Q Board
Q Board gathers multiple board game environments and provides Deep Q Learning models and algorithms.
Environments
All environments use only Pytorch tensors.
- Tic Tac Toe
Goal
Q Board aims to implement AIs using Deep Q Learning that play against themselves.
Files
- agents : Q Learning algorithms
- dqn : Q Networks
- envs : Environments
- log : Logger and utils to print / process text
- mem : Memories (replay buffers)
- main : App's entry
- utils : Extra functions for Pytorch
Depedencies
- Python 3 (written in 3.7)
- Pytorch