mikemccand commented on PR #15564: URL: https://github.com/apache/lucene/pull/15564#issuecomment-3745561885
Whoa, thanks @benwtrent and Claude or Gemini or Grok or ... something. I've found pair programming with a genai to be a weird inversion: rather than spending my human time writing code, it's the opposite. I spend gobbs of human time reviewing what the genai wrote, iterating/reprompting, then re-reviewing completely different code (since it seems to rewrite stuff all the time -- that temptation we human programmers have and try to suppress). So this (reading/reviewing, not writing) is a different muscle, and not as strong (I think?) for me, so I'm left with a less complete and more shaky understanding of what the two of us actually produced. It's such a strange thing, to produce tons of code by only reviewing it (and wee bit of prompting magic). > Lucene Util benchmarks (no rescoring) These tables are so hard to read ... I think `knnPerfTest.py` attempts to write out a graphical chart (ROC curves) of the runs (@msokolov added at some point!), maybe by default? Maybe look at that and see if it's possibly usable / visually communicative? -- 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]
