Package cc.mallet.extract

Examples of cc.mallet.extract.StringSpan

143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
      int start = buf.length ();
      buf.append (text);
      int end = buf.length ();
      buf.append (" ");

      StringSpan span = new StringSpan (buf, start, end);

      while (j < maxFeatureIdx) {
        span.setFeatureValue (toks[j].intern (), 1.0);
        j++;
      }

      if (includeTokenText) {
        span.setFeatureValue ((textFeaturePrefix+text).intern(), 1.0);
      }

      if (labelsAtEnd) {
        int firstLblIdx = j;
        while (j < toks.length) {
View Full Code Here
111
112
113
114
115
116
117
118
119
120
121
        lexer.setCharSequence (string.subSequence (textStart, textEnd));
        while (lexer.hasNext()) {
          lexer.next ();
          int tokStart = textStart + lexer.getStartOffset ();
          int tokEnd = textStart + lexer.getEndOffset ();
          dataTokens.add (new StringSpan (string, tokStart, tokEnd));
          targetTokens.add (new Token (tag));
        }
      }
      textStart = nextStart;
      tag = nextTag;
View Full Code Here
46
47
48
49
50
51
52
53
54
55
56
57
58
        int start = buf.length ();
        buf.append (token.getText());
        int end = buf.length();

        StringSpan span = new StringSpan (buf, start, end);
        span.setFeatures (token.getFeatures ());
        span.setProperties (token.getProperties ());

        spans.add (span);
        buf.append (" ");
      }
View Full Code Here
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
      } else
        nFeatures = tokens[l].length;
      int start = source.length();
      String word = makeText(tokens[l]);
      source.append(word + " ");
      Token tok = new StringSpan(source, start, source.length() - 1);
      if (setTokensAsFeatures) {
        for (int f = 0; f < nFeatures; f++)
          tok.setFeatureValue(tokens[l][f], 1.0);
      } else {
        for (int f = 1; f < nFeatures; f++)
          tok.setFeatureValue(tokens[l][f], 1.0);
      }
      ts.add(tok);
    }
    carrier.setData(ts);
    if (isTargetProcessing())
View Full Code Here
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
      if (saveSource) {
        if (word.equals ("-<S>-")) source.append ("\n\n");
        source.append (word); source.append (" ");
      }

      Token token = new StringSpan (source, start, source.length () - 1);

      // Word and tag unigram at current time
      if (doSpelling) {
        for (int j = 0; j < endings.length; j++) {
          ending[2][j] = ending[1][j];
          ending[1][j] = ending[0][j];
          ending[0][j] = endingPatterns[j].matcher(word).matches();
          if (ending[0][j]) token.setFeatureValue (endingNames[0][0][j], 1);
        }
      }

      if (doTags) {
        token.setFeatureValue ("T="+tag, 1);
      }

      if (doPhrases) {
        token.setFeatureValue ("P="+phrase, 1);
      }

      data.add (token);

      if (isTargetProcessing ()) {
View Full Code Here
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
      Label notstart = dict.lookupLabel ("notstart");

      boolean lastWasSpace = true;
      StringBuffer sb = new StringBuffer();
      for (int i = 0; i < ts.size(); i++) {
        StringSpan t = (StringSpan) ts.getSpan(i);
        if (t.getText().equals(" "))
          lastWasSpace = true;
        else {
          sb.append(t.getText());
          newTs.add(t);
          labelSeq.add(lastWasSpace ? "start" : "notstart");
          lastWasSpace = false;
        }
      }
View Full Code Here
63
64
65
66
67
68
69
70
71
72
    CharSequence string = (CharSequence) carrier.getData();
    lexer.setCharSequence (string);
    TokenSequence ts = new StringTokenization (string);
    while (lexer.hasNext()) {
      lexer.next();
      ts.add (new StringSpan (string, lexer.getStartOffset (), lexer.getEndOffset ()));
    }
    carrier.setData(ts);
    return carrier;
  }
View Full Code Here
111
112
113
114
115
116
117
118
119
120
121
        lexer.setCharSequence (string.subSequence (textStart, textEnd));
        while (lexer.hasNext()) {
          lexer.next ();
          int tokStart = textStart + lexer.getStartOffset ();
          int tokEnd = textStart + lexer.getEndOffset ();
          dataTokens.add (new StringSpan (string, tokStart, tokEnd));
          targetTokens.add (new Token (tag));
        }
      }
      textStart = nextStart;
      tag = nextTag;
View Full Code Here
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
      int start = buf.length ();
      buf.append (text);
      int end = buf.length ();
      buf.append (" ");

      StringSpan span = new StringSpan (buf, start, end);

      while (j < maxFeatureIdx) {
        span.setFeatureValue (toks[j].intern (), 1.0);
        j++;
      }

      if (includeTokenText) {
        span.setFeatureValue ((textFeaturePrefix+text).intern(), 1.0);
      }

      if (labelsAtEnd) {
        int firstLblIdx = j;
        while (j < toks.length) {
View Full Code Here
TOP

Related Classes of cc.mallet.extract.StringSpan

  • cc.mallet.extract.pipe.TokenSequence2Tokenization
  • cc.mallet.fst.tests.TestMEMM$TestMEMMTokenSequenceRemoveSpaces
  • cc.mallet.grmm.learning.GenericAcrfData2TokenSequence
  • cc.mallet.pipe.CharSequence2TokenSequence
  • cc.mallet.pipe.SGML2TokenSequence
  • cc.mallet.pipe.SimpleTaggerSentence2StringTokenization
  • cc.mallet.share.casutton.ner.ConllNer2003Sentence2TokenSequence
    • 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.