Different Gridworld implementations conforming to OpenAI gym interface. Used for developing Reinforcement Learning agents.
pip install git+https://github.com/drmaj/gridworldsgym#egg=gridworldsgym
$ import gym
$ import gridworldsgym
$ env = gym.make('GridWorld-v0')
- GridWorld-v0: Simple grid world from Artificial Intelligence: A modern approach by Russel and Norvig
- SlipperyGridWorld-v0: Simple grid world with slippery actions from Artificial Intelligence: A modern approach by Russel and Norvig
- WindyGridWorld-v0: Windy Grid World from Reinforcement Learning by Sutton and Barto
- CliffGridWorld-v0: Cliff Grid World from Reinforcement Learning by Sutton and Barto
- SlipperyCliffGridWorld-v0
This project is licensed under the GNU GENERAL PUBLIC LICENSE Version 3