Burhan
Burhan
https://github.com/ultralytics/assets/pull/65 has been merged.
I'm going to check this out and build the docs locally. I'll push any changes that are needed and then added the TODO label when I think we're ready. Probably...
Badge and icon looks good, thank you @Y-T-G! I updated a sentence to flow a little better and using ChatGPT + Google Translate, included the changes in the Chinese Readme...
Related to #13564
Here's a sample from VS Code of the type hinting and docstring preview 
All the tests failing 😆 I know this will go no where without being able to pass the CI tests, but like I mentioned, it's more of a conceptual starting...
@glenn-jocher as far as I'm aware, the type hinting is all done statically, so we can't use the YAML values. This was something I thought about and tried, but wasn't...
@Laughing-q my quick test `(coco128, epochs=5)` didn't seem to help at all, but I suspect that it could take longer for the benefits to come thru. Additionally there might be...
Tested with COCO validation ```sh yolo val model=yolov8s.pt data=coco.yaml ``` The mAP50-95 for Soft-NMS (all classes) was ~0.10 compared to normal NMS which was ~0.44. Made an attempt to reorder...
Local testing shows that failure occurs when using `model(SOURCE)` but not `model.predict(imgsz=32, save_txt=True, save_crop=True, augment=True)` which is strange. UPDATE: works on **first** predict call, no matter which method is used.