Github user WeichenXu123 commented on the issue:
https://github.com/apache/spark/pull/19988
I think we can discuss the following cases:
- When gradient non-zero, line-search failed, will the model always be
meaning-less ?
- When gradient nearly zero, and line-search failed. I think this is also
possible. This should get valid model.
- Sometimes, when computation crashes (e.g. overflow or some other reason)
cause the linesearch failed, this will surely get meaning-less model.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
