[ 
https://issues.apache.org/jira/browse/MADLIB-1166?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16273610#comment-16273610
 ] 

ASF GitHub Bot commented on MADLIB-1166:
----------------------------------------

GitHub user Swatisoni opened a pull request:

    https://github.com/apache/madlib/pull/206

    Feature: Allow NULL in rows for computing correlations and covariance

    JIRA: MADLIB-1166
    
    Additional Authors:
    Orhan Kislal <[email protected]>
    
    - Replace NULL values with the mean of corresponding column.

You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/Swatisoni/madlib feature/corr_null3

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/madlib/pull/206.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 #206
    
----
commit 5de8a1c9c877401552cde5b46e666af724867484
Author: Swatisoni <[email protected]>
Date:   2017-11-30T22:52:09Z

    Feature: Allow NULL in rows for computing correlations and covariance
    JIRA: MADLIB-1166
    
    Additional Authors:
    Orhan Kislal <[email protected]>
    
    - Replace NULL values with the mean of corresponding column.

----


> Correlation - do not drop whole row if a target column is NULL
> --------------------------------------------------------------
>
>                 Key: MADLIB-1166
>                 URL: https://issues.apache.org/jira/browse/MADLIB-1166
>             Project: Apache MADlib
>          Issue Type: Improvement
>          Components: Module: Descriptive Statistics
>            Reporter: Frank McQuillan
>            Priority: Minor
>             Fix For: v1.13
>
>
> http://madlib.apache.org/docs/latest/group__grp__correlation.html
> Current implementation ignores a row that contains NULL entirely. This means 
> any correlation in such a row (with NULLs) does not contribute to the final 
> answer.
> This improvement is not to drop the whole row if a target column is NULL, but 
> rather to do pairwise correlation for any target columns that exist, to make 
> most use of the data.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Reply via email to