Github user yanboliang commented on a diff in the pull request: https://github.com/apache/spark/pull/16646#discussion_r97016294 --- Diff: mllib/src/main/scala/org/apache/spark/ml/r/GaussianMixtureWrapper.scala --- @@ -124,7 +129,8 @@ private[r] object GaussianMixtureWrapper extends MLReadable[GaussianMixtureWrapp val rMetadataStr = sc.textFile(rMetadataPath, 1).first() val rMetadata = parse(rMetadataStr) val dim = (rMetadata \ "dim").extract[Int] - new GaussianMixtureWrapper(pipeline, dim, isLoaded = true) + val logLikelihood = (rMetadata \ "logLikelihood").extract[Double] + new GaussianMixtureWrapper(pipeline, dim, logLikelihood, isLoaded = true) --- End diff -- Yeah, it will break existing persisted model, but I think we don't guarantee mode persistent compatibility between different versions for SparkR. We are planing to make model persistence consistent between SparkR and MLlib, then there is no specific handling for SparkR and will let MLlib to handle all model persistent issue. However if we want to make model persistent compatibility for SparkR currently, I can add code to handle different versions here but will lead maintenance more complicated. What's your opinions?
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org