Examples of edu.stanford.nlp.trees.Tree

edu.stanford.nlp.trees.Tree
The abstract class Tree is used to collect all of the tree types, and acts as a generic extendable type. This is the standard implementation of inheritance-based polymorphism. All Tree objects support accessors for their children (a Tree[]), their label (a Label), and their score (a double). However, different concrete implementations may or may not include the latter two, in which case a default value is returned. The class Tree defines no data fields. The two abstract methods that must be implemented are: children(), and treeFactory(). Notes that setChildren(Tree[]) is now an optional operation, whereas it was previously required to be implemented. There is now support for finding the parent of a tree. This may be done by search from a tree root, or via a directly stored parent. The Tree class now implements the Collection interface: in terms of this, each node of the tree is an element of the collection; hence one can explore the tree by using the methods of this interface. A Tree is regarded as a read-only Collection (even though the Tree class has various methods that modify trees). Moreover, the implementation is not thread-safe: no attempt is made to detect and report concurrent modifications. @author Christopher Manning @author Dan Klein @author Sarah Spikes (sdspikes@cs.stanford.edu) - filled in types


  private Distribution<Integer> getSegmentedWordLengthDistribution(Treebank tb) {
    // CharacterLevelTagExtender ext = new CharacterLevelTagExtender();
    ClassicCounter<Integer> c = new ClassicCounter<Integer>();
    for (Iterator iterator = tb.iterator(); iterator.hasNext();) {
      Tree gold = (Tree) iterator.next();
      StringBuilder goldChars = new StringBuilder();
      ArrayList goldYield = gold.yield();
      for (Iterator wordIter = goldYield.iterator(); wordIter.hasNext();) {
        Word word = (Word) wordIter.next();
        goldChars.append(word);
      }
      List<HasWord> ourWords = segment(goldChars.toString());

View Full Code Here

      while (treeNodeMatchCandidateIterator.hasNext()) {
        nextTreeNodeMatchCandidate = treeNodeMatchCandidateIterator.next();
        if (myNode.descriptionMode == null) {
          // this is a backreference or link
          if (myNode.isLink) {
            Tree otherTree = namesToNodes.get(myNode.linkedName);
            if (otherTree != null) {
              String otherValue = myNode.basicCatFunction == null ? otherTree.value() : myNode.basicCatFunction.apply(otherTree.value());
              String myValue = myNode.basicCatFunction == null ? nextTreeNodeMatchCandidate.value() : myNode.basicCatFunction.apply(nextTreeNodeMatchCandidate.value());
              if (otherValue.equals(myValue)) {
                finished = false;
                break;
              }

View Full Code Here

   */
  private double depScoreTree(Tree tr) {
    // System.err.println("Here's our tree:");
    // tr.pennPrint();
    // System.err.println(Trees.toDebugStructureString(tr));
    Tree cwtTree = tr.deepCopy(new LabeledScoredTreeFactory(), new CategoryWordTagFactory());
    cwtTree.percolateHeads(binHeadFinder);
    // System.err.println("Here's what it went to:");
    // cwtTree.pennPrint();
    List<IntDependency> deps = MLEDependencyGrammar.treeToDependencyList(cwtTree, wordIndex, tagIndex);
    // System.err.println("Here's the deps:\n" + deps);
    return dg.scoreAll(deps);

View Full Code Here

  List<Tree> prune(List<Tree> treeList, Label label, int start, int end) {
    // get reference tree
    if (treeList.size() == 1) {
      return treeList;
    }
    Tree testTree = treeList.get(0).treeFactory().newTreeNode(label, treeList);
    Tree tempTree = parser.extractBestParse(label.value(), start, end);
    // parser.restoreUnaries(tempTree);
    Tree pcfgTree = debinarizer.transformTree(tempTree);
    Set<Constituent> pcfgConstituents = pcfgTree.constituents(new LabeledScoredConstituentFactory());
    // delete child labels that are not in reference but do not cross reference
    List<Tree> prunedChildren = new ArrayList<Tree>();
    int childStart = 0;
    for (int c = 0, numCh = testTree.numChildren(); c < numCh; c++) {
      Tree child = testTree.getChild(c);
      boolean isExtra = true;
      int childEnd = childStart + child.yield().size();
      Constituent childConstituent = new LabeledScoredConstituent(childStart, childEnd, child.label(), 0);
      if (pcfgConstituents.contains(childConstituent)) {
        isExtra = false;
      }
      if (childConstituent.crosses(pcfgConstituents)) {
        isExtra = false;
      }
      if (child.isLeaf() || child.isPreTerminal()) {
        isExtra = false;
      }
      if (pcfgTree.yield().size() != testTree.yield().size()) {
        isExtra = false;
      }
      if (!label.value().startsWith("NP^NP")) {
        isExtra = false;
      }
      if (isExtra) {
        System.err.println("Pruning: " + child.label() + " from " + (childStart + start) + " to " + (childEnd + start));
        System.err.println("Was: " + testTree + " vs " + pcfgTree);
        prunedChildren.addAll(child.getChildrenAsList());
      } else {
        prunedChildren.add(child);
      }
      childStart = childEnd;
    }

View Full Code Here

  }


  protected void tallyTreeIterator(Iterator<Tree> treeIterator, 
                                   Function<Tree, Tree> f, double weight) {
    while (treeIterator.hasNext()) {
      Tree tree = treeIterator.next();
      try {
        tree = f.apply(tree);
      } catch (Exception e) {
        if (op.testOptions.verbose) {
          e.printStackTrace();

