（转）Mahout使用入门

一、简介

Mahout 是 Apache Software Foundation（ASF）旗下的一个开源项目，提供一些可扩展的机器学习领域经典算法的实现，旨在帮助开发人员更加方便快捷地创建智能应用程序。Apache Mahout项目已经发展到了它的第三个年头，目前已经有了三个公共发行版本。Mahout包含许多实现，包括集群、分类、推荐过滤、频繁子项挖掘。此外，通过使用 Apache Hadoop 库，Mahout 可以有效地扩展到云中。

二、下载与准备

程序下载

下载hadoop http://labs.renren.com/apache-mirror/hadoop/common/下载适合版本的包（本文采用稳定版 hadoop-0.20.203.0rc1.tar.gz ）

下载mahout http://labs.renren.com/apache-mirror/mahout/

      （本文采用mahout-distribution-0.5.tar.gz） 

如需更多功能可能还需下载 maven 和 mahout-collections

数据下载

数据源：http://kdd.ics.uci.edu/databases/里面有大量经典数据提供下载

（本文使用synthetic_control 数据，synthetic_control.tar.gz）

三、安装与部署

为了不污染Linux root环境，本文采用在个人Home目录安装程序，程序目录为$HOME/local。

程序已经下载到$HOME/Downloads，使用tar命令解压：

tar zxvf hadoop-0.20.203.0rc1.tar.gz -C ~/local/

cd ~/local

mv hadoop-0.20.203.0 hadoop

 

tar zxvf mahout-distribution-0.5.tar.gz -C ~/local/

cd ~/local

mv mahout-distribution-0.5 mahout

 

修改.bash_profile / .bashrc

export HADOOP_HOME=$HOME/local/hadoop

export HADOOP_CONF_DIR=$HADOOP_HOME/conf

为方便使用程序命令，可把程序bin目录添加到$PATH下，或者直接alias 。

#Alias for apps

alias mahout='$HOME/local/mahout/mahout'

alias hdp='$HOME/local/hadoop/hdp'

 

测试

输入命令： mahout

预期结果：

Running on hadoop, using HADOOP_HOME=/home/username/local/hadoop

HADOOP_CONF_DIR=/home/username/local/hadoop/conf

An example program must be given as the first argument.

Valid program names are:

  arff.vector: : Generate Vectors from an ARFF file or directory

  canopy: : Canopy clustering

  cat: : Print a file or resource as the logistic regression models would see it

  cleansvd: : Cleanup and verification of SVD output

  clusterdump: : Dump cluster output to text

  dirichlet: : Dirichlet Clustering

  eigencuts: : Eigencuts spectral clustering

  evaluateFactorization: : compute RMSE of a rating matrix factorization against probes in memory

  evaluateFactorizationParallel: : compute RMSE of a rating matrix factorization against probes

  fkmeans: : Fuzzy K-means clustering

  fpg: : Frequent Pattern Growth

  itemsimilarity: : Compute the item-item-similarities for item-based collaborative filtering

  kmeans: : K-means clustering

  lda: : Latent Dirchlet Allocation

  ldatopics: : LDA Print Topics

  lucene.vector: : Generate Vectors from a Lucene index

  matrixmult: : Take the product of two matrices

  meanshift: : Mean Shift clustering

  parallelALS: : ALS-WR factorization of a rating matrix

  predictFromFactorization: : predict preferences from a factorization of a rating matrix

  prepare20newsgroups: : Reformat 20 newsgroups data

  recommenditembased: : Compute recommendations using item-based collaborative filtering

  rowid: : Map SequenceFile<Text,VectorWritable> to {SequenceFile<IntWritable,VectorWritable>, SequenceFile<IntWritable,Text>}

  rowsimilarity: : Compute the pairwise similarities of the rows of a matrix

  runlogistic: : Run a logistic regression model against CSV data

  seq2sparse: : Sparse Vector generation from Text sequence files

  seqdirectory: : Generate sequence files (of Text) from a directory

  seqdumper: : Generic Sequence File dumper

  seqwiki: : Wikipedia xml dump to sequence file

  spectralkmeans: : Spectral k-means clustering

  splitDataset: : split a rating dataset into training and probe parts

  ssvd: : Stochastic SVD

  svd: : Lanczos Singular Value Decomposition

  testclassifier: : Test Bayes Classifier

  trainclassifier: : Train Bayes Classifier

  trainlogistic: : Train a logistic regression using stochastic gradient descent

  transpose: : Take the transpose of a matrix

  vectordump: : Dump vectors from a sequence file to text

  wikipediaDataSetCreator: : Splits data set of wikipedia wrt feature like country

  wikipediaXMLSplitter: : Reads wikipedia data and creates ch

输入命令：hdp

预期结果：

Usage: hadoop [--config confdir] COMMAND

where COMMAND is one of:

