[
https://issues.apache.org/jira/browse/MATH-978?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13663248#comment-13663248
]
Phil Steitz commented on MATH-978:
----------------------------------
Thanks!
Can you please
a) either explain or point to some docs explaining why the aggregating formula
above is correct
b) add some unit tests, including a test to show consistency - i.e. that
aggregation yields close to the same result as direct computation.
It will make it a little easier / faster to commit if you provide the new code
and tests as patches attached to the ticket.
Thanks again for your contribution.
> StorelessCovariance to be map/reducable
> ---------------------------------------
>
> Key: MATH-978
> URL: https://issues.apache.org/jira/browse/MATH-978
> Project: Commons Math
> Issue Type: New Feature
> Reporter: Ajo Fod
> Original Estimate: 2h
> Remaining Estimate: 2h
>
> Current version of StorelessCovariance and StorelessBivariateCovariance don't
> allow adding another StorelessCovariance to an existing one. With this
> feature it is possible to run maps on different data sets and then do a final
> reduce to get the overall covariance.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
