6

I am trying to do a "clusterdump" on the output of mahout kmeans clustering example (synthetic_control example). But I am experiencing the following error:

> ~/MAHOUT/trunk/bin/mahout clusterdump --seqFileDir clusters-10-final --pointsDir clusteredPoints --output a1.txt

MAHOUT_LOCAL is not set; adding HADOOP_CONF_DIR to classpath.
Running on hadoop, using /usr/lib/hadoop/bin/hadoop and HADOOP_CONF_DIR=/usr/lib/hadoop/conf/
MAHOUT-JOB: /home/<username>/MAHOUT/trunk/examples/target/mahout-examples-0.8-SNAPSHOT-job.jar

12/06/21 22:43:18 WARN conf.Configuration: DEPRECATED: hadoop-site.xml found in the classpath. Usage of hadoop-site.xml is deprecated. Instead use core-site.xml, mapred-site.xml and hdfs-site.xml to override properties of core-default.xml, mapred-default.xml and hdfs-default.xml respectively

12/06/21 22:43:25 ERROR common.AbstractJob: Unexpected --seqFileDir while processing Job-Specific Options:
usage: <command> [Generic Options] [Job-Specific Options]
.....

So I guess there is no "seqFileDir" option for clusterdump but all the online tutorials (e.g https://cwiki.apache.org/MAHOUT/cluster-dumper.html) refer this option. Can you please suggest me the remedy or what I am missing ?

tuxdna
  • 8,257
  • 4
  • 43
  • 61
Aniruddha Basak
  • 67
  • 1
  • 6

1 Answers1

2

Had you tried to specify it as --input option?

Alex Ott
  • 80,552
  • 8
  • 87
  • 132
  • 1
    yes I did. When I replaced --seqFileDir by --input, an output file was generated , but that was EMPTY ! Do u know why the tutorials talk about --seqFileDir option ? – Aniruddha Basak Jun 22 '12 at 17:20
  • 2
    I was working on this problem. Magically I found the solution !!! Thanks for your suggestion --input worked inplace of --seqFileDir option. What I was doing wrong was, I did not realize clusterdump (with HADOOP_HOME set) reads from HDFS and writes output to local filesystem. Anyway, things are working fine now ! – Aniruddha Basak Jun 22 '12 at 18:57