Github user facaiy commented on the issue: https://github.com/apache/spark/pull/18288 In my opinion, `numFeatures` is vital for sparse data. Say our feature is 100-dim indeed, while in a small train data their maximum size is 990. It is dangerous (or wrong) to train a 990-dim model as it might fail in the coming test data. Hence, in most cases, `numFeatures` should be given by user.
--- 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