From 7f7be770989e3bc587bfb74fd649d9b5450b1b14 Mon Sep 17 00:00:00 2001
From: Nipun Batra <nipunbatra0@gmail.com>
Date: Mon, 11 Dec 2023 15:18:11 +0530
Subject: [PATCH 1/5] added new notebook on RL

---
 .DS_Store                   | Bin 8196 -> 8196 bytes
 posts/2023-Dec-11-gym.ipynb | 537 ++++++++++++++++++++++++++++++++++++
 2 files changed, 537 insertions(+)
 create mode 100644 posts/2023-Dec-11-gym.ipynb

diff --git a/.DS_Store b/.DS_Store
index defd813fa1e3ef1a1fda591f6476e6fa5a4d12b6..0736f0675d35d8c72799a749fbef0254fbd8f691 100644
GIT binary patch
delta 320
zcmZp1XmQveE-1V+sURn_xWvHV8Y2@k3o9EtJNslU!Dt^&&UgXI>S{wnGaUsJBjZ{f
zg=#}%GeaE(Gh?ILT22m8Wqs?Q`0SkAy!>tkFkoba&<wm#8b<X@z9i_RBqb#iFCd(h
zn4FznlwX>c;+&t8U*wcini`RroSzq*S(Q53UT9Z+ya0c3Mt->`P%NM*wJbBWJUYE7
zGbOknF*!3YUD^Hv#8hR602o69!Vq9khFPfW(ANNFa|keo<mKn3a&js=ctd457_=E2
z8N3*R88R5k8TuGzF|1+O!Elt}7Q-8cZ;X;GjAD#RjM|JQjJAySj2?`Ej6sYcjFF5{
Un>&PM7&kBGNnzg1F7cNg0NfQu`v3p{

delta 264
zcmZp1XmQveE-1V!sURn_xWvHV8Y2@k3o9EtJLhCA!Dw9$&UgW->S{v+BNH733yWGE
z1t{Clz}T|3mXkwNS>HM+K07BjFTZE<MnQST-pLmQowS9dW#R>dlM<7&(~I&;^HQAi
zbMlLva!OO<1^A0I^2<Gw^Ya3VQp++^%O{%(?c$8c1PTOaR;5m66c%OVoGdRa7PZNg
z0R%V%7(?>%^HMpL9Rk1t49X7P5C#JWgEoUBLoh=OLm@*0!*qtl3|kovGMr<0%<z@r
yKO--r7^4!SHlq%sDWf@~1EVLS7o!hjAY;(xc3~OD&1+@CnI|^zY-X4E%MJjuDnrWv

diff --git a/posts/2023-Dec-11-gym.ipynb b/posts/2023-Dec-11-gym.ipynb
new file mode 100644
index 0000000..aa1114c
--- /dev/null
+++ b/posts/2023-Dec-11-gym.ipynb
@@ -0,0 +1,537 @@
+{
+ "cells": [
+  {
+   "cell_type": "raw",
+   "id": "12a1fd37",
+   "metadata": {},
+   "source": [
+    "---\n",
+    "author: Nipun Batra\n",
+    "badges: true\n",
+    "categories:\n",
+    "- ML\n",
+    "date: '2023-12-11'\n",
+    "output-file: rl.html\n",
+    "title: Super Reinforcement Learning\n",
+    "toc: true\n",
+    "\n",
+    "---\n",
+    "\n"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "c576bebd",
+   "metadata": {},
+   "source": [
+    "### Basic Imports"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "id": "f41ca63d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "import torch\n",
+    "import gymnasium as gym\n",
+    "import numpy as np\n",
+    "\n",
+    "%matplotlib inline\n",
+    "%config InlineBackend.figure_format='retina'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['CartPole-v0',\n",
+       " 'CartPole-v1',\n",
+       " 'MountainCar-v0',\n",
+       " 'MountainCarContinuous-v0',\n",
+       " 'Pendulum-v1',\n",
+       " 'Acrobot-v1',\n",
+       " 'phys2d/CartPole-v0',\n",
+       " 'phys2d/CartPole-v1',\n",
+       " 'phys2d/Pendulum-v0',\n",
+       " 'LunarLander-v2',\n",
+       " 'LunarLanderContinuous-v2',\n",
+       " 'BipedalWalker-v3',\n",
+       " 'BipedalWalkerHardcore-v3',\n",
+       " 'CarRacing-v2',\n",
+       " 'Blackjack-v1',\n",
+       " 'FrozenLake-v1',\n",
+       " 'FrozenLake8x8-v1',\n",
+       " 'CliffWalking-v0',\n",
+       " 'Taxi-v3',\n",
+       " 'tabular/Blackjack-v0',\n",
+       " 'tabular/CliffWalking-v0',\n",
+       " 'Reacher-v2',\n",
+       " 'Reacher-v4',\n",
+       " 'Pusher-v2',\n",
+       " 'Pusher-v4',\n",
+       " 'InvertedPendulum-v2',\n",
+       " 'InvertedPendulum-v4',\n",
+       " 'InvertedDoublePendulum-v2',\n",
+       " 'InvertedDoublePendulum-v4',\n",
+       " 'HalfCheetah-v2',\n",
+       " 'HalfCheetah-v3',\n",
+       " 'HalfCheetah-v4',\n",
+       " 'Hopper-v2',\n",
+       " 'Hopper-v3',\n",
+       " 'Hopper-v4',\n",
+       " 'Swimmer-v2',\n",
+       " 'Swimmer-v3',\n",
+       " 'Swimmer-v4',\n",
+       " 'Walker2d-v2',\n",
+       " 'Walker2d-v3',\n",
+       " 'Walker2d-v4',\n",
+       " 'Ant-v2',\n",
+       " 'Ant-v3',\n",
+       " 'Ant-v4',\n",
+       " 'Humanoid-v2',\n",
+       " 'Humanoid-v3',\n",
+       " 'Humanoid-v4',\n",
+       " 'HumanoidStandup-v2',\n",
+       " 'HumanoidStandup-v4',\n",
+       " 'GymV21Environment-v0',\n",
+       " 'GymV26Environment-v0']"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# List of environments\n",
+    "list(gym.envs.registry.keys())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/nipun/miniconda3/lib/python3.9/site-packages/gymnasium/envs/registration.py:513: DeprecationWarning: \u001b[33mWARN: The environment CartPole-v0 is out of date. You should consider upgrading to version `v1`.\u001b[0m\n",
+      "  logger.deprecation(\n"
+     ]
+    }
+   ],
+   "source": [
+    "env = gym.make(\"CartPole-v0\", render_mode=\"human\")\n",
+    "observation, info = env.reset(seed=42)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Discrete(2)"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "env.action_space"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "env.action_space.sample()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "env.action_space.sample()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Box([-4.8000002e+00 -3.4028235e+38 -4.1887903e-01 -3.4028235e+38], [4.8000002e+00 3.4028235e+38 4.1887903e-01 3.4028235e+38], (4,), float32)\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(env.observation_space)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for _ in range(100):\n",
+    "   action = env.action_space.sample()  # this is where you would insert your policy\n",
+    "   observation, reward, terminated, truncated, info = env.step(action)\n",
+    "\n",
+    "   if terminated or truncated:\n",
+    "      observation, info = env.reset()\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Episode finished after 8 timesteps\n",
+      "Episode finished after 17 timesteps\n",
+      "Episode finished after 27 timesteps\n",
+      "Episode finished after 37 timesteps\n",
+      "Episode finished after 46 timesteps\n",
+      "Episode finished after 54 timesteps\n",
+      "Episode finished after 64 timesteps\n",
+      "Episode finished after 73 timesteps\n",
+      "Episode finished after 83 timesteps\n",
+      "Episode finished after 93 timesteps\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Now, let us take only the action 1\n",
+    "observation, info = env.reset(seed=42)\n",
+    "for i in range(100):\n",
+    "   action = 0  # this is where you would insert your policy\n",
+    "   observation, reward, terminated, truncated, info = env.step(action)\n",
+    "\n",
+    "   if terminated or truncated:\n",
+    "        print(\"Episode finished after {} timesteps\".format(i+1))\n",
+    "        observation, info = env.reset()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Create a q table\n",
+    "num_states = env.observation_space.shape[0]\n",
+    "num_actions = env.action_space.n\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 49,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Define the number of bins for each dimension\n",
+    "num_bins = [16, 16, 16, 16]  # Adjust these values based on your preference\n",
+    "\n",
+    "# Discretize the continuous state space\n",
+    "def discretize_state(state, num_bins):\n",
+    "    state_discrete = []\n",
+    "    for i in range(len(state)):\n",
+    "        bins = np.linspace(env.observation_space.low[i], env.observation_space.high[i], num_bins[i] + 1)[1:-1]\n",
+    "        state_discrete.append(np.digitize(state[i], bins))\n",
+    "    return tuple(state_discrete)\n",
+    "\n",
+    "\n",
+    "# Initialize Q-table with zeros\n",
+    "q_table = torch.zeros(num_bins + [env.action_space.n])\n",
+    "\n",
+    "def update_q_table(q_table, state, action, reward, next_state, learning_rate, discount_factor):\n",
+    "    state = tuple(state)\n",
+    "    next_state = tuple(next_state)\n",
+    "    q_table[state][action] = (1 - learning_rate) * q_table[state][action] + learning_rate * (reward + discount_factor * torch.max(q_table[next_state]))\n",
+    "\n",
+    "    return q_table"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 50,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Learn the q table\n",
+    "\n",
+    "# Hyperparameters\n",
+    "learning_rate = 0.1\n",
+    "discount_factor = 0.9\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Learn q-table without epsilon greedy approach and print rewards\n",
+    "num_episodes = 10\n",
+    "render_mode = False\n",
+    "\n",
+    "    "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 52,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "All possible states:\n",
+      "Box([-4.8000002e+00 -3.4028235e+38 -4.1887903e-01 -3.4028235e+38], [4.8000002e+00 3.4028235e+38 4.1887903e-01 3.4028235e+38], (4,), float32)\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Print all possible states\n",
+    "print(\"All possible states:\")\n",
+    "print(env.observation_space)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 53,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Sample State: [-8.2706535e-01 -2.8081686e+38 -3.0893192e-01 -9.7170763e+37]\n",
+      "Sample State: [ 1.9736170e+00 -2.1225781e+38 -1.8008137e-01  4.6366904e+36]\n",
+      "Sample State: [-1.7399893e+00  5.8333194e+37 -5.9940118e-02 -2.7254928e+38]\n",
+      "Sample State: [-3.8519826e+00  2.2531339e+38  6.3289993e-02  9.1026116e+37]\n",
+      "Sample State: [ 2.5397954e+00 -1.2115095e+38  2.4115197e-01 -1.3963663e+38]\n",
