SaveBugs.java

/** 
 * SaveBUGS.java
 * @author Fabio G. Cozman 
 * Copyright 1996 - 1999, Fabio G. Cozman,
 *          Carnergie Mellon University, Universidade de Sao Paulo
 * fgcozman@usp.br, http://www.cs.cmu.edu/~fgcozman/home.html
 *
 * The JavaBayes distribution is free software; you can
 * redistribute it and/or modify it under the terms of the GNU General
 * Public License as published by the Free Software Foundation (either
 * version 2 of the License or, at your option, any later version), 
 * provided that this notice and the name of the author appear in all 
 * copies. Upon request to the author, some of the packages in the 
 * JavaBayes distribution can be licensed under the GNU Lesser General
 * Public License as published by the Free Software Foundation (either
 * version 2 of the License, or (at your option) any later version).
 * If you're using the software, please notify fgcozman@usp.br so
 * that you can receive updates and patches. JavaBayes is distributed
 * "as is", in the hope that it will be useful, but WITHOUT ANY WARRANTY;
 * without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
 * PARTICULAR PURPOSE. See the GNU General Public License for more details.
 * You should have received a copy of the GNU General Public License
 * along with the JavaBayes distribution. If not, write to the Free
 * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 */

package BayesianNetworks;

import java.io.*;
import java.util.Date;

/**************************************************************************/

public class SaveBugs {
  BayesNet bn;

  /**
   * Default constructor for a SaveBUGS object.
   */
  public SaveBugs(BayesNet b_n) {
    bn = b_n;
  }

  /**
   * Save a BayesNet in a stream in the BUGS format.
   */
  public void save(PrintStream pstream) {
    pstream.println("# Bayesian Network in the BUGS Format");
    pstream.println("# Produced by BayesianNetworks package in JavaBayes");
    pstream.println("# Output created " + (new Date()));
    pstream.println("\n");

    save_model(pstream);

    save_variables(pstream);

    print_data_in(pstream);

    save_structure(pstream);

    save_data(pstream);
  }

  /**
   * Save the name of the network in BUGS format.
   */
  private void save_model(PrintStream pstream) {
    pstream.println("model " + bn.name + ";");
  }

  /**
   * Declare all the variables used in the network.
   */
  private void save_variables(PrintStream pstream) {
    ProbabilityFunction pf;
    ProbabilityVariable pv;
    int i, j;

    pstream.println("var");
    for (i = (bn.probability_variables.length - 1); i >= 0; i--) {
      pstream.println("\t" + bn.probability_variables[i].name + ",");
    }

    for (i = (bn.probability_functions.length - 1); i >= 0; i--) {
      pf = bn.probability_functions[i];
      pv = (ProbabilityVariable)(pf.variables[0]);
      pstream.print("\tp." + pv.name + "[" + pf.values.length + "]");
      if (i > 0) pstream.println(",");
    }
    pstream.println(";");
  }

  /**
   * Indicate which file contains the data; note that the data
   * is appended to the given pstream, so this space is left
   * to be filled by the user.
   */
  private void print_data_in(PrintStream pstream) {
    pstream.println("data in <user-defined-name-for-data-file>;");
  }

  /**
   * The parenthood relationships in the network.
   */
  private void save_structure(PrintStream pstream) {
    ProbabilityFunction pf;
    ProbabilityVariable pv;
    int i, j;

    pstream.println("{");
    for (i = (bn.probability_functions.length - 1); i >= 0; i--) {
      pf = bn.probability_functions[i];
      pv = (ProbabilityVariable)(pf.variables[0]);
      pstream.print(pv.name + "  ~  dcat(p." + pv.name + "[");
      for (j=1; j < pf.variables.length; j++) {
	    pstream.print(pf.variables[j].name);
	    pstream.print(",");
      }
      pstream.println("]);");
    }
    pstream.println("}\n\n");
  }

  /**
   * The numeric values for the distributions, saved in
   * the Splus format understood by BUGS. Note that the
   * user has to place this data in a separate file.
   */
  private void save_data(PrintStream pstream) {
    ProbabilityFunction pf;
    ProbabilityVariable pv;
    int i, j, k;
    int step;
    double value;

    pstream.println("list(");
    for (i = (bn.probability_functions.length - 1); i >= 0; i--) {
      pf = bn.probability_functions[i];
      pv = (ProbabilityVariable)(pf.variables[0]);
      /**** Put distribution values in the correct format. ****/
      pstream.print("\tp." + pv.name + "  = c(");
      step = 1;
      for (j = 1; j < pf.variables.length; j++)
        step *= pf.variables[j].values.length;
      for (j = 0; j < step; j++) {
        for (k = 0; k < pf.variables[0].values.length; k++) {
            value = pf.values[ k * step + j ];
            pstream.print(" " + value);
            if (k < (pf.variables[0].values.length - 1))
                pstream.print(",");
        }
        if (j < (step - 1))
            pstream.print(",");
      }
      pstream.print(")");
      if (i > 0) pstream.println(",");
    }
    pstream.println(")\n");
  }
}

/*******************************************************************/