Iva Eftimska

08/22/2023, 10:32 AM
What to do when you have a dataset of frequency of 5 min and you want to use predict with xgb model, but you want predictions for whole year that is approximately 40k samples, so inside the predict function is 40k and that takes too long. How to optimize this?

José Morales

08/22/2023, 4:08 PM
Hey. Can you provide some more details? • How many series do you have? • How long are they? • Which lag transforms are you using? • Do you have exogenous features? Also forecasting a horizon that long with the recursive method probably won't yield so good results, you may get a better performance from deep learning models, such as the N-HiTS