+      "Sample State: [3.6893094e+00 3.1494561e+37 2.1339690e-02 2.6938287e+38]\n",
+      "Sample State: [ 3.2560833e+00  1.3971532e+38  1.6714491e-02 -2.7917394e+38]\n",
+      "Sample State: [5.7502143e-02 1.3929552e+38 3.0328104e-01 3.1058507e+38]\n",
+      "Sample State: [ 7.9759490e-01 -2.2935894e+38 -4.7575418e-02 -9.9444867e+37]\n",
+      "Sample State: [-4.6630378e+00  4.8043669e+37 -1.4580821e-01  2.4613882e+38]\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Print some sample states\n",
+    "for _ in range(10):\n",
+    "    state = env.observation_space.sample()\n",
+    "    print(\"Sample State:\", state)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 54,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "TypeError",
+     "evalue": "'<' not supported between instances of 'dict' and 'dict'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:57\u001b[0m, in \u001b[0;36m_wrapfunc\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     56\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m---> 57\u001b[0m     \u001b[39mreturn\u001b[39;00m bound(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n\u001b[1;32m     58\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mTypeError\u001b[39;00m:\n\u001b[1;32m     59\u001b[0m     \u001b[39m# A TypeError occurs if the object does have such a method in its\u001b[39;00m\n\u001b[1;32m     60\u001b[0m     \u001b[39m# class, but its signature is not identical to that of NumPy's. This\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     64\u001b[0m     \u001b[39m# Call _wrapit from within the except clause to ensure a potential\u001b[39;00m\n\u001b[1;32m     65\u001b[0m     \u001b[39m# exception has a traceback chain.\u001b[39;00m\n",
+      "\u001b[0;31mTypeError\u001b[0m: '<' not supported between instances of 'dict' and 'dict'",
+      "\nDuring handling of the above exception, another exception occurred:\n",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 18\u001b[0m line \u001b[0;36m3\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m \u001b[39m# Training loop\u001b[39;00m\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=1'>2</a>\u001b[0m \u001b[39mfor\u001b[39;00m episode \u001b[39min\u001b[39;00m \u001b[39mrange\u001b[39m(num_episodes):\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=2'>3</a>\u001b[0m     state \u001b[39m=\u001b[39m discretize_state(env\u001b[39m.\u001b[39;49mreset(), num_bins)\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=4'>5</a>\u001b[0m     \u001b[39mwhile\u001b[39;00m \u001b[39mTrue\u001b[39;00m:\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=5'>6</a>\u001b[0m         \u001b[39m# Choose action using the current Q-table\u001b[39;00m\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=6'>7</a>\u001b[0m         action \u001b[39m=\u001b[39m torch\u001b[39m.\u001b[39margmax(q_table[state])\u001b[39m.\u001b[39mitem()\n",
+      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 18\u001b[0m line \u001b[0;36m9\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=6'>7</a>\u001b[0m \u001b[39mfor\u001b[39;00m i \u001b[39min\u001b[39;00m \u001b[39mrange\u001b[39m(\u001b[39mlen\u001b[39m(state)):\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=7'>8</a>\u001b[0m     bins \u001b[39m=\u001b[39m np\u001b[39m.\u001b[39mlinspace(env\u001b[39m.\u001b[39mobservation_space\u001b[39m.\u001b[39mlow[i], env\u001b[39m.\u001b[39mobservation_space\u001b[39m.\u001b[39mhigh[i], num_bins[i] \u001b[39m+\u001b[39m \u001b[39m1\u001b[39m)[\u001b[39m1\u001b[39m:\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m]\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=8'>9</a>\u001b[0m     state_discrete\u001b[39m.\u001b[39mappend(np\u001b[39m.\u001b[39;49mdigitize(state[i], bins))\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=9'>10</a>\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mtuple\u001b[39m(state_discrete)\n",
+      "File \u001b[0;32m<__array_function__ internals>:200\u001b[0m, in \u001b[0;36mdigitize\u001b[0;34m(*args, **kwargs)\u001b[0m\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/lib/function_base.py:5614\u001b[0m, in \u001b[0;36mdigitize\u001b[0;34m(x, bins, right)\u001b[0m\n\u001b[1;32m   5612\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mlen\u001b[39m(bins) \u001b[39m-\u001b[39m _nx\u001b[39m.\u001b[39msearchsorted(bins[::\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m], x, side\u001b[39m=\u001b[39mside)\n\u001b[1;32m   5613\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m-> 5614\u001b[0m     \u001b[39mreturn\u001b[39;00m _nx\u001b[39m.\u001b[39;49msearchsorted(bins, x, side\u001b[39m=\u001b[39;49mside)\n",
+      "File \u001b[0;32m<__array_function__ internals>:200\u001b[0m, in \u001b[0;36msearchsorted\u001b[0;34m(*args, **kwargs)\u001b[0m\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:1413\u001b[0m, in \u001b[0;36msearchsorted\u001b[0;34m(a, v, side, sorter)\u001b[0m\n\u001b[1;32m   1345\u001b[0m \u001b[39m@array_function_dispatch\u001b[39m(_searchsorted_dispatcher)\n\u001b[1;32m   1346\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39msearchsorted\u001b[39m(a, v, side\u001b[39m=\u001b[39m\u001b[39m'\u001b[39m\u001b[39mleft\u001b[39m\u001b[39m'\u001b[39m, sorter\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m):\n\u001b[1;32m   1347\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m   1348\u001b[0m \u001b[39m    Find indices where elements should be inserted to maintain order.\u001b[39;00m\n\u001b[1;32m   1349\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   1411\u001b[0m \n\u001b[1;32m   1412\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m-> 1413\u001b[0m     \u001b[39mreturn\u001b[39;00m _wrapfunc(a, \u001b[39m'\u001b[39;49m\u001b[39msearchsorted\u001b[39;49m\u001b[39m'\u001b[39;49m, v, side\u001b[39m=\u001b[39;49mside, sorter\u001b[39m=\u001b[39;49msorter)\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:66\u001b[0m, in \u001b[0;36m_wrapfunc\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     57\u001b[0m     \u001b[39mreturn\u001b[39;00m bound(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwds)\n\u001b[1;32m     58\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mTypeError\u001b[39;00m:\n\u001b[1;32m     59\u001b[0m     \u001b[39m# A TypeError occurs if the object does have such a method in its\u001b[39;00m\n\u001b[1;32m     60\u001b[0m     \u001b[39m# class, but its signature is not identical to that of NumPy's. This\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     64\u001b[0m     \u001b[39m# Call _wrapit from within the except clause to ensure a potential\u001b[39;00m\n\u001b[1;32m     65\u001b[0m     \u001b[39m# exception has a traceback chain.\u001b[39;00m\n\u001b[0;32m---> 66\u001b[0m     \u001b[39mreturn\u001b[39;00m _wrapit(obj, method, \u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:43\u001b[0m, in \u001b[0;36m_wrapit\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     41\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mAttributeError\u001b[39;00m:\n\u001b[1;32m     42\u001b[0m     wrap \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m\n\u001b[0;32m---> 43\u001b[0m result \u001b[39m=\u001b[39m \u001b[39mgetattr\u001b[39;49m(asarray(obj), method)(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n\u001b[1;32m     44\u001b[0m \u001b[39mif\u001b[39;00m wrap:\n\u001b[1;32m     45\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39misinstance\u001b[39m(result, mu\u001b[39m.\u001b[39mndarray):\n",
+      "\u001b[0;31mTypeError\u001b[0m: '<' not supported between instances of 'dict' and 'dict'"
+     ]
+    }
+   ],
+   "source": [
+    "# Training loop\n",
+    "for episode in range(num_episodes):\n",
+    "    state = discretize_state(env.reset(), num_bins)\n",
+    "\n",
+    "    while True:\n",
+    "        # Choose action using the current Q-table\n",
+    "        action = torch.argmax(q_table[state]).item()\n",
+    "\n",
+    "        # Take the chosen action and observe the next state and reward\n",
+    "        next_state, reward, done, _ = env.step(action)\n",
+    "        next_state = discretize_state(next_state, num_bins)\n",
+    "\n",
+    "        # Update the Q-table using the Q-learning update rule\n",
+    "        q_table = update_q_table(q_table, state, action, reward, next_state, learning_rate, discount_factor)\n",
+    "\n",
+    "        state = next_state\n",
+    "\n",
+    "        if done:\n",
+    "            break\n",
+    "\n",
+    "# Print the learned Q-table\n",
+    "print(\"Learned Q-table:\")\n",
+    "print(q_table)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 55,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "TypeError",
