zhengruifeng commented on pull request #32734: URL: https://github.com/apache/spark/pull/32734#issuecomment-853779208
> So you mean if the number of zeros is more in a vector column, then consider that vector column as sparse, and If all the vector columns are sparse, then proceed with the sparse based implementation. right? This is just an idea. I mean compute average nnz, if nnz_avg < numFeatues * factor(0.5?), the proceduce perfer the code path prefer sparse dataset, otherwise another path. -- 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]
