Github user srowen commented on the issue: https://github.com/apache/spark/pull/16038 Yes, but the result is inherently dense -- not going to be many zeroes in the gradient if any. There's no way around that. The discussion has been about the initial value, the zero vector, right? That you agree is sparse. I don't see what the problem is then. A huge sparse 0 vector is _not_ huge when serialized.
--- 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