Package edu.stanford.nlp.parser.lexparser

Examples of edu.stanford.nlp.parser.lexparser.EnglishTreebankParserParams

92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
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
    if(args.length < minArgs) {
      System.out.println(usage.toString());
      System.exit(-1);
    }

    TreebankLangParserParams tlpp = new EnglishTreebankParserParams();
    int maxGoldYield = Integer.MAX_VALUE;
    int maxGuessYield = Integer.MAX_VALUE;
    boolean VERBOSE = false;
    boolean skipGuess = false;
    boolean tagMode = false;
    String guessFile = null;
    String goldFile = null;

    for(int i = 0; i < args.length; i++) {

      if(args[i].startsWith("-")) {

        switch (args[i]) {
          case "-l":
            Language lang = Language.valueOf(args[++i].trim());
            tlpp = Languages.getLanguageParams(lang);

            break;
          case "-y":
            maxGoldYield = Integer.parseInt(args[++i].trim());

            break;
          case "-t":
            tagMode = true;

            break;
          case "-v":
            VERBOSE = true;

            break;
          case "-g":
            maxGuessYield = Integer.parseInt(args[++i].trim());
            skipGuess = true;

            break;
          default:
            System.out.println(usage.toString());
            System.exit(-1);
        }

      } else {
        //Required parameters
        goldFile = args[i++];
        guessFile = args[i];
        break;
      }
    }

    final PrintWriter pwOut = tlpp.pw();

    final Treebank guessTreebank = tlpp.diskTreebank();
    guessTreebank.loadPath(guessFile);
    pwOut.println("GUESS TREEBANK:");
    pwOut.println(guessTreebank.textualSummary());

    final Treebank goldTreebank = tlpp.diskTreebank();
    goldTreebank.loadPath(goldFile);
    pwOut.println("GOLD TREEBANK:");
    pwOut.println(goldTreebank.textualSummary());

    final String evalName = (tagMode) ? "TsarfatyTAG" : "TsarfatySEG";
    final TsarfatyEval eval = new TsarfatyEval(evalName, tagMode);

    final TreeTransformer tc = tlpp.collinizer();

    //PennTreeReader skips over null/malformed parses. So when the yields of the gold/guess trees
    //don't match, we need to keep looking for the next gold tree that matches.
    //The evalb ref implementation differs slightly as it expects one tree per line. It assigns
    //status as follows:
View Full Code Here
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
    if(args.length < minArgs) {
      System.out.println(usage.toString());
      System.exit(-1);
    }

    TreebankLangParserParams tlpp = new EnglishTreebankParserParams();
    DiskTreebank tb = null;
    String encoding = "UTF-8";
    Language lang = Language.English;

    for(int i = 0; i < args.length; i++) {
      if(args[i].startsWith("-")) {
        switch (args[i]) {
          case "-l":
            lang = Language.valueOf(args[++i].trim());
            tlpp = Languages.getLanguageParams(lang);

            break;
          case "-e":
            encoding = args[++i];

            break;
          default:
            System.out.println(usage.toString());
            System.exit(-1);
        }

      } else {
        if(tb == null) {
          if(tlpp == null) {
            System.out.println(usage.toString());
            System.exit(-1);
          } else {
            tlpp.setInputEncoding(encoding);
            tlpp.setOutputEncoding(encoding);
            tb = tlpp.diskTreebank();
          }
        }
        tb.loadPath(args[i]);
      }
    }

    PrintWriter pw = tlpp.pw();
    Options op = new Options();
    Options.LexOptions lexOptions = op.lexOptions;
    if(lang == Language.French) {
      lexOptions.useUnknownWordSignatures = 1;
      lexOptions.smartMutation = false;
      lexOptions.unknownSuffixSize = 2;
      lexOptions.unknownPrefixSize = 1;
    } else if(lang == Language.Arabic) {
      lexOptions.smartMutation = false;
      lexOptions.useUnknownWordSignatures = 9;
      lexOptions.unknownPrefixSize = 1;
      lexOptions.unknownSuffixSize = 1;
    }
    Index<String> wordIndex = new HashIndex<String>();
    Index<String> tagIndex = new HashIndex<String>();
    Lexicon lex = tlpp.lex(op, wordIndex, tagIndex);
   
