update

contents/5_Deep_Q_Network/DQN_modified.py

@@ -53,13 +53,14 @@ def __init__(
 
         t_params = tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES, scope='target_net')
         e_params = tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES, scope='eval_net')
-        self.target_replace_op = [tf.assign(t, e) for t, e in zip(t_params, e_params)]
+
+        with tf.variable_scope('soft_replacement'):
+            self.target_replace_op = [tf.assign(t, e) for t, e in zip(t_params, e_params)]
 
         self.sess = tf.Session()
 
         if output_graph:
             # $ tensorboard --logdir=logs
-            # tf.train.SummaryWriter soon be deprecated, use following
             tf.summary.FileWriter("logs/", self.sess.graph)
 
         self.sess.run(tf.global_variables_initializer())
@@ -77,16 +78,16 @@ def _build_net(self):
         # ------------------ build evaluate_net ------------------
         with tf.variable_scope('eval_net'):
             e1 = tf.layers.dense(self.s, 20, tf.nn.relu, kernel_initializer=w_initializer,
-                                 bias_initializer=b_initializer)
+                                 bias_initializer=b_initializer, name='e1')
             self.q_eval = tf.layers.dense(e1, self.n_actions, kernel_initializer=w_initializer,
-                                          bias_initializer=b_initializer)
+                                          bias_initializer=b_initializer, name='q')
 
         # ------------------ build target_net ------------------
         with tf.variable_scope('target_net'):
             t1 = tf.layers.dense(self.s_, 20, tf.nn.relu, kernel_initializer=w_initializer,
-                                 bias_initializer=b_initializer)
+                                 bias_initializer=b_initializer, name='t1')
             self.q_next = tf.layers.dense(t1, self.n_actions, kernel_initializer=w_initializer,
-                                          bias_initializer=b_initializer)
+                                          bias_initializer=b_initializer, name='t2')
 
         with tf.variable_scope('q_target'):
             q_target = self.r + self.gamma * tf.reduce_max(self.q_next, axis=1, name='Qmax_s_')    # shape=(None, )