hzl.rar_jxta开发

/* * 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. */ /* * PairedTTester.java * Copyright (C) 1999 University of Waikato, Hamilton, New Zealand * */ package weka.experiment; import weka.core.Attribute; import weka.core.FastVector; import weka.core.Instance; import weka.core.Instances; import weka.core.Option; import weka.core.OptionHandler; import weka.core.Range; import weka.core.RevisionHandler; import weka.core.RevisionUtils; import weka.core.Utils; import java.io.BufferedReader; import java.io.FileReader; import java.text.SimpleDateFormat; import java.util.Date; import java.util.Enumeration; import java.util.Vector; /** * Calculates T-Test statistics on data stored in a set of instances. <p/> * <!-- options-start --> * Valid options are: <p/> * * <pre> -D &lt;index,index2-index4,...&gt; * Specify list of columns that specify a unique * dataset. * First and last are valid indexes. (default none)</pre> * * <pre> -R &lt;index&gt; * Set the index of the column containing the run number</pre> * * <pre> -F &lt;index&gt; * Set the index of the column containing the fold number</pre> * * <pre> -G &lt;index1,index2-index4,...&gt; * Specify list of columns that specify a unique * 'result generator' (eg: classifier name and options). * First and last are valid indexes. (default none)</pre> * * <pre> -S &lt;significance level&gt; * Set the significance level for comparisons (default 0.05)</pre> * * <pre> -V * Show standard deviations</pre> * * <pre> -L * Produce table comparisons in Latex table format</pre> * * <pre> -csv * Produce table comparisons in CSV table format</pre> * * <pre> -html * Produce table comparisons in HTML table format</pre> * * <pre> -significance * Produce table comparisons with only the significance values</pre> * * <pre> -gnuplot * Produce table comparisons output suitable for GNUPlot</pre> * <!-- options-end --> * * @author Len Trigg (trigg@cs.waikato.ac.nz) * @version $Revision: 6431 $ */ public class PairedTTester implements OptionHandler, Tester, RevisionHandler { /** for serialization */ static final long serialVersionUID = 8370014624008728610L; /** The set of instances we will analyse */ protected Instances m_Instances; /** The index of the column containing the run number */ protected int m_RunColumn = 0; /** The option setting for the run number column (-1 means last) */ protected int m_RunColumnSet = -1; /** The option setting for the fold number column (-1 means none) */ protected int m_FoldColumn = -1; /** The column to sort on (-1 means default sorting) */ protected int m_SortColumn = -1; /** The sorting of the datasets (according to the sort column) */ protected int[] m_SortOrder = null; /** The sorting of the columns (test base is always first) */ protected int[] m_ColOrder = null; /** The significance level for comparisons */ protected double m_SignificanceLevel = 0.05; /** * The range of columns that specify a unique "dataset" * (eg: scheme plus configuration) */ protected Range m_DatasetKeyColumnsRange = new Range(); /** An array containing the indexes of just the selected columns */ protected int [] m_DatasetKeyColumns; /** The list of dataset specifiers */ protected DatasetSpecifiers m_DatasetSpecifiers = new DatasetSpecifiers(); /** * The range of columns that specify a unique result set * (eg: scheme plus configuration) */ protected Range m_ResultsetKeyColumnsRange = new Range(); /** An array containing the indexes of just the selected columns */ protected int [] m_ResultsetKeyColumns; /** An array containing the indexes of the datasets to display */ protected int[] m_DisplayedResultsets = null; /** Stores a vector for each resultset holding all instances in each set */ protected FastVector m_Resultsets = new FastVector(); /** Indicates whether the instances have been partitioned */ protected boolean m_ResultsetsValid; /** Indicates whether standard deviations should be displayed */ protected boolean m_ShowStdDevs = false; /** the instance of the class to produce the output. */ protected ResultMatrix m_ResultMatrix = new ResultMatrixPlainText(); /** A list of unique "dataset" specifiers that have been observed */ protected class DatasetSpecifiers implements RevisionHandler { /** the specifiers that have been observed */ FastVector m_Specifiers = new FastVector(); /** * Removes all specifiers. */ protected void removeAllSpecifiers() { m_Specifiers.removeAllElements(); } /** * Add an instance to the list of specifiers (if necessary) * * @param inst the instance to add */ protected void add(Instance inst) { for (int i = 0; i < m_Specifiers.size(); i++) { Instance specifier = (Instance)m_Specifiers.elementAt(i); boolean found = true; for (int j = 0; j < m_DatasetKeyColumns.length; j++) { if (inst.value(m_DatasetKeyColumns[j]) != specifier.value(m_DatasetKeyColumns[j])) { found = false; } } if (found) { return; } } m_Specifiers.addElement(inst); } /** * Get the template at the given position. * * @param i the index * @return the template */ protected Instance specifier(int i) { return (Instance)m_Specifiers.elementAt(i); } /** * Gets the number of specifiers. * * @return the current number of specifiers */ protected int numSpecifiers() { return m_Specifiers.size(); } /** * Returns the revision string. * * @return the revision */ public String getRevision() { return RevisionUtils.extract("$Revision: 6431 $"); } } /** Utility class to store the instances pertaining to a dataset */ protected class Dataset implements RevisionHandler { /** the template */ Instance m_Template; /** the dataset */ FastVector m_Dataset; /** * Constructor * * @param template the template */ public Dataset(Instance template) { m_Template = template; m_Dataset = new FastVector(); add(template); } /** * Returns true if the two instances match on those attributes that have * been designated key columns (eg: scheme name and scheme options) * * @param first the first instance * @return true if first and second match on the currently set key columns */ protected boolean matchesTemplate(Instance first) { for (int i = 0; i < m_DatasetKeyColumns.length; i++) { if (first.value(m_DatasetKeyColumns[i]) != m_Template.value(m_DatasetKeyColumns[i])) { return false; } } return true; } /** * Adds the given instance to the dataset * * @param inst the instance to add */ protected void add(Instance inst) { m_Dataset.addElement(inst); } /** * Returns a vector containing the instances in the dataset * * @return the current contents */ protected FastVector contents() { return m_Dataset; } /** * Sorts the instances in the dataset by the run number. *