zhengruifeng commented on issue #28078: [SPARK-31309][ML] Migrate the 
ChiSquareTest from MLlib to ML
URL: https://github.com/apache/spark/pull/28078#issuecomment-610818801
 
 
   @srowen This PR is mainly moving the code without bringing more 
simplification. But it maybe the first step to futher improvements, a possible 
point may be _returning results in rows instead a single row_ (in this PR the 
methods `chiSquaredDenseFeatures` and `chiSquaredSparseFeatures` are added to 
directly return rdd of test results, I thought migration is enough to open a 
PR, so I did not point it out).
   
   For other impls, if I want to make an improvements based on DF or DS, should 
I just use DF or DS in the .mllib side (since .mllib side are almost impled on 
RDD) or move it to .ml at first?
   
   
   > I'm sure there are cases of both unfortunately (? I'd have to check).
   
   Yes, at least GMM was implemented in both side.
   
   

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


With regards,
Apache Git Services

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

Reply via email to