A very simple shell-based maze game. Useful for traing simple reinforcement learning algorithms. To play:
$ python TinyMaze.py maze_size
where "maze_size" is an integer such that the maze will be maze_size x maze_size squares.
To use in reinforcement learning:
import TinyMaze
moves = ["a", "s", "w", "z"]
game = TinyMaze.TinyMazeEnv(maze_size)
status = game.step(moves[move_index])