    int computeAfter = (int) (0.50 * tb.size());
    Counter<String> vocab = new ClassicCounter<String>();
    Counter<String> unkCounter = new ClassicCounter<String>();
    int treeId = 0;
View Full Code Here
65
66
67
68
69
70
71
72
73
    case Spanish:
      tlpp = new SpanishTreebankParserParams();
      break;

    default:
      tlpp = new EnglishTreebankParserParams();
    }
    return tlpp;
  }
View Full Code Here
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
173
174
175
176
177
178
179
180
181
182
183
184
185
   * Run the Evalb scoring metric on guess/gold input. The default language is English.
   *
   * @param args
   */
  public static void main(String[] args) {
    TreebankLangParserParams tlpp = new EnglishTreebankParserParams();
    int maxGoldYield = Integer.MAX_VALUE;
    boolean VERBOSE = false;
    String encoding = "UTF-8";

    String guessFile = null;
    String goldFile = null;

    Map<String, String[]> argsMap = StringUtils.argsToMap(args, optionArgDefs);

    for(Map.Entry<String, String[]> opt : argsMap.entrySet()) {
      if(opt.getKey() == null) continue;
      if(opt.getKey().equals("-l")) {
        Language lang = Language.valueOf(opt.getValue()[0].trim());
        tlpp = Languages.getLanguageParams(lang);

      } else if(opt.getKey().equals("-y")) {
        maxGoldYield = Integer.parseInt(opt.getValue()[0].trim());

      } else if(opt.getKey().equals("-v")) {
        VERBOSE = true;

      } else if(opt.getKey().equals("-e")) {
        encoding = opt.getValue()[0];

      } else {
        System.err.println(usage.toString());
        System.exit(-1);
      }

      //Non-option arguments located at key null
      String[] rest = argsMap.get(null);
      if(rest == null || rest.length < minArgs) {
        System.err.println(usage.toString());
        System.exit(-1);
      }
      goldFile = rest[0];
      guessFile = rest[1];
    }

    tlpp.setInputEncoding(encoding);
    final PrintWriter pwOut = tlpp.pw();

    final Treebank guessTreebank = tlpp.diskTreebank();
    guessTreebank.loadPath(guessFile);
    pwOut.println("GUESS TREEBANK:");
    pwOut.println(guessTreebank.textualSummary());

    final Treebank goldTreebank = tlpp.diskTreebank();
    goldTreebank.loadPath(goldFile);
    pwOut.println("GOLD TREEBANK:");
    pwOut.println(goldTreebank.textualSummary());

    final UnlabeledAttachmentEval metric = new UnlabeledAttachmentEval("UAS LP/LR", true, tlpp.headFinder());

    final TreeTransformer tc = tlpp.collinizer();

    //The evalb ref implementation assigns status for each tree pair as follows:
    //
    //   0 - Ok (yields match)
    //   1 - length mismatch
View Full Code Here
29
30
31
32
33
34
35
36
37
38
39
40
    // the intermediate GrammaticalStructure instead
    SemanticGraph graph = SemanticGraphFactory.generateUncollapsedDependencies(tree);

    // Alternatively, this could have been the Chinese params or any
    // other language supported.  As of 2014, only English and Chinese
    TreebankLangParserParams params = new EnglishTreebankParserParams();
    GrammaticalStructureFactory gsf = params.treebankLanguagePack().grammaticalStructureFactory(params.treebankLanguagePack().punctuationWordRejectFilter(), params.typedDependencyHeadFinder());

    GrammaticalStructure gs = gsf.newGrammaticalStructure(tree);

    System.err.println(graph);
View Full Code Here
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
    if(args.length < minArgs) {
      System.out.println(usage.toString());
      System.exit(-1);
    }

