I get an error when running model.fit. It looks like the same error that I had with model.fit in both of the Labs for this week, and at least one other person reports having the same issue, so I suspect it is some type of system-level issue, not a mistake with my code.
What do you recommend? Please help ASAP.
Hi John! There seems to be a bug in Colab. We pushed a fix to the notebooks in the classroom. Kindly reopen it and paste your solutions there. It’s basically adding this line up top to avoid the error.
# Install this package to use Colab's GPU for training
!apt install --allow-change-held-packages libcudnn8=8.4.1.50-1+cuda11.6
Hope it works!
It works! and Colab is allowing me to use a GPU again, so I’ve been able to complete the assignment and labs. Thanks for your help. I see that this push went to many (all?) labs and assignments, so lots of work for you and team. Appreciate it.