update bots

main.py

@@ -3,7 +3,7 @@
 import socket
 import threading
 from src.SocialStudy import questions
-from src.Pi.PiBot.pibot2 import PiBot2
+from src.Pi.PiBot.PiBot2 import PiBot2
 """
 determine if main or pi
 """
@@ -29,7 +29,7 @@ def main_train_and_test():
         threading.Thread(target=questions.get_user_input, args=()).start()
     else:
         pibot = PiBot2()
-        steps = 250
+        steps = 1
         train.train_session(pibot, steps)
         # publish_data(actions)
 

outputs/actions.json

@@ -0,0 +1 @@
+{"count": "0"}

settings.py

@@ -15,7 +15,7 @@
     ACTION_FILE = r".\outputs\actions.txt"
     RESPONSE_FILE = r".\outputs\responses.json"
 else:
-    ACTION_FILE = "/home/pi/RealDilemma/outputs/actions.txt"
+    ACTION_FILE = "/home/pi/RealDilemma/outputs/actions.json"
     RESPONSE_FILE = r"/home/pi/RealDilemma/outputs/responses.json"
 
 QUESTIONS_FILE = "./outputs/Questions.json"

src/DRL/gym-pibot/gym_pibot/envs/pibot_env.py

@@ -1,7 +1,7 @@
 import gym
 from gym import error, spaces, utils
 from gym.utils import seeding
-from src.Pi.PiBot.pibot import PiBot
+from src.Pi.PiBot.PiBot import PiBot
 import numpy as np
 
 ENERGY_THRES = 4000# thres for total amount of different motors used in robo

src/DRL/gym-pibot/gym_pibot/envs/pibot_env2.py

@@ -1,6 +1,8 @@
 import gym
 from gym import error, spaces, utils
 import numpy as np
+from settings import ACTION_FILE
+import json
 
 ENERGY_THRES = 4000# thres for total amount of different motors used in robo
 MAX_REWARD = 1000
@@ -13,7 +15,7 @@ class PiBotEnv2(gym.Env):
   """
   metadata = {'render.modes': ['human']}
 
-  def __init__(self, PiBot, servo=False):
+  def __init__(self, PiBot, servo=True):
       super(PiBotEnv2, self).__init__()
       self.PiBot = PiBot
       self.SERVO = servo
@@ -25,6 +27,9 @@ def __init__(self, PiBot, servo=False):
           4: self.PiBot.stop,
           5: self.PiBot._servo
       }
+      # FOR RECORDING ACTIONS TAKEN #
+      self.ACTION = []
+      self._RECORD_ACTION = False
       if self.SERVO:
           #self.reward_range = (0, MAX_REWARD)
           self.action_space = spaces.MultiDiscrete([6,# action
@@ -71,6 +76,8 @@ def do_action(self, action):
       do = self.CONTROL_LOOKUP[action[0]]
       duty  = 100 - action[1]
       time = action[2]
+      if self._RECORD_ACTION:
+          self.ACTION.append(str([int(x) for x in action]))
       # perform action
       do(duty, time)
 
@@ -81,7 +88,6 @@ def _get_reward(self):
       """
       # maximize amount of movement, maximize distance in ultrasound sensor
       state = self._get_state()
-      # diff(X) + score
       reward = state[1] + state[2]
       return reward
 
@@ -102,8 +108,19 @@ def _get_state(self):
       :return:
       """
       return self.PiBot.get_state()
+  def _record_actions(self):
+      with open(ACTION_FILE, "r") as f:
+          data = json.load(f)
+      count = int(data["count"])
+      data[count] = self.ACTION
+      data["count"] = count+1
+      with open(ACTION_FILE, "w") as f:
+          json.dump(data, f)
 
   def close(self):
+      if self._RECORD_ACTION:
+        print("saving actions")
+        self._record_actions()
       del self.PiBot
 
 

src/DRL/train/train.py

@@ -33,10 +33,12 @@ def train_session(pibot, steps, train_dict=TRAIN_DICT):
 def train(steps, pibot, model, model_name) -> str:
     # If the environment don't follow the interface, an error will be thrown
     env = PiBotEnv2(pibot)
+    # record actions
+    env._RECORD_ACTION = True
     check_env(env, warn=True)
     # The algorithms require a vectorized environment to run
-    env = DummyVecEnv([partial(PiBotEnv2, PiBot=pibot)])
-    model = model(env=env)
+    d_env = DummyVecEnv([partial(PiBotEnv2, PiBot=pibot)])
+    model = model(env=d_env)
     model.learn(total_timesteps=steps)
     fpath = os.path.join(TRAIN_DIR, model_name)
     model.save(fpath)
@@ -45,6 +47,6 @@ def train(steps, pibot, model, model_name) -> str:
       policies[model_name] = fpath
     with open(POLICYF, "w") as f:
         json.dump(policies, f)
-    del pibot
-    del env
+    env._record_actions()
+
     return fpath

src/Pi/PiBot/PiBot2.py

@@ -2,7 +2,7 @@
 import time
 import numpy as np
 from threading import Thread
-from src.Pi.PiBot.pibot import PiBot
+from src.Pi.PiBot.PiBot import PiBot
 
 class PiBot2(PiBot):
     """