Invalid size error while performing backward in vanilla AlexNet architecture with FCN

[anil_batra]

The following link help me to get the exact error and now I am able to fix.

gdb python
catch throw
run <script name>
backtrace

Loss.backward() cause RuntimeError: invalid argument (in version 0.1.12 no error)

Hi all, I observe the same. it occurs in loss backward, if the batchsize at the end of the dataloader sampling becomes 1: loss torch.Size([1]) torch.Size([64, 3, 256, 256]) torch.Size([64, 1]) loss torch.Size([1]) torch.Size([64, 3, 256, 256]) torch.Size([64, 1]) loss torch.Size([1]) torch.Size([64, 3, 256, 256]) torch.Size([64, 1]) loss torch.Size([1]) torch.Size([64, 3, 256, 256]) torch.Size([64, 1]) loss torch.Size([1]) torch.Size([64, 3, 256, 256]) torch.Size([64, 1]) loss torch.Size(…

Thanks.