Github user mengxr commented on the pull request: https://github.com/apache/spark/pull/3637#issuecomment-72405143 About the metadata, I'm thinking of creating ML Attribute/VectorAttribute classes that stores feature information, which can be load from/saved to Spark SQL's metadata. It is similar to Weka's Attribute implementation. Since `RDD[LabeledPoint]` doesn't carry this extra information, could we make ML attributes as an input argument to the `train` method? For example ~~~ def train(dataset: RDD[LabeledPoint], attributes: (Attribute, VectorAttribute)) ~~~
--- 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