RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (torch.FloatTensor) should be the same

[quantumcode-martin]

I get this error when running the Training Loop.

I loaded the state like so:
netG.load_state_dict(torch.load("./checkpoints/trained_netG_original.pth"))

And the error happens here:
errG = BCE_loss(generated_pred, cartoon_labels) + content_loss(generated_data, real_data)

Here is the full log

RuntimeError                              Traceback (most recent call last)
Cell In [7], line 79
     77 print(type(generated_pred))
     78 print(type(cartoon_labels))
---> 79 errG = BCE_loss(generated_pred, cartoon_labels.to(device)) + content_loss(generated_data, real_data)
     81 # Calculate gradients for G
     82 errG.backward()

File e:\sylvain-chomet-GAN\env\lib\site-packages\torch\nn\modules\module.py:1130, in Module._call_impl(self, *input, **kwargs)
   1126 # If we don't have any hooks, we want to skip the rest of the logic in
   1127 # this function, and just call forward.
   1128 if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks
   1129         or _global_forward_hooks or _global_forward_pre_hooks):
-> 1130     return forward_call(*input, **kwargs)
   1131 # Do not call functions when jit is used
   1132 full_backward_hooks, non_full_backward_hooks = [], []

File e:\sylvain-chomet-GAN\utils\loss.py:40, in ContentLoss.forward(self, x1, x2)
     39 def forward(self, x1, x2):
---> 40     x1 = self.perception(x1)
     41     x2 = self.perception(x2)
     43     return self.omega * self.base_loss(x1, x2)

File e:\sylvain-chomet-GAN\env\lib\site-packages\torch\nn\modules\module.py:1130, in Module._call_impl(self, *input, **kwargs)
...
    452                     _pair(0), self.dilation, self.groups)
--> 453 return F.conv2d(input, weight, bias, self.stride,
    454                 self.padding, self.dilation, self.groups)

[quantumcode-martin]

Precision:
It seems the error is triggered by the call to content_loss.

[quantumcode-martin]

Managed to run the train.py script.
Still having those issues with the experiment.ipynb.