Update 04_ppo_with_sb3.ipynb

Author: daphne-cornelisse

examples/04_ppo_with_sb3.ipynb

@@ -130,7 +130,7 @@
     "\n",
     "Now take a look at information you've logged over training; did we learn?\n",
     "\n",
-    "One important metric for assess the effectiveness of your policy is the average cumulative reward per episode. In our case, the **maximum** achievable return per episode is approximately between 9 and 10 (it varies per traffic scene and per agent). With the configurations above, your policy should approach this value in 150,000 steps. Here, steps (the `global_step`) represents the total number of **frames** our policy network has seen, you can think of it as the accumulated experience."
+    "One important metric for assess the effectiveness of your policy is the average cumulative reward per episode. In our case, the **maximum** achievable return per episode is 1 per agent. With the configurations above, your policy should approach this value in 150,000 steps. Here, steps (the `global_step`) represents the total number of **frames** our policy network has seen, you can think of it as the accumulated experience."
    ]
   }
  ],