Hi team, Is there any reason this is an error and not just a warning?

Frequency chosen: M
Error in nixtlar::nixtla_client_forecast(df = timegpt_fit_tbl, h = prepared_data_list$horizon,  : 
  Your time series is too short. Please make sure that each of your series contains at least 36 observations.

Hey. If you're using finetuning or computing prediction intervals we need at least those samples to do that, that's why it's an error

Ok so if I’m not doing finetuning nor intervals, these limits won’t apply?

If you're just forecasting one sample per serie is enough, but keep in mind that it won't be as good as it would be with more samples

Understood. Thanks

Hi @Vidar Ingason what José mentioned is correct. After reviewing your use case, we also noticed that nixtlar had slightly stricter data requirements in certain cases compared to our Python SDK. We've updated these requirements recently and merged the changes into the main branch. If you experience further issues, please switch to the GitHub version of nixtlar. Thanks.

Ok awesome, I’ll check it out :)