Better to pass callback in fit_one_cycle
It’s better to pass callbacks to fine_tune() and fit_one_cycle() method instead of passing directly to the learner object, to avoid unwanted effects when using your models for inferencing. I found this solution from one of the posts written by Wayde Gilliam.
Is there anyway to call learn.get_preds() without triggering any of the callbacks?
Deep Learning Course Forums
Zach replied to my tweet, confirming it's correct
Copy link