Package edu.umd.cloud9.io.array

Examples of edu.umd.cloud9.io.array.ArrayListOfIntsWritable

107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
      String lang = p.getLanguage();
      int langID;

      ArrayListWritable<Text> sentences;
      ArrayListWritable<HMapSFW> vectors = new ArrayListWritable<HMapSFW>();
      ArrayListOfIntsWritable sentLengths = new ArrayListOfIntsWritable();
      // identify sentences in document, filter out ones below MinSentLength threshold
      // convert each sentence into a tf-idf vector, using general DF map for collection and a heuristic for avg. doc length
      // filter out sentences for which the vector has less than MinVectorTerms terms
  
      try {
        String article = p.getContent();
        if (lang.equals(eLang)) {
          sentences = helper.getESentences(article, vectors, sentLengths);   
          langID = CLIRUtils.E;
        }else if (lang.equals(fLang)){
          // Turkish Wiki articles' XML does not encode paragraph breaks
          if (lang.equals("tr")) {
            article = article.replaceAll("\\.", ". ");
          }
          sentences = helper.getFSentences(article, vectors, sentLengths);
          langID = CLIRUtils.F;
        }else {
          throw new RuntimeException("Unknown language: " + lang);
        }
      } catch (Exception e) {
        e.printStackTrace();
        throw new RuntimeException("Error in sentence detection for language: " + lang + ", helper: " + helper + " article title: " + p.getTitle());
      }

      // documents with no sentences (after we filter out some by length)
      if (sentences.size() == 0) {
        if (langID == CLIRUtils.E) {
          reporter.incrCounter(Docs.EEmpty, 1); 
        }else {
          reporter.incrCounter(Docs.FEmpty, 1);
        }
      }else{
        if (langID == CLIRUtils.E) {
          reporter.incrCounter(Docs.E, 1);
        }else {
          reporter.incrCounter(Docs.F, 1);
        }
      }  

      for (int i = 0; i < sentences.size(); i++) {
        if (langID == CLIRUtils.E) {
          if (helper.getEOOVRate(sentences.get(i).toString()) > maxOOVRate ) {
            reporter.incrCounter(Sentences.OOV, 1);
            continue;
          }         
          reporter.incrCounter(Sentences.ELength, sentLengths.get(i));
          reporter.incrCounter(Sentences.E, 1);   
        }else {
          if (helper.getFOOVRate(sentences.get(i).toString()) > maxOOVRate ) {
            reporter.incrCounter(Sentences.OOV, 1);
            continue;
          }
          reporter.incrCounter(Sentences.FLength, sentLengths.get(i));
          reporter.incrCounter(Sentences.F, 1);   
        }
        keyOut.set(docno, langID);     
        valOut.set(langID, sentences.get(i), vectors.get(i));
        output.collect(keyOut, valOut);
View Full Code Here
17
18
19
20
21
22
23
24
25
26
27
   * @param i
   *     length of permutation
   */
  public PermutationByBit(int i){
    rand = new Random();
    randPerm = new ArrayListOfIntsWritable();
    for(int j = 0; j < i; j++){
      randPerm.add(j);
    }
    length = i;
  }
View Full Code Here
32
33
34
35
36
37
38
      int i = rand.nextInt(length);
      int j = randPerm.get(i);
      randPerm.set(i, randPerm.get(k));
      randPerm.set(k, j);
    }
    return new ArrayListOfIntsWritable(randPerm);
  } 
View Full Code Here
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
  public static void main(String[] args){
    int SIZE = 1000;
    PermutationByBit p = new PermutationByBit(SIZE);
    for(int k=0;k<100;k++){
      ArrayListOfIntsWritable a = p.nextPermutation();

      //make sure the permutation is not out of bounds
      for(int i=0;i<SIZE;i++){
        assertTrue(i+"-->"+a.get(i),a.get(i)<SIZE && a.get(i)>=0);
      }

      //make sure each position is included in the permutation exactly once
      int[] positions = new int[SIZE];
      for(int i=0;i<SIZE;i++){
        if(positions[a.get(i)]==1){
          fail("Same position included twice: "+a.get(i));
        }
        positions[a.get(i)]=1;
      }
      for(int i=0;i<SIZE;i++){
        if(positions[i]==0){
          //          System.out.println(java.util.Arrays.binarySearch(positions, i));
          fail("Position not included: "+i);
View Full Code Here
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
          int fDocno = key.getRightElement();
          fDocno -= 1000000000;
          int eDocno = key.getLeftElement();
          if(langID == CLIRUtils.E){
            if(!pwsimMapping.containsKey(eDocno)){
              pwsimMapping.put(eDocno, new ArrayListOfIntsWritable());
            }
            pwsimMapping.get(eDocno).add(fDocno);   // we add 1000000000 to foreign docnos to distinguish them during pwsim algo
          }else{
            if(!pwsimMapping.containsKey(fDocno)){
              pwsimMapping.put(fDocno, new ArrayListOfIntsWritable());
            }
            pwsimMapping.get(fDocno).add(eDocno);   // we add 1000000000 to foreign docnos to distinguish them during pwsim algo
          }
          key = (PairOfInts) reader.getKeyClass().newInstance();
          value = (IntWritable) reader.getValueClass().newInstance();
View Full Code Here
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
      String lang = p.getLanguage();
      int langID;

