[GitHub] [spark] zhengruifeng commented on a diff in pull request #37918: [SPARK-40476][ML][SQL] Reduce the shuffle size of ALS
zhengruifeng commented on code in PR #37918:
URL: https://github.com/apache/spark/pull/37918#discussion_r977466819
##
sql/core/src/main/scala/org/apache/spark/sql/functions.scala:
##
@@ -367,6 +367,9 @@ object functions {
*/
def collect_set(columnName: String): Column = collect_set(Column(columnName))
+ private[spark] def collect_top_k(e: Column, num: Int, reverse: Boolean):
Column =
Review Comment:
I don't know, I also think it's useful and may further use it in
Pandas-API-on-Spark.
But I don't know whether it is suitable to be public @cloud-fan @HyukjinKwon
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
[GitHub] [spark] zhengruifeng commented on a diff in pull request #37918: [SPARK-40476][ML][SQL] Reduce the shuffle size of ALS
zhengruifeng commented on code in PR #37918:
URL: https://github.com/apache/spark/pull/37918#discussion_r974838677
##
mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala:
##
@@ -496,18 +499,23 @@ class ALSModel private[ml] (
.iterator.map { j => (srcId, dstIds(j), scores(j)) }
}
}
- }
-// We'll force the IDs to be Int. Unfortunately this converts IDs to Int
in the output.
-val topKAggregator = new TopByKeyAggregator[Int, Int, Float](num,
Ordering.by(_._2))
-val recs = ratings.as[(Int, Int,
Float)].groupByKey(_._1).agg(topKAggregator.toColumn)
- .toDF("id", "recommendations")
+ }.toDF(srcOutputColumn, dstOutputColumn, ratingColumn)
+
+val aggFunc = CollectOrdered(struct(ratingColumn, dstOutputColumn).expr,
num, true)
+ .toAggregateExpression(false)
Review Comment:
sure, I think we don't want to expose it, so mark it `private[spark]`
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