+     "evalue": "'<' not supported between instances of 'dict' and 'dict'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:57\u001b[0m, in \u001b[0;36m_wrapfunc\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     56\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m---> 57\u001b[0m     \u001b[39mreturn\u001b[39;00m bound(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n\u001b[1;32m     58\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mTypeError\u001b[39;00m:\n\u001b[1;32m     59\u001b[0m     \u001b[39m# A TypeError occurs if the object does have such a method in its\u001b[39;00m\n\u001b[1;32m     60\u001b[0m     \u001b[39m# class, but its signature is not identical to that of NumPy's. This\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     64\u001b[0m     \u001b[39m# Call _wrapit from within the except clause to ensure a potential\u001b[39;00m\n\u001b[1;32m     65\u001b[0m     \u001b[39m# exception has a traceback chain.\u001b[39;00m\n",
+      "\u001b[0;31mTypeError\u001b[0m: '<' not supported between instances of 'dict' and 'dict'",
+      "\nDuring handling of the above exception, another exception occurred:\n",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 19\u001b[0m line \u001b[0;36m1\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X65sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m discretize_state(env\u001b[39m.\u001b[39;49mreset(), num_bins)\n",
+      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 19\u001b[0m line \u001b[0;36m9\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X65sZmlsZQ%3D%3D?line=6'>7</a>\u001b[0m \u001b[39mfor\u001b[39;00m i \u001b[39min\u001b[39;00m \u001b[39mrange\u001b[39m(\u001b[39mlen\u001b[39m(state)):\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X65sZmlsZQ%3D%3D?line=7'>8</a>\u001b[0m     bins \u001b[39m=\u001b[39m np\u001b[39m.\u001b[39mlinspace(env\u001b[39m.\u001b[39mobservation_space\u001b[39m.\u001b[39mlow[i], env\u001b[39m.\u001b[39mobservation_space\u001b[39m.\u001b[39mhigh[i], num_bins[i] \u001b[39m+\u001b[39m \u001b[39m1\u001b[39m)[\u001b[39m1\u001b[39m:\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m]\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X65sZmlsZQ%3D%3D?line=8'>9</a>\u001b[0m     state_discrete\u001b[39m.\u001b[39mappend(np\u001b[39m.\u001b[39;49mdigitize(state[i], bins))\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X65sZmlsZQ%3D%3D?line=9'>10</a>\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mtuple\u001b[39m(state_discrete)\n",
+      "File \u001b[0;32m<__array_function__ internals>:200\u001b[0m, in \u001b[0;36mdigitize\u001b[0;34m(*args, **kwargs)\u001b[0m\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/lib/function_base.py:5614\u001b[0m, in \u001b[0;36mdigitize\u001b[0;34m(x, bins, right)\u001b[0m\n\u001b[1;32m   5612\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mlen\u001b[39m(bins) \u001b[39m-\u001b[39m _nx\u001b[39m.\u001b[39msearchsorted(bins[::\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m], x, side\u001b[39m=\u001b[39mside)\n\u001b[1;32m   5613\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m-> 5614\u001b[0m     \u001b[39mreturn\u001b[39;00m _nx\u001b[39m.\u001b[39;49msearchsorted(bins, x, side\u001b[39m=\u001b[39;49mside)\n",
+      "File \u001b[0;32m<__array_function__ internals>:200\u001b[0m, in \u001b[0;36msearchsorted\u001b[0;34m(*args, **kwargs)\u001b[0m\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:1413\u001b[0m, in \u001b[0;36msearchsorted\u001b[0;34m(a, v, side, sorter)\u001b[0m\n\u001b[1;32m   1345\u001b[0m \u001b[39m@array_function_dispatch\u001b[39m(_searchsorted_dispatcher)\n\u001b[1;32m   1346\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39msearchsorted\u001b[39m(a, v, side\u001b[39m=\u001b[39m\u001b[39m'\u001b[39m\u001b[39mleft\u001b[39m\u001b[39m'\u001b[39m, sorter\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m):\n\u001b[1;32m   1347\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m   1348\u001b[0m \u001b[39m    Find indices where elements should be inserted to maintain order.\u001b[39;00m\n\u001b[1;32m   1349\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   1411\u001b[0m \n\u001b[1;32m   1412\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m-> 1413\u001b[0m     \u001b[39mreturn\u001b[39;00m _wrapfunc(a, \u001b[39m'\u001b[39;49m\u001b[39msearchsorted\u001b[39;49m\u001b[39m'\u001b[39;49m, v, side\u001b[39m=\u001b[39;49mside, sorter\u001b[39m=\u001b[39;49msorter)\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:66\u001b[0m, in \u001b[0;36m_wrapfunc\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     57\u001b[0m     \u001b[39mreturn\u001b[39;00m bound(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwds)\n\u001b[1;32m     58\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mTypeError\u001b[39;00m:\n\u001b[1;32m     59\u001b[0m     \u001b[39m# A TypeError occurs if the object does have such a method in its\u001b[39;00m\n\u001b[1;32m     60\u001b[0m     \u001b[39m# class, but its signature is not identical to that of NumPy's. This\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     64\u001b[0m     \u001b[39m# Call _wrapit from within the except clause to ensure a potential\u001b[39;00m\n\u001b[1;32m     65\u001b[0m     \u001b[39m# exception has a traceback chain.\u001b[39;00m\n\u001b[0;32m---> 66\u001b[0m     \u001b[39mreturn\u001b[39;00m _wrapit(obj, method, \u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:43\u001b[0m, in \u001b[0;36m_wrapit\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     41\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mAttributeError\u001b[39;00m:\n\u001b[1;32m     42\u001b[0m     wrap \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m\n\u001b[0;32m---> 43\u001b[0m result \u001b[39m=\u001b[39m \u001b[39mgetattr\u001b[39;49m(asarray(obj), method)(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n\u001b[1;32m     44\u001b[0m \u001b[39mif\u001b[39;00m wrap:\n\u001b[1;32m     45\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39misinstance\u001b[39m(result, mu\u001b[39m.\u001b[39mndarray):\n",
+      "\u001b[0;31mTypeError\u001b[0m: '<' not supported between instances of 'dict' and 'dict'"
+     ]
+    }
+   ],
+   "source": [
+    "discretize_state(env.reset(), num_bins)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "tensor([[ 0.0982, -0.0016],\n",
+       "        [ 0.0100,  0.0040],\n",
+       "        [-0.0006,  0.0108],\n",
+       "        [ 0.0028,  0.0059]])"
+      ]
+     },
+     "execution_count": 36,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "q_table"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "IndexError",
+     "evalue": "tensors used as indices must be long, int, byte or bool tensors",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
+      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 22\u001b[0m line \u001b[0;36m1\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X60sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m q_table[state]\n",
+      "\u001b[0;31mIndexError\u001b[0m: tensors used as indices must be long, int, byte or bool tensors"
+     ]
+    }
+   ],
+   "source": [
+    "q_table[state]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.15"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

From cd0a4bf69f326ed1d751325fd8ddf615919cba76 Mon Sep 17 00:00:00 2001
From: Nipun Batra <nipunbatra0@gmail.com>
Date: Mon, 11 Dec 2023 15:56:50 +0530
Subject: [PATCH 2/5] added

---
 posts/2023-Dec-11-gym.ipynb | 535 +++++++++++++++++++++++++++++-------
 1 file changed, 435 insertions(+), 100 deletions(-)

diff --git a/posts/2023-Dec-11-gym.ipynb b/posts/2023-Dec-11-gym.ipynb
index aa1114c..a443131 100644
--- a/posts/2023-Dec-11-gym.ipynb
+++ b/posts/2023-Dec-11-gym.ipynb
@@ -273,12 +273,42 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 49,
+   "execution_count": 59,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "i = 1\n",
+    "bins = np.linspace(env.observation_space.low[i], env.observation_space.high[i], num_bins[i] + 1)[1:-1]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 63,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([7])"
+      ]
+     },
+     "execution_count": 63,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "np.digitize([env.observation_space.low[2]], bins)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 99,
    "metadata": {},
    "outputs": [],
    "source": [
     "# Define the number of bins for each dimension\n",
-    "num_bins = [16, 16, 16, 16]  # Adjust these values based on your preference\n",
+    "num_bins = [3, 3, 3, 3]  # Adjust these values based on your preference\n",
     "\n",
     "# Discretize the continuous state space\n",
     "def discretize_state(state, num_bins):\n",
@@ -290,7 +320,7 @@
     "\n",
     "\n",
     "# Initialize Q-table with zeros\n",
-    "q_table = torch.zeros(num_bins + [env.action_space.n])\n",
+    "q_table = 1e-2*torch.randn(num_bins + [env.action_space.n])\n",
     "\n",
     "def update_q_table(q_table, state, action, reward, next_state, learning_rate, discount_factor):\n",
     "    state = tuple(state)\n",
@@ -302,7 +332,143 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 50,
+   "execution_count": 100,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "tensor([[[[[ 0.0003, -0.0164],\n",
+       "           [-0.0066, -0.0123],\n",
+       "           [-0.0020, -0.0066]],\n",
+       "\n",
+       "          [[ 0.0140,  0.0090],\n",
+       "           [-0.0116,  0.0231],\n",
+       "           [ 0.0031, -0.0010]],\n",
