evaluation.html

数据挖掘的最常用工具。由于开源

<A NAME="constructor_detail"><!-- --></A>
<TABLE BORDER="1" WIDTH="100%" CELLPADDING="3" CELLSPACING="0" SUMMARY="">
<TR BGCOLOR="#CCCCFF" CLASS="TableHeadingColor">
<TH ALIGN="left" COLSPAN="1"><FONT SIZE="+2">
<B>Constructor Detail</B></FONT></TH>
</TR>
</TABLE>

<A NAME="Evaluation(weka.core.Instances)"><!-- --></A><H3>
Evaluation</H3>
<PRE>
public <B>Evaluation</B>(<A HREF="../../weka/core/Instances.html" title="class in weka.core">Instances</A>&nbsp;data)
           throws java.lang.Exception</PRE>
<DL>
<DD>Initializes all the counters for the evaluation. 
 Use <code>useNoPriors()</code> if the dataset is the test set and you
 can't initialize with the priors from the training set via 
 <code>setPriors(Instances)</code>.
<P>
<DL>
<DT><B>Parameters:</B><DD><CODE>data</CODE> - set of training instances, to get some header 
                        information and prior class distribution information
<DT><B>Throws:</B>
<DD><CODE>java.lang.Exception</CODE> - if the class is not defined<DT><B>See Also:</B><DD><A HREF="../../weka/classifiers/Evaluation.html#useNoPriors()"><CODE>useNoPriors()</CODE></A>, 
<A HREF="../../weka/classifiers/Evaluation.html#setPriors(weka.core.Instances)"><CODE>setPriors(Instances)</CODE></A></DL>
</DL>
<HR>

<A NAME="Evaluation(weka.core.Instances, weka.classifiers.CostMatrix)"><!-- --></A><H3>
Evaluation</H3>
<PRE>
public <B>Evaluation</B>(<A HREF="../../weka/core/Instances.html" title="class in weka.core">Instances</A>&nbsp;data,
                  <A HREF="../../weka/classifiers/CostMatrix.html" title="class in weka.classifiers">CostMatrix</A>&nbsp;costMatrix)
           throws java.lang.Exception</PRE>
<DL>
<DD>Initializes all the counters for the evaluation and also takes a
 cost matrix as parameter.
 Use <code>useNoPriors()</code> if the dataset is the test set and you
 can't initialize with the priors from the training set via 
 <code>setPriors(Instances)</code>.
<P>
<DL>
<DT><B>Parameters:</B><DD><CODE>data</CODE> - set of training instances, to get some header 
                        information and prior class distribution information<DD><CODE>costMatrix</CODE> - the cost matrix---if null, default costs will be used
<DT><B>Throws:</B>
<DD><CODE>java.lang.Exception</CODE> - if cost matrix is not compatible with 
                        data, the class is not defined or the class is numeric<DT><B>See Also:</B><DD><A HREF="../../weka/classifiers/Evaluation.html#useNoPriors()"><CODE>useNoPriors()</CODE></A>, 
<A HREF="../../weka/classifiers/Evaluation.html#setPriors(weka.core.Instances)"><CODE>setPriors(Instances)</CODE></A></DL>
</DL>

<!-- ============ METHOD DETAIL ========== -->

<A NAME="method_detail"><!-- --></A>
<TABLE BORDER="1" WIDTH="100%" CELLPADDING="3" CELLSPACING="0" SUMMARY="">
<TR BGCOLOR="#CCCCFF" CLASS="TableHeadingColor">
<TH ALIGN="left" COLSPAN="1"><FONT SIZE="+2">
<B>Method Detail</B></FONT></TH>
</TR>
</TABLE>

<A NAME="areaUnderROC(int)"><!-- --></A><H3>
areaUnderROC</H3>
<PRE>
public double <B>areaUnderROC</B>(int&nbsp;classIndex)</PRE>
<DL>
<DD>Returns the area under ROC for those predictions that have been collected
 in the evaluateClassifier(Classifier, Instances) method. Returns 
 Instance.missingValue() if the area is not available.
<P>
<DD><DL>
</DL>
</DD>
<DD><DL>
<DT><B>Parameters:</B><DD><CODE>classIndex</CODE> - the index of the class to consider as "positive"
<DT><B>Returns:</B><DD>the area under the ROC curve or not a number</DL>
</DD>
</DL>
<HR>

