youngjae-you

@tensorturtle Oh, the supplementary explanation is missing, the first result is forced to `half=false` in the code, and the second result is the result of turning the original code

> Faiss does not L2 normalize either the query, not the database vectors. Have you done this normalization on the vectors yourself before adding them to the index and querying?...

It's not experiencing the issue with **_supervision 0.19.0_**, so which one is the bug?

> Hi, @YoungjaeDev 👋🏻 do you have a tracker plugged into your pipeline? Yes, of course I'm using it. It's the same code, it works for 0.19.0, but not 0.20.0.

@LinasKo @SkalskiP `supervision==0.21.0rc5` seems to smooth out the behavior What was the cause?

@SkalskiP Just category_id. Image_id had no problem for importing data

@SkalskiP cvat.apps.dataset_manager.bindings.CvatImportError: Image C00_012_0007_000000: can't import annotation #0 (bbox): annotation has no label. This is error log

@SkalskiP No, the `image_id` starts with 1, which is a CVAT requirement, and Supervision does that. However, CVAT requires the **_category_id_** to start with 1, and Supervision starts with 0.

Yes, thank you

@ryouchinsa I analyzed the `json2yolo` conversion code and am curious about the principle. Why do ti divide by k=0, k=1? In the end, the first and last points of each...