Package joshua.corpus

Examples of joshua.corpus.Phrase

72
73
74
75
76
77
78
79
80
81
82
  }
 
  @Test(dependsOnMethods = {"basicPattern"})
  public void copiedPattern() {
   
    Phrase phrase = pattern;
   
    Pattern copiedPattern = new Pattern(phrase);
   
    Assert.assertEquals(copiedPattern.getWordIDs().length, words.length);
    for (int i=0; i<words.length; i++) {
View Full Code Here
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
   
    Iterator<Map.Entry<Phrase,Integer>> i = map.entrySet().iterator();
    Assert.assertNotNull(i);
   
    Map.Entry<Phrase, Integer> entry;
    Phrase phrase;
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertEquals(phrase.toString(), "the");
 
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertEquals(phrase.toString(), "of");
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertEquals(phrase.toString(), "to");
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertEquals(phrase.toString(), "that");
   
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertEquals(phrase.toString(), ".");
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertEquals(phrase.toString(), "chromosome");

    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertEquals(phrase.toString(), "genes");
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertEquals(phrase.toString(), ",");
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertEquals(phrase.toString(), "is");
   
    Assert.assertFalse(i.hasNext());
   
  }
View Full Code Here
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
    LinkedHashMap<Phrase,Short> ranks = frequentPhrases.getRanks();
    Iterator<Map.Entry<Phrase,Short>> i = ranks.entrySet().iterator();
    Assert.assertNotNull(i);
   
    Map.Entry<Phrase,Short> entry;
    Phrase phrase;
    short rank;
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertNotNull(phrase);
    Assert.assertEquals(phrase.toString(), "the");
    rank = entry.getValue();
    Assert.assertEquals(rank, 0);
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertNotNull(phrase);
    Assert.assertEquals(phrase.toString(), "of");
    rank = entry.getValue();
    Assert.assertEquals(rank, 1);
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertNotNull(phrase);
    Assert.assertEquals(phrase.toString(), "to");
    rank = entry.getValue();
    Assert.assertEquals(rank, 2);
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertNotNull(phrase);
    Assert.assertEquals(phrase.toString(), "that");
    rank = entry.getValue();
    Assert.assertEquals(rank, 3);
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertNotNull(phrase);
    Assert.assertEquals(phrase.toString(), ".");
    rank = entry.getValue();
    Assert.assertEquals(rank, 4);
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertNotNull(phrase);
    Assert.assertEquals(phrase.toString(), "chromosome");
    rank = entry.getValue();
    Assert.assertEquals(rank, 5);
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertNotNull(phrase);
    Assert.assertEquals(phrase.toString(), "genes");
    rank = entry.getValue();
    Assert.assertEquals(rank, 6);

    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertNotNull(phrase);
    Assert.assertEquals(phrase.toString(), ",");
    rank = entry.getValue();
    Assert.assertEquals(rank, 7);
   
    Assert.assertTrue(i.hasNext());
    entry = i.next();
    phrase = entry.getKey();
    Assert.assertNotNull(phrase);
    Assert.assertEquals(phrase.toString(), "is");
    rank = entry.getValue();
    Assert.assertEquals(rank, 8);
   
    Assert.assertFalse(i.hasNext());
  }
View Full Code Here
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
   
    // Adam Lopez's example...
    String corpusString = "it makes him and it mars him , it sets him on and it takes him off .";

    vocab = new Vocabulary();
    Phrase exampleSentence = new BasicPhrase(corpusString, vocab);
   
    exampleSentence = new BasicPhrase(corpusString, vocab);
    int[] sentences = new int[1];
    sentences[0] = 0;
    int[] corpus = new int[exampleSentence.size()];
    for(int i = 0; i < exampleSentence.size(); i++) {
      corpus[i] = exampleSentence.getWordID(i);
    }
   
    CorpusArray corpusArray = new CorpusArray(corpus, sentences, vocab);
   
    if (binaryFileName==null || binaryFileName.trim().length()==0)
View Full Code Here
137
138
139
140
141
142
143
144
145
146
147
  @Test
  public void findPhrase() {
   
    // Look up phrase "it makes him"
   
    Phrase phrase = new BasicPhrase("it makes him", vocab);
    int[] bounds = suffixArray.findPhrase(phrase);
   
    int expectedSuffixArrayStartIndex = 0;
    int expectedSuffixArrayEndIndex = 0;
   
View Full Code Here
212
213
214
215
216
217
218
219
220
221
222
    }
    List<Phrase> phrases = new ArrayList<Phrase>();
   
    for (int i = 0; i < size(); i++) {
      for (int j = i + 1; (j <= size()) && (j - i <= maxLength); j++) {
        Phrase subPhrase = subPhrase(i,j);
        phrases.add(subPhrase);
      }
    }
    return phrases;
  }
View Full Code Here
158
159
160
161
162
163
164
165
166
167
168
169
       
        // Verify that the sentence end position in the memory-mapped corpus and the in-memory corpus have the same value
        Assert.assertEquals(mmCorpus.getSentenceEndPosition(i), corpus.getSentenceEndPosition(i));
       
        // Verify that the phrase corresponding to this sentence is the same
        Phrase sentence = corpus.getSentence(i);
        Phrase mmSentence = mmCorpus.getSentence(i);
        Assert.assertNotNull(sentence);
        Assert.assertNotNull(mmSentence);
        Assert.assertEquals(mmSentence, sentence);
      }
     
View Full Code Here
TOP

Related Classes of joshua.corpus.Phrase

  • joshua.corpus.CorpusArrayTest
  • joshua.corpus.suffix_array.BasicPhrase
  • joshua.corpus.suffix_array.FrequentClassesTest
  • joshua.corpus.suffix_array.FrequentPhrases
  • joshua.corpus.suffix_array.HierarchicalPhrase
  • joshua.corpus.suffix_array.PatternTest
  • joshua.corpus.suffix_array.SuffixArrayTest
  • joshua.prefix_tree.LMAdaptingRuleExtractor
  • joshua.subsample.BiCorpus
  • joshua.subsample.Subsampler
    • 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.