This message was deleted.
# neural-forecast
s
This message was deleted.
j
Do you want to run the cross validation after finding the best model with the auto classes?
The auto models save the best model under the
model
attribute, so for those you'd need something like:
nf.models[0].model.trainer_kwargs = {"max_steps": 0, ...}
You're not accessing the
model
attribute in your code, just the Auto model
If you don't need anything else from the auto model you can just replace it with the regular version, e.g.
nf.models[0] = nf.models[0].model