Examples with CoreLabel edu.stanford.nlp.ling.CoreLabel used on opensource projects

Example 11 with CoreLabel

use of edu.stanford.nlp.ling.CoreLabel in project CoreNLP by stanfordnlp.

the class SpanishXMLTreeReader method process.

/**
   * Read trees from the given file and output their processed forms to
   * standard output.
   */
public static void process(File file, TreeReader tr, Pattern posPattern, Pattern wordPattern, boolean plainPrint) throws IOException {
    Tree t;
    int numTrees = 0, numTreesRetained = 0;
    String canonicalFileName = file.getName().substring(0, file.getName().lastIndexOf('.'));
    while ((t = tr.readTree()) != null) {
        numTrees++;
        if (!shouldPrintTree(t, posPattern, wordPattern))
            continue;
        numTreesRetained++;
        String ftbID = ((CoreLabel) t.label()).get(CoreAnnotations.SentenceIDAnnotation.class);
        String output = toString(t, plainPrint);
        System.out.printf("%s-%s\t%s%n", canonicalFileName, ftbID, output);
    }
    System.err.printf("%s: %d trees, %d matched and printed%n", file.getName(), numTrees, numTreesRetained);
}

Example 12 with CoreLabel

use of edu.stanford.nlp.ling.CoreLabel in project CoreNLP by stanfordnlp.

the class SpanishXMLTreeReader method readTree.

public Tree readTree() {
    Tree t = null;
    while (t == null && sentences != null && sentIdx < sentences.getLength()) {
        int thisSentenceId = sentIdx++;
        Node sentRoot = sentences.item(thisSentenceId);
        t = getTreeFromXML(sentRoot);
        if (t != null) {
            t = treeNormalizer.normalizeWholeTree(t, treeFactory);
            if (t.label() instanceof CoreLabel)
                ((CoreLabel) t.label()).set(CoreAnnotations.SentenceIDAnnotation.class, Integer.toString(thisSentenceId));
        }
    }
    return t;
}

Example 13 with CoreLabel

use of edu.stanford.nlp.ling.CoreLabel in project CoreNLP by stanfordnlp.

the class NERFeatureFactoryITest method testSloppyGazette.

public void testSloppyGazette() {
    List<CoreLabel> sentence = SentenceUtils.toCoreLabelList("For three years , John Bauer has worked at Stanford .".split(" +"));
    PaddedList<CoreLabel> paddedSentence = new PaddedList<CoreLabel>(sentence, new CoreLabel());
    Properties props = new Properties();
    props.setProperty("useGazettes", "true");
    props.setProperty("sloppyGazette", "true");
    props.setProperty("gazette", "projects/core/data/edu/stanford/nlp/ie/test_gazette.txt");
    SeqClassifierFlags flags = new SeqClassifierFlags(props);
    NERFeatureFactory<CoreLabel> factory = new NERFeatureFactory<CoreLabel>();
    factory.init(flags);
    Set<String> features;
    features = new HashSet<String>(factory.featuresC(paddedSentence, 4));
    checkFeatures(features, "BAR-GAZ", "BAZ-GAZ", "FOO-GAZ", "BAR-GAZ2", "BAZ-GAZ2", "FOO-GAZ1", "John-WORD");
    features = new HashSet<String>(factory.featuresC(paddedSentence, 5));
    checkFeatures(features, "BAR-GAZ", "BAZ-GAZ", "BAR-GAZ2", "BAZ-GAZ2", "Bauer-WORD");
    features = new HashSet<String>(factory.featuresC(paddedSentence, 6));
    checkFeatures(features, "has-WORD");
}

Example 14 with CoreLabel

use of edu.stanford.nlp.ling.CoreLabel in project CoreNLP by stanfordnlp.

the class CRFClassifierITest method adapt2.

/** adapt2 changes from {@code Pair<List<CoreLabel>, Double>} to {@code Pair<List<String>, Double>} to make printout
   *  better.
   */
private static List<Pair<List<String>, Double>> adapt2(List<Pair<List<CoreLabel>, Double>> in) {
    List<Pair<List<String>, Double>> ret = new ArrayList<>();
    for (Pair<List<CoreLabel>, Double> pair : in) {
        List<String> strs = new ArrayList<>();
        for (CoreLabel c : pair.first()) {
            String label = c.getString(CoreAnnotations.AnswerAnnotation.class);
            int max = Math.min(3, label.length());
            strs.add(label.substring(0, max));
        }
        ret.add(new Pair<>(strs, pair.second()));
    }
    return ret;
}

Example 15 with CoreLabel

use of edu.stanford.nlp.ling.CoreLabel in project CoreNLP by stanfordnlp.

the class SpanishTokenizerAnnotatorITest method testSpanish.

public void testSpanish() {
    Annotation ann = new Annotation("Damelo");
    Properties props = new Properties();
    props.setProperty("annotators", "tokenize");
    props.setProperty("tokenize.language", "es");
    StanfordCoreNLP pipeline = new StanfordCoreNLP(props);
    pipeline.annotate(ann);
    Iterator<String> it = spanishTokens.iterator();
    for (CoreLabel word : ann.get(CoreAnnotations.TokensAnnotation.class)) {
        assertEquals("Bung token in new CoreLabel usage", it.next(), word.get(CoreAnnotations.TextAnnotation.class));
    }
    assertFalse("Too few tokens in new CoreLabel usage", it.hasNext());
}