<A NAME="confusionMatrix()"><!-- --></A><H3>
confusionMatrix</H3>
<PRE>
public double[][] <B>confusionMatrix</B>()</PRE>
<DL>
<DD>Returns a copy of the confusion matrix.
<P>
<DD><DL>
</DL>
</DD>
<DD><DL>

<DT><B>Returns:</B><DD>a copy of the confusion matrix as a two-dimensional array</DL>
</DD>
</DL>
<HR>

<A NAME="crossValidateModel(weka.classifiers.Classifier, weka.core.Instances, int, java.util.Random, java.lang.Object...)"><!-- --></A><H3>
crossValidateModel</H3>
<PRE>
public void <B>crossValidateModel</B>(<A HREF="../../weka/classifiers/Classifier.html" title="class in weka.classifiers">Classifier</A>&nbsp;classifier,
                               <A HREF="../../weka/core/Instances.html" title="class in weka.core">Instances</A>&nbsp;data,
                               int&nbsp;numFolds,
                               java.util.Random&nbsp;random,
                               java.lang.Object...&nbsp;forPredictionsPrinting)
                        throws java.lang.Exception</PRE>
<DL>
<DD>Performs a (stratified if class is nominal) cross-validation 
 for a classifier on a set of instances. Now performs
 a deep copy of the classifier before each call to 
 buildClassifier() (just in case the classifier is not
 initialized properly).
<P>
<DD><DL>
</DL>
</DD>
<DD><DL>
<DT><B>Parameters:</B><DD><CODE>classifier</CODE> - the classifier with any options set.<DD><CODE>data</CODE> - the data on which the cross-validation is to be 
 performed<DD><CODE>numFolds</CODE> - the number of folds for the cross-validation<DD><CODE>random</CODE> - random number generator for randomization<DD><CODE>forPredictionsString</CODE> - varargs parameter that, if supplied, is
 expected to hold a StringBuffer to print predictions to, 
 a Range of attributes to output and a Boolean (true if the distribution
 is to be printed)
<DT><B>Throws:</B>
<DD><CODE>java.lang.Exception</CODE> - if a classifier could not be generated 
 successfully or the class is not defined</DL>
</DD>
</DL>
<HR>

<A NAME="crossValidateModel(java.lang.String, weka.core.Instances, int, java.lang.String[], java.util.Random)"><!-- --></A><H3>
crossValidateModel</H3>
<PRE>
public void <B>crossValidateModel</B>(java.lang.String&nbsp;classifierString,
                               <A HREF="../../weka/core/Instances.html" title="class in weka.core">Instances</A>&nbsp;data,
                               int&nbsp;numFolds,
                               java.lang.String[]&nbsp;options,
                               java.util.Random&nbsp;random)
                        throws java.lang.Exception</PRE>
<DL>
<DD>Performs a (stratified if class is nominal) cross-validation 
 for a classifier on a set of instances.
<P>
<DD><DL>
</DL>
</DD>
<DD><DL>
<DT><B>Parameters:</B><DD><CODE>classifierString</CODE> - a string naming the class of the classifier<DD><CODE>data</CODE> - the data on which the cross-validation is to be 
 performed<DD><CODE>numFolds</CODE> - the number of folds for the cross-validation<DD><CODE>options</CODE> - the options to the classifier. Any options<DD><CODE>random</CODE> - the random number generator for randomizing the data
 accepted by the classifier will be removed from this array.
<DT><B>Throws:</B>
<DD><CODE>java.lang.Exception</CODE> - if a classifier could not be generated 
 successfully or the class is not defined</DL>
</DD>
</DL>
<HR>

<A NAME="evaluateModel(java.lang.String, java.lang.String[])"><!-- --></A><H3>
evaluateModel</H3>
<PRE>
public static java.lang.String <B>evaluateModel</B>(java.lang.String&nbsp;classifierString,
                                             java.lang.String[]&nbsp;options)
                                      throws java.lang.Exception</PRE>
<DL>
<DD>Evaluates a classifier with the options given in an array of
 strings. <p/>

