Github user thvasilo commented on the pull request:
https://github.com/apache/flink/pull/861#issuecomment-114896540
>The idea is to emulate a discrete probability distribution, which it does.
For the continuous case, we're emulating a continuous probability distribution,
which means we can effectively add any value we want.
OK so we don't provide a online discrete histogram in the general sense,
but rather one that specifically aimed at modeling a discrete probability
distribution. That does kind of limit the applicability of the code to the
specific use case of the decision tree, doesn't it? Do you think it's possible
to generalize the discrete histogram code without breaking the decision tree
functionality?
---
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.
---