  namenode -format     format the DFS filesystem

  secondarynamenode    run the DFS secondary namenode

  namenode             run the DFS namenode

  datanode             run a DFS datanode

  dfsadmin             run a DFS admin client

  mradmin              run a Map-Reduce admin client

  fsck                 run a DFS filesystem checking utility

  fs                   run a generic filesystem user client

  balancer             run a cluster balancing utility

  fetchdt              fetch a delegation token from the NameNode

  jobtracker           run the MapReduce job Tracker node

  pipes                run a Pipes job

  tasktracker          run a MapReduce task Tracker node

  historyserver        run job history servers as a standalone daemon

  job                  manipulate MapReduce jobs

  queue                get information regarding JobQueues

  version              print the version

  jar <jar>            run a jar file

  distcp <srcurl> <desturl> copy file or directories recursively

  archive -archiveName NAME -p <parent path> <src>* <dest> create a hadoop archive

  classpath            prints the class path needed to get the

                       Hadoop jar and the required libraries

  daemonlog            get/set the log level for each daemon

 or

  CLASSNAME            run the class named CLASSNAME

Most commands print help when invoked w/o parameters.

 

五、运行

步骤一：

通过这个命令可以查看mahout提供了哪些算法,以及如何使用

mahout --help

 

mahout kmeans --input /user/hive/warehouse/tmp_data/complex.seq   --clusters  5 --output  /home/hadoopuser/1.txt

 

mahout下处理的文件必须是SequenceFile格式的，所以需要把txtfile转换成sequenceFile。SequenceFile是hadoop中的一个类，允许我们向文件中写入二进制的键值对，具体介绍请看

eyjian写的http://www.hadoopor.com/viewthread.php?tid=144&highlight=sequencefile

 

mahout中提供了一种将指定文件下的文件转换成sequenceFile的方式。

（You may find Tika (http://lucene.apache.org/tika) helpful in converting binary documents to text.）

使用方法如下：

 

$MAHOUT_HOME/mahout seqdirectory \

--input <PARENT DIR WHERE DOCS ARE LOCATED> --output <OUTPUT DIRECTORY> \

<-c <CHARSET NAME OF THE INPUT DOCUMENTS> {UTF-8|cp1252|ascii...}> \

<-chunk <MAX SIZE OF EACH CHUNK in Megabytes> 64> \

<-prefix <PREFIX TO ADD TO THE DOCUMENT ID>>

 

举个例子：

mahout seqdirectory --input /hive/hadoopuser/ --output /mahout/seq/ --charset UTF-8

 

步骤二：

运行kmeans的简单的例子：

 

1：将样本数据集放到hdfs中指定文件下,应该在testdata文件夹下

$HADOOP_HOME/hdp fs -put <PATH TO DATA> testdata

例如：

dap fs   -put ~/datasetsynthetic_controltest/synthetic_control.data  ~/local/mahout/testdata/

 

2：使用kmeans算法

hdp jar $MAHOUT_HOME/examples/target/mahout-examples-$MAHOUT_VERSION.job org.apache.mahout.clustering.syntheticcontrol.kmeans.Job

例如：

hdp jar /home/hadoopuser/mahout-0.3/mahout-examples-0.1.job org.apache.mahout.clustering.syntheticcontrol.kmeans.Job

 

3：使用canopy算法

hdp jar $MAHOUT_HOME/examples/target/mahout-examples-$MAHOUT_VERSION.job org.apache.mahout.clustering.syntheticcontrol.canopy.Job

例如：

hdp jar /home/hadoopuser/mahout-0.3/mahout-examples-0.1.job org.apache.mahout.clustering.syntheticcontrol.canopy.Job

 

4：使用dirichlet 算法

mahout jar $MAHOUT_HOME/examples/target/mahout-examples-$MAHOUT_VERSION.job org.apache.mahout.clustering.syntheticcontrol.dirichlet.Job

 

5：使用meanshift算法

meanshift :

hdp jar $MAHOUT_HOME/examples/target/mahout-examples-$MAHOUT_VERSION.job org.apache.mahout.clustering.syntheticcontrol.meanshift.Job

 

6：查看一下结果吧

mahout vectordump --seqFile /user/hadoopuser/output/data/part-00000

这个直接把结果显示在控制台上。

 

可以到hdfs中去看看数据是什么样子的

上面跑的例子大多以testdata作为输入和输出文件夹名

可以使用 hdp fs -lsr 来查看所有的输出结果

 

KMeans 方法的输出结果在  output/points

Canopy 和 MeanShift 结果放在了 output/clustered-points