In the example, logits needs requires_grad=True

[aaronrmm]

I'm using torch==1.11.0+cu113 and the example for Poly1 Cross-Entropy Loss throws an error RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn.

I just have to change line 7 to logits = torch.rand([batch_size, num_classes], requires_grad=True) for it to work.