Training on different data sets

[denwolff]

Hello,

This is more of a beginner question. I'm using MLForecast to predict a target time series based on three other 'input' time series / features (same time points/samples for features and target). I manage to do this with one data set.

But now I would like to train the model on multiple datasets stemming from different 'sites'. More specifically, I would like to train the model on datasets from sites 1 to N-1 and test the performance on dataset N for example.

What would be a good way to do this training on multiple independent datasets in MLForecast?

I would be grateful for any hint or advice.