org.apache.mahout.classifier
Class BayesFileFormatter

java.lang.Object
  org.apache.mahout.classifier.BayesFileFormatter

public final class BayesFileFormatter
extends Object

Flatten a file into format that can be read by the Bayes M/R job.

One document per line, first token is the label followed by a tab, rest of the line are the terms.

Method Summary

static void	collapse(String label, org.apache.lucene.analysis.Analyzer analyzer, File inputDir, Charset charset, File outputFile) Collapse all the files in the inputDir into a single file in the proper Bayes format, 1 document per line
static void	format(String label, org.apache.lucene.analysis.Analyzer analyzer, File input, Charset charset, File outDir) Write the input files to the outdir, one output file per input file
static void	main(String[] args) Run the FileFormatter
static String[]	readerToDocument(org.apache.lucene.analysis.Analyzer analyzer, Reader reader) Convert a Reader to a vector

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

collapse

public static void collapse(String label,
                            org.apache.lucene.analysis.Analyzer analyzer,
                            File inputDir,
                            Charset charset,
                            File outputFile)
                     throws IOException

Collapse all the files in the inputDir into a single file in the proper Bayes format, 1 document per line

Parameters:

label - The label

analyzer - The analyzer to use

inputDir - The input Directory

charset - The charset of the input files

outputFile - The file to collapse to

Throws:

IOException

format

public static void format(String label,
                          org.apache.lucene.analysis.Analyzer analyzer,
                          File input,
                          Charset charset,
                          File outDir)
                   throws IOException

Write the input files to the outdir, one output file per input file

Parameters:

label - The label of the file

analyzer - The analyzer to use

input - The input file or directory. May not be null

charset - The Character set of the input files

outDir - The output directory. Files will be written there with the same name as the input file

Throws:

IOException

readerToDocument

public static String[] readerToDocument(org.apache.lucene.analysis.Analyzer analyzer,
                                        Reader reader)
                                 throws IOException

Convert a Reader to a vector

Parameters:

analyzer - The Analyzer to use

reader - The reader to feed to the Analyzer

Returns:

An array of unique tokens

Throws:

IOException

main

public static void main(String[] args)
                 throws Exception

Run the FileFormatter

Parameters:

args - The input args. Run with -h to see the help

Throws:

ClassNotFoundException - if the Analyzer can't be found

IllegalAccessException - if the Analyzer can't be constructed

InstantiationException - if the Analyzer can't be constructed

IOException - if the files can't be dealt with properly

Exception