vigyasharma commented on PR #15621: URL: https://github.com/apache/lucene/pull/15621#issuecomment-3809126263
That's okay, I can make the changes if needed. :) I was thinking about this, and while cosine gets mathematically broken with 0 vectors, I'm not sure if zero vectors are meaningful for other functions either. They'll always return 0 values for dot product and Max inner product, and search won't really be able to differentiate based on those similarity scores. It will silently affect scores, graph geometry and result set, which feels trappy? Are there meaningful scenarios where we want to allow zero vectors? I'm not sure if bit vectors need them? Are hamming distance or jaccard sim meaningful when all bits are 0? -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
