2.1.1.1.1.4. emicroml.modelling.cbed.distortion.estimation.load_ml_model_from_state_dict

load_ml_model_from_state_dict(ml_model_state_dict, device_name=None)

Load a machine learning model from a dictionary

The current function loads machine learning (ML) models, represented by the class emicroml.modelling.cbed.distortion.estimation.MLModel, from dictionary representations of said ML models.

Dictionary representations of ML models can be generated via the method emicroml.modelling.cbed.distortion.estimation.MLModel.state_dict().

Moreover, dictionary representations of ML models can be generated then saved to files via the method emicroml.modelling.cbed.distortion.estimation.MLModelTrainer.train_ml_model() of the class emicroml.modelling.cbed.distortion.estimation.MLModelTrainer. For further details see the documentation for said method.

Let ml_model_state_dict_filename be the relative or absolute path to a file storing a dictionary representation of an ML model of interest. One can load said dictionary representation via the function torch.load(), where the function parameter f should be set to ml_model_state_dict_filename. In this case, the function torch.load() should return the dictionary representation. For further details see the documentation for the function torch.load().

Parameters:

ml_model_state_dictdict

The dictionary representation of the ML model to load.

device_namestr | None, optional

This parameter specifies the device in which to store the ML model. If device_name is a string, then it is the name of the device to be used, e.g. ”cuda” or ”cpu”. If device_name is set to None and a GPU device is available, then a GPU device is to be used. Otherwise, the CPU is used.

Returns:

ml_modelemicroml.modelling.cbed.distortion.estimation.MLModel

The ML model represented by the dictionary ml_model_state_dict.