    TreebankLangParserParams tlpp = new EnglishTreebankParserParams();
    DiskTreebank tb = null;
    String encoding = "UTF-8";

    for(int i = 0; i < args.length; i++) {
      if(args[i].startsWith("-")) {
        switch (args[i]) {
          case "-l":
            Language lang = Language.valueOf(args[++i].trim());
            tlpp = Languages.getLanguageParams(lang);

            break;
          case "-e":
            encoding = args[++i];

            break;
          default:
            System.out.println(usage.toString());
            System.exit(-1);
        }

      } else {
        if(tb == null) {
          if(tlpp == null) {
            System.out.println(usage.toString());
            System.exit(-1);
          } else {
            tlpp.setInputEncoding(encoding);
            tlpp.setOutputEncoding(encoding);
            tb = tlpp.diskTreebank();
          }
        }
        tb.loadPath(args[i]);
      }
    }

    Counter<String> vocab = new ClassicCounter<String>();
    for(Tree t : tb) {
      List<Label> yield = t.yield();
      for(Label word : yield)
        vocab.incrementCount(word.value());
    }

    List<String> biggestKeys = new ArrayList<String>(vocab.keySet());
    Collections.sort(biggestKeys, Counters.toComparatorDescending(vocab));

    PrintWriter pw = tlpp.pw();
    for(String wordType : biggestKeys)
      pw.printf("%s\t%d%n", wordType,(int) vocab.getCount(wordType));
    pw.close();
  }
View Full Code Here
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
    if(args.length < minArgs) {
      System.out.println(usage.toString());
      System.exit(-1);
    }

    TreebankLangParserParams tlpp = new EnglishTreebankParserParams();
    DiskTreebank tb = null;
    String encoding = "UTF-8";
    String puncTag = null;

    for(int i = 0; i < args.length; i++) {
      if(args[i].startsWith("-")) {
        switch (args[i]) {
          case "-l":
            Language lang = Language.valueOf(args[++i].trim());
            tlpp = Languages.getLanguageParams(lang);

            break;
          case "-e":
            encoding = args[++i];

            break;
          default:
            System.out.println(usage.toString());
            System.exit(-1);
        }

      } else {
        puncTag = args[i++];
       
        if(tb == null) {
          if(tlpp == null) {
            System.out.println(usage.toString());
            System.exit(-1);
          } else {
            tlpp.setInputEncoding(encoding);
            tlpp.setOutputEncoding(encoding);
            tb = tlpp.diskTreebank();
          }
        }
        tb.loadPath(args[i]);
      }
    }

    Counter<String> puncTypes = new ClassicCounter<String>();
    for(Tree t : tb) {
      List<CoreLabel> yield = t.taggedLabeledYield();
      for(CoreLabel word : yield)
        if(word.tag().equals(puncTag))
          puncTypes.incrementCount(word.word());
    }

    List<String> biggestKeys = new ArrayList<String>(puncTypes.keySet());
    Collections.sort(biggestKeys, Counters.toComparatorDescending(puncTypes));

    PrintWriter pw = tlpp.pw();
    for(String wordType : biggestKeys)
      pw.printf("%s\t%d%n", wordType,(int) puncTypes.getCount(wordType));
    pw.close();
  }
View Full Code Here
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
    if(args.length < minArgs) {
      System.out.println(usage.toString());
      System.exit(-1);
    }

    TreebankLangParserParams tlpp = new EnglishTreebankParserParams();
    int maxGoldYield = Integer.MAX_VALUE;
    boolean VERBOSE = false;
    String encoding = "UTF-8";

    String guessFile = null;
    String goldFile = null;

    Map<String, String[]> argsMap = StringUtils.argsToMap(args, optionArgDefs);

    for(Map.Entry<String, String[]> opt : argsMap.entrySet()) {
      if(opt.getKey() == null) continue;
      if(opt.getKey().equals("-l")) {
        Language lang = Language.valueOf(opt.getValue()[0].trim());
        tlpp = Languages.getLanguageParams(lang);

      } else if(opt.getKey().equals("-y")) {
        maxGoldYield = Integer.parseInt(opt.getValue()[0].trim());

      } else if(opt.getKey().equals("-v")) {
        VERBOSE = true;

      } else if(opt.getKey().equals("-c")) {
        TaggingEval.doCatLevelEval = true;

