Github user jkbradley commented on the issue:
https://github.com/apache/spark/pull/16722
Hi all, I can try to track this work now.
> This patch maintains the meaning of minInstancesPerNode, in that the
parameter still corresponds to raw, unweighted counts. It also adds a new
parameter minWeightFractionPerNode which requires that nodes must contain at
least minWeightFractionPerNode * weightedNumExamples total weight.
I think minInstancesPerNode should use sample weights, not unweighted
counts. IMHO that matches the semantics of minInstancesPerNode and sample
weights better: a sample with weight 2 is worth 2 samples with weight 1, so
minInstancesPerNode should treat those cases in the same way.
If we do that, then the old BaggedPoint should work.
---
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]
