You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mahout.apache.org by "Suneel Marthi (JIRA)" <ji...@apache.org> on 2013/06/28 05:57:20 UTC

[jira] [Commented] (MAHOUT-1271) classify-20newsgroups.sh fails during the seqdirectory step

    [ https://issues.apache.org/jira/browse/MAHOUT-1271?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13695245#comment-13695245 ] 

Suneel Marthi commented on MAHOUT-1271:
---------------------------------------

This issue is only seen when running with Complimentary Naive Bayes or Naive Bayes options.

seqdirectory now executes as a MapReduce job (in light of the changes in Mahout-833) unless the sequential option is explicitly specified.  

classify-20newsgroups.sh creates 20new-seq/ directory on the first run, but subsequent runs fail as the directory already exists from the previous runs.

The fix was to specify --overwrite flag when invoking seqdirectory.
                
> classify-20newsgroups.sh fails during the seqdirectory step
> -----------------------------------------------------------
>
>                 Key: MAHOUT-1271
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-1271
>             Project: Mahout
>          Issue Type: Bug
>          Components: Examples
>    Affects Versions: 0.8
>            Reporter: Suneel Marthi
>            Assignee: Suneel Marthi
>             Fix For: 0.8
>
>
> classify-20newsgroups.sh fails during the seqdirectory step as it fails to overwrite the output folder.
> {Code}
> Exception in thread "main" org.apache.hadoop.mapred.FileAlreadyExistsException: Output directory /tmp/mahout-work-jenkins/20news-seq already exists
> {Code}

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira