Compile Pytorch model for iOS

Hi @L1onKing!

As I see you already dealt with that problem. The issue described by you in Tune the model in iOS is a next level of progress with iOS tuning.

But anyway the issue with transferring python objects between process on MacOS is still present in tvm. That’s a wee known issue with python on MacOS. I’ve just check it with latest TVM main branch and it really cannot pass default_module_loader_mgr object into subprocess.

Could you please share your patch with fix/workaround?