Hi everyone, If I am using LGBMRegressor in MLForecast and training 100 unique_id, would the model learn the whole 100 time series in 1 model or would there be separate model for each id? If it's combined as 1 model, how would MLForecast handles the difference of seasonality and statistic between different time series?

It would learn one global model and you need to add relevant features that allows the model to capture seasonalities etc. In automl some standard features are tested based on given seasonal length.

I suppose those features are lag and lag transform, is that correct?

And date features also

okay, so as long as I included them, they are already calculated in the model and may separate the patterns. Thank you