Package edu.umd.cloud9.io

Examples of edu.umd.cloud9.io.SequenceFileUtils


    List<Integer> sortedIndices = new ArrayList<Integer>();
    HMapIF index2ProbMap = new HMapIF();

    float sumOfProbs = 0.0f;    //only extract the top K<15 if the mass prob. exceeds MAX_probThreshold
    while(!topTrans.isEmpty() && sumOfProbs < cumProbThreshold){
      PairOfFloatString e = topTrans.pollLast();
      String term = e.getRightElement();
      float pr = e.getLeftElement()/cumProb;    // normalize
      logger.debug(term+"-->"+pr);
      int trgIndex = trgVocab.addOrGet(term);
      sumOfProbs += e.getLeftElement();         // keep track of unnormalized cumulative prob for determining cutoff
      sortedIndices.add(trgIndex);
      index2ProbMap.put(trgIndex, pr);
    }

    // to enable faster access with binary search, we sort entries by vocabulary index.
View Full Code Here


          int e2 = eVocabTrg.get(eTerm);        

          float prob2 = f2e_Probs.get(f2, e2);
          float prob = prob1*prob2;
          sumOfProbs += prob;
          topTrans.add(new PairOfFloatString(prob, fTerm));
        }
        logger.info("Adding "+eTerm);
        addToTable(e1, topTrans, sumOfProbs, table, fVocabTrg, 1.0f, stats);     
      }
      logger.info(stats);
View Full Code Here

    if (featSet > 2) {
      // uppercase token matching features : find uppercased tokens that exactly appear on both sides
      // lack of this evidence does not imply anything, but its existence might indicate parallel
//      fSentence.replaceAll("([',:;.?%!])", " $1 ");
//      eSentence.replaceAll("([',:;.?%!])", " $1 ");
      PairOfFloats pair = getUppercaseRatio(fTokenizer.processContent(fSentence), eTokenizer.processContent(eSentence));
      features.add("uppercaseratio1=" + pair.getLeftElement() );
      features.add("uppercaseratio2=" + pair.getRightElement() );
    }

    if (featSet > 3) {
      // future work = count number of single/double letter words in src and trg side
View Full Code Here

    // now, read tokens in first sentence and keep track of sequences of uppercased tokens in buffer
    HashSet<String> upperCaseMap1 = getUppercaseParts(tokens1);
    HashSet<String> upperCaseMap2 = getUppercaseParts(tokens2);
    float cntUpperRatio1 = getRatio(upperCaseMap1, upperCaseMap2);
    float cntUpperRatio2 = getRatio(upperCaseMap2, upperCaseMap1);
    PairOfFloats result = new PairOfFloats(cntUpperRatio1, cntUpperRatio2);
    return result;
  }
View Full Code Here

                String term = m2.group(1);
                if ( !term.equals("NULL") ) {
                  float prob = Float.parseFloat(m2.group(2));
                  int engIndex = trgVocab.addOrGet(term);
                  logger.debug("Added: "+term+" with index: "+engIndex+" and prob:"+prob);
                  indexProbPairs.add(new PairOfIntFloat(engIndex, prob));
                  sumOfProbs += prob;
                }
              }
            }
            // if number of translations not set, we never cut-off, so all cases are long tails
View Full Code Here

                String term = m2.group(1);
                if (!term.equals("NULL")) {
                  float prob = Float.parseFloat(m2.group(2));
                  int engIndex = trgVocab.addOrGet(term);
                  logger.debug("Added: "+term+" with index: "+engIndex+" and prob:"+prob);
                  indexProbPairs.add(new PairOfIntFloat(engIndex, prob));
                  sumOfProbs+=prob;
                }
              }
            }
            if(sumOfProbs > probThreshold){
View Full Code Here

    @Override
    public void reduce(Text key, Iterable<PairOfIntLong> values, Context context)
        throws IOException, InterruptedException {
      String term = key.toString();
      Iterator<PairOfIntLong> iter = values.iterator();
      PairOfIntLong p = iter.next();
      int df = p.getLeftElement();
      long cf = p.getRightElement();
      WritableUtils.writeVInt(dfByTermOut, df);
      WritableUtils.writeVLong(cfByTermOut, cf);
      if (iter.hasNext()) {
        throw new RuntimeException("More than one record for term: " + term);
      }
View Full Code Here

      // map from the id back to text
      // sLogger.info("termid: " + key);
      String term = mTermIdMap.getTerm(key.get());
      // sLogger.info("term: " + term);
      PairOfIntLong pair = gs.getStats(term);

      if (pair == null) {
        p.setCf(-1);
        p.setDf(-1);
      } else {
        p.setCf(pair.getRightElement());
        p.setDf(pair.getLeftElement());
      }

      output.collect(key, p);
    }
View Full Code Here

      float sumProb2 = 0;
      for (Entry<String> entry : probDist.entrySet()) {
        float pr = entry.getValue() / sumProb;
        if (pr > lexProbThreshold) {
          sumProb2 += pr;
          sortedFilteredProbDist.add(new PairOfStringFloat(entry.getKey(), pr));
        }
      }

      // re-normalize values after removal of low-prob entries
      float cumProb = 0;
      int cnt = 0;
      while (cnt < maxNumTrans && cumProb < cumProbThreshold && !sortedFilteredProbDist.isEmpty()) {
        PairOfStringFloat entry = sortedFilteredProbDist.pollLast();
        float pr = entry.getValue() / sumProb2;
        cumProb += pr;
        normProbDist.put(entry.getKey(), pr);
        cnt++;
      }

      probMap.put(sourceTerm, normProbDist);
    }
View Full Code Here

        String[] parts = rule.split("\\|\\|\\|");
        String[] lhs = parts[0].trim().split(" ");
        String[] rhs = parts[1].trim().split(" ");;
        for (String l : lhs) {
          for (String r : rhs) {
            pairsInSCFG.add(new PairOfStrings(l, r));
          }
        }
      }
    } catch (UnsupportedEncodingException e) {
      e.printStackTrace();
View Full Code Here

TOP

Related Classes of edu.umd.cloud9.io.SequenceFileUtils

Copyright © 2018 www.massapicom. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.