option to retrain old model

generate_model.py

@@ -10,9 +10,11 @@ def __int__(self):
         self.input_frames = tf.placeholder(dtype=tf.float32, shape=[None, 84, 84, 4], name='input_frames')
         self.target_frame = tf.placeholder(dtype=tf.float32, shape=[None, 84, 84, 1], name='target_frame')
         self.action_performed = tf.placeholder(dtype=tf.int32, shape=[None, 4], name='action_performed')
+        self.n_epochs = 100
         self.learning_rate = 1e-4
         self.batch_size = 32
         self.momentum = 0.9
+        self.logdir = './Results/prediction_model'
 
     def model(self, x, action, reuse=False):
         if reuse:
@@ -65,6 +67,20 @@ def train(self):
         # TODO: Currently only shows latest input frame
         tf.summary.image(name='Input_images', tensor=self.input_frames[:, :, :, 0])
 
+        saver = tf.train.Saver()
 
+        init = tf.global_variables_initializer()
 
+        with tf.Session() as sess:
+            sess.run(init)
+
+            file_writer = tf.summary.FileWriter(logdir=self.logdir, graph=sess.graph)
+
+            # TODO: Train on 1 step prediction objective later extend
+            # for e in range(self.n_epochs):
+                # n_batches =
+
+
+    def get_next_batch(self):
+        pass
 

mission_control_breakout.py

@@ -29,7 +29,7 @@
 ########################################################################################################################
 # Control
 train_model = False
-load_trained_model = False
+load_trained_model = True
 show_ui = True
 show_action = False
 

play_acrobat.py

@@ -10,9 +10,11 @@
 import ops
 import sys
 import itertools
+from gym.wrappers import Monitor
 
 # Setup the environment
 env = gym.make('Acrobot-v1')
+env = Monitor(env=env, directory="./Results/Videos/Acrobat", resume=True)
 
 # Placeholders
 # TODO: make the shape of X_input generalized

play_breakout.py

@@ -11,9 +11,11 @@
 import mission_control_breakout as mc
 import ops
 import matplotlib.pyplot as plt
+from gym.wrappers import Monitor
 
 # Setup the environment
 env = gym.make('BreakoutDeterministic-v4')
+env = Monitor(env=env, directory="./Results/Videos/Breakout", resume=True)
 
 # Placeholders
 X_input = tf.placeholder(dtype=tf.float32, shape=[None, 84, 84, 4], name='Observations')
@@ -167,7 +169,7 @@ def play(sess, agent, no_plays, log_dir=None, show_ui=False, show_action=False):
         while not done:
             if show_ui:
                 env.render()
-            if np.random.rand() < 0.01:
+            if np.random.rand() < 0.05:
                 action = env.action_space.sample()
             else:
                 action = np.argmax(sess.run(agent, feed_dict={X_input: state}))

play_cartpole.py

@@ -10,9 +10,11 @@
 import ops
 import sys
 import itertools
+from gym.wrappers import Monitor
 
 # Setup the environment
 env = gym.make('CartPole-v1')
+env = Monitor(env=env, directory="./Results/Videos/CartPole", resume=True)
 
 # Placeholders
 # TODO: make the shape of X_input generalized