This message was deleted.

hey @Mads Jensen! Thanks for using neuralforecast. We have experienced that behavior due to the huge amount of logs output by ray, is that the case? The workaround is simply to put %%capture at the beginning of the cell that runs

fit

or

cross_validation

Hey, Thanks I will try that, it does indeed sounds like that could be the reason.

Cool! Let me know what happens, we are still trying to figure out how to get rid of those prints xD

Thanks, I will do that. It is running now 🙂 There is also a lot of outputs, so I can why you wanna do that or at least control it someway.