Tried running my ML model on CPU vs GPU last week and the gap was wild

Trained a small image classifier on 50k images, CPU took around 14 hours, GPU finished in 22 minutes on the same box. Am I the only one who underestimated how much the hardware shift matters, or is CPU training still fine for certain kinds of projects?

0 comments

Log in to join the discussion

Log In

0 Comments

No comments yet

Be the first to share your thoughts on this discussion.