Github user srowen commented on a diff in the pull request:
https://github.com/apache/spark/pull/14923#discussion_r77357355
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/stat/MultivariateOnlineSummarizer.scala
---
@@ -231,9 +231,9 @@ class MultivariateOnlineSummarizer extends
MultivariateStatisticalSummary with S
*/
@Since("1.1.0")
override def numNonzeros: Vector = {
- require(totalWeightSum > 0, s"Nothing has been added to this
summarizer.")
+ require(totalCnt > 0, s"Nothing has been added to this summarizer.")
- Vectors.dense(weightSum)
+ Vectors.dense(nnz.map(_.asInstanceOf[Double]))
--- End diff --
The former is a cast, which is 'wrong' because a Long is not a Double, but
maybe works out with implicits. I assume that's slower. Really you mean "take
this long and convert it to a double" so the latter is better.
---
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
