You mentioned a future update with classification model support in your Readme and I was wondering if we could have some information about this update (estimated release date, difficulties encountered, etc.).
I'm very interested because I'm trying to implement a Yolov8-cls in C/C++ for a project myself, with no success so far.
I've tried using and adapting Yolov8-TensorRT-CPP and this repo but can seem to get a correct inference result (it's either not a number in the first case or the resulting class is always 65535 in the second one).
Perhaps we could work together to develop this new feature?
I'm also looking forward to yolo classification, especially as secondary inference engines. In the meantime I will be trying to adapt it from the NVIDIA YOLO repo.
Hi Marcos,
You mentioned a future update with classification model support in your Readme and I was wondering if we could have some information about this update (estimated release date, difficulties encountered, etc.).
I'm very interested because I'm trying to implement a Yolov8-cls in C/C++ for a project myself, with no success so far.
I've tried using and adapting Yolov8-TensorRT-CPP and this repo but can seem to get a correct inference result (it's either not a number in the first case or the resulting class is always 65535 in the second one).
Perhaps we could work together to develop this new feature?
Kind regards,
Ludmilla