New Computer Vision model released!

Great to hear! I do worry that existing ID errors caused in part by the unavailability of hybrids in the CV model for recent observations might cause experiments to see hybrid accuracy as underperforming, compared to how it will be after the CV had had hybrids in it for a while. Is there a way to control for that? Maybe doing the experiment using specific hybrids you expect to already have a lower ID error rate/have been cleaned up?

1 Like

Hi @alex. That’s very encouraging. Hopefully you can find an approach that works well for both hybrid plants and hybrid birds!

3 Likes

have hybrids been reintroduced to the CV? Offered as a suggestion at this observation: https://www.inaturalist.org/observations/57190587

3 Likes

Next CV update coming mid-June ?

See https://forum.inaturalist.org/t/cv-suggesting-some-hybrid-taxa-but-hybrid-taxon-pages-do-not-display-computer-vision-model-included-badge/53129/8