      ArrayListWritable<Text> sentences;
      ArrayListWritable<HMapSFW> vectors = new ArrayListWritable<HMapSFW>();
      ArrayListOfIntsWritable sentLengths = new ArrayListOfIntsWritable();
      try {
        // identify sentences in document, filter out ones below MinSentLength threshold
        // convert each sentence into a tf-idf vector, using general DF map for collection and a heuristic for avg. doc length
        // filter out sentences for which the vector has less than MinVectorTerms terms
        String article = p.getContent();
        if (lang.equals("en")) {
          sentences = helper.getESentences(article, vectors, sentLengths);   
          langID = CLIRUtils.E;
        }else {
          // Turkish Wiki articles' XML does not encode paragraph breaks
          if (lang.equals("tr")) {
            article = article.replaceAll("\\.", ". ");
          }
          sentences = helper.getFSentences(article, vectors, sentLengths);
          langID = CLIRUtils.F;
        }
      } catch (Exception e) {
        e.printStackTrace();
        throw new RuntimeException(e);
      }

      // documents with no sentences (after we filter out some by length)
      if (sentences.size() == 0) {
        if (langID == CLIRUtils.E) {
          reporter.incrCounter(Docs.EEmpty, 1); 
        }else {
          reporter.incrCounter(Docs.FEmpty, 1);
        }
      }else{
        if (langID == CLIRUtils.E) {
          reporter.incrCounter(Docs.E, 1);
        }else {
          reporter.incrCounter(Docs.F, 1);
        }
      }  

      for (int i = 0; i < sentences.size(); i++) {
        if (langID == CLIRUtils.E) {
          if (helper.getEInVocabRate(sentences.get(i).toString()) < minInVocabRate ) {
            reporter.incrCounter(Sentences.OOV, 1);
            return;
          }         
          reporter.incrCounter(Sentences.ELength, sentLengths.get(i));
          reporter.incrCounter(Sentences.E, 1);   
        }else {
          if (helper.getFInVocabRate(sentences.get(i).toString()) < minInVocabRate ) {
            reporter.incrCounter(Sentences.OOV, 1);
            return;
          }
          reporter.incrCounter(Sentences.FLength, sentLengths.get(i));
          reporter.incrCounter(Sentences.F, 1);   
        }
        keyOut.set(docno, langID);     
        valOut.set(langID, sentences.get(i), vectors.get(i));
        output.collect(keyOut, valOut);
View Full Code Here
243
244
245
246
247
248
249
250
251
252
253
    }
    SequenceFile.Writer writer = SequenceFile.createWriter(fs, job, new Path(randomPermFile),
        IntWritable.class, ArrayListOfIntsWritable.class);
    Permutation p = new PermutationByBit(numBits);
    for (int i = 0; i < numOfPermutations; i++) {
      ArrayListOfIntsWritable perm = p.nextPermutation();
      writer.append(new IntWritable(i), perm);
      sLogger.debug(i + ":" + perm);
    }
    writer.close();
    sLogger.info("Random permutations written.");
View Full Code Here
40
41
42
43
44
45
46
47
48
49
50
    System.out.println(s);

    int loopcnt = 0;
    MinhashSignature permutedS = new MinhashSignature(D);
    while (loopcnt++ < 10) {
      ArrayListOfIntsWritable a = p.nextPermutation();
      s.perm(a, permutedS);
      for (int i = 0; i < s.size(); i++) {
        assertTrue(permutedS.containsTerm(s.get(i)));
      }
      assertTrue(permutedS.size() == s.size());
View Full Code Here
108
109
110
111
112
113
114
115
116
117
        if (!anchor.isExternalInLink() && !anchor.isInternalInLink()) {
          continue;
        }

        keyOut.set(anchor.getText());
        anchorTextTarget.setSources(new ArrayListOfIntsWritable(anchor.getDocuments()));
        anchorTextTarget.setWeight(weightingScheme.getWeight(key.get(), anchor));
        output.collect(keyOut, anchorTextTarget);
      }
    }
View Full Code Here
34
35
36
37
38
39
40
41
42
43
44
    }
    System.out.println(s);

    PermutationByBit p = new PermutationByBit(64);
    SixtyFourBitSignature permutedS = new SixtyFourBitSignature(64);
    ArrayListOfIntsWritable a;
    int loopcnt = 0;
    while (loopcnt++ < 1000000) {
        a = p.nextPermutation();
        s.perm(a, permutedS);
       
View Full Code Here
TOP

Related Classes of edu.umd.cloud9.io.array.ArrayListOfIntsWritable

  • bak.pcj.IntIterator
  • cc.mrlda.InformedPrior
  • edu.umd.cloud9.example.bfs.BfsNodeTest
  • edu.umd.cloud9.example.pagerank.PageRankNodeTest
  • edu.umd.cloud9.io.array.ArrayListOfIntsWritable
  • edu.umd.cloud9.io.fastutil.Int2FloatOpenHashMapWritableTest
  • edu.umd.cloud9.io.fastutil.Int2IntOpenHashMapWritableTest
  • edu.umd.cloud9.io.fastutil.String2FloatOpenHashMapWritableTest
  • edu.umd.cloud9.io.fastutil.String2IntOpenHashMapWritableTest
  • edu.umd.cloud9.io.map.HMapIDW
    • 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.