Regarding model evaluation - boxsuppress function

[Sal1141]

How to construct my own mask, when evaluating the scores here scores(any([-1 0 1 0 ; 0 -1 0 1] * boxes < 0)) = -inf ; in your code you are using boxes with size 6464 -->[-1 0 1 0 ; 0 -1 0 1] and i am using 50100 boxes --> [ ????]

how to reconstruct it ??