Error encountered during evaluation

[jkini]

ValueError: Expected more than 1 value per channel when training, got input size [1, 512].

Please let me know the environment details.
pytorch 1.1.0, cuda10.2, and torchvision is 0.6.0 on python 3.7

[zhanghaobucunzai]

请问下，该问题是怎么解决的啊？Excuse me, how is this problem solved?

[synsin0]

I have encountered the same question, seemingly the batch size can not be 1 because group normalization is used. but when I change the batch size to 2, another problem appears, maybe the author may solve the problem.

problem is as this:
return torch.stack(batch, 0, out=out)
RuntimeError: stack expects each tensor to be equal size, but got [5] at entry 0 and [4] at entry 1

[Ri-Bai]

I have encountered the same question, seemingly the batch size can not be 1 because group normalization is used. but when I change the batch size to 2, another problem appears, maybe the author may solve the problem.

problem is as this:
return torch.stack(batch, 0, out=out)
RuntimeError: stack expects each tensor to be equal size, but got [5] at entry 0 and [4] at entry 1

are your solve it? I have the same question

[0-ing]

I have encountered the same question, seemingly the batch size can not be 1 because group normalization is used. but when I change the batch size to 2, another problem appears, maybe the author may solve the problem.
problem is as this:
return torch.stack(batch, 0, out=out)
RuntimeError: stack expects each tensor to be equal size, but got [5] at entry 0 and [4] at entry 1

are your solve it? I have the same question

I also encountered this problem. Have you solved it？