You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As someone who actively makes money through the use of in house neural nets (that provides a valuable service to my customers) trained on relatively small amounts of data and commodity hardware I'd like to tell you this:
Not only are you incredibly entitled (and make all consumers of open source look bad) but you're also dead wrong about deep learning.
This is research code, we'll try to add more comments but it's of course not production ready. It's for enthusiasts and researchers to build on top of, our priority was to get all ideas from the paper out in the code immediately for people to use in other work.
Contributions welcome if you want to improve it :)
Would be nice to use nbdev notebooks from fastAI.
Also use named tensors. I heard they added this in pytorch.
The text was updated successfully, but these errors were encountered: