Github user MLnick commented on a diff in the pull request:
https://github.com/apache/spark/pull/20164#discussion_r161535696
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/classification/OneVsRest.scala ---
@@ -170,21 +170,24 @@ final class OneVsRestModel private[ml] (
newDataset.persist(StorageLevel.MEMORY_AND_DISK)
}
+ // temporary column to store intermediate raw prediction
+ val tmpRawPredictionColName = "rawPrediction_" +
UUID.randomUUID().toString
+
// update the accumulator column with the result of prediction of
models
val aggregatedDataset =
models.zipWithIndex.foldLeft[DataFrame](newDataset) {
case (df, (model, index)) =>
- val rawPredictionCol = model.getRawPredictionCol
- val columns = origCols ++ List(col(rawPredictionCol),
col(accColName))
+ val columns = origCols ++ List(col(tmpRawPredictionColName),
col(accColName))
--- End diff --
This line doesn't need to be in the `foldLeft` block any longer?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
