Functional MicroGrad: gist.github.com/cghawthorne/68
I really enjoyed 's recent video about building an autograd engine from scratch. To better understand it, and because I like JAX, I had some fun writing a version that's more "functional" in flavor.
Conversation
The "Neural Networks: Zero to Hero" series is the best explanation I've seen for the details behind neural network training. The content was familiar, but it really helped me connect a lot of dots. Highly recommend watching it!
