Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/2595#issuecomment-58460850
@chouqin @mengxr @manishamde I updated the timing results above. Sorry
it took this long, but I kept running into various issues. (The problems with
the original results were from mixed-up row orders + additional running time
from missing Hadoop logging configuration files.)
The trends I see are:
* more features => better speedups from dist-agg
* more trees => fewer features => worse speedups
Conclusion: For all of the larger problems (with more features), dist-agg
helps.
---
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]
