choice of metrics used to measure detection models performances on benchmarks

[nikokks]

hello =)

I would like to know why you use precision and recall to measure the performance of your detection models?

Indeed, there are metrics that more appropriately measure the correct detection of texts: for example MaP (Mean Average Precision).

thanks for your work !