databaseresultproducer.java

Java 编写的多种数据挖掘算法 包括聚类、分类、预处理等

/*
 *    This program 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.
 *
 *    This program is distributed 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 this program; if not, write to the Free Software
 *    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 */

/*
 *    DatabaseResultProducer.java
 *    Copyright (C) 1999 Len Trigg
 *
 */


package weka.experiment;

import weka.core.AdditionalMeasureProducer;
import weka.core.Instances;
import weka.core.Option;
import weka.core.OptionHandler;
import weka.core.Utils;

import java.util.Enumeration;
import java.util.Vector;

/**
 <!-- globalinfo-start -->
 * Examines a database and extracts out the results produced by the specified ResultProducer and submits them to the specified ResultListener. If a result needs to be generated, the ResultProducer is used to obtain the result.
 * <p/>
 <!-- globalinfo-end -->
 *
 <!-- options-start -->
 * Valid options are: <p/>
 * 
 * <pre> -F &lt;field name&gt;
 *  The name of the database field to cache over.
 *  eg: "Fold" (default none)</pre>
 * 
 * <pre> -W &lt;class name&gt;
 *  The full class name of a ResultProducer.
 *  eg: weka.experiment.CrossValidationResultProducer</pre>
 * 
 * <pre> 
 * Options specific to result producer weka.experiment.CrossValidationResultProducer:
 * </pre>
 * 
 * <pre> -X &lt;number of folds&gt;
 *  The number of folds to use for the cross-validation.
 *  (default 10)</pre>
 * 
 * <pre> -D
 * Save raw split evaluator output.</pre>
 * 
 * <pre> -O &lt;file/directory name/path&gt;
 *  The filename where raw output will be stored.
 *  If a directory name is specified then then individual
 *  outputs will be gzipped, otherwise all output will be
 *  zipped to the named file. Use in conjuction with -D. (default splitEvalutorOut.zip)</pre>
 * 
 * <pre> -W &lt;class name&gt;
 *  The full class name of a SplitEvaluator.
 *  eg: weka.experiment.ClassifierSplitEvaluator</pre>
 * 
 * <pre> 
 * Options specific to split evaluator weka.experiment.ClassifierSplitEvaluator:
 * </pre>
 * 
 * <pre> -W &lt;class name&gt;
 *  The full class name of the classifier.
 *  eg: weka.classifiers.bayes.NaiveBayes</pre>
 * 
 * <pre> -C &lt;index&gt;
 *  The index of the class for which IR statistics
 *  are to be output. (default 1)</pre>
 * 
 * <pre> -I &lt;index&gt;
 *  The index of an attribute to output in the
 *  results. This attribute should identify an
 *  instance in order to know which instances are
 *  in the test set of a cross validation. if 0
 *  no output (default 0).</pre>
 * 
 * <pre> -P
 *  Add target and prediction columns to the result
 *  for each fold.</pre>
 * 
 * <pre> 
 * Options specific to classifier weka.classifiers.rules.ZeroR:
 * </pre>
 * 
 * <pre> -D
 *  If set, classifier is run in debug mode and
 *  may output additional info to the console</pre>
 * 
 <!-- options-end -->
 * 
 * @author Len Trigg ([e-mail])
 * @version $Revision: 1.16 $
 */
public class DatabaseResultProducer 
  extends DatabaseResultListener
  implements ResultProducer, OptionHandler, AdditionalMeasureProducer {

  /** for serialization */
  static final long serialVersionUID = -5620660780203158666L;
  
  /** The dataset of interest */
  protected Instances m_Instances;

  /** The ResultListener to send results to */
  protected ResultListener m_ResultListener = new CSVResultListener();

  /** The ResultProducer used to generate results */
  protected ResultProducer m_ResultProducer
    = new CrossValidationResultProducer();

  /** The names of any additional measures to look for in SplitEvaluators */
  protected String [] m_AdditionalMeasures = null;

  /**
   * Returns a string describing this result producer
   * @return a description of the result producer suitable for
   * displaying in the explorer/experimenter gui
   */
  public String globalInfo() {
    return "Examines a database and extracts out "
      +"the results produced by the specified ResultProducer "
      +"and submits them to the specified ResultListener. If a result needs "
      +"to be generated, the ResultProducer is used to obtain the result.";
  }

  /**
   * Creates the DatabaseResultProducer, letting the parent constructor do
   * it's thing.
   *
   * @throws Exception if an error occurs
   */
  public DatabaseResultProducer() throws Exception {

    super();
  }
  
  /**
   * Gets the keys for a specified run number. Different run
   * numbers correspond to different randomizations of the data. Keys
   * produced should be sent to the current ResultListener
   *
   * @param run the run number to get keys for.
   * @throws Exception if a problem occurs while getting the keys
   */
  public void doRunKeys(int run) throws Exception {

    if (m_ResultProducer == null) {
      throw new Exception("No ResultProducer set");
    }
    if (m_ResultListener == null) {
      throw new Exception("No ResultListener set");
    }
    if (m_Instances == null) {
      throw new Exception("No Instances set");
    }

    // Tell the resultproducer to send results to us
    m_ResultProducer.setResultListener(this);
    m_ResultProducer.setInstances(m_Instances);
    m_ResultProducer.doRunKeys(run);
  }

