zhengruifeng commented on pull request #29018: URL: https://github.com/apache/spark/pull/29018#issuecomment-658066066
@viirya Thanks for reviewing! > This win only happens when maxBins is less Yes, but in most cases, maxBin(default=32) < 128 > the perf regression happens for all cases Yes, I think so. > I'm also not sure how often memory is an issue when training the model It will make sense if there is no enough memory for orginal treePoint(Array[Int]). I personally think it maybe worthwhile if the regression is small enough, but I am not sure whether current performance results are OK. ---------------------------------------------------------------- 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]
