Examples with FeatureVector edu.illinois.cs.cogcomp.sl.util.FeatureVector used on opensource projects

Example 1 with FeatureVector

use of edu.illinois.cs.cogcomp.sl.util.FeatureVector in project cogcomp-nlp by CogComp.

the class PreExtractor method consumeInstance.

protected void consumeInstance(SenseInstance x, SenseStructure y) throws Exception {
    countFeatures(x);
    synchronized (buffer) {
        FeatureVector fv = x.getCachedFeatureVector();
        assert fv != null;
        buffer.add(new PreExtractRecord(x.getPredicateLemma(), y.getLabel(), fv));
    }
    if (buffer.size() > 10000) {
        synchronized (buffer) {
            if (buffer.size() > 10000) {
                for (PreExtractRecord r : buffer) {
                    cacheDB.put(r.lemma, r.label, r.features);
                }
                buffer.clear();
            }
        }
    }
}

Example 2 with FeatureVector

use of edu.illinois.cs.cogcomp.sl.util.FeatureVector in project cogcomp-nlp by CogComp.

the class PruningPreExtractor method consume.

@Override
protected void consume(Pair<SenseInstance, SenseStructure> input) {
    SenseInstance x = input.getFirst();
    SenseStructure y = input.getSecond();
    FeatureVector features = x.getCachedFeatureVector();
    ModelInfo modelInfo = manager.getModelInfo();
    Lexicon lexicon = modelInfo.getLexicon();
    int threshold = manager.getPruneSize();
    Pair<int[], float[]> pair = lexicon.pruneFeaturesByCount(features.getIdx(), features.getValue(), threshold);
    features = new FeatureVector(pair.getFirst(), pair.getSecond());
    synchronized (buffer) {
        buffer.add(new PreExtractRecord(x.getPredicateLemma(), y.getLabel(), features));
    }
    if (buffer.size() > 10000) {
        synchronized (buffer) {
            if (buffer.size() > 10000) {
                for (PreExtractRecord r : buffer) {
                    try {
                        cache.put(r.lemma, r.label, r.features);
                    } catch (Exception e) {
                        throw new RuntimeException(e);
                    }
                }
                buffer.clear();
            }
        }
    }
    counter.incrementAndGet();
}

Example 3 with FeatureVector

use of edu.illinois.cs.cogcomp.sl.util.FeatureVector in project cogcomp-nlp by CogComp.

the class SenseInstance method cacheFeatureVector.

public void cacheFeatureVector(Set<Feature> features) {
    Map<String, Float> featureMap = new HashMap<>();
    for (Feature f : features) {
        featureMap.put(f.getName(), f.getValue());
    }
    ModelInfo modelInfo = manager.getModelInfo();
    Pair<int[], float[]> feats = modelInfo.getLexicon().getFeatureVector(featureMap);
    this.cacheFeatureVector(new FeatureVector(feats.getFirst(), feats.getSecond()));
}

Example 4 with FeatureVector

use of edu.illinois.cs.cogcomp.sl.util.FeatureVector in project cogcomp-nlp by CogComp.

the class PreExtractor method countFeatures.

/**
     * This is where actual feature extraction is taking place. The features are defined in the
     * <b>features.fex</b> file and are read by {@link FeatureExtractor}
     * 
     * @param x The predicate to extract features from.
     * @throws EdisonException
     */
public void countFeatures(SenseInstance x) throws EdisonException {
    ModelInfo modelInfo = manager.getModelInfo();
    Set<Feature> feats = modelInfo.fex.getFeatures(x.getConstituent());
    // This is the only place where a new feature can be added to the lexicon.
    List<Integer> ids = new ArrayList<>();
    List<Float> values = new ArrayList<>();
    synchronized (lexicon) {
        for (Feature f : feats) {
            if (addNewFeatures) {
                if (!lexicon.contains(f.getName())) {
                    lexicon.previewFeature(f.getName());
                }
            } else if (!lexicon.contains(f.getName())) {
                continue;
            }
            int featureId = lexicon.lookupId(f.getName());
            lexicon.countFeature(featureId);
            ids.add(featureId);
            values.add(f.getValue());
        }
    }
    x.cacheFeatureVector(new FeatureVector(ArrayUtilities.asIntArray(ids), ArrayUtilities.asFloatArray(values)));
}