def load(run_uuid: str, checkpoint: Optional[int] = None):
r"""
Loads and starts the experiment from a previous checkpoint.
Keyword Arguments:
run_uuid (str): if provided the experiment will start from
a saved state in the run with UUID ``run_uuid``
checkpoint (str, optional): if provided the experiment will start from
given checkpoint. Otherwise it will start from the last checkpoint.
"""
_experiment_singleton().start(run_uuid=run_uuid, checkpoint=checkpoint)
def calculate_configs(configs: Optional[BaseConfigs],
configs_dict: Optional[Dict[str, any]] = None,
run_order: Optional[List[Union[List[str], str]]] = None):
r"""
Calculate configurations
Arguments:
configs (Configs, optional): configurations object
configs_dict (Dict[str, any], optional): a dictionary of
configs to be overridden
run_order (List[Union[str, List[str]]], optional): list of
configs to be calculated and the order in which they should be
calculated. If ``None`` all configs will be calculated.
"""
if configs_dict is None:
configs_dict = {}
_experiment_singleton().calc_configs(configs, configs_dict, run_order)
def save_numpy(name: str, array: np.ndarray):
r"""
Saves a single numpy array. This is used to save processed data.
"""
numpy_path = Path(_experiment_singleton().run.numpy_path)
if not numpy_path.exists():
numpy_path.mkdir(parents=True)
file_name = name + ".npy"
np.save(str(numpy_path / file_name), array)
def save_checkpoint():
_experiment_singleton().save_checkpoint()
def start():
r"""
Starts the experiment.
"""
_experiment_singleton().start()