nealrichardson commented on pull request #9615: URL: https://github.com/apache/arrow/pull/9615#issuecomment-833924620
I'm a little skeptical, with the exception of the big change on the data.frames of factor columns, that this isn't just noise. I don't think there's been any other changes in the data.frame to Arrow code between latest master and where this branch is based. For the sake of argument, let's assume that "a little better or a little worse" is really just no change. I'm more surprised that there seems only to be that one improvement. The fannie mae dataset has 31 columns: with 8 cores, why is essentially the same performance as before/with 1 core? -- 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]
