Github user sachingoel0101 commented on the pull request:
https://github.com/apache/flink/pull/861#issuecomment-133707118
@tillrohrmann, thanks for the brilliant suggestions. Using a `TreeMap` and
`PriorityQueue` with invalidation, I've managed to bring down the complexity of
the `add` and `merge` operations to logarithmic time. Further, `quantile` and
`count` are also linear only, as they should be.
Further, I've decided to put both the Histograms in the `accumulator`
package since they're supposed to work like one anyway. There already was a
*discrete* histogram in the `accumulator` package. The *continuous* one now
resides in the same place.
Also, the `DataSetUtils` class now contains functions to create histograms,
providing access to these classes from the Java api itself instead of the ML
library. That was needed to be done sooner or later. Flink-2274 actually asks
for that.
@thvasilo @chiwanpark
---
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.
---
