Github user sethah commented on the issue: https://github.com/apache/spark/pull/20188 cc @srowen @holdenk The MLlib counterparts actually make mention of this, but for some reason the note never got ported over to ML package. The only caveat I can think of is that this doc is the same for all algos that inherit it, but a new algo could potentially not ignore it, but throw an error or manually set the checkpoint dir. For now, ALS, LDA, and GBT all use it and ignore it in the case the checkpoint dir is not set.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org