Examples of edu.stanford.nlp.util.CoreMap

edu.stanford.nlp.util.CoreMap

Base type for all annotatable core objects. Should usually be instantiated as {@link ArrayCoreMap}. Many common key definitions live in {@link edu.stanford.nlp.ling.CoreAnnotations}, but others may be defined elsewhere. See {@link edu.stanford.nlp.ling.CoreAnnotations} for details.

Note that implementations of this interface must take care to implement equality correctly: by default, two CoreMaps are .equal if they contain the same keys and all corresponding values are .equal. Subclasses that wish to change this behavior (such as {@link HashableCoreMap}) must make sure that all other CoreMap implementations have a special case in their .equals to use that equality definition when appropriate. Similarly, care must be taken when defining hashcodes. The default hashcode is 37 * sum of all keys' hashcodes plus the sum of all values' hashcodes. However, use of this class as HashMap keys is discouraged because the hashcode can change over time. Consider using a {@link HashableCoreMap}.
@author dramage @author rafferty

    if(sents != null){
      for(CoreMap sent: sents){
        if(! (sent instanceof Annotation)){
          throw new RuntimeException("ERROR: Sentences must instantiate Annotation!");
        }
        CoreMap newSent = sentenceDeepMentionCopy((Annotation) sent);
        newSents.add(newSent);
      }
    }


    addSentences(newDataset, newSents);

View Full Code Here

      assert doc.get(CoreAnnotations.SentencesAnnotation.class).size() > 0;
      List<CoreMap> timexAnnotations = doc.get(TimeAnnotations.TimexAnnotations.class);
      if (timexAnnotations.size() > 1) {
        throw new RuntimeException("Too many timexes for '" + str + "'");
      }
      CoreMap timex = timexAnnotations.get(0);


      return timex.get(TimeExpression.Annotation.class).getTemporal();
    } catch (Exception e) {
      SUTimeSimpleParser.SUTimeParsingError parsingError = new SUTimeSimpleParser.SUTimeParsingError(str);
      parsingError.initCause(e);
      throw parsingError;
    }

View Full Code Here

        mentionCount.incrementCount(m.goldCorefClusterID);
      }
    }


    for(int i = 0 ; i<sentences.size(); i++) {
      CoreMap sentence = sentences.get(i);
      List<Mention> mentions = allMentions.get(i);


      String[] tokens = sentence.get(CoreAnnotations.TextAnnotation.class).split(" ");
      String sent = "";
      List<CoreLabel> t = sentence.get(CoreAnnotations.TokensAnnotation.class);
      if(previousOffset+2 < t.get(0).get(CoreAnnotations.CharacterOffsetBeginAnnotation.class)) sent += "\n";
      previousOffset = t.get(t.size()-1).get(CoreAnnotations.CharacterOffsetEndAnnotation.class);
      Counter<Integer> startCounts = new ClassicCounter<Integer>();
      Counter<Integer> endCounts = new ClassicCounter<Integer>();
      Map<Integer, Set<Integer>> endID = Generics.newHashMap();

View Full Code Here

    Counter<String> correct = new ClassicCounter<String>();
    Counter<String> predicted = new ClassicCounter<String>();
    Counter<String> gold = new ClassicCounter<String>();


    for (int i = 0; i < goldStandard.size(); i++) {
      CoreMap goldSent = goldStandard.get(i);
      CoreMap sysSent = extractorOutput.get(i);
      String sysText = sysSent.get(TextAnnotation.class);
      String goldText = goldSent.get(TextAnnotation.class);


      if (verbose) {
        System.err.println("SCORING THE FOLLOWING SENTENCE:");
        System.err.println(sysSent.get(CoreAnnotations.TokensAnnotation.class));
      }


      HashSet<String> matchedGolds = new HashSet<String>();
      List<EntityMention> goldEntities = goldSent
          .get(MachineReadingAnnotations.EntityMentionsAnnotation.class);
      if (goldEntities == null) {
        goldEntities = new ArrayList<EntityMention>();
      }


      for (EntityMention m : goldEntities) {
        String label = makeLabel(m);
        if (excludedClasses != null && excludedClasses.contains(label))
          continue;
        gold.incrementCount(label);
      }


      List<EntityMention> sysEntities = sysSent
          .get(MachineReadingAnnotations.EntityMentionsAnnotation.class);
      if (sysEntities == null) {
        sysEntities = new ArrayList<EntityMention>();
      }
      for (EntityMention m : sysEntities) {

View Full Code Here

          tokenOffset += tokens.size();
          continue;
        }
      }


      CoreMap sentence = new Annotation(textContent.toString());
      sentence.set(CoreAnnotations.DocIDAnnotation.class, docId);
      sentence.set(CoreAnnotations.TokensAnnotation.class, words);
      logger.info("Reading sentence: \"" + textContent + "\"");


