Petting Zoo environment for LBF

lbforaging.py

@@ -4,7 +4,7 @@
 import time
 import gym
 import numpy as np
-import lbforaging
+import lbforaging.gym_env
 
 
 logger = logging.getLogger(__name__)

lbforaging/__init__.py

@@ -1,25 +0,0 @@
-from gym.envs.registration import registry, register, make, spec
-from itertools import product
-
-sizes = range(5, 20)
-players = range(2, 20)
-foods = range(1, 10)
-coop = [True, False]
-grid_observation = [True, False]
-
-for s, p, f, c, grid_obs in product(sizes, players, foods, coop, grid_observation):
-    for sight in range(1, s + 1):
-        register(
-            id="Foraging{5}{4}-{0}x{0}-{1}p-{2}f{3}-v2".format(s, p, f, "-coop" if c else "", "" if sight == s else f"-{sight}s", "-grid" if grid_obs else ""),
-            entry_point="lbforaging.foraging:ForagingEnv",
-            kwargs={
-                "players": p,
-                "max_player_level": 3,
-                "field_size": (s, s),
-                "max_food": f,
-                "sight": sight,
-                "max_episode_steps": 50,
-                "force_coop": c,
-                "grid_observation": grid_obs,
-            },
-        )

lbforaging/agents/hba.py

@@ -1,5 +1,5 @@
 from . import QAgent
-from foraging import Env
+from gym_env import Env
 import random
 import numpy as np
 from agents import H1, H2, H3, H4

lbforaging/agents/heuristic_agent.py

@@ -1,7 +1,7 @@
 import random
 import numpy as np
-from foraging import Agent
-from foraging.environment import Action
+from gym_env import Agent
+from gym_env.environment import Action
 
 
 class HeuristicAgent(Agent):

lbforaging/agents/monte_carlo.py

@@ -8,7 +8,7 @@
 import plotly.graph_objs as go
 from networkx.drawing.nx_pydot import graphviz_layout
 
-from foraging import Agent, Env
+from gym_env import Agent, Env
 
 MCTS_DEPTH = 15
 

lbforaging/agents/nn_agent.py

@@ -1,6 +1,6 @@
 import random
 
-from foraging import Agent
+from gym_env import Agent
 
 
 class NNAgent(Agent):

lbforaging/agents/q_agent.py

@@ -5,8 +5,8 @@
 import pandas as pd
 
 from agents import H1
-from lbforaging import Agent, Env
-from lbforaging.environment import Action
+from gym_env import Agent, Env
+from gym_env.environment import Action
 
 _CACHE = None
 

lbforaging/agents/random_agent.py

@@ -1,6 +1,6 @@
 import random
 
-from lbforaging import Agent
+from gym_env import Agent
 
 
 class RandomAgent(Agent):

lbforaging/gym_env/__init__.py

@@ -0,0 +1,26 @@
+from lbforaging.gym_env.environment import ForagingEnv
+from gym.envs.registration import registry, register, make, spec
+from itertools import product
+
+sizes = range(5, 20)
+players = range(2, 20)
+foods = range(1, 10)
+coop = [True, False]
+grid_observation = [True, False]
+
+for s, p, f, c, grid_obs in product(sizes, players, foods, coop, grid_observation):
+    for sight in range(1, s + 1):
+        register(
+            id="Foraging{5}{4}-{0}x{0}-{1}p-{2}f{3}-v2".format(s, p, f, "-coop" if c else "", "" if sight == s else f"-{sight}s", "-grid" if grid_obs else ""),
+            entry_point="lbforaging.foraging:ForagingEnv",
+            kwargs={
+                "players": p,
+                "max_player_level": 3,
+                "field_size": (s, s),
+                "max_food": f,
+                "sight": sight,
+                "max_episode_steps": 50,
+                "force_coop": c,
+                "grid_observation": grid_obs,
+            },
+        )

lbforaging/petting_zoo/__init__.py

@@ -0,0 +1 @@
+from .environment import env, parallel_env