imatiach-msft commented on issue #23818: [SPARK-19591][ML][PYSPARK][FOLLOWUP] 
Add sample weights to decision trees
URL: https://github.com/apache/spark/pull/23818#issuecomment-464973632
 
 
   GBTClassifier and GBTRegressor don't support weight column yet.  There is a 
separate JIRA for random forest:
   https://issues.apache.org/jira/browse/SPARK-9478
   and another one for GBTs:
   https://issues.apache.org/jira/browse/SPARK-9612
   In GBTClassifier/Reg we currently don't pass weights:
   
https://github.com/apache/spark/blob/master/mllib/src/main/scala/org/apache/spark/ml/classification/GBTClassifier.scala#L204
   Unfortunately GBTClassifier/Reg uses the same params as 
DecisionTreeClassifier/Reg, so this is kind of a bug now - since we expose this 
param but it doesn't actually do anything.  Thanks for finding this!
   We should try to fix this soon at the scala layer but maybe in python for 
now we shouldn't expose it?

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to