hi guys! first, I would like to thank for this awesome package. I have a question: when I use a model in MLForecast with multiple time series, does the model fit for each series separately, or uses all series at the same time to estimate parameters?

Hey. It uses all series to train a single model. If you want one model per serie we recently introduced sklearn models in statsforecast

thank you!

If we want to add one-hot encoded features for the unique_id we just do that manually, right?

You can use a pipeline like here but for the unique_id. Note that in order to include it you have to set it as a static feature, e.g.

static_features=['unique_id']