mob-ai edited a comment on issue #25909: [SPARK-29224]Implement Factorization Machines as a ml-pipeline component URL: https://github.com/apache/spark/pull/25909#issuecomment-540466862 > @mob-ai Thanks for this work! > But before you continue, I guess you can refer to previous dicsussion [SPARK-7008](https://issues.apache.org/jira/browse/SPARK-7008). I think you should provide some information like convergence curves on common datasets, and proof that mini-batch SGD is a good choice as an efficient solver. > As to the PR itself, besides owen's comments, I think there should be `FMClassifier` & `FMRegressor` respectively. Thank you for your suggestion. I will test FM on public datasets. Actually, this PR also implements AdamW solver to optimize FM. AdamW is more efficient than normal SGD, so default solver is AdamW. I will compare normal SGD and AdamW on public datasets.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
