neat python output export to pytorch, generate a network definition to train on the gpu with examples

[pax7]

Is your feature request related to a problem? Please describe.
I'd like to be able to export a neat-python network to pytorch so it can be run on the gpu.

Describe the solution you'd like
Use case 1:
Export to .pt such that the model can then be loaded in a python pytorch script, and inference to happen on the gpu.

Wish use case 2:
Generate a network definition using neat, that can then be trained on the gpu using examples, and then subsequently be run on the gpu.

Describe alternatives you've considered
I failed to do this myself.

Additional context
N/A