[
https://issues.apache.org/jira/browse/SPARK-21028?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ajay Saini updated SPARK-21028:
-------------------------------
Description: Right now the Scala implementation of the one vs. rest
algorithm allows for parallelism but does not allow for the amount of
parallelism to be tuned. Adding a tunable parameter for the number of jobs to
run in parallel at a time would be useful because it would allow the user to
adjust the level of parallelism to be optimal for their task. (was: Adding a
class for a parallel one vs. rest implementation to the ml package in Spark.)
> Parallel One vs. Rest Classifier Scala
> --------------------------------------
>
> Key: SPARK-21028
> URL: https://issues.apache.org/jira/browse/SPARK-21028
> Project: Spark
> Issue Type: New Feature
> Components: ML
> Affects Versions: 2.2.0, 2.2.1
> Reporter: Ajay Saini
>
> Right now the Scala implementation of the one vs. rest algorithm allows for
> parallelism but does not allow for the amount of parallelism to be tuned.
> Adding a tunable parameter for the number of jobs to run in parallel at a
> time would be useful because it would allow the user to adjust the level of
> parallelism to be optimal for their task.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
