Java CategoryWordTag类代码示例

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
public static void main(String[] args) {
  CategoryWordTag.printWordTag = false;
  String path = args[0];
  List<Tree> trees = TreebankAnnotator.getTrees(path, 200, 219, 0, 10);
  List<Tree> annotatedTrees = new TreebankAnnotator(new Options(), path).annotateTrees(trees);
  for (Tree tree : annotatedTrees) {
    System.out.println("ORIGINAL:\n");
    tree.pennPrint();
    System.out.println("CNFed:\n");
    Tree cnfTree = new ToCNFTransformer().transformTree(tree);
    cnfTree.pennPrint();
    System.out.println("UnCNFed:\n");
    Tree unCNFTree = new FromCNFTransformer().transformTree(cnfTree);
    unCNFTree.pennPrint();
    System.out.println("\n\n");
  }
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
public static void printTrainTree(PrintWriter pw, String message, Tree t) {
  PrintWriter myPW;
  if (pw == null) {
    myPW = new PrintWriter(System.out, true);
  } else {
    myPW = pw;
  }
  if (message != null && pw == null) {
    // hard coded to not print message if using file output!
    myPW.println(message);
  }
  boolean previousState = CategoryWordTag.printWordTag;
  CategoryWordTag.printWordTag = false;
  t.pennPrint(myPW);
  CategoryWordTag.printWordTag = previousState;
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
public static void printTrainTree(PrintWriter pw, String message, Tree t) {
  PrintWriter myPW;
  if (pw == null) {
    myPW = new PrintWriter(System.out, true);
  } else {
    myPW = pw;
  }
  if (message != null && pw == null) {
    // hard coded to not print message if using file output!
    myPW.println(message);
  }
  // TODO FIXME:  wtf is this shit
  boolean previousState = CategoryWordTag.printWordTag;
  CategoryWordTag.printWordTag = false;
  t.pennPrint(myPW);
  CategoryWordTag.printWordTag = previousState;
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
public static void printTrainTree(PrintWriter pw, String message, Tree t) {
  PrintWriter myPW;
  if (pw == null) {
    myPW = new PrintWriter(System.out, true);
  } else {
    myPW = pw;
  }
  if (message != null && pw == null) {
    // hard coded to not print message if using file output!
    logger.trace(message);
  }
  // TODO FIXME:  wtf is this shit
  boolean previousState = CategoryWordTag.printWordTag;
  CategoryWordTag.printWordTag = false;
  t.pennPrint(myPW);
  CategoryWordTag.printWordTag = previousState;
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
public Tree transformTree(Tree tree) {
  Label lab = tree.label();
  if (tree.isLeaf()) {
    Tree leaf = tf.newLeaf(lab);
    leaf.setScore(tree.score());
    return leaf;
  }
  String s = lab.value();
  s = treebankLanguagePack().basicCategory(s);
  int numKids = tree.numChildren();
  List<Tree> children = new ArrayList<Tree>(numKids);
  for (int cNum = 0; cNum < numKids; cNum++) {
    Tree child = tree.getChild(cNum);
    Tree newChild = transformTree(child);
    // cdm 2007: for just subcategory stripping, null shouldn't happen
    // if (newChild != null) {
    children.add(newChild);
    // }
  }
  // if (children.isEmpty()) {
  //   return null;
  // }
  CategoryWordTag newLabel = new CategoryWordTag(lab);
  newLabel.setCategory(s);
  if (lab instanceof HasTag) {
    String tag = ((HasTag) lab).tag();
    tag = treebankLanguagePack().basicCategory(tag);
    newLabel.setTag(tag);
  }
  Tree node = tf.newTreeNode(newLabel, children);
  node.setScore(tree.score());
  return node;
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
public static void main(String[] args) {
  CategoryWordTag.printWordTag = false;
  String path = args[0];
  List<Tree> trees = getTrees(path, 200, 219, 0, 10);
  trees.iterator().next().pennPrint();
  Options op = new Options();
  List<Tree> annotatedTrees = TreebankAnnotator.removeDependencyRoots(new TreebankAnnotator(op, path).annotateTrees(trees));
  annotatedTrees.iterator().next().pennPrint();
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
/**
 * Changes the ROOT label, and adds a Lexicon.BOUNDARY daughter to it.
 * This is needed for the dependency parser.
 * <i>Note:</i> This is a destructive operation on the tree passed in!!
 *
 * @param t The current tree into which a boundary is inserted
 */
public void addRoot(Tree t) {
  if (t.isLeaf()) {
    System.err.println("Warning: tree is leaf: " + t);
    t = tf.newTreeNode(tlp.startSymbol(), Collections.singletonList(t));
  }
  t.setLabel(new CategoryWordTag(tlp.startSymbol(), Lexicon.BOUNDARY, Lexicon.BOUNDARY_TAG));
  List<Tree> preTermChildList = new ArrayList<Tree>();
  Tree boundaryTerm = tf.newLeaf(new StringLabel(Lexicon.BOUNDARY));//CategoryWordTag(Lexicon.BOUNDARY,Lexicon.BOUNDARY,""));
  preTermChildList.add(boundaryTerm);
  Tree boundaryPreTerm = tf.newTreeNode(new CategoryWordTag(Lexicon.BOUNDARY_TAG, Lexicon.BOUNDARY, Lexicon.BOUNDARY_TAG), preTermChildList);
  List<Tree> childList = t.getChildrenAsList();
  childList.add(boundaryPreTerm);
  t.setChildren(childList);
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
/**
 * Go through trees and determine their heads and print them.
