[BUG] TypeError: __deepcopy__() takes 1 positional argument but 2 were given when using darts.models.forecasting.regression_ensemble_model.RegressionEnsembleModel

[omerfguzel]

Describe the bug
I'm trying to use Ensemble model but I got this error. Any suggestions?

System:

• Python version: 3.7
• darts version 0.15.0

Additional context
full error line: TypeError: deepcopy() takes 1 positional argument but 2 were given

[dennisbader]

Hi, could you please send a code snippet so we can reproduce the issue? Thanks.

[omerfguzel]

Solved and here is how it is solved:
h3ik0th/Darts_ensembleFC#1 (comment)

[dennisbader]

When creating a model object all parameters are saved.
Deepcopying fitted/trained forecasting_models with EnsembleModels fails.

You can reset your models with models_untrained = [m.untrained_model() for m in models]