Merge pull request #4 from Unity-Technologies/ppo-timestep

Fix timestep for PPO.ipynb

python/PPO.ipynb

    "outputs": [],
    "source": [
     "### General parameters\n",
-    "max_steps = 10000 # Set maximum number of steps to run environment.\n",
+    "max_steps = 5e5 # Set maximum number of steps to run environment.\n",
     "run_path = \"ppo\" # The sub-directory name for model and summary statistics\n",
     "load_model = False # Whether to load a saved model.\n",
     "train_model = True # Whether to train the model.\n",