Using RLlib for a custom multi-agent gym environment - reinforcement-learning

Question

I am trying to set up a custom multi-agent environment using RLlib, but either I am using the once available online or I am making one, I am being encountered by the same errors as mentioned below. Please help me out.
I have installed whatever they have asked from me in step (a)
I am registering my environment using
def env_creator(env_config):
return SimpleCorridor(env_config)
register_env("corridor", env_creator)
if __name__ == "__main__":
ray.shutdown()
ray.init()
tune.run(
"PPO",
stop={
"timesteps_total": 10000,
},
config={
"env": "corridor", # <--- This works fine!
"env_config": {
"corridor_length": 5,
},
},
)
(pid=266728) Try one of the following:
(pid=266728) a) For Atari support: `pip install gym[atari] atari_py`.
(pid=266728) For VizDoom support: Install VizDoom
(pid=266728) (https://github.com/mwydmuch/ViZDoom/blob/master/doc/Building.md) and
(pid=266728) `pip install vizdoomgym`.
(pid=266728) For PyBullet support: `pip install pybullet`.
(pid=266728) b) To register your custom env, do `from ray import tune;
(pid=266728) tune.register('[name]', lambda cfg: [return env obj from here using cfg])`.
(pid=266728) Then in your config, do `config['env'] = [name]`.
(pid=266728) c) Make sure you provide a fully qualified classpath, e.g.:
(pid=266728) `ray.rllib.examples.env.repeat_after_me_env.RepeatAfterMeEnv`
Is there something else I should be taking care of? This is just a basic environment from the examples I saw. Even the environment I am customizing is facing the same problem. I have initialized the observation space as a tuple so I am not able to use stable baselines for evaluation.
Please Please help me out.