Load a :class:`~fairseq.models.FairseqModel` from a pre-trained model file. Downloads and caches the pre-trained model file if needed. The base implementation returns a :class:`~fairseq.hub_utils.GeneratorHubInterface`, which can be used to generate translat
(cls, model_name_or_path, checkpoint_file='model.pt', data_name_or_path='.', **kwargs)
source not stored for this graph (policy: none)
nothing calls this directly
no test coverage detected