[
https://issues.apache.org/jira/browse/SPARK-7014?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14503476#comment-14503476
]
Ram Sriharsha commented on SPARK-7014:
--------------------------------------
good point, the priority is minor.
the plan is to have it as a separate library code like spark-afro for example.
it would be good to know who else is using VW formatted data for example.,
maybe there isn't enough usage to warrant a parser.
> Support loading VW formatted data.
> ----------------------------------
>
> Key: SPARK-7014
> URL: https://issues.apache.org/jira/browse/SPARK-7014
> Project: Spark
> Issue Type: Improvement
> Components: MLlib
> Reporter: Ram Sriharsha
> Assignee: Ram Sriharsha
> Priority: Minor
> Original Estimate: 96h
> Remaining Estimate: 96h
>
> Support loading data in VW format. VW format is used fairly widely and has
> support for namespaces, importance weighting and multi label and cost
> sensitive multi class classification formats.
> We can support this just as we support avro and csv formats today .
> It probably belongs to a new package say spark-vw but this JIRA is simply to
> track and discuss the issue of supporting VW format.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
