tgravescs commented on issue #25983: [SPARK-29327][MLLIB]Support specifying features via multiple columns URL: https://github.com/apache/spark/pull/25983#issuecomment-545514423 What do you mean by can take n columns instead of 1? The api currently allows for one parameter. yes its being used in practice in XGBoost on GPU. Any of the algorithms could either do the vector assembly internally (thus making it less complex for the user), or if the algorithm knows what to do with the feature columns it can skip the creation of the vector altogether. We had some CPU code for XGboost that did this as well. @firestarman can you provide an example.
---------------------------------------------------------------- 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] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
