Nixtla team, do y’all have suggested best practices on finetuning? like, iterate in steps of 10 b/w 0 and 100 e.g. 0, 10, 20, 30, … 100? Also, are you able to share any more color on what exactly

clean_ex_first=True

does with exogenous variables?

Your approach for fine-tuning is good. You should not need to finetune for too many steps. Also, you might overfit by fine-tuning too much. So doing 0, 10, 20, etc. is good.

clean_ex_first

simply controls the order in which the model processes exogenous features. If True, it processes the features first, if False, it processes the features at the end. The default value of True was shown in our own experiments to give empirically better results.