Objection Detection prediction does work with Test Set. EmptyLabel is not subscriptable

Here is my Coco Kaggle Kernel running with the Coco_Tiny dataset. I am able to run everything fine but when I try to get prediction for my test set, it runs into an error.

I believe the problem is with EmptyLabel because self.data = 0 and the code expect a list.

Or does anyone knows how to replace the EmptyLabel class with your own? I do see that data_block.py’s add_test_folder has a label parameter. Or if I am doing something wrong, any help would be appreciated.