 Valid options are: <p/>

 -t filename <br/>
 Name of the file with the training data. (required) <p/>

 -T filename <br/>
 Name of the file with the test data. If missing a cross-validation 
 is performed. <p/>

 -c index <br/>
 Index of the class attribute (1, 2, ...; default: last). <p/>

 -x number <br/>
 The number of folds for the cross-validation (default: 10). <p/>

 -no-cv <br/>
 No cross validation.  If no test file is provided, no evaluation
 is done. <p/>
 
 -split-percentage percentage <br/>
 Sets the percentage for the train/test set split, e.g., 66. <p/>
 
 -preserve-order <br/>
 Preserves the order in the percentage split instead of randomizing
 the data first with the seed value ('-s'). <p/>

 -s seed <br/>
 Random number seed for the cross-validation and percentage split
 (default: 1). <p/>

 -m filename <br/>
 The name of a file containing a cost matrix. <p/>

 -l filename <br/>
 Loads classifier from the given file. In case the filename ends with
 ".xml" the options are loaded from XML. <p/>

 -d filename <br/>
 Saves classifier built from the training data into the given file. In case 
 the filename ends with ".xml" the options are saved XML, not the model. <p/>

 -v <br/>
 Outputs no statistics for the training data. <p/>

 -o <br/>
 Outputs statistics only, not the classifier. <p/>
 
 -i <br/>
 Outputs detailed information-retrieval statistics per class. <p/>

 -k <br/>
 Outputs information-theoretic statistics. <p/>

 -p range <br/>
 Outputs predictions for test instances (or the train instances if no test
 instances provided  and -no-cv is used), along with the attributes in the specified range (and 
  nothing else). Use '-p 0' if no attributes are desired. <p/>

 -distribution <br/>
 Outputs the distribution instead of only the prediction
 in conjunction with the '-p' option (only nominal classes). <p/>

 -r <br/>
 Outputs cumulative margin distribution (and nothing else). <p/>

 -g <br/> 
 Only for classifiers that implement "Graphable." Outputs
 the graph representation of the classifier (and nothing
 else). <p/>

 -xml filename | xml-string <br/>
 Retrieves the options from the XML-data instead of the command line. <p/>
 
 -threshold-file file <br/>
 The file to save the threshold data to.
 The format is determined by the extensions, e.g., '.arff' for ARFF
 format or '.csv' for CSV. <p/>
         
 -threshold-label label <br/>
 The class label to determine the threshold data for
 (default is the first label) <p/>
<P>
<DD><DL>
</DL>
</DD>
<DD><DL>
<DT><B>Parameters:</B><DD><CODE>classifierString</CODE> - class of machine learning classifier as a string<DD><CODE>options</CODE> - the array of string containing the options
<DT><B>Returns:</B><DD>a string describing the results
<DT><B>Throws:</B>
<DD><CODE>java.lang.Exception</CODE> - if model could not be evaluated successfully</DL>
</DD>
</DL>
<HR>

<A NAME="main(java.lang.String[])"><!-- --></A><H3>
main</H3>
<PRE>
public static void <B>main</B>(java.lang.String[]&nbsp;args)</PRE>
<DL>
<DD>A test method for this class. Just extracts the first command line
 argument as a classifier class name and calls evaluateModel.
<P>
<DD><DL>
</DL>
</DD>
<DD><DL>
<DT><B>Parameters:</B><DD><CODE>args</CODE> - an array of command line arguments, the first of which
 must be the class name of a classifier.</DL>
</DD>
</DL>
<HR>

<A NAME="evaluateModel(weka.classifiers.Classifier, java.lang.String[])"><!-- --></A><H3>
evaluateModel</H3>
<PRE>
public static java.lang.String <B>evaluateModel</B>(<A HREF="../../weka/classifiers/Classifier.html" title="class in weka.classifiers">Classifier</A>&nbsp;classifier,
                                             java.lang.String[]&nbsp;options)
                                      throws java.lang.Exception</PRE>
<DL>
<DD>Evaluates a classifier with the options given in an array of
 strings. <p/>

 Valid options are: <p/>

 -t name of training file <br/>
 Name of the file with the training data. (required) <p/>

 -T name of test file <br/>
 Name of the file with the test data. If missing a cross-validation 
 is performed. <p/>

 -c class index <br/>