liujiayi771 opened a new pull request, #5175:
URL: https://github.com/apache/incubator-gluten/pull/5175

   ## What changes were proposed in this pull request?
   
   The calculation formula of `corr` in Velox is slightly different from that 
in Spark, which causes some differences in the accuracy of the final 
calculation results, but these differences are within the allowed error range. 
Previously, the inconsistent test cases were deleted, but now it has been found 
that other statistics-related agg functions such as `skewness`, `stddev`, and 
`variance` also have differences in calculation accuracy. We cannot delete all 
of these test cases that have differences. A better approach is to modify the 
content in the result file.
   
   In this PR, we restore the `corr` test cases in group-by.sql and 
udf-group-by.sql, and modify the correct results in .out file.
   
   ## How was this patch tested?
   
   Exists CI.
   
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to