org.apache.mahout.utils

Class SplitInputJob

java.lang.Object

org.apache.mahout.utils.SplitInputJob

public final class SplitInputJob
extends Object

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	SplitInputJob.SplitInputComparator Randomly permute key value pairs
static class	SplitInputJob.SplitInputMapper Mapper which downsamples the input by downsamplingFactor
static class	SplitInputJob.SplitInputReducer Reducer which uses MultipleOutputs to randomly allocate key value pairs between test and training outputs

Method Summary

Methods

Modifier and Type	Method and Description
static void	run(org.apache.hadoop.conf.Configuration initialConf, org.apache.hadoop.fs.Path inputPath, org.apache.hadoop.fs.Path outputPath, int keepPct, float randomSelectionPercent) Run job to downsample, randomly permute and split data into test and training sets.

Methods inherited from class java.lang.Object

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

Method Detail

run

public static void run(org.apache.hadoop.conf.Configuration initialConf,
       org.apache.hadoop.fs.Path inputPath,
       org.apache.hadoop.fs.Path outputPath,
       int keepPct,
       float randomSelectionPercent)
                throws IOException,
                       ClassNotFoundException,
                       InterruptedException

Run job to downsample, randomly permute and split data into test and training sets. This job takes a SequenceFile as input and outputs two SequenceFiles test-r-00000 and training-r-00000 which contain the test and training sets respectively

Parameters:

initialConf -

inputPath - path to input data SequenceFile

outputPath - path for output data SequenceFiles

keepPct - percentage of key value pairs in input to keep. The rest are discarded

randomSelectionPercent - percentage of key value pairs to allocate to test set. Remainder are allocated to training set

Throws:

IOException

ClassNotFoundException

InterruptedException