> I understand that PyTorch is an awesome tool for researchers, but it doesn't necessarily fit into a prod environment. You need to write some infrastructure a

> I understand that PyTorch is an awesome tool for researchers, but it doesn't necessarily fit into a prod environment. You need to write some infrastructure around PyTorch to make it work. Something like a key/mapping in each checkpoint that says which architecture to choose with which parameters. It sure could be easier, but is saving the model's code into the checkpoint enough? Things like the data pre-processing expected by the model would also have to be included for it to really be self-contained.