manager.trainer.load_iteration

manager.trainer.load_iteration(name, iteration, model_dir=None)

Load a saved model at a given iteration.

Parameters

Name	Type	Description	Default
name	str	Model name (e.g. `"lgbm"`).	required
iteration	int	Iteration of the model.	required
model_dir	Optional[Union[str, Path]]	Directory where models are stored. If None, defaults to :func:`get_cache_home`.	`None`

Returns

Name	Type	Description
	Optional[Any]	The loaded model instance, or None if the file does not exist.

Examples

>>> import tempfile
>>> from spotforecast2_safe.manager.trainer import load_iteration
>>> with tempfile.TemporaryDirectory() as tmpdir:
...     result = load_iteration("lgbm", 99, model_dir=tmpdir)
...     result is None
True