Multi Threaded Gym Environment - Extract

play.py

if __name__ == '__main__':
  env = gym.make('Mario-Kart-Luigi-Raceway-Multi-v0')
  obs = env.reset()
  env.render()

  while not end_episode:
    # Action should be multi-threaded + setting agent
    for i in range(num_agents):
        agent = i+1

        # Set current agent
        CurrentAgent.set_current_agent(agent)

        # CReate Thread
        thread = AgentThread(target=get_action, args=(agent, obs,actors[i],))
        thread.start()

        # Get action
        action = thread.join()
        # cprint('[Gym Thread] Got action %s for agent %i' %((action,agent)),'red')
        obs, reward, end_episode, info = env.step(action)

  env.render()
  #self.queue.put('render')
  total_reward += reward

Hello, I was wondering if you could share where the modified multiplayer environment code used is being pulled from? I had a look in gym-mupen64plus and while I see the MarioKart environment I don't believe I'm finding the multi environment you're using.

I'm looking for the same thing. Also end_episode is not defined.