jtibshirani commented on issue #10665: URL: https://github.com/apache/lucene/issues/10665#issuecomment-1297299587
Thanks @benwtrent, it's great to see that PyLucene works well and has low overhead! It feels more solid than what we were doing before. +1 to preparing a new version. As I mentioned on the PR, we made a big correction to the algorithm in Lucene 9.2 (https://issues.apache.org/jira/browse/LUCENE-10527). It would be too bad if results came out and Lucene did much worse than hnswlib because it was missing this fix! It'd also be great to compare the results against hnswlib as part of the submission. We can double-check that recall is the same for a given set of parameters. This would give confidence we've interpreted all the parameters right in the ann-benchmarks runner. I'm happy to help with this since I have hnswlib set up locally. -- 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
