[CONF] Apache Lucene Mahout > Partial Implementation

[co...@apache.org]

Space: Apache Lucene Mahout (http://cwiki.apache.org/confluence/display/MAHOUT)
Page: Partial Implementation (http://cwiki.apache.org/confluence/display/MAHOUT/Partial+Implementation)


Edited by abdelHakim Deneche:
---------------------------------------------------------------------
h1. Introduction

This quick start page shows how to build a decision forest using the partial implementation. This tutorial also explains how to use the decision forest to classify new data.
Partial Decision Forests is a mapreduce implementation where each mapper builds a subset of the forest using only the data available in its partition. This allows building forests using large datasets as long as each partition can be loaded in-memory.

h1. Steps
h2. Download the data
* The current implementation is compatible with the UCI repository file format. In this example we'll use the NSL-KDD dataset because its large enough to show the performances of the partial implementation.
You can download the dataset here http://nsl.cs.unb.ca/NSL-KDD/
You can either download the full training set "KDDTrain+.ARFF", or a 20% subset "KDDTrain+_20Percent.ARFF" (we'll use the full dataset in this tutorial) and the test set "KDDTest+.ARFF".
* Open the train and test files and remove all the lines that begin with '@'. All those lines are at the top of the files. Actually you can keep those lines somewhere, because they'll help us describe the dataset to Mahout
* Put the data in HDFS: {code}$HADOOP_HOME/bin/hadoop fs -put <PATH TO DATA> testdata{code}

h2. Build the Job files
* In $MAHOUT_HOME/ run: {code}mvn install -DskipTests{code}

h2. Generate a file descriptor for the dataset: 
run the following command:
{code}
$HADOOP_HOME/bin/hadoop jar $MAHOUT_HOME/core/target/mahout-core-<VERSION>.job org.apache.mahout.df.tools.Describe -p testdata/KDDTrain+.arff -f testdata/KDDTrain+.info -d N 3 C 2 N C 4 N C 8 N 2 C 19 N L
{code}
The "N 3 C 2 N C 4 N C 8 N 2 C 19 N L" string indicates the nature of the variables. which means 1 numerical(N) attribute, followed by 3 numerical(Categorical) attributes, ...L indicates the label, and you can use I to ignore some attributes

h2. Run the example

For now there are two implementations, one "mapred" that works with Hadoop old API (pre 0.20) and "mapreduce" that works with the new Hadoop API (0.20). You should only use the "mapreduce" version, the "mapred" version should be discarded in the future.

{code}
$HADOOP_HOME/hadoop jar $MAHOUT_HOME/examples/target/mahout-examples-<version>.job org.apache.mahout.df.mapreduce.BuildForest -Dmapred.max.split.size=1874231 -oob -d testdata/KDDTrain+.arff -ds testdata/KDDTrain+.info -sl 5 -p -t 100 -o nsl-forest
{code}
which builds 100 trees (-t argument) using the partial implementation (-p). Each tree is built using 5 random selected attribute per node (-sl argument) the example computes the out-of-bag error (-oob) and outputs the decision tree in the "nsl-forest" directory (-o)
The number of partitions is controlled by the -Dmapred.max.split.size argument that indicates to Hadoop the max. size of each partition, in this case 1/10 of the size of the dataset. Thus 10 partitions will be used.
IMPORTANT: using less partitions should give better classification results, but needs a lot of memory. So if you are the Jobs are failing, try increasing Hadoop's childs HeapSize or the number of partitions.
* The example outputs the Build Time and the oob error estimation

{code}
10/03/13 17:57:29 INFO mapreduce.BuildForest: Build Time: 0h 7m 43s 582
10/03/13 17:57:33 INFO mapreduce.BuildForest: oob error estimate : 0.002325895231517865
10/03/13 17:57:33 INFO mapreduce.BuildForest: Storing the forest in: nsl-forest/forest.seq
{code}

h2. Using the Decision Forest to Classify new data
run the following command:
{code}
$HADOOP_HOME/hadoop jar $MAHOUT_HOME/target/mahout-examples-<version>-SNAPSHOT.job org.apache.mahout.df.mapreduce.TestForest -i nsl-kdd/KDDTest+.info -ds nsl-kdd/KDDTrain+.info -m nsl-forest -a -o predictions.txt
{code}
This will compute the predictions of "KDDTest+.arff" dataset (-i argument) using the same data descriptor generated for the training dataset (-ds) and the decision forest built previously (-m). Optionally (if the test dataset contains the labels of the tuples) run the analyzer to compute the confusion matrix (-a), and you can also store the predictions in a text file (-o).
* The example should output the classification time and the confusion matrix

{code}
10/03/13 18:08:56 INFO mapreduce.TestForest: Classification Time: 0h 0m 6s 355
10/03/13 18:08:56 INFO mapreduce.TestForest: =======================================================
Summary
-------------------------------------------------------
Correctly Classified Instances          :      17657	   78.3224%
Incorrectly Classified Instances        :       4887	   21.6776%
Total Classified Instances              :      22544

=======================================================
Confusion Matrix
-------------------------------------------------------
a    	b    	<--Classified as
9459 	252  	 |  9711  	a     = normal
4635 	8198 	 |  12833 	b     = anomaly
Default Category: unknown: 2
{code}

h2. Known Issues and limitations
The "Decision Forest" code is still "a work in progress", many features are still missing. Here is a list of some known issues:
* The input dataset must be a single file. Multiple input files are not, yet, supported
* The tree building is done when each mapper.close() method is called. Because the mappers don't refresh their state, the job can fail when the dataset is big and you try to build a large number of trees.

Change your notification preferences: http://cwiki.apache.org/confluence/users/viewnotifications.action