      } else if(opt.getKey().equals("-e")) {
        encoding = opt.getValue()[0];

      } else {
        System.err.println(usage.toString());
        System.exit(-1);
      }

      //Non-option arguments located at key null
      String[] rest = argsMap.get(null);
      if(rest == null || rest.length < minArgs) {
        System.err.println(usage.toString());
        System.exit(-1);
      }
      goldFile = rest[0];
      guessFile = rest[1];
    }

    tlpp.setInputEncoding(encoding);
    final PrintWriter pwOut = tlpp.pw();

    final Treebank guessTreebank = tlpp.diskTreebank();
    guessTreebank.loadPath(guessFile);
    pwOut.println("GUESS TREEBANK:");
    pwOut.println(guessTreebank.textualSummary());

    final Treebank goldTreebank = tlpp.diskTreebank();
    goldTreebank.loadPath(goldFile);
    pwOut.println("GOLD TREEBANK:");
    pwOut.println(goldTreebank.textualSummary());

    final TaggingEval metric = new TaggingEval("Tagging LP/LR");

    final TreeTransformer tc = tlpp.collinizer();

    //The evalb ref implementation assigns status for each tree pair as follows:
    //
    //   0 - Ok (yields match)
    //   1 - length mismatch
View Full Code Here
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
    if(args.length < minArgs) {
      System.out.println(usage.toString());
      System.exit(-1);
    }

    TreebankLangParserParams tlpp = new EnglishTreebankParserParams();
    DiskTreebank tb = null;
    String encoding = "UTF-8";
    TregexPattern rootMatch = null;
   
    for(int i = 0; i < args.length; i++) {
      if(args[i].startsWith("-")) {
        switch (args[i]) {
          case "-l":
            Language lang = Language.valueOf(args[++i].trim());
            tlpp = Languages.getLanguageParams(lang);

            break;
          case "-e":
            encoding = args[++i];

            break;
          default:
            System.out.println(usage.toString());
            System.exit(-1);
        }

      } else {
        rootMatch = TregexPattern.compile("@" + args[i++]);

        if(tb == null) {
          if(tlpp == null) {
            System.out.println(usage.toString());
            System.exit(-1);
          } else {
            tlpp.setInputEncoding(encoding);
            tlpp.setOutputEncoding(encoding);
            tb = tlpp.diskTreebank();
          }
        }
        tb.loadPath(args[i++]);
      }
    }

    Counter<String> rhsCounter = new ClassicCounter<String>();
    for(Tree t : tb) {
      TregexMatcher m = rootMatch.matcher(t);
      while(m.findNextMatchingNode()) {
        Tree match = m.getMatch();
        StringBuilder sb = new StringBuilder();
        for(Tree kid : match.children())
          sb.append(kid.value()).append(" ");
        rhsCounter.incrementCount(sb.toString().trim());
      }
    }

    List<String> biggestKeys = new ArrayList<String>(rhsCounter.keySet());
    Collections.sort(biggestKeys, Counters.toComparatorDescending(rhsCounter));

    PrintWriter pw = tlpp.pw();
    for(String rhs : biggestKeys)
      pw.printf("%s\t%d%n", rhs,(int) rhsCounter.getCount(rhs));
    pw.close();
  }
View Full Code Here
TOP

Related Classes of edu.stanford.nlp.parser.lexparser.EnglishTreebankParserParams

  • edu.stanford.nlp.international.Languages
  • edu.stanford.nlp.parser.eval.UNKPrinter
  • edu.stanford.nlp.parser.metrics.TaggingEval
  • edu.stanford.nlp.parser.metrics.TsarfatyEval
  • edu.stanford.nlp.parser.metrics.UnlabeledAttachmentEval
  • edu.stanford.nlp.parser.tools.PunctFrequencyDist
  • edu.stanford.nlp.parser.tools.RHSFrequency
  • edu.stanford.nlp.parser.tools.VocabFrequency
  • edu.stanford.nlp.semgraph.semgrex.demo.SemgrexDemo
  • edu.stanford.nlp.io.RuntimeIOException
    • 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.