+       "\n",
+       "          [[ 0.0121,  0.0208],\n",
+       "           [-0.0023, -0.0126],\n",
+       "           [ 0.0163,  0.0042]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 0.0097, -0.0119],\n",
+       "           [-0.0123,  0.0099],\n",
+       "           [-0.0013,  0.0010]],\n",
+       "\n",
+       "          [[ 0.0209,  0.0026],\n",
+       "           [-0.0195,  0.0041],\n",
+       "           [-0.0099, -0.0078]],\n",
+       "\n",
+       "          [[-0.0035,  0.0008],\n",
+       "           [ 0.0115, -0.0144],\n",
+       "           [-0.0160,  0.0070]]],\n",
+       "\n",
+       "\n",
+       "         [[[-0.0112, -0.0223],\n",
+       "           [-0.0097,  0.0071],\n",
+       "           [ 0.0129,  0.0088]],\n",
+       "\n",
+       "          [[ 0.0082,  0.0146],\n",
+       "           [-0.0008, -0.0076],\n",
+       "           [ 0.0041, -0.0039]],\n",
+       "\n",
+       "          [[ 0.0014,  0.0015],\n",
+       "           [ 0.0046,  0.0244],\n",
+       "           [-0.0038,  0.0122]]]],\n",
+       "\n",
+       "\n",
+       "\n",
+       "        [[[[ 0.0010,  0.0007],\n",
+       "           [-0.0169, -0.0045],\n",
+       "           [-0.0164, -0.0050]],\n",
+       "\n",
+       "          [[-0.0079, -0.0085],\n",
+       "           [ 0.0078, -0.0072],\n",
+       "           [-0.0204, -0.0249]],\n",
+       "\n",
+       "          [[ 0.0093,  0.0161],\n",
+       "           [ 0.0042,  0.0019],\n",
+       "           [-0.0062, -0.0085]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 0.0038,  0.0045],\n",
+       "           [ 0.0106,  0.0087],\n",
+       "           [ 0.0012, -0.0024]],\n",
+       "\n",
+       "          [[ 0.0148,  0.0127],\n",
+       "           [ 0.0059,  0.0148],\n",
+       "           [ 0.0075, -0.0032]],\n",
+       "\n",
+       "          [[-0.0160,  0.0056],\n",
+       "           [-0.0156, -0.0129],\n",
+       "           [ 0.0055, -0.0089]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 0.0071,  0.0224],\n",
+       "           [-0.0014, -0.0146],\n",
+       "           [ 0.0055, -0.0037]],\n",
+       "\n",
+       "          [[ 0.0101,  0.0052],\n",
+       "           [ 0.0103, -0.0091],\n",
+       "           [ 0.0132,  0.0055]],\n",
+       "\n",
+       "          [[-0.0008,  0.0241],\n",
+       "           [-0.0195,  0.0115],\n",
+       "           [-0.0165, -0.0068]]]],\n",
+       "\n",
+       "\n",
+       "\n",
+       "        [[[[ 0.0028,  0.0051],\n",
+       "           [ 0.0056, -0.0025],\n",
+       "           [ 0.0075, -0.0114]],\n",
+       "\n",
+       "          [[-0.0013, -0.0074],\n",
+       "           [ 0.0086, -0.0125],\n",
+       "           [-0.0140, -0.0217]],\n",
+       "\n",
+       "          [[-0.0125,  0.0022],\n",
+       "           [-0.0010,  0.0142],\n",
+       "           [ 0.0019, -0.0038]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 0.0041,  0.0007],\n",
+       "           [-0.0042,  0.0151],\n",
+       "           [ 0.0067, -0.0067]],\n",
+       "\n",
+       "          [[ 0.0072, -0.0263],\n",
+       "           [-0.0084,  0.0122],\n",
+       "           [-0.0015, -0.0004]],\n",
+       "\n",
+       "          [[-0.0149,  0.0111],\n",
+       "           [ 0.0030, -0.0091],\n",
+       "           [-0.0121,  0.0127]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 0.0050, -0.0017],\n",
+       "           [-0.0040,  0.0099],\n",
+       "           [-0.0021, -0.0062]],\n",
+       "\n",
+       "          [[-0.0026, -0.0048],\n",
+       "           [-0.0002,  0.0143],\n",
+       "           [-0.0086,  0.0055]],\n",
+       "\n",
+       "          [[ 0.0176,  0.0029],\n",
+       "           [-0.0067,  0.0101],\n",
+       "           [ 0.0047,  0.0129]]]]])"
+      ]
+     },
+     "execution_count": 100,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "q_table"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 101,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -315,12 +481,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 51,
+   "execution_count": 102,
    "metadata": {},
    "outputs": [],
    "source": [
     "# Learn q-table without epsilon greedy approach and print rewards\n",
-    "num_episodes = 10\n",
+    "num_episodes = 50\n",
     "render_mode = False\n",
     "\n",
     "    "
@@ -328,7 +494,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 52,
+   "execution_count": 103,
    "metadata": {},
    "outputs": [
     {
@@ -348,23 +514,23 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 53,
+   "execution_count": 104,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Sample State: [-8.2706535e-01 -2.8081686e+38 -3.0893192e-01 -9.7170763e+37]\n",
-      "Sample State: [ 1.9736170e+00 -2.1225781e+38 -1.8008137e-01  4.6366904e+36]\n",
-      "Sample State: [-1.7399893e+00  5.8333194e+37 -5.9940118e-02 -2.7254928e+38]\n",
-      "Sample State: [-3.8519826e+00  2.2531339e+38  6.3289993e-02  9.1026116e+37]\n",
-      "Sample State: [ 2.5397954e+00 -1.2115095e+38  2.4115197e-01 -1.3963663e+38]\n",
-      "Sample State: [3.6893094e+00 3.1494561e+37 2.1339690e-02 2.6938287e+38]\n",
-      "Sample State: [ 3.2560833e+00  1.3971532e+38  1.6714491e-02 -2.7917394e+38]\n",
-      "Sample State: [5.7502143e-02 1.3929552e+38 3.0328104e-01 3.1058507e+38]\n",
-      "Sample State: [ 7.9759490e-01 -2.2935894e+38 -4.7575418e-02 -9.9444867e+37]\n",
-      "Sample State: [-4.6630378e+00  4.8043669e+37 -1.4580821e-01  2.4613882e+38]\n"
+      "Sample State: [ 2.7051659e+00 -1.5201922e+38 -2.6240057e-01 -7.0476830e+37]\n",
+      "Sample State: [ 5.9647828e-01  1.7090080e+38 -1.7759679e-01  1.2014864e+38]\n",
+      "Sample State: [3.5260351e+00 3.2534387e+38 2.8506801e-01 2.4451416e+38]\n",
+      "Sample State: [ 2.9073255e+00 -2.6790161e+38 -2.9842880e-01 -9.2181645e+37]\n",
+      "Sample State: [ 3.7392182e+00 -2.4017081e+38  2.4199644e-01  2.9771261e+38]\n",
+      "Sample State: [-1.6355207e+00 -2.4831722e+38  3.3985817e-01 -3.2439041e+38]\n",
+      "Sample State: [-1.1353507e+00  1.9392829e+38  3.3289015e-01  3.0479645e+38]\n",
+      "Sample State: [3.0383275e+00 2.9051530e+38 3.7706238e-01 1.9133415e+38]\n",
+      "Sample State: [-1.2916708e+00  5.8141520e+37 -1.1080918e-01 -6.8264896e+37]\n",
+      "Sample State: [ 2.1536710e+00 -3.1940789e+38 -4.0808445e-01 -8.4556874e+37]\n"
      ]
     }
    ],
@@ -377,52 +543,262 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 54,
+   "execution_count": 107,
    "metadata": {},
    "outputs": [
     {
-     "ename": "TypeError",
-     "evalue": "'<' not supported between instances of 'dict' and 'dict'",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:57\u001b[0m, in \u001b[0;36m_wrapfunc\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     56\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m---> 57\u001b[0m     \u001b[39mreturn\u001b[39;00m bound(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n\u001b[1;32m     58\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mTypeError\u001b[39;00m:\n\u001b[1;32m     59\u001b[0m     \u001b[39m# A TypeError occurs if the object does have such a method in its\u001b[39;00m\n\u001b[1;32m     60\u001b[0m     \u001b[39m# class, but its signature is not identical to that of NumPy's. This\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     64\u001b[0m     \u001b[39m# Call _wrapit from within the except clause to ensure a potential\u001b[39;00m\n\u001b[1;32m     65\u001b[0m     \u001b[39m# exception has a traceback chain.\u001b[39;00m\n",
-      "\u001b[0;31mTypeError\u001b[0m: '<' not supported between instances of 'dict' and 'dict'",
-      "\nDuring handling of the above exception, another exception occurred:\n",
-      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
-      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 18\u001b[0m line \u001b[0;36m3\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m \u001b[39m# Training loop\u001b[39;00m\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=1'>2</a>\u001b[0m \u001b[39mfor\u001b[39;00m episode \u001b[39min\u001b[39;00m \u001b[39mrange\u001b[39m(num_episodes):\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=2'>3</a>\u001b[0m     state \u001b[39m=\u001b[39m discretize_state(env\u001b[39m.\u001b[39;49mreset(), num_bins)\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=4'>5</a>\u001b[0m     \u001b[39mwhile\u001b[39;00m \u001b[39mTrue\u001b[39;00m:\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=5'>6</a>\u001b[0m         \u001b[39m# Choose action using the current Q-table\u001b[39;00m\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=6'>7</a>\u001b[0m         action \u001b[39m=\u001b[39m torch\u001b[39m.\u001b[39margmax(q_table[state])\u001b[39m.\u001b[39mitem()\n",
