Ayush Chaurasia
Ayush Chaurasia
No currently you can only finetune official yolov5 models for detection and segmentation. For classification you can finetune on yolov5-cls models or any models available on torch hub
@thepycoder might be able to help you with this
@thepycoder Just seeing this! I missed the previous ping. Glad that its working as intended now
Yes, it is. After exporting the model to torchscript, you can use C++ to infer the segmentation in C++. You can find more info on using torchscript models with c++...
@SkalskiP can you try again? it should not return None. It should return a generator
@SkalskiP @DCCCXIX Yeah there have been temp solutions to fix memory overloading issues. We are working on a better stable solution to provide an interface to outputs. You can follow...
@LaserRider what's the version that you're using. You might want to update to the latest version by running `pip install --upgrade ultralytics`
@glenn-jocher @Laughing-q are you guys aware of this issue?
@Zengyf-CVer hey it could be because of this line https://github.com/ultralytics/ultralytics/blob/main/ultralytics/yolo/engine/results.py#L260-L260 as the segments are calculated in reverse order. Can you please confirm this removing this fixes your issue and I'll...
@Zengyf-CVer okay we'll fix it