 0) {
UTIL_THROW(util::Exception, " alignment information empty\ncount-tables need to include alignment information for computation of lexical weights.\nUse --phrase-word-alignment during training; for on-disk tables, also set -alignment-info when creating on-disk tables.");
}
try {
pair<vector< set<size_t> >, vector< set<size_t> > > alignment = GetAlignmentsForLexWeights(src, static_cast<const Phrase&>(*statistics->targetPhrase), statistics->targetPhrase->GetAlignTerm());
vector< set<size_t> > alignedToT = alignment.first;
vector< set<size_t> > alignedToS = alignment.second;
double lexst = ComputeWeightedLexicalTranslation(static_cast<const Phrase&>(*statistics->targetPhrase), src, alignedToS, m_lexTable_e2f, multimodelweights[1], false );
double lexts = ComputeWeightedLexicalTranslation(src, static_cast<const Phrase&>(*statistics->targetPhrase), alignedToT, m_lexTable_f2e, multimodelweights[3], true );
Scores scoreVector(5);
scoreVector[0] = FloorScore(TransformScore(m_combineFunction(statistics->fst, statistics->ft, multimodelweights[0])));
scoreVector[1] = FloorScore(TransformScore(lexst));
scoreVector[2] = FloorScore(TransformScore(m_combineFunction(statistics->fst, fs, multimodelweights[2])));
scoreVector[3] = FloorScore(TransformScore(lexts));
scoreVector[4] = FloorScore(TransformScore(2.718));
statistics->targetPhrase->GetScoreBreakdown().Assign(this, scoreVector);
statistics->targetPhrase->Evaluate(src, GetFeaturesToApply());
} catch (AlignmentException& e) {
continue;
}
ret->Add(new TargetPhrase(*statistics->targetPhrase));
}
RemoveAllInMap(*allStats);
delete allStats;
return ret;
}
void PhraseDictionaryFuzzyMatch::InitializeForInput(InputType const& inputSentence)
{
char dirName[] = "/tmp/moses.XXXXXX";
char *temp = mkdtemp(dirName);
UTIL_THROW_IF2(temp == NULL,
"Couldn't create temporary directory " << dirName);
string dirNameStr(dirName);
string inFileName(dirNameStr + "/in");
ofstream inFile(inFileName.c_str());
for (size_t i = 1; i < inputSentence.GetSize() - 1; ++i) {
inFile << inputSentence.GetWord(i);
}
inFile << endl;
inFile.close();
long translationId = inputSentence.GetTranslationId();
string ptFileName = m_FuzzyMatchWrapper->Extract(translationId, dirNameStr);
// populate with rules for this sentence
PhraseDictionaryNodeMemory &rootNode = m_collection[translationId];
FormatType format = MosesFormat;
// data from file
InputFileStream inStream(ptFileName);
// copied from class LoaderStandard
PrintUserTime("Start loading fuzzy-match phrase model");
const StaticData &staticData = StaticData::Instance();
const std::string& factorDelimiter = staticData.GetFactorDelimiter();
string lineOrig;
size_t count = 0;
while(getline(inStream, lineOrig)) {
const string *line;
if (format == HieroFormat) { // reformat line
UTIL_THROW(util::Exception, "Cannot be Hiero format");
//line = ReformatHieroRule(lineOrig);
} else {
// do nothing to format of line
line = &lineOrig;
}
vector<string> tokens;
vector<float> scoreVector;
TokenizeMultiCharSeparator(tokens, *line , "|||" );
if (tokens.size() != 4 && tokens.size() != 5) {
stringstream strme;
strme << "Syntax error at " << ptFileName << ":" << count;
UserMessage::Add(strme.str());
abort();
}
const string &sourcePhraseString = tokens[0]
, &targetPhraseString = tokens[1]
, &scoreString = tokens[2]
, &alignString = tokens[3];
bool isLHSEmpty = (sourcePhraseString.find_first_not_of(" \t", 0) == string::npos);
if (isLHSEmpty && !staticData.IsWordDeletionEnabled()) {
TRACE_ERR( ptFileName << ":" << count << ": pt entry contains empty target, skipping\n");
continue;
}
Tokenize<float>(scoreVector, scoreString);
const size_t numScoreComponents = GetNumScoreComponents();
if (scoreVector.size() != numScoreComponents) {
stringstream strme;
strme << "Size of scoreVector != number (" << scoreVector.size() << "!="
<< numScoreComponents << ") of score components on line " << count;
UserMessage::Add(strme.str());
abort();
}
UTIL_THROW_IF2(scoreVector.size() != numScoreComponents,
"Number of scores incorrectly specified");
// parse source & find pt node
// constituent labels
Word *sourceLHS;
Word *targetLHS;
// source
Phrase sourcePhrase( 0);
sourcePhrase.CreateFromString(Input, m_input, sourcePhraseString, factorDelimiter, &sourceLHS);
// create target phrase obj
TargetPhrase *targetPhrase = new TargetPhrase();
targetPhrase->CreateFromString(Output, m_output, targetPhraseString, factorDelimiter, &targetLHS);
// rest of target phrase
targetPhrase->SetAlignmentInfo(alignString);
targetPhrase->SetTargetLHS(targetLHS);
//targetPhrase->SetDebugOutput(string("New Format pt ") + line);
// component score, for n-best output
std::transform(scoreVector.begin(),scoreVector.end(),scoreVector.begin(),TransformScore);
std::transform(scoreVector.begin(),scoreVector.end(),scoreVector.begin(),FloorScore);
targetPhrase->GetScoreBreakdown().Assign(this, scoreVector);
targetPhrase->Evaluate(sourcePhrase, GetFeaturesToApply());
TargetPhraseCollection &phraseColl = GetOrCreateTargetPhraseCollection(rootNode, sourcePhrase, *targetPhrase, sourceLHS);
phraseColl.Add(targetPhrase);
count++;
if (format == HieroFormat) { // reformat line
delete line;
} else {
// do nothing
}
}
// sort and prune each target phrase collection
SortAndPrune(rootNode);
//removedirectoryrecursively(dirName);
}