-      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 18\u001b[0m line \u001b[0;36m9\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=6'>7</a>\u001b[0m \u001b[39mfor\u001b[39;00m i \u001b[39min\u001b[39;00m \u001b[39mrange\u001b[39m(\u001b[39mlen\u001b[39m(state)):\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=7'>8</a>\u001b[0m     bins \u001b[39m=\u001b[39m np\u001b[39m.\u001b[39mlinspace(env\u001b[39m.\u001b[39mobservation_space\u001b[39m.\u001b[39mlow[i], env\u001b[39m.\u001b[39mobservation_space\u001b[39m.\u001b[39mhigh[i], num_bins[i] \u001b[39m+\u001b[39m \u001b[39m1\u001b[39m)[\u001b[39m1\u001b[39m:\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m]\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=8'>9</a>\u001b[0m     state_discrete\u001b[39m.\u001b[39mappend(np\u001b[39m.\u001b[39;49mdigitize(state[i], bins))\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X63sZmlsZQ%3D%3D?line=9'>10</a>\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mtuple\u001b[39m(state_discrete)\n",
-      "File \u001b[0;32m<__array_function__ internals>:200\u001b[0m, in \u001b[0;36mdigitize\u001b[0;34m(*args, **kwargs)\u001b[0m\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/lib/function_base.py:5614\u001b[0m, in \u001b[0;36mdigitize\u001b[0;34m(x, bins, right)\u001b[0m\n\u001b[1;32m   5612\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mlen\u001b[39m(bins) \u001b[39m-\u001b[39m _nx\u001b[39m.\u001b[39msearchsorted(bins[::\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m], x, side\u001b[39m=\u001b[39mside)\n\u001b[1;32m   5613\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m-> 5614\u001b[0m     \u001b[39mreturn\u001b[39;00m _nx\u001b[39m.\u001b[39;49msearchsorted(bins, x, side\u001b[39m=\u001b[39;49mside)\n",
-      "File \u001b[0;32m<__array_function__ internals>:200\u001b[0m, in \u001b[0;36msearchsorted\u001b[0;34m(*args, **kwargs)\u001b[0m\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:1413\u001b[0m, in \u001b[0;36msearchsorted\u001b[0;34m(a, v, side, sorter)\u001b[0m\n\u001b[1;32m   1345\u001b[0m \u001b[39m@array_function_dispatch\u001b[39m(_searchsorted_dispatcher)\n\u001b[1;32m   1346\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39msearchsorted\u001b[39m(a, v, side\u001b[39m=\u001b[39m\u001b[39m'\u001b[39m\u001b[39mleft\u001b[39m\u001b[39m'\u001b[39m, sorter\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m):\n\u001b[1;32m   1347\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m   1348\u001b[0m \u001b[39m    Find indices where elements should be inserted to maintain order.\u001b[39;00m\n\u001b[1;32m   1349\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   1411\u001b[0m \n\u001b[1;32m   1412\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m-> 1413\u001b[0m     \u001b[39mreturn\u001b[39;00m _wrapfunc(a, \u001b[39m'\u001b[39;49m\u001b[39msearchsorted\u001b[39;49m\u001b[39m'\u001b[39;49m, v, side\u001b[39m=\u001b[39;49mside, sorter\u001b[39m=\u001b[39;49msorter)\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:66\u001b[0m, in \u001b[0;36m_wrapfunc\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     57\u001b[0m     \u001b[39mreturn\u001b[39;00m bound(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwds)\n\u001b[1;32m     58\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mTypeError\u001b[39;00m:\n\u001b[1;32m     59\u001b[0m     \u001b[39m# A TypeError occurs if the object does have such a method in its\u001b[39;00m\n\u001b[1;32m     60\u001b[0m     \u001b[39m# class, but its signature is not identical to that of NumPy's. This\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     64\u001b[0m     \u001b[39m# Call _wrapit from within the except clause to ensure a potential\u001b[39;00m\n\u001b[1;32m     65\u001b[0m     \u001b[39m# exception has a traceback chain.\u001b[39;00m\n\u001b[0;32m---> 66\u001b[0m     \u001b[39mreturn\u001b[39;00m _wrapit(obj, method, \u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:43\u001b[0m, in \u001b[0;36m_wrapit\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     41\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mAttributeError\u001b[39;00m:\n\u001b[1;32m     42\u001b[0m     wrap \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m\n\u001b[0;32m---> 43\u001b[0m result \u001b[39m=\u001b[39m \u001b[39mgetattr\u001b[39;49m(asarray(obj), method)(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n\u001b[1;32m     44\u001b[0m \u001b[39mif\u001b[39;00m wrap:\n\u001b[1;32m     45\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39misinstance\u001b[39m(result, mu\u001b[39m.\u001b[39mndarray):\n",
-      "\u001b[0;31mTypeError\u001b[0m: '<' not supported between instances of 'dict' and 'dict'"
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Episode: 0\n",
+      "Episode reward: 8.0\n",
+      "Episode: 1\n",
+      "Episode reward: 9.0\n",
+      "Episode: 2\n",
+      "Episode reward: 10.0\n",
+      "Episode: 3\n",
+      "Episode reward: 10.0\n",
+      "Episode: 4\n",
+      "Episode reward: 10.0\n",
+      "Episode: 5\n",
+      "Episode reward: 9.0\n",
+      "Episode: 6\n",
+      "Episode reward: 9.0\n",
+      "Episode: 7\n",
+      "Episode reward: 10.0\n",
+      "Episode: 8\n",
+      "Episode reward: 9.0\n",
+      "Episode: 9\n",
+      "Episode reward: 10.0\n",
+      "Episode: 10\n",
+      "Episode reward: 10.0\n",
+      "Episode: 11\n",
+      "Episode reward: 9.0\n",
+      "Episode: 12\n",
+      "Episode reward: 9.0\n",
+      "Episode: 13\n",
+      "Episode reward: 10.0\n",
+      "Episode: 14\n",
+      "Episode reward: 10.0\n",
+      "Episode: 15\n",
+      "Episode reward: 9.0\n",
+      "Episode: 16\n",
+      "Episode reward: 8.0\n",
+      "Episode: 17\n",
+      "Episode reward: 9.0\n",
+      "Episode: 18\n",
+      "Episode reward: 9.0\n",
+      "Episode: 19\n",
+      "Episode reward: 9.0\n",
+      "Episode: 20\n",
+      "Episode reward: 8.0\n",
+      "Episode: 21\n",
+      "Episode reward: 10.0\n",
+      "Episode: 22\n",
+      "Episode reward: 8.0\n",
+      "Episode: 23\n",
+      "Episode reward: 8.0\n",
+      "Episode: 24\n",
+      "Episode reward: 10.0\n",
+      "Episode: 25\n",
+      "Episode reward: 9.0\n",
+      "Episode: 26\n",
+      "Episode reward: 8.0\n",
+      "Episode: 27\n",
+      "Episode reward: 8.0\n",
+      "Episode: 28\n",
+      "Episode reward: 10.0\n",
+      "Episode: 29\n",
+      "Episode reward: 9.0\n",
+      "Episode: 30\n",
+      "Episode reward: 8.0\n",
+      "Episode: 31\n",
+      "Episode reward: 10.0\n",
+      "Episode: 32\n",
+      "Episode reward: 9.0\n",
+      "Episode: 33\n",
+      "Episode reward: 10.0\n",
+      "Episode: 34\n",
+      "Episode reward: 9.0\n",
+      "Episode: 35\n",
+      "Episode reward: 11.0\n",
+      "Episode: 36\n",
+      "Episode reward: 10.0\n",
+      "Episode: 37\n",
+      "Episode reward: 8.0\n",
+      "Episode: 38\n",
+      "Episode reward: 10.0\n",
+      "Episode: 39\n",
+      "Episode reward: 8.0\n",
+      "Episode: 40\n",
+      "Episode reward: 10.0\n",
+      "Episode: 41\n",
+      "Episode reward: 9.0\n",
+      "Episode: 42\n",
+      "Episode reward: 10.0\n",
+      "Episode: 43\n",
+      "Episode reward: 8.0\n",
+      "Episode: 44\n",
+      "Episode reward: 9.0\n",
+      "Episode: 45\n",
+      "Episode reward: 10.0\n",
+      "Episode: 46\n",
+      "Episode reward: 9.0\n",
+      "Episode: 47\n",
+      "Episode reward: 9.0\n",
+      "Episode: 48\n",
+      "Episode reward: 9.0\n",
+      "Episode: 49\n",
+      "Episode reward: 9.0\n",
+      "Learned Q-table:\n",
+      "tensor([[[[[ 2.5774e-04, -1.6386e-02],\n",
+      "           [-6.6059e-03, -1.2301e-02],\n",
+      "           [-1.9540e-03, -6.5953e-03]],\n",
+      "\n",
+      "          [[ 1.4049e-02,  8.9660e-03],\n",
+      "           [-1.1570e-02,  2.3123e-02],\n",
+      "           [ 3.0566e-03, -1.0416e-03]],\n",
+      "\n",
+      "          [[ 1.2075e-02,  2.0824e-02],\n",
+      "           [-2.3481e-03, -1.2590e-02],\n",
+      "           [ 1.6281e-02,  4.2098e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 9.7180e-03, -1.1887e-02],\n",
+      "           [-1.2298e-02,  9.9179e-03],\n",
+      "           [-1.3349e-03,  9.6772e-04]],\n",
+      "\n",
+      "          [[ 2.0868e-02,  2.6027e-03],\n",
+      "           [-1.9472e-02,  4.0829e-03],\n",
+      "           [-9.9453e-03, -7.8108e-03]],\n",
+      "\n",
+      "          [[-3.5114e-03,  8.3449e-04],\n",
+      "           [ 1.1458e-02, -1.4358e-02],\n",
+      "           [-1.6030e-02,  7.0141e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[-1.1180e-02, -2.2336e-02],\n",
+      "           [-9.6797e-03,  7.0699e-03],\n",
+      "           [ 1.2940e-02,  8.7505e-03]],\n",
+      "\n",
+      "          [[ 8.1591e-03,  1.4591e-02],\n",
+      "           [-8.2323e-04, -7.5956e-03],\n",
+      "           [ 4.1017e-03, -3.9041e-03]],\n",
+      "\n",
+      "          [[ 1.3982e-03,  1.5428e-03],\n",
+      "           [ 4.5702e-03,  2.4447e-02],\n",
+      "           [-3.8499e-03,  1.2206e-02]]]],\n",
+      "\n",
+      "\n",
+      "\n",
+      "        [[[[ 9.5825e-04,  6.5356e-04],\n",
+      "           [-1.6917e-02, -4.4886e-03],\n",
+      "           [-1.6378e-02, -4.9732e-03]],\n",
+      "\n",
+      "          [[-7.8593e-03, -8.5154e-03],\n",
+      "           [ 7.7686e-03, -7.1885e-03],\n",
+      "           [-2.0450e-02, -2.4888e-02]],\n",
+      "\n",
