No Output from MOT Evaluation about bdd100k

[supperted825]

I've tested the code with the original BDD annotations, and it works. So, the problem seems to be with my result files.

I'm following the following format from the bdd100k docs, but I'm getting the following errors. Any help is appreciated!

pydantic.error_wrappers.ValidationError: 10 validation errors for Frame
labels -> 0 -> box2d
value is not a valid dict (type=type_error.dict)
labels -> 1 -> box2d
value is not a valid dict (type=type_error.dict)

My result JSON format, according to BDD100k documentation:

- videoName: str, name of the current sequence,
- name: str, name of the current frame,
- framIndex: int, index of the current frame within the sequence,
- labels []:
    - id: str, unique instance id of the prediction in the current sequence,
    - category: str, name of the predicted category,
    - box2d []:
        - x1: float,
        - y1: float,
        - x2: float,
        - y2: float

[supperted825]

I've figured out the problem. In the result template, box2d should be a dictionary instead of a list. In addition, there has been a typo on the documentations. framIndex should be frameIndex!

With this, it works as expected now!