What a lot of machine learning courses & books teach: - Load clean data - Train an MNIST-* classifier with 99% accuracy What they should actually be teaching: - Process, clean, load your own data - Fail many experiments and research/find/understand ways to improve your ML model
-
With total respect to
#deoldify's business model it looks like the toughest part of machine learning is model deployment and maintainence which is why they license their model rather than build all the infra around it. (If I'm wrong please correct me).
Well...really I'd characterize it as just one less tough problem we'd rather not have. But yes, tough!
-
-
I only mentioned it because I really admire your story. You had some technical background and decided to take the Fastai course. After lots of experiments you settled on colorization. Finally you managed to license your tech to MyHeritage. I am excited to see what's next.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.