Package com.aliasi.corpus

Examples of com.aliasi.corpus.Parser


  Logger logger = Logger.getLogger(TrainPOSTagger.class.getName());
  logger.debug("Started POS tagged model generation");

  //*-- set up parser with estimator as handler
  HmmCharLmEstimator estimator = new HmmCharLmEstimator(N_GRAM, NUM_CHARS, LAMBDA_FACTOR);
  Parser parser = new BrownPosParser();
  parser.setHandler(estimator);

  //*-- train on files in data directory ending in "txt"
  if (!TRAINING_DIR.isDirectory())
  { logger.fatal("Could not find training directory=" + TRAINING_DIR); }
  File[] files = TRAINING_DIR.listFiles(new FileExtensionFilter("txt"));
  for (int i = 0; i < files.length; ++i)
  { logger.debug("Training on file: " + files[i]); parser.parse(files[i]); }

  //*-- write output to file
  File modelFile = new File(MUSTRU_HOME + File.separator + "data" + File.separator + "training" + File.separator + "pos" + File.separator + "pos_tagger");
  ObjectOutputStream objOut = new ObjectOutputStream(new FileOutputStream(modelFile));
  estimator.compileTo(objOut);
View Full Code Here

TOP

Related Classes of com.aliasi.corpus.Parser

Copyright © 2018 www.massapicom. All rights reserved.
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.