The human-centric platform for production ML & AI

Outerbounds

Hey all, is it possible to retrieve model artifacts I log using `current.torch.run`, `@torchrun`, `@checkpoint`, logging my Trainer artifacts to `current.checkpoint.directory`?

Would someone be willing to whip up a quick example, or give me some help and I can provide one? I'm training a model in a distributed step and then want to load the model into `self.model` in my join step.

Looks like this may be what I was looking for <https://github.com/outerbounds/metaflow-checkpoint-examples/blob/master/cifar_distributed/flow.py>