Super excited to be part of fast.ai. I just finished the first week and I am re-watching it again. One question that I think Jeremy answered and has been answered on the forum but I still wanted to ask again. I have access to a CPU HPC with 96 cores (Xeon), 512GB RAM. Can I possibly modify the fastai pytorch code to run it on this CPU based HPC instead of GPU. I get it for free so I am thinking of leveraging that instead of spending on GPUs for initial prototype and learning.
Any thoughts on how I can do it? Is it possible?