Examples of ArrayListOfInts

edu.umd.cloud9.util.array.ArrayListOfInts
Object representing a list of ints, backed by an resizable-array.

Examples of edu.umd.cloud9.util.array.ArrayListOfInts

    }


    // in SCFG rule such as a b X1 X2 c --> X1 d e X2 f, we want to find out the src/trg tokens that are aligned to some trg/src token, ignoring the X variable
    // we can then decide if we want to include it as a multi-token phrase in our query representation based on various heuristics (e.g., only include if no X in between of tokens)
    String fPhrase = "";
    ArrayListOfInts sourceTokenIds = new ArrayListOfInts();      
    ArrayListOfInts targetTokenIds = new ArrayListOfInts();
    int f=0;
    for (; f < lhs.length; f++) {
      String fTerm = lhs[f];
      if (queryLangTokenizer.isStopWord(fTerm) || fTerm.matches("\\[X,\\d+\\]") || fTerm.matches("<s>") || fTerm.matches("</s>")) {
        continue;
      }


      srcTokenCnt.increment(fTerm);
      sourceTokenIds.add(f);


      ArrayListOfInts ids;
      if (isPassThrough){
        ids = new ArrayListOfInts();
        ids.add(0);
      }else {
        ids = one2manyAlign.get(f);
      }


      if (ids == null || (isOne2Many == 0 && ids.size() > 1)) {
        continue;
      }


      // find phrase in LHS and match to phrase in RHS
      if (isMany2Many) {
        fPhrase += fTerm + " ";
        targetTokenIds = targetTokenIds.mergeNoDuplicates(ids);        
      }


      String eTerm = null;
      for (int e : ids) {
        eTerm = rhs[e];


        // assumption: if this is pass-through rule, re-stem token in doc-language
        if (isPassThrough || (unknownWords != null && unknownWords.contains(fTerm))) {
          eTerm = stemmed2Stemmed.get(eTerm);
        }


        if (eTerm == null || docLangTokenizer.isStopWord(eTerm)) {
          //          LOG.info("Skipped trg token " + eTerm);
          eTerm = null;
          continue;      
        }
        bagOfTargetTokens.add(eTerm);
        if (isOne2Many <= 1) {
          if (probDist.containsKey(fTerm)) {
            HMapSFW eToken2Prob = probDist.get(fTerm);
            eToken2Prob.increment(eTerm, weight);
          }else {
            HMapSFW eToken2Prob = new HMapSFW();
            eToken2Prob.put(eTerm, weight);
            probDist.put(fTerm, eToken2Prob);
          }
        }
      }


      if (isOne2Many == 2) {
        // if ids.size() > 1 eTerm is a multi-token expression
        // even if eTerm is overwritten here, we need to do above loop to update bagOfTargetTokens
        if (ids.size() > 1) {
          eTerm = isConsecutiveWithStopwords(ids, rhs, docLangTokenizer);     // <---- heuristic
        }


        // no proper translation on target-side (e.g., stopword OR non-consecutive multi-word translation), let's skip
        if (eTerm == null) {

View Full Code Here

Examples of edu.umd.cloud9.util.array.ArrayListOfInts

      String[] alPair = alignment.split("-");
      int f = Integer.parseInt(alPair[0]);
      int e = Integer.parseInt(alPair[1]);


      if(!one2manyAlign.containsKey(f)){
        one2manyAlign.put(f, new ArrayListOfInts());  
      }
      one2manyAlign.get(f).add(e);
    }


    // for each source token id, sort ids of its translations in ascending order
    for(Integer f : one2manyAlign.keySet()) {
      ArrayListOfInts lst = one2manyAlign.get(f);
      lst.sort();
      one2manyAlign.put(f, lst);
    }


    return one2manyAlign;
  }

View Full Code Here

Examples of edu.umd.cloud9.util.array.ArrayListOfInts


      // Remember, token position is numbered started from one...
      if (positions.containsKey(term)) {
        positions.get(term).add(i + 1);
      } else {
        ArrayListOfInts l = new ArrayListOfInts();
        l.add(i + 1);
        positions.put(term, l);
      }
    }


    int doclength = 0;
    Iterator<Map.Entry<String, ArrayListOfInts>> it = positions.entrySet().iterator();
    Map.Entry<String, ArrayListOfInts> e;
    ArrayListOfInts positionsList;
    while (it.hasNext()) {
      e = it.next();
      positionsList = e.getValue();


      // We're storing tfs as shorts, so check for overflow...
      if (positionsList.size() >= TF_CUT) {
        // There are a few ways to handle this... If we're getting such a high tf, then it most
        // likely means that this is a junk doc.
        LOG.warn("Error: tf of " + e.getValue()
            + " will overflow max short value. docno=" + doc.getDocid() + ", term="
            + e.getKey());
        it.remove();
      } else {
        positionsList.trimToSize();
        doclength += positionsList.size();
      }
    }


    if ( positions.size() == 0 ) {
      return positions;
    }


    positions.put("", new ArrayListOfInts(new int[] { doclength }));
    return positions;
  }

View Full Code Here

Examples of edu.umd.cloud9.util.array.ArrayListOfInts

        String[] alignments = parts[4].trim().split(" ");
        HMapIV<ArrayListOfInts> one2manyAlign = readAlignments(alignments);


