Github user sethah commented on the issue:
https://github.com/apache/spark/pull/19680
Mima failures :) These APIs were all marked as experimental, which does
give us some freedom to move them, though I know we prefer to avoid it. It's
mainly complaining that we changed these classes from
`org.apache.spark.ml.classification` to `org.apache.spark.ml.summary`. We could
keep the new traits in the `summary` package and keep the actual
implementations in the `classficiation/regression/clustering` packages but that
is less desirable IMO. I'll wait for some feedback from other reviewers.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
