Skip to content

Latest commit

 

History

History
6 lines (4 loc) · 419 Bytes

README.md

File metadata and controls

6 lines (4 loc) · 419 Bytes

BQN-grad

An experiment implementing reverse-mode auto-differentiation, and a (very slow) neural network in BQN.

Inspired by micrograd and Andrej Karpathy's exellent tutorial on backpropagation.

The qmnist dataset used to test can be found here.