jtibshirani commented on PR #796: URL: https://github.com/apache/lucene/pull/796#issuecomment-1118021852
Super interesting, looking forward to hearing more! I do hope we can stick with a prefiltering-like approach (and just improve its performance), since it feels easier to work with for users. If you request `k` documents, you always get `k` back -- there's no guessing about how many candidates you need to retrieve as in post-filtering. This doesn't sound like what you're talking about, but I did notice that prefiltering can be expensive when the filter matches a lot of documents. Unlike in postfiltering, the filter is not allowed skip any of the docs and you need to convert all the matches into a bit set, which is not cheap. -- 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
