[D] PyTorch to Desktop App Deployment

I have trained a model through PyTorch and wish to bring it into production through a Windows desktop app for real-time inference. My current plan is to convert my PyTorch model to a TorchScript. Then in actual C inference, I'll then use LibTorch to call the TorchScript-formatted model. My goal is to integrate this into a desktop app and run it in real-time on a stream of data. Also, I plan on only running on CPU, not GPU.