When using nhits and making predictions my outputs blow up with the infomation of making predicitons: Predictiing dataloader. As I make predictions for a full year based on earlier trained model. Is there a way to have no output when doing the predictions

Thanks so much for using NF! Could you please share some code, screenshots and errors you are seeing? In that way we can help you better.

@Jelte Bottema I had this issue today as well, it comes from the trainer kwargs passed into BaseModel: https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/common/_base_model.py#L88-L123 which are then passed into

pl.Trainer

, so if when you create the model you pass in

enable_progress_bar=False

it should eliminate this logging. If you're loading in a model you can do this:

nf.models[0].trainer_kwargs['enable_progress_bar'] = False

if you want to reduce the logger noise elsewhere do:

logging.getLogger("pytorch_lightning").setLevel(logging.ERROR)
logging.getLogger("lightning.pytorch.utilities.rank_zero").setLevel(logging.ERROR)

I'm also trying to do many predictions, but the library only allows you to do one at a time. If I make progress on that to reduce noise, I'll post about it here

Thanks for the reply I will try to make it work

I did it with AutoNBEATSx, I'd suggest stepping through with a debugger

then it should work I guess, thanks again for the info, will try to get it working

I think it might only work if you have loaded the model from disk. It might work if you modify the

trainer

attribute directly