Added averageRewards.png and reformatted code

DeepQLearning/agentMemoryDQN.py

-import numpy as np
 import matplotlib.pyplot as plt
+import numpy as np
+
 
 class agentMemory(object):
     def __init__(self, memSize, input_shape, action_shape):

DeepQLearning/dqnAgent.py

 import time
+
 import numpy as np
-from agentMemoryDQN import agentMemory
-from tensorflow.keras.layers import Dense, Activation, Conv2D, Flatten
+from tensorflow.keras.layers import Dense
 from tensorflow.keras.models import Sequential, load_model
 from tensorflow.keras.optimizers import Adam
 
+from agentMemoryDQN import agentMemory
+
+
 def qFunctionNN(lr, outputs, inputs):
     QFunction = Sequential()
     QFunction.add(Dense(24, activation='relu', input_dim=inputs))
@@ -13,6 +16,7 @@ def qFunctionNN(lr, outputs, inputs):
     QFunction.compile(optimizer=Adam(lr=lr), loss='mean_squared_error')
     return QFunction
 
+
 class dqnAgent(object):
     def __init__(self, lr, gamma, actions, vareps, bSize, observations,
                  epsDec=0.0, epsMin=0.01, memSize=10000, name='Alan', bins=7):
@@ -55,7 +59,7 @@ class dqnAgent(object):
                     target = (1.0 - 0.1) * r[i] + 0.1 * self.gamma * next_action
                 y = self.Q.predict(state[i])
                 y[0] = target
-                history = self.Q.fit(x=state[i], y=y, verbose=0, epochs=1)
+                self.Q.fit(x=state[i], y=y, verbose=0, epochs=1)
 
             self.steps += 1
         print("learn time: ", time.time() - start)

DeepQLearning/main.py

+import time
+
+import gym
 import numpy as np
 from tqdm import tqdm
-from dqnAgent import dqnAgent
-import gym
-import time
 
-from config import *
+from dqnAgent import dqnAgent
 
 env = gym.make('BipedalWalker-v3')
 nmb_of_actions = env.action_space.shape[0]
@@ -13,7 +13,8 @@ marvin = dqnAgent(gamma=0.99, vareps=1.0, lr=0.001,
                   observations=nmb_of_obs, actions=nmb_of_actions, memSize=25000,
                   epsMin=0.05, bSize=16, epsDec=0.999, bins=7)
 
-rewards = []; epsHistory = []
+rewards = [];
+epsHistory = []
 avg_rewards = []
 steps = 0
 verbose = False
@@ -80,6 +81,3 @@ while not done:
     # marvin.addMemory(observation, action_ind, reward, obs, int(done))
     env.render()
     observation = obs
-
-
-  
\ No newline at end of file

DeepQLearning/show_agent.py

+import gym
+import matplotlib.pyplot as plt
 import numpy as np
-from tqdm import tqdm
+
 from dqnAgent import dqnAgent
-import matplotlib.pyplot as plt
-import gym
 
 env = gym.make('BipedalWalker-v3')
 nmb_of_actions = env.action_space.shape[0]
 observation = env.reset()
 marvin = dqnAgent(gamma=0.99, vareps=0.0, lr=0.001,
                   observations=len(observation), actions=nmb_of_actions, memSize=25000,
-                  epsMin=0.02, bSize=32, replace=1000, epsDec=0.001)
+                  epsMin=0.02, bSize=32, epsDec=0.001)
 marvin.loadCNNs()
 
-
 total_rewards = np.load("total_rewards.npy", allow_pickle=True)
 eps = np.load("eps.npy", allow_pickle=True)
 avg_rewards = np.load("avg_rewards.npy", allow_pickle=True)