MultiModalPredictor.load

classmethod MultiModalPredictor.load(path: str, resume: bool | None = False, verbosity: int | None = 3)[source]

Load a predictor object from a directory specified by path. The to-be-loaded predictor can be completely or partially trained by .fit(). If a previous training has completed, it will load the checkpoint model.ckpt. Otherwise, if a previous training accidentally collapses in the middle, it can load the last.ckpt checkpoint by setting resume=True. It also supports loading one specific checkpoint given its path.

Warning

autogluon.multimodal.MultiModalPredictor.load() uses pickle module implicitly, which is known to be insecure. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling. Never load data that could have come from an untrusted source, or that could have been tampered with. Only load data you trust.

Parameters:

  • path – The directory to load the predictor object.

  • resume – Whether to resume training from last.ckpt. This is useful when a training was accidentally broken during the middle, and we want to resume the training from the last saved checkpoint.

  • verbosity – Verbosity levels range from 0 to 4 and control how much information is printed. Higher levels correspond to more detailed print statements. You can set verbosity = 0 to suppress warnings.

Return type:

The loaded predictor object.