      List<AceEntityMention> entityMentions = aceDocument.getEntityMentions(sentenceIndex);
      List<AceRelationMention> relationMentions = aceDocument.getRelationMentions(sentenceIndex);
      List<AceEventMention> eventMentions = aceDocument.getEventMentions(sentenceIndex);

View Full Code Here

        w.set(CoreAnnotations.IndexAnnotation.class, i+1);
        w.set(CoreAnnotations.UtteranceAnnotation.class, 0);
        if(i>0) textContent.append(" ");
        textContent.append(w.getString(CoreAnnotations.TextAnnotation.class));
      }
      CoreMap sentCoreMap = new Annotation(textContent.toString());
      allSentences.add(sentCoreMap);
      sentCoreMap.set(CoreAnnotations.TokensAnnotation.class, sentence);
    }


    // assign goldCorefClusterID
    Map<Integer, Mention> idMention = Generics.newHashMap();    // temporary use
    for (List<Mention> goldMentions : allGoldMentions) {

View Full Code Here

    tokens.get(1).set(CoreAnnotations.NamedEntityTagAnnotation.class, "PERSON");
    tokens.get(2).set(CoreAnnotations.NamedEntityTagAnnotation.class, "PERSON");
    tokens.get(5).set(CoreAnnotations.NamedEntityTagAnnotation.class, "LOCATION");
    tokens.get(7).set(CoreAnnotations.NamedEntityTagAnnotation.class, "LOCATION");


    CoreMap sentence = new ArrayCoreMap();
    sentence.set(CoreAnnotations.TokensAnnotation.class, tokens);


    List<CoreMap> sentences = new ArrayList<CoreMap>();
    sentences.add(sentence);


    Annotation corpus = new Annotation("President Barack Obama lives in Chicago, Illinois," +

View Full Code Here

    tokens.get(3).set(CoreAnnotations.NamedEntityTagAnnotation.class, "LOCATION");
    tokens.get(9).set(CoreAnnotations.NamedEntityTagAnnotation.class, "ORGANIZATION");
    tokens.get(10).set(CoreAnnotations.NamedEntityTagAnnotation.class, "ORGANIZATION");
    tokens.get(11).set(CoreAnnotations.NamedEntityTagAnnotation.class, "ORGANIZATION");


    CoreMap sentence = new ArrayCoreMap();
    sentence.set(CoreAnnotations.TokensAnnotation.class, tokens);


    List<CoreMap> sentences = new ArrayList<CoreMap>();
    sentences.add(sentence);


    Annotation corpus = new Annotation("I like Ontario Place, and I like the Native" +

View Full Code Here

    String str = "Christianity is of higher regex priority than Early Christianity . ";
    String[] split = str.split(" ");


    List<CoreLabel> tokens = Sentence.toCoreLabelList(split);


    CoreMap sentence = new ArrayCoreMap();
    sentence.set(CoreAnnotations.TokensAnnotation.class, tokens);


    List<CoreMap> sentences = new ArrayList<CoreMap>();
    sentences.add(sentence);


    Annotation corpus = new Annotation("Christianity is of higher regex priority than Early " +

View Full Code Here


  private static void checkNer(String message, String[][][] expected, CoreMap coremap, String coremapOutput) {
    List<CoreMap> sentences = coremap.get(CoreAnnotations.SentencesAnnotation.class);
    assertEquals(message + ": number of sentences for\n" + coremapOutput, expected.length, sentences.size());
    for (int i = 0; i < expected.length; i++) {
      CoreMap sentence = sentences.get(i);
      List<CoreLabel> tokens = sentence.get(CoreAnnotations.TokensAnnotation.class);
      assertEquals(message + ": number of tokens for sentence " + (i+1) + "\n" + coremapOutput, expected[i].length, tokens.size());
      for (int j = 0; j < expected[i].length; j++) {
        String text = expected[i][j][0];
        String ner = expected[i][j][1];
        String debug = "sentence " + (i+1) + ", token " + (j+1);

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of edu.stanford.nlp.util.CoreMap

edu.stanford.nlp.dcoref.ACEMentionExtractor

edu.stanford.nlp.dcoref.CoNLLMentionExtractor

edu.stanford.nlp.dcoref.Document

edu.stanford.nlp.dcoref.MUCMentionExtractor

edu.stanford.nlp.dcoref.RuleBasedCorefMentionFinder

edu.stanford.nlp.dcoref.SieveCoreferenceSystem

edu.stanford.nlp.ie.crf.TestSequenceModel

edu.stanford.nlp.ie.machinereading.BasicRelationFeatureFactory

edu.stanford.nlp.ie.machinereading.domains.ace.AceReader

edu.stanford.nlp.ie.machinereading.EntityExtractorResultsPrinter

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.