  /**
   * Gets the results for a specified run number. Different run
   * numbers correspond to different randomizations of the data. Results
   * produced should be sent to the current ResultListener
   *
   * @param run the run number to get results for.
   * @throws Exception if a problem occurs while getting the results
   */
  public void doRun(int run) throws Exception {

    if (m_ResultProducer == null) {
      throw new Exception("No ResultProducer set");
    }
    if (m_ResultListener == null) {
      throw new Exception("No ResultListener set");
    }
    if (m_Instances == null) {
      throw new Exception("No Instances set");
    }

    // Tell the resultproducer to send results to us
    m_ResultProducer.setResultListener(this);
    m_ResultProducer.setInstances(m_Instances);
    m_ResultProducer.doRun(run);

  }
  
  /**
   * Prepare for the results to be received.
   *
   * @param rp the ResultProducer that will generate the results
   * @throws Exception if an error occurs during preprocessing.
   */
  public void preProcess(ResultProducer rp) throws Exception {

    super.preProcess(rp);
    if (m_ResultListener == null) {
      throw new Exception("No ResultListener set");
    }
    m_ResultListener.preProcess(this);
  }

  /**
   * When this method is called, it indicates that no more results
   * will be sent that need to be grouped together in any way.
   *
   * @param rp the ResultProducer that generated the results
   * @throws Exception if an error occurs
   */
  public void postProcess(ResultProducer rp) throws Exception {

    super.postProcess(rp);
    m_ResultListener.postProcess(this);
  }
  
  /**
   * Prepare to generate results. The ResultProducer should call
   * preProcess(this) on the ResultListener it is to send results to.
   *
   * @throws Exception if an error occurs during preprocessing.
   */
  public void preProcess() throws Exception {
    
    if (m_ResultProducer == null) {
      throw new Exception("No ResultProducer set");
    }
    m_ResultProducer.setResultListener(this);
    m_ResultProducer.preProcess();
  }
  
  /**
   * When this method is called, it indicates that no more requests to
   * generate results for the current experiment will be sent. The
   * ResultProducer should call preProcess(this) on the
   * ResultListener it is to send results to.
   *
   * @throws Exception if an error occurs
   */
  public void postProcess() throws Exception {

    m_ResultProducer.postProcess();
  }
    
  /**
   * Accepts results from a ResultProducer.
   *
   * @param rp the ResultProducer that generated the results
   * @param key an array of Objects (Strings or Doubles) that uniquely
   * identify a result for a given ResultProducer with given compatibilityState
   * @param result the results stored in an array. The objects stored in
   * the array may be Strings, Doubles, or null (for the missing value).
   * @throws Exception if the result could not be accepted.
   */
  public void acceptResult(ResultProducer rp, Object [] key, Object [] result)
    throws Exception {

    if (m_ResultProducer != rp) {
      throw new Error("Unrecognized ResultProducer sending results!!");
    }
    //    System.err.println("DBRP::acceptResult");

    // Is the result needed by the listener?
    boolean isRequiredByListener = m_ResultListener.isResultRequired(this,
								     key);
    // Is the result already in the database?
    boolean isRequiredByDatabase = super.isResultRequired(rp, key);

    // Insert it into the database here
    if (isRequiredByDatabase) {
      // We could alternatively throw an exception if we only want values
      // that are already in the database
      if (result != null) {

	// null result could occur from a chain of doRunKeys calls
	super.acceptResult(rp, key, result);
      }
    }

    // Pass it on
    if (isRequiredByListener) {
      m_ResultListener.acceptResult(this, key, result);
    }
  }

  /**
   * Determines whether the results for a specified key must be
   * generated.
   *
   * @param rp the ResultProducer wanting to generate the results
   * @param key an array of Objects (Strings or Doubles) that uniquely
   * identify a result for a given ResultProducer with given compatibilityState
   * @return true if the result should be generated
   * @throws Exception if it could not be determined if the result 
   * is needed.
   */
  public boolean isResultRequired(ResultProducer rp, Object [] key) 
    throws Exception {

    if (m_ResultProducer != rp) {
      throw new Error("Unrecognized ResultProducer sending results!!");
    }
    //    System.err.println("DBRP::isResultRequired");

    // Is the result needed by the listener?
    boolean isRequiredByListener = m_ResultListener.isResultRequired(this,
								     key);
    // Is the result already in the database?
    boolean isRequiredByDatabase = super.isResultRequired(rp, key);

    if (!isRequiredByDatabase && isRequiredByListener) {
      // Pass the result through to the listener
      Object [] result = getResultFromTable(m_ResultsTableName,
					    rp, key);
      System.err.println("Got result from database: "
			 + DatabaseUtils.arrayToString(result));
      m_ResultListener.acceptResult(this, key, result);
      return false;
    }

    return (isRequiredByListener || isRequiredByDatabase);
  }

  /**
   * Gets the names of each of the columns produced for a single run.
   *
   * @return an array containing the name of each column
   * @throws Exception if something goes wrong.
   */
  public String [] getKeyNames() throws Exception {

    return m_ResultProducer.getKeyNames();
  }

  /**
   * Gets the data types of each of the columns produced for a single run.
   * This method should really be static.
   *