Previously we had just been adding the historical ...
# timegpt
m
Previously we had just been adding the historical exogenous variables to our dataframe modeldf and thought the model was using them automatically until we noticed our models with and without historical exogenous variables were giving us the same forecasts. Were there any changes to the model recently? Is there another way to implement historical exogenous variables?
m
Hello! Can you upgrade the to the latest version of
nixtla
? It should be v.0.6.6. Let me know if that fixes the problem!
m
thanks Marco! We will try that and report back.
so in speaking with our team, I realized we are actually using TimeGen-1 Azure deployment. Is there a way to upgrade there too?
m
Ah, thanks for the precision! At the moment, TimeGEN does not support historical exogenous variables. We are updating it and we'll let you know when it's available!
m
thanks for the update! any idea when it might be available? just to clarify, this feature used to work but doesn't now temporarily? I swear it was working a couple weeks ago.