gym-cfd

Trying to integrate the CFD simulation of 6sigmaRoom into gym in order to realize reinforcement learning algoritms.

Installation

First Install gym with pip install gym. Then enter the directory where you put 'gym-cfd' file. Installed your package with pip install -e gym-cfd, you can create an instance of the environment with gym.make('gym_cfd:cfd_simple-v0') or gym.make('gym_cfd:cfd_complex-v0')

Test

Once you finished installment, you can use codes below to see whether it works:

import gym
env = gym.make('gym_cfd:cfd_simple-v0')
env.reset()

for i in range(1000):
    action = env.action_space.sample() # take a random action
    state, reward, done, info = env.step(action) 
    # def : state = (temperature_new, load_new, temperature_setpoint)
    print(state, reward, done, info)
env.close()

Development

This project is far from completed. Welcome further discussion. Embarrassing is that my software has already expired...

Citation

If you find this package really helpful and are interested in citing it here's how you do it:

@Misc{,
    author = {Shu Li},
    title = {{Gym-CFD}: An implement of combining reinforcement learning  CFD simulation(6SigmaRoom) based on libraries from OpenAI},
    year = {2019--},
    url = " https://github.com/Li-Shu14/gym-cfd/"
}