Closed jpcsmith closed 2 years ago
Thanks! hopefully it was pretty easy. Did you optimize for recall
or r-precision
when training models?
It depended on the model, as they were utilised as presented in the prior work.
On Thu, 31 Mar 2022 at 15:57, Jordan Holland @.***> wrote:
Thanks! hopefully it was pretty easy. Did you optimize for recall or r-precision when training models?
— Reply to this email directly, view it on GitHub https://github.com/nprint/benchmarks/pull/3#issuecomment-1084624677, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAM4YXAKLXEREA7YFJDODW3VCWVMLANCNFSM5SEMECMQ . You are receiving this because you authored the thread.Message ID: @.***>
2 leaderboards, each containing the models that were optimized for the metric associated with the leaderboard.
When you have a moment, let's have a brief call.
Updated with suggested changes.
Hey! I incorporated the feedback from #2 except for point 1. Do you have a suggestion for that point?