Minor bug fix

Added missing declaration for rewards_list, otherwise it will fail the first time you try to train it

Author: jsztompka

DQN/Space Invaders/DQN Atari Space Invaders.ipynb

@@ -681,6 +681,7 @@
    "source": [
     "# Saver will help us to save our model\n",
     "saver = tf.train.Saver()\n",
+    "rewards_list=[]\n",
     "\n",
     "if training == True:\n",
     "    with tf.Session() as sess:\n",