Examples of cc.mallet.types.IDSorter

cc.mallet.types.IDSorter
This class is contains a comparator for use in sorting integers that have associated floating point values. One example would be sorting words by probability in a Naive Bayes model. Ties are broken by the ID. IDSorter[] sortedIDs = new IDSorter[n]; for (int i=0; i<n; i++) { sortedIDs[i] = new IDSorter(i, weights[i]); } Arrays.sort(sortedIDs); for (int i=0; i<10; i++) { @author David Mimno

    int[] topicCounts = new int[ numTopics ];


    IDSorter[] sortedTopics = new IDSorter[ numTopics ];
    for (int topic = 0; topic < numTopics; topic++) {
      // Initialize the sorters with dummy values
      sortedTopics[topic] = new IDSorter(topic, topic);
    }


    if (max < 0 || max > numTopics) {
      max = numTopics;
    }

View Full Code Here

        wordCount+=count;
        
        IDSorter[] sortedTypes = new IDSorter[numTypes];
        //sort word in topic k
        for (int type=0; type < numTypes; type++) {
          sortedTypes[type] = new IDSorter(type, nkt.get(k)[type]);
        }
        
        Arrays.sort(sortedTypes);
      
        Alphabet alphabet = instances.getDataAlphabet();

View Full Code Here

          wordCount+=nk.get(k);
          
          IDSorter[] sortedTypes = new IDSorter[numTypes];
          //sort word in topic k
          for (int type=0; type < numTypes; type++) {
            sortedTypes[type] = new IDSorter(type, nkt.get(k)[type]);
          }
          
          Arrays.sort(sortedTypes);
        
          Alphabet alphabet = instances.getDataAlphabet();

View Full Code Here

            DataBag topic = BagFactory.getInstance().newDefaultBag();
            
            // Get the set of keywords with weights for this topic and add them as tuples
            // to the databag used to represent this topic
            while (iterator.hasNext() && topic.size() < numKeywords) {
                IDSorter info = iterator.next();
                Tuple weightedWord = TupleFactory.getInstance().newTuple(2);
                String wordToken = model.alphabet.lookupObject(info.getID()).toString(); // get the actual term text
                weightedWord.set(0, wordToken);
                weightedWord.set(1, info.getWeight()); // the raw weight of the term
                topic.add(weightedWord);
            }
            Tuple topicTuple = TupleFactory.getInstance().newTuple(2);
            topicTuple.set(0, topicNum);
            topicTuple.set(1, topic);

View Full Code Here

TOP

Related Classes of cc.mallet.types.IDSorter

cc.mallet.topics.HDP

cc.mallet.topics.LDAStream

varaha.topic.LDATopics

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.