first step towards simple make interface

studies/sat_mad.py

@@ -1,11 +1,12 @@
-from environments.factory import rooms
+from environments.factory import make
 import random
 from gym.wrappers import FrameStack
 
-env = rooms(n_agents=2)
+n_agents = 4
+env = make('DirtyFactory-v0', n_agents=n_agents)
 env = FrameStack(env, num_stack=3)
 state, *_ = env.reset()
 
 for i in range(1000):
-    state, *_ = env.step([random.randint(0, 9), random.randint(0, 9)])
+    state, *_ = env.step([env.unwrapped.action_space.sample() for _ in range(n_agents)])
     env.render()