[
https://issues.apache.org/jira/browse/MAHOUT-1464?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13968603#comment-13968603
]
Dmitriy Lyubimov commented on MAHOUT-1464:
------------------------------------------
[~pferrel] if you look inside the Sebastian's patch, you will find it is
hardcoded to use "local" Spark master. The master you specify to Client only
tells which cluster to ship code to, not which master for the application to
use. Which is why i think this Client thing is a little bit raw idea. Either
way, it will not work with Sebastian's app. Instead, I'd suggest you to run
Sebastian script directly from IDEA as a first step, after hacking master url
in this line
{code}
implicit val sc = mahoutSparkContext(masterUrl = "local", appName =
"MahoutLocalContext",
+ customJars = Traversable.empty[String])
{code}
or making the script to accept it from environment or app params.
(the convention in Spark example programs is that they accept master url as the
first parameter).
> Cooccurrence Analysis on Spark
> ------------------------------
>
> Key: MAHOUT-1464
> URL: https://issues.apache.org/jira/browse/MAHOUT-1464
> Project: Mahout
> Issue Type: Improvement
> Components: Collaborative Filtering
> Environment: hadoop, spark
> Reporter: Pat Ferrel
> Assignee: Sebastian Schelter
> Fix For: 1.0
>
> Attachments: MAHOUT-1464.patch, MAHOUT-1464.patch, MAHOUT-1464.patch,
> MAHOUT-1464.patch, MAHOUT-1464.patch, MAHOUT-1464.patch, run-spark-xrsj.sh
>
>
> Create a version of Cooccurrence Analysis (RowSimilarityJob with LLR) that
> runs on Spark. This should be compatible with Mahout Spark DRM DSL so a DRM
> can be used as input.
> Ideally this would extend to cover MAHOUT-1422. This cross-cooccurrence has
> several applications including cross-action recommendations.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
