Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/2607#issuecomment-61000703
It's a good point about the sequential nature of boosting models being
important when doing approximate predictions (using only some of the weak
hypotheses); I could imagine that being useful. Perhaps the generic
WeightedEnsembleModel could be subclassed in order to support that kind of
extended functionality in the future.
Distributed models sound useful to me, though I suspect applying a
sparsifying step (like running Lasso on the outputs of the many trees to choose
a subset of trees) might be faster and almost as accurate in many cases.
---
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]
