Closed ZealACMer closed 5 years ago
@ZealACMer Are you sure that the number of samples is 16? 3161088 is not equal to batch96162828. I think it's better to check the shape of "y" after 2D tensor and figure out the problem. Let me know if you still didn't find the problem.
I tried to change 2D tensor into 3D tensor, so it can be feed into 3D Conv. The code are shown as follows: y = self.con1x1(x) # 2D tensor y = y.view((-1, 96, 16) + y.size()[2:]) #2D tensor changed to 3D tensor
error: Thank you so much for you kind help.