GitHub user iyerr3 opened a pull request:
https://github.com/apache/incubator-madlib/pull/142
DT: Include NULL rows in count for termination check
When the primary split feature for a node is computed, the statistics of
rows going to the true and false side don't include the rows that have
NULL value for this split feature. These "NULL" rows can only be
included in the statistics during the next pass when surrogates have
been trained. This commit ensures that in the presence of NULL rows, we
don't terminate prematurely by comparing with a lower count.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/iyerr3/incubator-madlib
bugfix/dt_accurate_termination
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/incubator-madlib/pull/142.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 #142
----
commit 7213d3008d657df323a577111355aae6354ef663
Author: Rahul Iyer <[email protected]>
Date: 2017-06-21T06:31:06Z
DT: Include NULL rows in count for termination check
When the primary split feature for a node is computed, the statistics of
rows going to the true and false side don't include the rows that have
NULL value for this split feature. These "NULL" rows can only be
included in the statistics during the next pass when surrogates have
been trained. This commit ensures that in the presence of NULL rows, we
don't terminate prematurely by comparing with a lower count.
----
---
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.
---
