Github user srowen commented on the issue:
https://github.com/apache/spark/pull/17306
Have a look at http://spark.apache.org/contributing.html first
So the idea is to search over pipelines with different components,
potentially, not just one set of components but varying its parameters?
What you have here amounts to running n pipeline evaluations with n grid
searches. That could just be done with the existing machinery, run n times. In
this new model it seems hard to work out which pipeline was selected? except by
inspecting it.
It makes some sense but the alternative isn't much code either, to just
combine the results of n grid searches.
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
