Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/12066#discussion_r61474448
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/classification/Classifier.scala ---
@@ -116,22 +165,7 @@ abstract class ClassificationModel[FeaturesType, M <:
ClassificationModel[Featur
}
predictUDF(col(getFeaturesCol))
}
- // The label column for base binary classifier of OneVsRest will not
be retained during
- // model transformation, so we should not handle label column
metadata as well.
- if (dataset.schema.fieldNames.contains($(labelCol))) {
- // determine number of classes either from metadata if provided.
- val labelSchema = dataset.schema($(labelCol))
- // extract label metadata from label column if present, or create
a nominal attribute
- // to output the number of labels
- val labelAttribute = Attribute.fromStructField(labelSchema) match {
- case _: NumericAttribute | UnresolvedAttribute =>
-
NominalAttribute.defaultAttr.withName("label").withNumValues(numClasses)
- case attr: Attribute => attr
- }
- outputData = outputData.withColumn(getPredictionCol, predUDF,
labelAttribute.toMetadata)
- } else {
- outputData = outputData.withColumn(getPredictionCol, predUDF)
- }
+ outputData = outputData.withColumn(getPredictionCol, predUDF,
predictionMetadata)
--- End diff --
Set ```predictionMetadata.withName(getPredictionCol)``` here in case user
updated predictionCol name.
---
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]
