Skip to content

Latest commit

 

History

History
2 lines (2 loc) · 380 Bytes

README.md

File metadata and controls

2 lines (2 loc) · 380 Bytes

In this repository I implement a minimalistic version of automatic differentiation software with abilities like backpropagation. There is also a PyTorch Like API on top of that for creating a Neural Network and some test code as well. This work is heavily inspired (+99%) by Andrej Karpathy's MicroGrad library, so anyone interested should check the original source here @karpathy