I was checking my dataset to see something totally unrelated to this and I stumbled upon this: part of the annotations made by me where wrong and had None category as it's Bounding Box class.
It surprised me because I don't know how could it work, and I would like to know if anyone knows the behaviour of darknet upon this. It just ignores it or can it impact negatively to the training of the net? Is this suposed to happen? Is this a bug? I don't know...
I was checking my dataset to see something totally unrelated to this and I stumbled upon this: part of the annotations made by me where wrong and had None category as it's Bounding Box class.
It surprised me because I don't know how could it work, and I would like to know if anyone knows the behaviour of darknet upon this. It just ignores it or can it impact negatively to the training of the net? Is this suposed to happen? Is this a bug? I don't know...
Thank you in advance.