ray.train.data_parallel_trainer.DataParallelTrainer.can_restore#

classmethod DataParallelTrainer.can_restore(path: Union[str, os.PathLike], storage_filesystem: Optional[pyarrow.fs.FileSystem] = None) bool#

Checks whether a given directory contains a restorable Train experiment.

Parameters

path – The path to the experiment directory of the Train experiment. This can be either a local directory (e.g., ~/ray_results/exp_name) or a remote URI (e.g., s3://bucket/exp_name).

Returns

Whether this path exists and contains the trainer state to resume from

Return type

bool