Building a chess engine using neural networks and nbdev :)
Building it from the ground up and as code along with George Hotz and his twitchchess see: https://youtu.be/RFaFmkCEGEs
TODOs
- Establish nbdev
- Build dataset and generate data.
- Establish a neural network
- Roll out search beyond 1 move
- Train on more data
- [] Add RL self learning support
It is a simple look ahead neural network value function. The trained net is in nets/value.pth. It takes in a serialized board and outputs a range from -1 to 1. -1 mean black wins, 1 means white wins, 0 is a draw.