Github user wzhfy commented on the issue:
https://github.com/apache/spark/pull/15637
With (value, freq) pairs, i.e. equi-width histogram, given a predicate
col<100 and col>50, we can know exactly how many records are in that range,
which is the estimated rowcount after the filter. And we can also know the
exact ndv after the filter, which is important for following estimation, e.g.,
in estimation of output row count of agg(group by).
---
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]