+      "          [[ 9.2768e-03,  1.6148e-02],\n",
+      "           [ 4.1603e-03,  1.9358e-03],\n",
+      "           [-6.2031e-03, -8.5290e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 3.7850e-03,  4.5296e-03],\n",
+      "           [ 7.5043e+00,  8.7035e-03],\n",
+      "           [ 1.1955e-03, -2.3893e-03]],\n",
+      "\n",
+      "          [[ 1.4752e-02,  1.2745e-02],\n",
+      "           [ 5.8729e-03,  8.4895e+00],\n",
+      "           [ 7.5059e-03, -3.2212e-03]],\n",
+      "\n",
+      "          [[-1.5970e-02,  5.5587e-03],\n",
+      "           [-1.5641e-02, -1.2862e-02],\n",
+      "           [ 5.4951e-03, -8.9383e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 7.0806e-03,  2.2394e-02],\n",
+      "           [-1.3594e-03, -1.4554e-02],\n",
+      "           [ 5.4938e-03, -3.7047e-03]],\n",
+      "\n",
+      "          [[ 1.0082e-02,  5.1565e-03],\n",
+      "           [ 1.0319e-02, -9.1123e-03],\n",
+      "           [ 1.3201e-02,  5.4954e-03]],\n",
+      "\n",
+      "          [[-7.8168e-04,  2.4071e-02],\n",
+      "           [-1.9452e-02,  1.1501e-02],\n",
+      "           [-1.6482e-02, -6.7994e-03]]]],\n",
+      "\n",
+      "\n",
+      "\n",
+      "        [[[[ 2.7583e-03,  5.0975e-03],\n",
+      "           [ 5.6140e-03, -2.4850e-03],\n",
+      "           [ 7.4893e-03, -1.1398e-02]],\n",
+      "\n",
+      "          [[-1.3201e-03, -7.3656e-03],\n",
+      "           [ 8.5875e-03, -1.2533e-02],\n",
+      "           [-1.3983e-02, -2.1707e-02]],\n",
+      "\n",
+      "          [[-1.2484e-02,  2.2155e-03],\n",
+      "           [-1.0199e-03,  1.4230e-02],\n",
+      "           [ 1.8682e-03, -3.7607e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 4.1444e-03,  6.7304e-04],\n",
+      "           [-4.1720e-03,  1.5125e-02],\n",
+      "           [ 6.7321e-03, -6.7075e-03]],\n",
+      "\n",
+      "          [[ 7.2062e-03, -2.6330e-02],\n",
+      "           [-8.4435e-03,  1.2237e-02],\n",
+      "           [-1.4670e-03, -4.3704e-04]],\n",
+      "\n",
+      "          [[-1.4862e-02,  1.1071e-02],\n",
+      "           [ 2.9910e-03, -9.0951e-03],\n",
+      "           [-1.2098e-02,  1.2734e-02]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 4.9769e-03, -1.7016e-03],\n",
+      "           [-3.9668e-03,  9.9020e-03],\n",
+      "           [-2.0652e-03, -6.1593e-03]],\n",
+      "\n",
+      "          [[-2.5633e-03, -4.7999e-03],\n",
+      "           [-2.4177e-04,  1.4281e-02],\n",
+      "           [-8.6366e-03,  5.5457e-03]],\n",
+      "\n",
+      "          [[ 1.7575e-02,  2.9216e-03],\n",
+      "           [-6.7075e-03,  1.0145e-02],\n",
+      "           [ 4.7358e-03,  1.2916e-02]]]]])\n"
      ]
     }
    ],
    "source": [
+    "rewards = []  # List to store rewards for each episode\n",
+    "\n",
     "# Training loop\n",
     "for episode in range(num_episodes):\n",
-    "    state = discretize_state(env.reset(), num_bins)\n",
+    "    print(\"Episode:\", episode)\n",
+    "    state, info = env.reset(seed=episode)\n",
+    "    state = discretize_state(state, num_bins)\n",
+    "    episode_reward = 0\n",
     "\n",
     "    while True:\n",
     "        # Choose action using the current Q-table\n",
     "        action = torch.argmax(q_table[state]).item()\n",
     "\n",
     "        # Take the chosen action and observe the next state and reward\n",
-    "        next_state, reward, done, _ = env.step(action)\n",
+    "        next_state, reward, terminated, truncated, info = env.step(action)\n",
     "        next_state = discretize_state(next_state, num_bins)\n",
     "\n",
     "        # Update the Q-table using the Q-learning update rule\n",
     "        q_table = update_q_table(q_table, state, action, reward, next_state, learning_rate, discount_factor)\n",
     "\n",
+    "        episode_reward += reward\n",
     "        state = next_state\n",
     "\n",
-    "        if done:\n",
+    "        if truncated or terminated:\n",
     "            break\n",
+    "    rewards.append(episode_reward)\n",
+    "    print(\"Episode reward:\", episode_reward)\n",
     "\n",
     "# Print the learned Q-table\n",
     "print(\"Learned Q-table:\")\n",
@@ -431,78 +807,37 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 55,
-   "metadata": {},
-   "outputs": [
-    {
-     "ename": "TypeError",
-     "evalue": "'<' not supported between instances of 'dict' and 'dict'",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:57\u001b[0m, in \u001b[0;36m_wrapfunc\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     56\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m---> 57\u001b[0m     \u001b[39mreturn\u001b[39;00m bound(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n\u001b[1;32m     58\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mTypeError\u001b[39;00m:\n\u001b[1;32m     59\u001b[0m     \u001b[39m# A TypeError occurs if the object does have such a method in its\u001b[39;00m\n\u001b[1;32m     60\u001b[0m     \u001b[39m# class, but its signature is not identical to that of NumPy's. This\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     64\u001b[0m     \u001b[39m# Call _wrapit from within the except clause to ensure a potential\u001b[39;00m\n\u001b[1;32m     65\u001b[0m     \u001b[39m# exception has a traceback chain.\u001b[39;00m\n",
-      "\u001b[0;31mTypeError\u001b[0m: '<' not supported between instances of 'dict' and 'dict'",
-      "\nDuring handling of the above exception, another exception occurred:\n",
-      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
-      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 19\u001b[0m line \u001b[0;36m1\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X65sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m discretize_state(env\u001b[39m.\u001b[39;49mreset(), num_bins)\n",
-      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 19\u001b[0m line \u001b[0;36m9\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X65sZmlsZQ%3D%3D?line=6'>7</a>\u001b[0m \u001b[39mfor\u001b[39;00m i \u001b[39min\u001b[39;00m \u001b[39mrange\u001b[39m(\u001b[39mlen\u001b[39m(state)):\n\u001b[1;32m      <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X65sZmlsZQ%3D%3D?line=7'>8</a>\u001b[0m     bins \u001b[39m=\u001b[39m np\u001b[39m.\u001b[39mlinspace(env\u001b[39m.\u001b[39mobservation_space\u001b[39m.\u001b[39mlow[i], env\u001b[39m.\u001b[39mobservation_space\u001b[39m.\u001b[39mhigh[i], num_bins[i] \u001b[39m+\u001b[39m \u001b[39m1\u001b[39m)[\u001b[39m1\u001b[39m:\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m]\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X65sZmlsZQ%3D%3D?line=8'>9</a>\u001b[0m     state_discrete\u001b[39m.\u001b[39mappend(np\u001b[39m.\u001b[39;49mdigitize(state[i], bins))\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X65sZmlsZQ%3D%3D?line=9'>10</a>\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mtuple\u001b[39m(state_discrete)\n",
-      "File \u001b[0;32m<__array_function__ internals>:200\u001b[0m, in \u001b[0;36mdigitize\u001b[0;34m(*args, **kwargs)\u001b[0m\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/lib/function_base.py:5614\u001b[0m, in \u001b[0;36mdigitize\u001b[0;34m(x, bins, right)\u001b[0m\n\u001b[1;32m   5612\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mlen\u001b[39m(bins) \u001b[39m-\u001b[39m _nx\u001b[39m.\u001b[39msearchsorted(bins[::\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m], x, side\u001b[39m=\u001b[39mside)\n\u001b[1;32m   5613\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m-> 5614\u001b[0m     \u001b[39mreturn\u001b[39;00m _nx\u001b[39m.\u001b[39;49msearchsorted(bins, x, side\u001b[39m=\u001b[39;49mside)\n",
-      "File \u001b[0;32m<__array_function__ internals>:200\u001b[0m, in \u001b[0;36msearchsorted\u001b[0;34m(*args, **kwargs)\u001b[0m\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:1413\u001b[0m, in \u001b[0;36msearchsorted\u001b[0;34m(a, v, side, sorter)\u001b[0m\n\u001b[1;32m   1345\u001b[0m \u001b[39m@array_function_dispatch\u001b[39m(_searchsorted_dispatcher)\n\u001b[1;32m   1346\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39msearchsorted\u001b[39m(a, v, side\u001b[39m=\u001b[39m\u001b[39m'\u001b[39m\u001b[39mleft\u001b[39m\u001b[39m'\u001b[39m, sorter\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m):\n\u001b[1;32m   1347\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"\u001b[39;00m\n\u001b[1;32m   1348\u001b[0m \u001b[39m    Find indices where elements should be inserted to maintain order.