[
https://issues.apache.org/jira/browse/MAHOUT-850?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13134780#comment-13134780
]
Deneche A. Hakim commented on MAHOUT-850:
-----------------------------------------
Does this problem appears also when using the latest version from svn ?
If no, and if you want to give it a try please note that because of some recent
changes the "-oob" parameter is no longer recognized so you should remove it
when calling BuildForest. You'll also need to regenerate the dataset
description files using Describe because the previous one wont work anymore.
> Random Forest Partial implementation in training stage throw EOF exception
> --------------------------------------------------------------------------
>
> Key: MAHOUT-850
> URL: https://issues.apache.org/jira/browse/MAHOUT-850
> Project: Mahout
> Issue Type: Bug
> Components: Classification
> Affects Versions: 0.5
> Environment: Java 1.6
> Mac OS 10.6.8
> Hadoop 0.20.205
> Reporter: Wangda Tan
> Labels: eof, forest, random
> Attachments: PartialBuilder.patch
>
>
> when running next command line to training:
> bin/hadoop jar mahout-examples-0.5-gphdce-1.1.0.0-job.jar
> org.apache.mahout.df.mapred.BuildForest -Dmapred.max.split.size=1874231 -oob
> -d rf_input/KDDTrain_20Percent.arff -ds rf_input/KDDTrain_20Percent.info -sl
> 5 -p -t 100
> it will throw an EOF exception.
> --
> Reason:
> In previous stage, _SUCCESS and _logs file will in the output folder, when
> parse the output folder, _SUCCESS and _logs will cause an EOF exception
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
