Closed gdalle closed 11 months ago
Thank you, Guillaume. Very well spotted. My supervisor and I both missed these.
Table 1: why 2 columns for XGBoost and SIRUS?
Because I thought it was clearer but I was wrong I think. I've now removed it in https://github.com/rikhuijzer/SIRUS.jl/pull/36/commits/9017cf58ec621b7309a41b43870733cb51a54f25.
L31: "perform poor" => perform poorly L80: "will most likely survive" => is more likely to survive (the proba from the first rule is not > 0.5) L136: repetition of "models"
Table 1: why 2 columns for XGBoost and SIRUS?
https://github.com/openjournals/joss-reviews/issues/5786