View Full Code Here

      //Silently ignore
    }
  }


  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;
  }

View Full Code Here

        lemmas = null;
      }


      //Terminals can have multiple tokens (MWEs). Make these into a
      //flat structure for now.
      Tree t = null;
      List<Tree> kids = new ArrayList<Tree>();
      if(leafToks.size() > 1) {
        for (int i = 0; i < leafToks.size(); ++i) {
          String tok = leafToks.get(i);
          String s = treeNormalizer.normalizeTerminal(tok);
          List<Tree> leafList = new ArrayList<Tree>();
          Tree leafNode = treeFactory.newLeaf(s);
          if(leafNode.label() instanceof HasWord)
            ((HasWord) leafNode.label()).setWord(s);
          if (leafNode.label() instanceof CoreLabel && lemmas != null) {
            ((CoreLabel) leafNode.label()).setLemma(lemmas.get(i));
          }
          if(leafNode.label() instanceof HasContext) {
            ((HasContext) leafNode.label()).setOriginalText(morph);
          }
          if (leafNode.label() instanceof HasCategory) {
            ((HasCategory) leafNode.label()).setCategory(subcat);
          }
          leafList.add(leafNode);


          Tree posNode = treeFactory.newTreeNode(MISSING_POS, leafList);
          if(posNode.label() instanceof HasTag)
            ((HasTag) posNode.label()).setTag(MISSING_POS);


          kids.add(posNode);
        }
        t = treeFactory.newTreeNode(MISSING_PHRASAL, kids);


      } else {
        String leafStr = treeNormalizer.normalizeTerminal(leafToks.get(0));
        Tree leafNode = treeFactory.newLeaf(leafStr);
        if (leafNode.label() instanceof HasWord)
          ((HasWord) leafNode.label()).setWord(leafStr);
        if (leafNode.label() instanceof CoreLabel && lemmas != null) {
          ((CoreLabel) leafNode.label()).setLemma(lemmas.get(0));
        }
        if (leafNode.label() instanceof HasContext) {
          ((HasContext) leafNode.label()).setOriginalText(morph);
        }
        if (leafNode.label() instanceof HasCategory) {
          ((HasCategory) leafNode.label()).setCategory(subcat);
        }
        kids.add(leafNode);


        t = treeFactory.newTreeNode(posStr, kids);
        if (t.label() instanceof HasTag) ((HasTag) t.label()).setTag(posStr);
      }


      return t;
    }


    List<Tree> kids = new ArrayList<Tree>();
    for(Node childNode = eRoot.getFirstChild(); childNode != null; childNode = childNode.getNextSibling()) {
      if(childNode.getNodeType() != Node.ELEMENT_NODE) continue;
      Tree t = getTreeFromXML(childNode);
      if(t == null) {
        System.err.printf("%s: Discarding empty tree (root: %s)%n", this.getClass().getName(),childNode.getNodeName());
      } else {
        kids.add(t);
      }
    }


    // MWEs have a label with a
    String rootLabel = eRoot.getNodeName().trim();
    boolean isMWE = rootLabel.equals("w") && eRoot.hasAttribute(ATTR_POS);
    if(isMWE)
      rootLabel = eRoot.getAttribute(ATTR_POS).trim();


    Tree t = (kids.size() == 0) ? null : treeFactory.newTreeNode(treeNormalizer.normalizeNonterminal(rootLabel), kids);


    if(t != null && isMWE)
      t = postProcessMWE(t);


    return t;

View Full Code Here

    Set<String> morphAnalyses = Generics.newHashSet();
    try {
      for(File file : fileList) {
        TreeReader tr = trf.newTreeReader(new BufferedReader(new InputStreamReader(new FileInputStream(file),"UTF-8")));


        Tree t;
        int numTrees;
        String canonicalFileName = file.getName().substring(0, file.getName().lastIndexOf('.'));


        for(numTrees = 0; (t = tr.readTree()) != null; numTrees++) {
          String ftbID = ((CoreLabel) t.label()).get(CoreAnnotations.SentenceIDAnnotation.class);
          System.out.printf("%s-%s\t%s%n",canonicalFileName, ftbID, t.toString());
          List<Label> leaves = t.yield();
          for(Label label : leaves) {
            if(label instanceof CoreLabel)
              morphAnalyses.add(((CoreLabel) label).originalText());
          }
        }

View Full Code Here

      if (annotation.containsKey(CoreAnnotations.SentencesAnnotation.class)) {
        // int sentNum = 0;
        for (CoreMap sentence: annotation.get(CoreAnnotations.SentencesAnnotation.class)) {
          List<CoreLabel> tokens = sentence.get(CoreAnnotations.TokensAnnotation.class);
          sentences.add(tokens);
          Tree tree = sentence.get(TreeCoreAnnotations.TreeAnnotation.class);
          trees.add(tree);


          if (!hasSpeakerAnnotations) {
            // check for speaker annotations
            for (CoreLabel t:tokens) {

View Full Code Here

   * matching node (that is, the tree node matching the root node of
   * the pattern) differs from the previous matching node.
   * @return true iff another matching node is found.
   */
  public boolean findNextMatchingNode() {
    Tree lastMatchingNode = getMatch();
    while(find()) {
      if(getMatch() != lastMatchingNode)
        return true;
    }
    return false;

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of edu.stanford.nlp.trees.Tree

com.music.service.text.SentimentAnalyzer

edu.stanford.nlp.dcoref.CoNLLMentionExtractor

edu.stanford.nlp.dcoref.MentionExtractor

edu.stanford.nlp.dcoref.RuleBasedCorefMentionFinder

edu.stanford.nlp.ie.machinereading.BasicRelationFeatureFactory

edu.stanford.nlp.ie.machinereading.GenericDataSetReader

edu.stanford.nlp.international.french.scripts.SplitCanditoTrees

edu.stanford.nlp.ling.CoreLabel

edu.stanford.nlp.ling.CyclicCoreLabel

edu.stanford.nlp.ling.HasWord

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.