Github user imatiach-msft commented on a diff in the pull request:
https://github.com/apache/spark/pull/16441#discussion_r95431048
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/classification/GBTClassifier.scala ---
@@ -159,14 +157,21 @@ class GBTClassifier @Since("1.4.0") (
val numFeatures = oldDataset.first().features.size
val boostingStrategy =
super.getOldBoostingStrategy(categoricalFeatures, OldAlgo.Classification)
- val instr = Instrumentation.create(this, oldDataset)
+ val numClasses: Int = getNumClasses(dataset)
--- End diff --
right, I removed it for now, but ideally the user would preprocess the data
and make the label column categorical. Either they would do that through the
string indexer, or if they know it ahead of time, they would just add the
metadata themselves (although unfortunately currently only advanced users would
be able to do this, there is no transform that will allow they to pre-specify
the labels if they know ahead of time what the labels are)
---
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]
