Github user NarineK commented on a diff in the pull request:
https://github.com/apache/spark/pull/12836#discussion_r62759544
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/r/MapPartitionsRWrapper.scala
---
@@ -25,6 +25,21 @@ import org.apache.spark.sql.Row
import org.apache.spark.sql.types.{BinaryType, StructField, StructType}
/**
+ * A function wrapper that applies the given R function to each partition
of each group.
+ */
+private[sql] case class MapGroupsRWrapper(
--- End diff --
Thank you for the explanation, @sun-rui !
As I understand from previous notes:
We want to pass multiple groups to a R worker and call the R function on
each group separately, right ?
I'm trying to understand what is the major difference between this approach
and repartition .
Because conceptually repartition was also passing multiple groups to a R
worker (guaranteed that each group will be in one partition, but there can be
multiple groups in a partition).
I'm fine with this optimization, just trying to compare.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
