Github user mgaido91 commented on the issue: https://github.com/apache/spark/pull/22763 Actually I don't see at all the need to revert this. Even without deprecating `computeCost`, using it for evaluating a model is a very poor approach. We should anyway advice the users to a better approach which is available, ie. using the `ClusteringEvaluator` to evaluate their models.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org