ssd.pytorch
ssd.pytorch copied to clipboard
questions in detection.py
In line 61
flt[(rank < self.top_k).unsqueeze(-1).expand_as(flt)].fill_(0)
why rank < self.top_k
? I think it should be rank > self.top_k
. The position not in top_k need to be filled with value 0.
Am I right? Could you give me some suggestions?