Github user srowen commented on a diff in the pull request:
https://github.com/apache/spark/pull/14335#discussion_r72011821
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/clustering/LDAOptimizer.scala ---
@@ -472,12 +473,13 @@ final class OnlineLDAOptimizer extends LDAOptimizer {
gammaPart = gammad :: gammaPart
}
Iterator((stat, gammaPart))
- }
+ }.persist(StorageLevel.MEMORY_AND_DISK)
val statsSum: BDM[Double] =
stats.map(_._1).treeAggregate(BDM.zeros[Double](k, vocabSize))(
_ += _, _ += _)
- expElogbetaBc.unpersist()
val gammat: BDM[Double] = breeze.linalg.DenseMatrix.vertcat(
stats.map(_._2).flatMap(list =>
list).collect().map(_.toDenseMatrix): _*)
--- End diff --
Yes, I'm wondering why not do this in a distributed way; am I missing why
it's only done serially on the driver? is it that the dense matrix class
doesn't serialize or serialize well?
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
