ray.rllib.algorithms.algorithm.Algorithm.save

Algorithm.save(checkpoint_dir: Optional[str] = None, prevent_upload: bool = False) → str

Saves the current model state to a checkpoint.

Subclasses should override save_checkpoint() instead to save state. This method dumps additional metadata alongside the saved path.

If a remote checkpoint dir is given, this will also sync up to remote storage.

Parameters

• checkpoint_dir – Optional dir to place the checkpoint.

• prevent_upload – If True, will not upload the saved checkpoint to cloud.

Returns

The given or created checkpoint directory.

Note the return path should match up with what is expected of restore().