How do you recommend saving and loading trained models?

[johnandersen777]

For example in

https://github.com/IntelPython/daal4py/blob/a579737bcebf42b81bca5db2e0b6203682a9d98d/examples/linear_regression_streaming.py#L69

should we just pickle train_result.model?

[johnandersen777]

This is what DFFML has done:

        self.lm_trained = self.lm.finalize().model
        self.joblib.dump(self.lm_trained, self.path)

[PivovarA]

Hi @pdxjohnny
In daal each model has its own serializer. Therefore, these serializers are called during serialization.
Therefore joblib should be enough for model saving