vigyasharma commented on PR #14009: URL: https://github.com/apache/lucene/pull/14009#issuecomment-2889130972
> The graph is a bit confusing, but the dots are the oversample (from 1 to 5). This is the piece I was missing! Can we edit the [benchmark comment](https://github.com/apache/lucene/pull/14009#issuecomment-2502665806) and add a line saying each dot represent oversample factor (from 1 to 6, i count 6 dots). So 4x oversample with 1_bit outperforms 2x oversample with 7_bit, in both latency and recall. And 3x oversample with 4_bit has better recall and latency than 7_bit with no oversample. Sorry I'm late to the party. Will look at the PR and benchmark more closely and try to think of latency laggers. -- 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: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
