gan.py template fails to run with GPU

[MarkoOrescanin]

Common bugs:

1. Tensorboard not showing in Jupyter-notebook see issue 79.
2. PyTorch 1.1.0 vs 1.2.0 support see FAQ

Describe the bug
When running the gan.py script with the only change in script

trainer = pl.Trainer(max_nb_epochs=10, gpus=1, distributed_backend='dp')
or with gpus=[0]

script fails with error on the loss function not being on GPU
File "/home/marko/anaconda2/envs/py36_torch13/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in call
result = self.forward(*input, **kwargs)
File "/home/marko/anaconda2/envs/py36_torch13/lib/python3.6/site-packages/pytorch_lightning/pt_overrides/override_data_parallel.py", line 58, in forward
return self.module.training_step(*inputs[0], **kwargs[0])
File "gan.py", line 116, in training_step
g_loss = self.adversarial_loss(self.discriminator(self.generated_imgs), valid)
File "gan.py", line 89, in adversarial_loss
return F.binary_cross_entropy(y_hat, y)
File "/home/marko/anaconda2/envs/py36_torch13/lib/python3.6/site-packages/torch/nn/functional.py", line 2065, in binary_cross_entropy
input, target, weight, reduction_enum)
RuntimeError: Expected object of device type cuda but got device type cpu for argument #2 'target' in call to _thnn_binary_cross_entropy_forward
Epoch 1: 0%| | 0/938 [00:00<?, ?batch/s]

To Reproduce
Steps to reproduce the behavior:

1. Go to gan.py
2. Change the trainer= pl.Trainer() to above on a GPU system

Expected behavior
A clear and concise description of what you expected to happen.

Desktop (please complete the following information):

• OS: Ubuntu 16.04

[williamFalcon]

Fixed!