GitHub user srowen opened a pull request:
https://github.com/apache/spark/pull/12299
[SPARK-14533] [MLLIB] RowMatrix.computeCovariance inaccurate when values
are very large
## What changes were proposed in this pull request?
For dense rows, compute covariance more accurately by centering in
computeCovariance (sparse case remains untouched). Also compute column means
more accurately with computeColumnSummaryStatistics
CC @mengxr -- am I missing anything here? seems OK for the dense case.
I'd like to back-port to 1.6 or even 1.5 if this looks OK.
## How was this patch tested?
Jenkins test, including new test mentioned in the JIRA.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/srowen/spark SPARK-14533
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/12299.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 #12299
----
commit 86313b55f8c9985e6257b3686f3f01f4992b02f3
Author: Sean Owen <[email protected]>
Date: 2016-04-11T12:36:21Z
For dense rows, compute covariance more accurately by centering in
computeCovariance (sparse case remains untouched). Also compute column means
more accurately with computeColumnSummaryStatistics
----
---
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]
