Top-k method for selecting candidate bags

[soHardToHaveAName]

You've mentioned the Top-k method for selecting candidate bags in your paper, but it seems that there is only soft center prior method in this repo.

How can I change this repo to use the Top-k method ?

And, if I use Top-k method to select candidate bag, the calculation will be much smaller, Isn't it? Since I only need to computer the weights of bboxes inside bags, the number of weights can be reduced from num_points * num_gts to num_points. Is that so?