\u001b[39;00m\n\u001b[1;32m   1349\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   1411\u001b[0m \n\u001b[1;32m   1412\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m-> 1413\u001b[0m     \u001b[39mreturn\u001b[39;00m _wrapfunc(a, \u001b[39m'\u001b[39;49m\u001b[39msearchsorted\u001b[39;49m\u001b[39m'\u001b[39;49m, v, side\u001b[39m=\u001b[39;49mside, sorter\u001b[39m=\u001b[39;49msorter)\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:66\u001b[0m, in \u001b[0;36m_wrapfunc\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     57\u001b[0m     \u001b[39mreturn\u001b[39;00m bound(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwds)\n\u001b[1;32m     58\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mTypeError\u001b[39;00m:\n\u001b[1;32m     59\u001b[0m     \u001b[39m# A TypeError occurs if the object does have such a method in its\u001b[39;00m\n\u001b[1;32m     60\u001b[0m     \u001b[39m# class, but its signature is not identical to that of NumPy's. This\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     64\u001b[0m     \u001b[39m# Call _wrapit from within the except clause to ensure a potential\u001b[39;00m\n\u001b[1;32m     65\u001b[0m     \u001b[39m# exception has a traceback chain.\u001b[39;00m\n\u001b[0;32m---> 66\u001b[0m     \u001b[39mreturn\u001b[39;00m _wrapit(obj, method, \u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/numpy/core/fromnumeric.py:43\u001b[0m, in \u001b[0;36m_wrapit\u001b[0;34m(obj, method, *args, **kwds)\u001b[0m\n\u001b[1;32m     41\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mAttributeError\u001b[39;00m:\n\u001b[1;32m     42\u001b[0m     wrap \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m\n\u001b[0;32m---> 43\u001b[0m result \u001b[39m=\u001b[39m \u001b[39mgetattr\u001b[39;49m(asarray(obj), method)(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwds)\n\u001b[1;32m     44\u001b[0m \u001b[39mif\u001b[39;00m wrap:\n\u001b[1;32m     45\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39misinstance\u001b[39m(result, mu\u001b[39m.\u001b[39mndarray):\n",
-      "\u001b[0;31mTypeError\u001b[0m: '<' not supported between instances of 'dict' and 'dict'"
-     ]
-    }
-   ],
-   "source": [
-    "discretize_state(env.reset(), num_bins)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 36,
+   "execution_count": 108,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "tensor([[ 0.0982, -0.0016],\n",
-       "        [ 0.0100,  0.0040],\n",
-       "        [-0.0006,  0.0108],\n",
-       "        [ 0.0028,  0.0059]])"
+       "[<matplotlib.lines.Line2D at 0x158a98310>]"
       ]
      },
-     "execution_count": 36,
+     "execution_count": 108,
      "metadata": {},
      "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "q_table"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 37,
-   "metadata": {},
-   "outputs": [
+    },
     {
-     "ename": "IndexError",
-     "evalue": "tensors used as indices must be long, int, byte or bool tensors",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
-      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 22\u001b[0m line \u001b[0;36m1\n\u001b[0;32m----> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X60sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m q_table[state]\n",
-      "\u001b[0;31mIndexError\u001b[0m: tensors used as indices must be long, int, byte or bool tensors"
-     ]
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "image/png": {
+       "height": 413,
+       "width": 556
+      }
+     },
+     "output_type": "display_data"
     }
    ],
    "source": [
-    "q_table[state]"
+    "plt.plot(rewards)"
    ]
   },
   {

From bb29bb2aab8254b1b5ef0198bc02b851ad16c92c Mon Sep 17 00:00:00 2001
From: Nipun Batra <nipunbatra0@gmail.com>
Date: Tue, 12 Dec 2023 10:53:07 +0530
Subject: [PATCH 3/5] added the notebook more details with eps-greedy

---
 posts/2023-Dec-11-gym.ipynb | 2324 ++++++++++++++++++++++++++++++-----
 1 file changed, 2013 insertions(+), 311 deletions(-)

diff --git a/posts/2023-Dec-11-gym.ipynb b/posts/2023-Dec-11-gym.ipynb
index a443131..5c6a90a 100644
--- a/posts/2023-Dec-11-gym.ipynb
+++ b/posts/2023-Dec-11-gym.ipynb
@@ -12,7 +12,7 @@
     "- ML\n",
     "date: '2023-12-11'\n",
     "output-file: rl.html\n",
-    "title: Super Reinforcement Learning\n",
+    "title: Reinforcement Learning\n",
     "toc: true\n",
     "\n",
     "---\n",
@@ -30,7 +30,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 1,
    "id": "f41ca63d",
    "metadata": {},
    "outputs": [],
@@ -211,6 +211,152 @@
     "print(env.observation_space)"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([ 0.0273956 , -0.00611216,  0.03585979,  0.0197368 ], dtype=float32)"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "observation"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1\n"
+     ]
+    }
+   ],
+   "source": [
+    "action = env.action_space.sample()\n",
+    "print(action)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "observation, reward, terminated, truncated, info = env.step(action)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[ 0.02727336  0.18847767  0.03625453 -0.26141977] 1.0 False False {}\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(observation, reward, terminated, truncated, info)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[ 0.03104291  0.38306385  0.03102613 -0.5424507 ] 1.0 False False {}\n"
+     ]
+    }
+   ],
+   "source": [
+    "observation, reward, terminated, truncated, info = env.step(action)\n",
+    "print(observation, reward, terminated, truncated, info)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[ 0.02727336  0.18847767  0.03625453 -0.26141977] 1.0 False False {}\n",
+      "[ 0.03104291  0.38306385  0.03102613 -0.5424507 ] 1.0 False False {}\n",
+      "[ 0.03870419  0.5777363   0.02017712 -0.8251987 ] 1.0 False False {}\n",
+      "[ 0.05025892  0.7725766   0.00367314 -1.111468  ] 1.0 False False {}\n",
+      "[ 0.06571045  0.96765006 -0.01855621 -1.4029963 ] 1.0 False False {}\n",
+      "[ 0.08506345  1.1629975  -0.04661614 -1.7014222 ] 1.0 False False {}\n",
+      "[ 0.1083234   1.3586243  -0.08064459 -2.0082438 ] 1.0 False False {}\n",
+      "[ 0.13549589  1.554488   -0.12080947 -2.3247683 ] 1.0 False False {}\n",
+      "[ 0.16658565  1.7504818  -0.16730483 -2.652048  ] 1.0 False False {}\n",
+      "[ 0.20159529  1.9464185  -0.22034578 -2.9908078 ] 1.0 True False {}\n"
+     ]
+    }
+   ],
+   "source": [
+    "observation, info = env.reset(seed=42)\n",
+    "for i in range(100):\n",
+    "    action = 1\n",
+    "    observation, reward, terminated, truncated, info = env.step(action)\n",
+    "    print(observation, reward, terminated, truncated, info)\n",
+    "    if terminated:\n",
+    "        break\n",
+    "    "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[ 0.02727336 -0.20172954  0.03625453  0.32351476] 1.0 False False {}\n",
+      "[ 0.02323877 -0.39734846  0.04272482  0.62740684] 1.0 False False {}\n",
+      "[ 0.0152918  -0.5930399   0.05527296  0.9332334 ] 1.0 False False {}\n",
+      "[ 0.003431   -0.7888622   0.07393762  1.2427603 ] 1.0 False False {}\n",
+      "[-0.01234624 -0.9848512   0.09879284  1.5576583 ] 1.0 False False {}\n",
+      "[-0.03204326 -1.1810076   0.129946    1.8794562 ] 1.0 False False {}\n",
+      "[-0.05566342 -1.3772845   0.16753513  2.209486  ] 1.0 False False {}\n",
+      "[-0.0832091  -1.573571    0.21172485  2.5488186 ] 1.0 True False {}\n"
+     ]
+    }
+   ],
+   "source": [
+    "observation, info = env.reset(seed=42)\n",
+    "for i in range(100):\n",
+    "    action = 0\n",
+    "    observation, reward, terminated, truncated, info = env.step(action)\n",
+    "    print(observation, reward, terminated, truncated, info)\n",
+    "    if terminated:\n",
+    "        break\n",
+    "    "
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 8,
@@ -303,12 +449,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 99,
+   "execution_count": 26,
    "metadata": {},
    "outputs": [],
    "source": [
     "# Define the number of bins for each dimension\n",
-    "num_bins = [3, 3, 3, 3]  # Adjust these values based on your preference\n",
+    "num_bins = [4, 4, 4, 4]  # Adjust these values based on your preference\n",
     "\n",
     "# Discretize the continuous state space\n",
     "def discretize_state(state, num_bins):\n",
@@ -332,132 +478,352 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 100,
+   "execution_count": 27,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "tensor([[[[[ 0.0003, -0.0164],\n",
-       "           [-0.0066, -0.0123],\n",
-       "           [-0.0020, -0.0066]],\n",
+       "tensor([[[[[-1.1406e-02, -1.0509e-02],\n",
+       "           [ 1.3085e-03,  1.3717e-03],\n",
+       "           [-6.0147e-03,  1.3680e-02],\n",
+       "           [-1.2355e-02, -1.0242e-02]],\n",
+       "\n",
+       "          [[-7.4378e-03,  1.3998e-03],\n",
+       "           [ 8.4903e-03,  2.0094e-03],\n",
+       "           [ 3.5405e-03, -2.5198e-03],\n",
+       "           [ 6.9709e-03,  1.0338e-02]],\n",
+       "\n",
+       "          [[-6.8366e-03,  9.4885e-03],\n",
+       "           [ 7.2773e-03, -6.8727e-03],\n",
+       "           [ 1.7096e-02,  1.0054e-02],\n",
+       "           [-9.2244e-03,  2.8881e-05]],\n",
+       "\n",
+       "          [[ 5.0180e-03, -1.0984e-02],\n",
+       "           [ 6.4403e-03,  1.6484e-02],\n",
+       "           [ 6.2945e-03, -1.0362e-02],\n",
+       "           [ 1.0791e-03, -1.1073e-02]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 1.0317e-02, -3.5725e-03],\n",
+       "           [-3.7283e-03,  5.4095e-03],\n",
+       "           [ 1.6884e-02, -2.2501e-03],\n",
