weecology
Update recommended citation
We are currently recommending that Weinstein et al 2019 be cited instead of the actual DeepForest paper.
We talked about this in the past, the backbone is from that first paper, but the API is from the MEE paper. We opted to do the earlier one. Want to switch? I'm completely open to either.
Ah, got it. In that case I think it probably makes sense to list both and just be explicit about what should be cited in what circumstances, which in most cases will be both. E.g., if you totally ignored the backbone and trained your own model from scratch - then just the software paper. If you just used the model but not the software - just the backbone paper. Otherwise both
Sounds good. I think this also gets us towards a place where we need to cite models produced by others.
On Wed, Nov 8, 2023 at 6:02 PM Ethan White @.***> wrote:
Ah, got it. In that case I think it probably makes sense to list both and just be explicit about what should be cited in what circumstances, which in most cases will be both. E.g., if you totally ignored the backbone and trained your own model from scratch - then just the software paper. If you just used the model but not the software - just the backbone paper. Otherwise both
[image: tony-stark-too-much-to-ask-for-both] https://user-images.githubusercontent.com/744427/281590556-a3d7a27c-e55a-409a-8825-57f7225f27e4.gif
— Reply to this email directly, view it on GitHub https://github.com/weecology/DeepForest/issues/539#issuecomment-1803046772, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAJHBLBJXRZUBCGASNTYO6LYDQ2RVAVCNFSM6AAAAAA67PAC7GVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMBTGA2DMNZXGI . You are receiving this because you commented.Message ID: @.***>
-- Ben Weinstein, Ph.D. Research Scientist University of Florida