LuciferYang commented on pull request #29857: URL: https://github.com/apache/spark/pull/29857#issuecomment-698819722
Synchronize the test result: https://github.com/apache/spark/blob/0bc0e91e4015eb98bd2f4bf17da2ec7135b520a9/mllib/src/main/scala/org/apache/spark/ml/tree/impl/RandomForest.scala#L194-L202 - **Test A：** Change to use `LinkedHashMap` to store topNodesForGroup and use `nodesForGroup.keys.toSeq.sorted` to init it was no essential impact. - **Test B：** Change to use `LinkedHashMap` to store `topNodesForGroup` and `nodesForGroup` seems that the goal of unified behavior in Scala 2.12 and 2.13 can be achieved, but need change the tol in the last test to 0.77(the result is 0.771) - **Test C：** Only change the tol in the last test to 0.75 can workaround Which do you recommend better @srowen , B or C? ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