 * Just for debuggin'. <br>
 * Usage: <code>
 * java edu.stanford.nlp.trees.CollinsHeadFinder 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 CollinsHeadFinder();
  treebank.apply(new TreeVisitor() {
    public void visitTree(Tree pt) {
      pt.percolateHeads(chf);
      pt.pennPrint();
      System.out.println();
    }
  });
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
@Override
public Tree transformTree(Tree tree) {
  Label lab = tree.label();
  if (tree.isLeaf()) {
    Tree leaf = tf.newLeaf(lab);
    leaf.setScore(tree.score());
    return leaf;
  }
  String s = lab.value();
  s = treebankLanguagePack().basicCategory(s);
  int numKids = tree.numChildren();
  List<Tree> children = new ArrayList<Tree>(numKids);
  for (int cNum = 0; cNum < numKids; cNum++) {
    Tree child = tree.getChild(cNum);
    Tree newChild = transformTree(child);
    // cdm 2007: for just subcategory stripping, null shouldn't happen
    // if (newChild != null) {
    children.add(newChild);
    // }
  }
  // if (children.isEmpty()) {
  //   return null;
  // }
  CategoryWordTag newLabel = new CategoryWordTag(lab);
  newLabel.setCategory(s);
  if (lab instanceof HasTag) {
    String tag = ((HasTag) lab).tag();
    tag = treebankLanguagePack().basicCategory(tag);
    newLabel.setTag(tag);
  }
  Tree node = tf.newTreeNode(newLabel, children);
  node.setScore(tree.score());
  return node;
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
/**
 * Changes the ROOT label, and adds a Lexicon.BOUNDARY daughter to it.
 * This is needed for the dependency parser.
 * <i>Note:</i> This is a destructive operation on the tree passed in!!
 *
 * @param t The current tree into which a boundary is inserted
 */
public void addRoot(Tree t) {
  if (t.isLeaf()) {
    System.err.println("Warning: tree is leaf: " + t);
    t = tf.newTreeNode(tlp.startSymbol(), Collections.singletonList(t));
  }
  t.setLabel(new CategoryWordTag(tlp.startSymbol(), Lexicon.BOUNDARY, Lexicon.BOUNDARY_TAG));
  List<Tree> preTermChildList = new ArrayList<Tree>();
  Tree boundaryTerm = tf.newLeaf(new Word(Lexicon.BOUNDARY));//CategoryWordTag(Lexicon.BOUNDARY,Lexicon.BOUNDARY,""));
  preTermChildList.add(boundaryTerm);
  Tree boundaryPreTerm = tf.newTreeNode(new CategoryWordTag(Lexicon.BOUNDARY_TAG, Lexicon.BOUNDARY, Lexicon.BOUNDARY_TAG), preTermChildList);
  List<Tree> childList = t.getChildrenAsList();
  childList.add(boundaryPreTerm);
  t.setChildren(childList);
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
/**
 * Go through trees and determine their heads and print them.
 * Just for debuggin'. <br>
 * Usage: <code>
 * java edu.stanford.nlp.trees.CollinsHeadFinder 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 CollinsHeadFinder();
  treebank.apply(new TreeVisitor() {
    @Override
    public void visitTree(Tree pt) {
      pt.percolateHeads(chf);
      pt.pennPrint();
      System.out.println();
    }
  });
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
/**
 * Go through trees and determine their heads and print them.
 * Just for debugging. <br>
 * Usage: <code>
 * java edu.stanford.nlp.trees.DybroFrenchHeadFinder 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 DybroFrenchHeadFinder();
  treebank.apply(new TreeVisitor() {
    public void visitTree(Tree pt) {
      pt.percolateHeads(chf);
      pt.pennPrint();
      System.out.println();
    }
  });
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
/**
 * 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(new TreeVisitor() {
    public void visitTree(Tree pt) {
      pt.percolateHeads(chf);
      pt.pennPrint();
      System.out.println();
    }
  });
}
 

import edu.stanford.nlp.ling.CategoryWordTag; //导入依赖的package包/类
public static void main(String[] args) {
  // simple testing code
  Treebank treebank = new DiskTreebank();
  CategoryWordTag.suppressTerminalDetails = true;
  treebank.loadPath(args[0]);
  final HeadFinder chf = new NoPunctuationHeadFinder();
  treebank.apply(new TreeVisitor() {
    public void visitTree(Tree pt) {
      pt.percolateHeads(chf);
      pt.pennPrint();
      System.out.println();
    }
  });
}