Github user acompa commented on the issue:
https://github.com/apache/spark/pull/16618
I'm also curious about this @MLnick. Seems like there was a lot of movement
earlier this year, but this PR has gotten stale.
I can also contribute if @Kornel cannot for any reason.
---
Github user Kornel commented on the issue:
https://github.com/apache/spark/pull/16618
@MLnick I'm wondering what's the status of this issue: seems closed, have
you any plans on picking it up again?
I might pick it up, but I'm not sure what's left: move from package mllib
to
Github user MLnick commented on the issue:
https://github.com/apache/spark/pull/16618
@daniloascione any update?
---
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
Github user MLnick commented on the issue:
https://github.com/apache/spark/pull/16618
@daniloascione are you able to update this? I'd like to target for `2.3`.
But, can we do the following:
1. Port over ranking metrics (udfs) to `ml` with the `RankingEvaluator` as
Github user MLnick commented on the issue:
https://github.com/apache/spark/pull/16618
The basic direction looks right - I won't have time to review immediately.
Spark 2.2 QA code freeze will happen shortly so this will wait until 2.3 dev
cycle starts
---
If your project is set up
Github user daniloascione commented on the issue:
https://github.com/apache/spark/pull/16618
I rewrote the ranking metrics from the mllib package as UDFs (as suggested
