mahout kmeans 例子

痛定思痛。 2022-08-18 00:44 173阅读 0赞

**一、mahout 简单例子测试**

--------------------

mahout 安装配置可以参考：[mahout安装配置][mahout]

1、kmeans 聚类算法测试数据来源：

地址：[http://archive.ics.uci.edu/ml/databases/synthetic\_control/synthetic\_control.data][http_archive.ics.uci.edu_ml_databases_synthetic_control_synthetic_control.data]

2、下载数据，把数据存放到hdfs上（hadoop2.6.1 已经启动）

创建测试目录testdata，并把数据导入到这个tastdata目录中(这里的目录的名字只能是testdata)

$ hdfs dfs -mkdir testdata
        $ hdfs dfs -put /home/lin/hadoop/mahout-distribution-0.10.0/test.data  testdata

3、执行kmeans算法，等待运行结果

$ hadoop jar /home/lin/hadoop/mahout-distribution-0.10.0/mahout-examples-0.10.0-job.jar org.apache.mahout.clustering.syntheticcontrol.kmeans.Job

4、运行成功查看运行结果

hdfs dfs -ls output

显示如下结果证明运行成功：

lin@lin162:~/hadoop/hadoop-2.6.1/etc/hadoop$ hdfs dfs -ls output         
        Found 15 items
        -rw-r--r--   2 lin supergroup        194 2015-12-01 12:27 output/_policy
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:27 output/clusteredPoints
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:22 output/clusters-0
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:23 output/clusters-1
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:27 output/clusters-10-final
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:23 output/clusters-2
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:24 output/clusters-3
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:24 output/clusters-4
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:25 output/clusters-5
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:25 output/clusters-6
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:25 output/clusters-7
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:26 output/clusters-8
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:26 output/clusters-9
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:22 output/data
        drwxr-xr-x   - lin supergroup          0 2015-12-01 12:22 output/random-seeds

--------------------

[mahout]: http://blog.csdn.net/linlinv3/article/details/50127921
[http_archive.ics.uci.edu_ml_databases_synthetic_control_synthetic_control.data]: http://archive.ics.uci.edu/ml/databases/synthetic_control/synthetic_control.data