TypeError: must be real number, not NoneType

[stu0440]

Annotations /root/autodl-tmp/datasets/coco/annotations/instancesonly_filtered_train.json: 4%|███▏ | 714/18732 [00:02<01:09, 257.80it/s]
Traceback (most recent call last):
File "general_json2yolo.py", line 396, in
convert_coco_json(
File "general_json2yolo.py", line 315, in convert_coco_json
file.write(("%g " * len(line)).rstrip() % line + "\n")
TypeError: must be real number, not NoneType

[UltralyticsAssistant]

👋 Hello @stu0440, thank you for submitting a ultralytics/JSON2YOLO 🚀 issue!

To help us address your concern efficiently, please provide the following information:

1. For bug reports:
   • A clear and concise description of the bug 🐛
   • A minimum reproducible example (MRE) that demonstrates the issue
   • Your environment details (OS, Python version, package versions)
   • Expected behavior vs. actual behavior
   • Any error messages or logs related to the issue – you've already included some of this, thank you!

Make sure to check if your issue has already been reported in existing issues to avoid duplicates.

🏷️ If you need to add more information, please comment on this issue. An Ultralytics engineer will assist you soon.

Thank you for contributing to the improvement of our project!

[huizhenlai]

I meet the same problem. I guess this is because the type of the dataset to be converted is inconsistent with COCO's native type.
I solved this problem by adding the argument "cls91to88 = False" to the function "convert_coco"

[pderrenger]

@huizhenlai thank you for sharing your solution. If the issue persists, please ensure you're using the latest version of the package. If you have further questions, feel free to ask.