Mahout Installation and configuration __mahout installation and configuration under Hadoop platform

First, download the binary file click on the Open link

Second, extract the file

    TAR-ZXVF mahout-distribution-0.9.tar.gz-c/usr

Third, configure environment variables: in/etc/profile, add mahout_home environment variables

Export mahout_home=/usr/apache-mahout-distribution-0.12.2

Export path= $PATH: $HADOOP _home/bin: $MAHOUT _home/bin

Export classpath=.: $JAVA _home/lib: $MAHOUT _home/lib: $JRE _home/lib: $CLASSPATH

Note: Be sure to execute the command-source/etc/profile after modifying the environment variable

Four, start Hadoop

start-all.sh

V. View the mahout version

Mahout--help

Vi. Use of Mahout

A. Download a file Synthetic_control.data, download the address Http://archive.ics.uci.edu/ml/databases/synthetic_control/synthetic_ Control.data, and put this file in the $mahout_home directory.

B. Start HADOOP: $HADOOP _home/bin/start-all.sh

C. Create test directory input and import data into this input directory

root@master# Hadoop Fs-mkdir Input 

root@master:~/$ Hadoop fs-put/home/zc/desktop/synthetic_control.data Input

D. Use the Kmeans algorithm (which will run for a few minutes)

root@master~/$ Hadoop Jar/usr/apache-mahout-distribution-0.12.2/mahout-examples-0.12.2.jar Org.apache.mahout.clustering.syntheticcontrol.kmeans.Job

E. Viewing results

Hadoop fs-cat/output/data/part-m-00000