Quesiton about required file structure

[JFMeyer2k]

Imagine I have a video I would like to work with, and there are 5 labels = {1,2,3,4,5}

The structure I would need to pass to det_dir, should be structured like this:

video_name
-test
--1
---frame_001.txt
---frame_002.txt
---frame_003.txt
--2
---frame_001.txt
---frame_003.txt
--3
---frame_005.txt
---frame_006.txt
---frame_007.txt
--4
--5
-val
--1
--2
--3
--4
--5

In my case, only some labels appear on all images, and the frames can contain any label at any time.
I looked at #5, and the content of one file seems to be structured as:

frame_number,x1,y2,x2,y2,confidence_score

are my assumptions correct?