About Input Resolution

[duzw9311]

Dear author, Thanks for your work. When I input the feature maps with sizes of [torch.Size([1, 256, 320, 208]), torch.Size([1, 512, 160, 104]), torch.Size([1, 1024, 80, 52]), torch.Size([1, 2048, 40, 26])] into FPT_a.py, i got the following error:

self.bn_reduction(self.str_conv3x3(x_low)) + x_att) RuntimeError: The size of tensor a (52) must match the size of tensor b (26) at non-singleton dimension 3"

Could you help me with this problem?

[XiongXin1207]

Same error.The resolution of P3 is four times that of P5,however the str_conv3x3 just downsample two times.The resolution is unmatch.Have you deal with it yet?