WeichenXu123 commented on pull request #30471:
URL: https://github.com/apache/spark/pull/30471#issuecomment-738013420


   @srowen 
   
   > Would this have any effect for code that already extends these classes? I 
like the cleanup and refactoring, just looking for anything this would break.
   
   I think won't break.
   
   * For interface, it only change CrossValidator/TrainValidateSplit/OneVsRest 
to inherit MLReadable/MLWritable instead of inherit 
JavaMLReadable/JavaMLWritable . This won't break anything except user run code 
like `isinstance(crossvalidator, JavaMLReadable)`, which is meaningless.
   
   * For save/load backward compatibility, also won't break anything. Any model 
which can be saved successfully by old code, in new code it will also be saved 
exactly the same, it will go _to_java().save() route. For python backend 
estimator/evaluator cases, old code will directly raise error but new code it 
will go python writer/reader route to save/load.


----------------------------------------------------------------
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]



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

Reply via email to