public TreeReaderFactory treeReaderFactory() { return (readPennFormat) ? new FrenchTreeReaderFactory() : new FrenchXMLTreeReaderFactory(false); }
/** {@inheritDoc} */ public HeadFinder headFinder() { return new FrenchHeadFinder(this); }
public TreeReader newTreeReader(Reader in) { return new FrenchXMLTreeReader(in, ccTagset); } }
public FrenchTreeNormalizer(boolean ccTagset) { super(new FrenchTreebankLanguagePack()); rootLabel = tlp.startSymbol(); this.ccTagset = ccTagset; aOverAFilter = new FrenchAOverAFilter(); emptyFilter = new Predicate<Tree>() { private static final long serialVersionUID = -22673346831392110L; public boolean test(Tree tree) { if(tree.isPreTerminal() && (tree.firstChild().value().equals("") || tree.firstChild().value().equals("-NONE-"))) { return false; } return true; } }; }
public FrenchHeadFinder() { this(new FrenchTreebankLanguagePack()); }
/** * Read parse trees from a Reader. * * @param in The <code>Reader</code> */ public FrenchXMLTreeReader(Reader in, boolean ccTagset) { this(in, new LabeledScoredTreeFactory(), new FrenchTreeNormalizer(ccTagset)); }
nonTerminalInfo.put(tlp.startSymbol(), new String[][]{{"left", "VN", "V", "NP", "Srel", "Ssub", "Sint"}}); nonTerminalInfo.put("SENT", new String[][]{{"left", "VN", "V", "NP", "Srel", "Ssub", "Sint"}});
@Override public HeadFinder headFinder() { if(headFinder == null) headFinder = new DybroFrenchHeadFinder(treebankLanguagePack()); //Superior for vanilla PCFG over Arun's headfinding rules return headFinder; }
public FTBDataset() { super(); //Need to use a MemoryTreebank so that we can compute gross corpus //stats for MWE pre-processing // The treebank may be reset if setOptions changes CC_TAGSET treebank = new MemoryTreebank(new FrenchXMLTreeReaderFactory(CC_TAGSET), FrenchTreebankLanguagePack.FTB_ENCODING); treeFileExtension = "xml"; }
/** * Extract the lemma attribute. * * @param node */ private List<String> getLemma(Element node) { String lemma = node.getAttribute(ATTR_LEMMA); if (lemma == null || lemma.equals("")) return null; return getWordString(lemma); }
public Tree readTree() { Tree t = null; while(t == null && sentences != null && sentIdx < sentences.getLength()) { Node sentRoot = sentences.item(sentIdx++); t = getTreeFromXML(sentRoot); if(t != null) { t = treeNormalizer.normalizeWholeTree(t, treeFactory); if(t.label() instanceof CoreLabel) { String ftbId = ((Element) sentRoot).getAttribute(ATTR_NUMBER); ((CoreLabel) t.label()).set(CoreAnnotations.SentenceIDAnnotation.class, ftbId); } } } return t; }
/** * Sets POS for punctuation to the punctuation token (like the PTB). * * @param t */ private String normalizePreterminal(Tree t) { if (ccTagset) { replacePOSTag(t, morpho); } if(tlp.isPunctuationWord(t.firstChild().value())) return tlp.punctuationTags()[0].intern(); //Map to a common tag // return t.firstChild().value();//Map to the punctuation item return t.value(); }
/** * Go through trees and determine their heads and print them. * Just for debugging. <br> * Usage: <code> * java edu.stanford.nlp.trees.FrenchHeadFinder treebankFilePath * </code> * * @param args The treebankFilePath */ public static void main(String[] args) { Treebank treebank = new DiskTreebank(); CategoryWordTag.suppressTerminalDetails = true; treebank.loadPath(args[0]); final HeadFinder chf = new AbishekFrenchHeadFinder(); treebank.apply(pt -> { pt.percolateHeads(chf); pt.pennPrint(); System.out.println(); }); }
public AbishekFrenchHeadFinder() { this(new FrenchTreebankLanguagePack()); }
/** {@inheritDoc} */ public HeadFinder typedDependencyHeadFinder() { return new FrenchHeadFinder(this); }
public TreeReader newTreeReader(Reader in) { return new PennTreeReader(in, new LabeledScoredTreeFactory(), new FrenchTreeNormalizer(false), new PennTreebankTokenizer(in)); } }
nonTerminalInfo.put(tlp.startSymbol(), new String[][]{{"left", "VN", "NP"}, {"left"}}); nonTerminalInfo.put("SENT", new String[][]{{"left", "VN", "NP"}, {"left"}});
public DybroFrenchHeadFinder() { this(new FrenchTreebankLanguagePack()); }
/** * Go through trees and determine their heads and print them. * Just for debugging. <br> * Usage: <code> * java edu.stanford.nlp.trees.FrenchHeadFinder treebankFilePath * </code> * * @param args The treebankFilePath */ public static void main(String[] args) { Treebank treebank = new DiskTreebank(); CategoryWordTag.suppressTerminalDetails = true; treebank.loadPath(args[0]); final HeadFinder chf = new FrenchHeadFinder(); treebank.apply(pt -> { pt.percolateHeads(chf); pt.pennPrint(); System.out.println(); }); }
public FrenchTreebankParserParams() { super(new FrenchTreebankLanguagePack()); setInputEncoding("UTF-8"); optionsString = new StringBuilder(); optionsString.append("FrenchTreebankParserParams\n"); initializeAnnotationPatterns(); }