Deepcopy base model

Open dhpitt opened this issue 1 year ago • 2 comments

The models in neuralop.models subclass neuralop.models.BaseModel, which adds lots of useful functionality but does not implement the __deepcopy__ method necessary to copy a torch module's architecture and state. This PR adds a method to do that.

Mar 13 '24 17:03 dhpitt

Wouldn't the default deepcopy work out of the box?

Mar 15 '24 18:03 JeanKossaifi

Wouldn't the default deepcopy work out of the box? No, it throws an error - there's no way to access the internal module state from base_model by default.

Mar 15 '24 18:03 dhpitt

Was this resolved @dhpitt ?

Apr 01 '24 19:04 JeanKossaifi