Tensorflow Support #2
Description
Right now the model is generated using PyTorch. This was simply because I've used PyTorch more and wanted to get something working fast. It would be good to also offer the option to use Tensorflow.
Ideally we could have a feature flag that allows the user to choose which backend they want use.