-
Notifications
You must be signed in to change notification settings - Fork 244
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add FCOS #265
Comments
Thanks for the issue @jdsgomes , would you like to submit a PR? |
FCOS uses the C++ ops of TorchVision like nms. I'm not sure we can add it on TorchHub without creating fallback ops in standard PyTorch. We know that such fallback ops are likely to be slower than the C++/CUDA equivalents but because these were implemented several years ago, we don't know how much slower they will be. It might be worth benchmarking this on the future and assess if fallback ops is a viable strategy to expose our models via TorchHub. I think for now, we are probably going to close the ticket as nofix? |
Thanks for raising those important points @datumbox. Closing the ticket. |
FCOS model has been recently added to TorchVision, therefore should also be added here
pytorch/vision#4961
The text was updated successfully, but these errors were encountered: