lwdoubles

I noticed that focal loss has been achieved in loss.py, I want to know how to use it in yolov3

关于l2norm中的 out = self.weight.unsqueeze(0).unsqueeze(2).unsqueeze(3).expand_as(x) * x

1

这里为何要用unsqueeze呀

one-stage可视化有问题，模型没有返回三个变量

1

one-stage可视化有问题，模型没有返回三个变量，希望能给个联系方式详细请教下