Feedback: Saving a DDP-Wrapped Model

[Samirkar05]

There is the following issue on this page: https://docs.pytorch.org/tutorials/beginner/saving_loading_models.html

Issue: I am referring to the section https://docs.pytorch.org/tutorials/beginner/saving_loading_models.html#saving-torch-nn-dataparallel-models. Contrary to this other tutorial https://docs.pytorch.org/tutorials/intermediate/ddp_tutorial.html it does not include (in the "save" section) that given a DDP wrapped model DDPmodel one needs to access its .load_state_dict method.

cc @wconstab @osalpekar @H-Huang @kwen2501 @mikaylagawarecki