        // append all target tokens that are aligned to some source token (other than nonterminals and sentence boundary markers)
        String fPhrase = "";
        ArrayListOfInts sourceTokenIds = new ArrayListOfInts();
        ArrayListOfInts targetTokenIds = new ArrayListOfInts();
        int f = 0;
        for (; f < lhs.length; f++) {
          String fTerm = lhs[f];
          if (fTerm.matches("\\[X,\\d+\\]") || fTerm.matches("<s>") || fTerm.matches("</s>")) {
            continue;
          }
          sourceTokenIds.add(f);
          ArrayListOfInts ids = one2manyAlign.get(f);


          // word-to-word translations
          if (!isPhrase) {
            // source token should be aligned to single target token
            if (ids == null || ids.size() != 1) continue;
            for (int e : ids) {
              String eTerm = rhs[e];
              if (docLangTokenizer.isStemmedStopWord(eTerm))  continue;
              if (scfgDist.containsKey(fTerm)) {
                HMapSFW eToken2Prob = scfgDist.get(fTerm);

View Full Code Here

Examples of edu.umd.cloud9.util.array.ArrayListOfInts

    for(String alignment : alignments){
      String[] alPair = alignment.split("-");
      int f = Integer.parseInt(alPair[0]);
      int e = Integer.parseInt(alPair[1]);
      if(!one2manyAlign.containsKey(f)){
        one2manyAlign.put(f, new ArrayListOfInts());  
      }
      one2manyAlign.get(f).add(e);
    }
    return one2manyAlign;
  }

View Full Code Here

Examples of edu.umd.cloud9.util.array.ArrayListOfInts

   *     length of permutation
   */
  public PermutationByBlock(int i){    
    blocks = new ArrayListOfInts[NUM_BLOCKS];
    for(int k=0;k<5;k++){
      blocks[k] = new ArrayListOfInts();
      int len=BLOCK_SIZE;
      if(k==(NUM_BLOCKS-1)){
        len = BLOCK_SIZE-1;
      }
      for(int j = (k*BLOCK_SIZE); j < (k*BLOCK_SIZE)+len; j++){

View Full Code Here

Examples of edu.umd.cloud9.util.array.ArrayListOfInts

    }
    return 0.0f;
  }
  
  public int[] getTranslations(float probThreshold){
    ArrayListOfInts words = new ArrayListOfInts(); 
    if (_useBinSearch) {
      for (int i=0; i < _data.length; i++) {
        if (_data[i] > probThreshold) {
          words.add(_indices[i]);
        }
      }
    }else{
      for (int i=0; i < _data.length; i++) {
        if (_data[i] > probThreshold) {
          words.add(i);
        }
      }
    }
    words.trimToSize();
    return words.getArray();
  }

View Full Code Here

Examples of edu.umd.cloud9.util.array.ArrayListOfInts


      context.getCounter(ReachableNodes.ReachableInMapper).increment(1);
      // Retain distance to self.
      map.put(nid.get(), node.getDistance());


      ArrayListOfInts adj = node.getAdjacenyList();
      int dist = node.getDistance() + 1;
      // Keep track of shortest distance to neighbors.
      for (int i = 0; i < adj.size(); i++) {
        int neighbor = adj.get(i);


        // Keep track of distance if it's shorter than previously
        // encountered, or if we haven't encountered this node.
        if ((map.containsKey(neighbor) && dist < map.get(neighbor)) || !map.containsKey(neighbor)) {
          map.put(neighbor, dist);

View Full Code Here

Examples of edu.umd.cloud9.util.array.ArrayListOfInts

   *     length of permutation
   */
  public PermutationByBlock(int i){    
    blocks = new ArrayListOfInts[NUM_BLOCKS];
    for(int k=0;k<5;k++){
      blocks[k] = new ArrayListOfInts();
      int len=BLOCK_SIZE;
      if(k==(NUM_BLOCKS-1)){
        len = BLOCK_SIZE-1;
      }
      for(int j = (k*BLOCK_SIZE); j < (k*BLOCK_SIZE)+len; j++){

View Full Code Here

Examples of edu.umd.cloud9.util.array.ArrayListOfInts

        String[] alignments = parts[4].trim().split(" ");
        HMapIV<ArrayListOfInts> one2manyAlign = readAlignments(alignments);


        // append all target tokens that are aligned to some source token (other than nonterminals and sentence boundary markers)
        String fPhrase = "";
        ArrayListOfInts sourceTokenIds = new ArrayListOfInts();
        ArrayListOfInts targetTokenIds = new ArrayListOfInts();
        int f = 0;
        for (; f < lhs.length; f++) {
          String fTerm = lhs[f];
          if (fTerm.matches("\\[X,\\d+\\]") || fTerm.matches("<s>") || fTerm.matches("</s>")) {
            continue;
          }
          sourceTokenIds.add(f);
          ArrayListOfInts ids = one2manyAlign.get(f);


          // word-to-word translations
          if (!isPhrase) {
            // source token should be aligned to single target token
            if (ids == null || ids.size() != 1) continue;
            for (int e : ids) {
              String eTerm = rhs[e];
              if (docLangTokenizer.isStopWord(eTerm))  continue;
              if (scfgDist.containsKey(fTerm)) {
                HMapSFW eToken2Prob = scfgDist.get(fTerm);

View Full Code Here

0 1 2

TOP

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.