Github user haying commented on the issue:
https://github.com/apache/incubator-madlib/pull/75
I am not so sure about the technical details about convergence behavior of
this algorithm. But it seems to me that it could be beneficial to expose the
gradient so that users can do convergence tests and stepsize/batchsize tuning
more easily. Of course, it would be even better if we can do more like
providing a warning if we observe the model may not converge, etc.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
