Reinforcement learning on a quantum maze
Reference for the gym environment creation
https://medium.com/@apoddar573/making-your-own-custom-environment-in-gym-c3b65ff8cdaa
and
https://github.com/MattChanTK/gym-maze/tree/master/gym_maze
In particular, after creating a new environment, you have to register it from its folder with
pip install -e .
To test the correct installation of the environment run
import gym
from gym_quantum_maze.envs import quantum_maze_env
env = gym.make('quantum-maze-v0')
env.reset()
for _ in range(3):
env.render()
action = env.action_space.sample()
print('action=', action)
env.step(action) # take a random action
env.close()