Tia Guo
10/31/2022, 7:26 PMKevin Kho
10/31/2022, 8:05 PMlog_artifact
?Tia Guo
10/31/2022, 8:29 PMKevin Kho
10/31/2022, 8:31 PMTia Guo
10/31/2022, 8:32 PMKevin Kho
11/01/2022, 3:11 AMTia Guo
11/01/2022, 1:25 PMKevin Kho
11/01/2022, 7:27 PMstatsforecast
is you can either use .forecast()
which is a straight forecasing or .fit()
and .predict()
. If you want to decouple training and prediction time, just call the .fit()
method and then you can save that pickle object to be used later.
The pickle should be able to be used with exogenous regressions because it saves the model weights and info. That is how it was done with sklearn before also. I don’t believe statsforecast
has it built in yet but you can just make your own saving and loading functions pretty easily.mlflow.sklearn.load_model
just uses pickle under the hood as well 🙂Tia Guo
11/01/2022, 8:48 PMKevin Kho
11/01/2022, 8:51 PMTia Guo
11/01/2022, 8:54 PMKevin Kho
11/01/2022, 8:54 PMload_dict
. The one above download_artifacts
.