diff --git a/Reinforcement (Q-)Learning with PyTorch.ipynb b/Reinforcement (Q-)Learning with PyTorch.ipynb
index 30cd93c59ff..6b05a78e1fe 100644
--- a/Reinforcement (Q-)Learning with PyTorch.ipynb	
+++ b/Reinforcement (Q-)Learning with PyTorch.ipynb	
@@ -60,7 +60,8 @@
     "import torch.nn.functional as F\n",
     "import torchvision.transforms as T\n",
     "\n",
-    "env = gym.make('CartPole-v0')"
+    "env = gym.make('CartPole-v0')\n",
+    "env = env.unwrapped"
    ]
   },
   {