GitHub user srowen opened a pull request:
https://github.com/apache/spark/pull/16129
[SPARK-18678][ML] Skewed feature subsampling in Random forest
## What changes were proposed in this pull request?
Fix reservoir sampling bias for small k. An off-by-one error meant that the
probability of replacement was slightly too high -- k/(l-1) after l element
instead of k/l, which matters for small k.
## How was this patch tested?
Existing test plus new test case.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/srowen/spark SPARK-18678
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/16129.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #16129
----
commit 8ac5dee8f9c0165da7a16d83d79f2f5080edb3ec
Author: Sean Owen <so...@cloudera.com>
Date: 2016-12-03T09:32:00Z
Fix reservoir sampling bias for small k
----
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