+       "           [-4.3716e-03, -8.7110e-03]],\n",
+       "\n",
+       "          [[-4.2364e-03,  1.4712e-02],\n",
+       "           [ 1.1050e-03,  1.4572e-02],\n",
+       "           [-3.4920e-03, -1.4701e-02],\n",
+       "           [ 1.1567e-02,  2.2020e-02]],\n",
+       "\n",
+       "          [[-8.6948e-04, -2.7715e-02],\n",
+       "           [-2.0818e-03,  9.2057e-03],\n",
+       "           [-2.1844e-03, -5.6509e-03],\n",
+       "           [-1.4649e-02, -2.0757e-02]],\n",
+       "\n",
+       "          [[ 1.4552e-02, -1.1368e-03],\n",
+       "           [ 1.7741e-03,  1.2114e-03],\n",
+       "           [ 3.1583e-03,  7.2406e-03],\n",
+       "           [-4.3346e-04, -1.6101e-02]]],\n",
+       "\n",
+       "\n",
+       "         [[[-2.5550e-04,  1.7958e-02],\n",
+       "           [-5.8764e-03, -7.2153e-04],\n",
+       "           [-4.4413e-03,  2.5568e-03],\n",
+       "           [-8.1304e-03,  4.7102e-03]],\n",
+       "\n",
+       "          [[-1.2885e-03, -5.7140e-03],\n",
+       "           [ 4.2125e-03, -6.7077e-03],\n",
+       "           [ 8.2713e-04,  1.6447e-02],\n",
+       "           [ 4.4602e-03, -6.7215e-03]],\n",
+       "\n",
+       "          [[ 2.2286e-04, -5.4333e-03],\n",
+       "           [-5.5148e-03,  7.2770e-03],\n",
+       "           [-7.1460e-03,  2.2344e-02],\n",
+       "           [ 9.4010e-03, -1.0193e-04]],\n",
+       "\n",
+       "          [[-1.2127e-02, -1.0535e-02],\n",
+       "           [-1.7398e-02,  7.0977e-03],\n",
+       "           [-5.5518e-03,  1.7752e-02],\n",
+       "           [ 4.2644e-03,  6.5037e-04]]],\n",
+       "\n",
+       "\n",
+       "         [[[-1.7868e-03, -4.7429e-03],\n",
+       "           [ 3.5122e-03, -3.5208e-03],\n",
+       "           [-7.4866e-03, -1.0636e-02],\n",
+       "           [ 1.0040e-02, -1.1361e-02]],\n",
+       "\n",
+       "          [[-2.1231e-03, -5.4856e-04],\n",
+       "           [-2.5714e-04,  1.5240e-03],\n",
+       "           [ 1.1099e-02,  7.3871e-03],\n",
+       "           [-1.9287e-02,  1.2078e-02]],\n",
+       "\n",
+       "          [[-7.6529e-03, -8.6546e-03],\n",
+       "           [ 8.2139e-03,  1.1296e-02],\n",
+       "           [ 1.9106e-02, -2.2947e-03],\n",
+       "           [ 2.8244e-03,  1.3176e-02]],\n",
+       "\n",
+       "          [[-4.5813e-03,  7.6768e-04],\n",
+       "           [ 3.7834e-03, -3.0948e-03],\n",
+       "           [-1.2757e-02,  8.3703e-03],\n",
+       "           [ 5.4267e-04,  4.7438e-04]]]],\n",
+       "\n",
+       "\n",
+       "\n",
+       "        [[[[ 2.9412e-03,  8.9573e-03],\n",
+       "           [ 1.7503e-02, -7.0822e-03],\n",
+       "           [-1.3228e-02,  8.8431e-03],\n",
+       "           [-1.8440e-02,  3.7296e-03]],\n",
+       "\n",
+       "          [[ 7.3911e-03,  8.4983e-03],\n",
+       "           [-4.7013e-03, -2.8349e-02],\n",
+       "           [ 4.0318e-03,  2.3796e-02],\n",
+       "           [-1.3484e-02,  1.8661e-02]],\n",
+       "\n",
+       "          [[ 1.1991e-02,  2.6918e-05],\n",
+       "           [ 1.3168e-02, -1.1984e-03],\n",
+       "           [-2.1404e-02, -2.9017e-03],\n",
+       "           [ 7.4044e-03,  5.5205e-03]],\n",
+       "\n",
+       "          [[-5.1570e-03, -7.7488e-03],\n",
+       "           [-1.0394e-02,  5.5803e-03],\n",
+       "           [-4.9909e-04,  2.5824e-02],\n",
+       "           [-9.5470e-03,  1.2008e-03]]],\n",
+       "\n",
+       "\n",
+       "         [[[-5.0071e-03,  3.1824e-03],\n",
+       "           [ 6.2639e-03, -1.1144e-02],\n",
+       "           [ 7.6656e-04,  1.6723e-02],\n",
+       "           [-6.5249e-03, -7.0401e-03]],\n",
+       "\n",
+       "          [[-6.7663e-03, -1.2750e-03],\n",
+       "           [ 2.2510e-03,  4.3926e-03],\n",
+       "           [-1.6809e-03,  8.1863e-03],\n",
+       "           [-1.5351e-02, -1.1513e-03]],\n",
        "\n",
-       "          [[ 0.0140,  0.0090],\n",
-       "           [-0.0116,  0.0231],\n",
-       "           [ 0.0031, -0.0010]],\n",
+       "          [[ 2.3877e-03, -2.0612e-03],\n",
+       "           [ 1.1877e-02,  8.4327e-03],\n",
+       "           [-1.1223e-03, -4.2275e-03],\n",
+       "           [-1.5361e-03, -1.3486e-03]],\n",
        "\n",
-       "          [[ 0.0121,  0.0208],\n",
-       "           [-0.0023, -0.0126],\n",
-       "           [ 0.0163,  0.0042]]],\n",
+       "          [[-3.0515e-03, -8.0040e-03],\n",
+       "           [ 1.0299e-03, -1.3834e-04],\n",
+       "           [-1.9414e-02, -7.4649e-03],\n",
+       "           [ 9.8406e-03,  1.7082e-02]]],\n",
        "\n",
        "\n",
-       "         [[[ 0.0097, -0.0119],\n",
-       "           [-0.0123,  0.0099],\n",
-       "           [-0.0013,  0.0010]],\n",
+       "         [[[-7.0127e-03, -1.6560e-03],\n",
+       "           [-2.8100e-03, -5.4193e-03],\n",
+       "           [-1.6859e-02,  1.1542e-02],\n",
+       "           [-1.1415e-02,  1.4802e-02]],\n",
        "\n",
-       "          [[ 0.0209,  0.0026],\n",
-       "           [-0.0195,  0.0041],\n",
-       "           [-0.0099, -0.0078]],\n",
+       "          [[ 6.0110e-03,  1.5143e-02],\n",
+       "           [ 4.0649e-03, -8.9268e-03],\n",
+       "           [-1.8577e-03, -2.3410e-02],\n",
+       "           [-3.9312e-03,  1.2311e-03]],\n",
        "\n",
-       "          [[-0.0035,  0.0008],\n",
-       "           [ 0.0115, -0.0144],\n",
-       "           [-0.0160,  0.0070]]],\n",
+       "          [[ 1.8359e-03, -3.4620e-03],\n",
+       "           [ 1.3212e-02, -3.8299e-03],\n",
+       "           [-1.0617e-02, -9.8063e-03],\n",
+       "           [ 7.7875e-03, -1.1448e-02]],\n",
        "\n",
+       "          [[-9.5362e-03,  1.5079e-02],\n",
+       "           [-6.8809e-03,  8.7390e-04],\n",
+       "           [ 8.9715e-03,  1.6310e-03],\n",
+       "           [-1.3025e-02,  6.7466e-03]]],\n",
        "\n",
-       "         [[[-0.0112, -0.0223],\n",
-       "           [-0.0097,  0.0071],\n",
-       "           [ 0.0129,  0.0088]],\n",
        "\n",
-       "          [[ 0.0082,  0.0146],\n",
-       "           [-0.0008, -0.0076],\n",
-       "           [ 0.0041, -0.0039]],\n",
+       "         [[[ 2.5851e-03,  6.6429e-03],\n",
+       "           [ 1.2671e-02, -8.8480e-03],\n",
+       "           [-3.8258e-03,  1.5023e-03],\n",
+       "           [ 9.2233e-03,  6.4037e-03]],\n",
        "\n",
-       "          [[ 0.0014,  0.0015],\n",
-       "           [ 0.0046,  0.0244],\n",
-       "           [-0.0038,  0.0122]]]],\n",
+       "          [[ 5.9047e-03, -6.3933e-03],\n",
+       "           [-8.4737e-03,  8.7749e-03],\n",
+       "           [ 7.2186e-03, -1.0333e-02],\n",
+       "           [-7.0906e-03,  2.4680e-02]],\n",
        "\n",
+       "          [[ 2.5566e-02, -2.3371e-03],\n",
+       "           [-5.3729e-03, -4.7599e-03],\n",
+       "           [ 1.9879e-02,  1.8122e-02],\n",
+       "           [-3.3057e-03, -1.3735e-02]],\n",
        "\n",
+       "          [[ 4.4050e-03, -1.1497e-02],\n",
+       "           [-1.1812e-02,  9.4849e-03],\n",
+       "           [ 1.0993e-02, -1.5682e-02],\n",
+       "           [ 6.7096e-03,  2.5279e-02]]]],\n",
        "\n",
-       "        [[[[ 0.0010,  0.0007],\n",
-       "           [-0.0169, -0.0045],\n",
-       "           [-0.0164, -0.0050]],\n",
        "\n",
-       "          [[-0.0079, -0.0085],\n",
-       "           [ 0.0078, -0.0072],\n",
-       "           [-0.0204, -0.0249]],\n",
        "\n",
-       "          [[ 0.0093,  0.0161],\n",
-       "           [ 0.0042,  0.0019],\n",
-       "           [-0.0062, -0.0085]]],\n",
+       "        [[[[ 5.1784e-03, -1.4195e-03],\n",
+       "           [-5.3990e-03, -1.0834e-02],\n",
+       "           [ 1.0641e-02,  6.1596e-03],\n",
+       "           [ 8.7506e-03,  2.9205e-03]],\n",
        "\n",
+       "          [[-1.6737e-02,  5.6834e-03],\n",
+       "           [ 6.5168e-03, -3.1135e-03],\n",
+       "           [ 2.9460e-03, -1.2015e-03],\n",
+       "           [-1.1608e-02,  7.6579e-03]],\n",
        "\n",
-       "         [[[ 0.0038,  0.0045],\n",
-       "           [ 0.0106,  0.0087],\n",
-       "           [ 0.0012, -0.0024]],\n",
+       "          [[ 1.0303e-02, -1.0884e-02],\n",
+       "           [ 9.5722e-03, -4.6254e-03],\n",
+       "           [ 4.6554e-03,  1.0371e-02],\n",
+       "           [-2.4556e-03,  4.5003e-03]],\n",
        "\n",
-       "          [[ 0.0148,  0.0127],\n",
-       "           [ 0.0059,  0.0148],\n",
-       "           [ 0.0075, -0.0032]],\n",
+       "          [[-5.7525e-03, -6.7471e-03],\n",
+       "           [ 4.3139e-03, -5.5204e-03],\n",
+       "           [ 9.1538e-03,  4.8076e-03],\n",
+       "           [ 9.0984e-03, -2.3925e-03]]],\n",
        "\n",
-       "          [[-0.0160,  0.0056],\n",
-       "           [-0.0156, -0.0129],\n",
-       "           [ 0.0055, -0.0089]]],\n",
        "\n",
+       "         [[[ 3.8985e-03, -7.3933e-03],\n",
+       "           [-5.1886e-03,  1.1477e-03],\n",
+       "           [ 6.4889e-03, -3.6765e-03],\n",
+       "           [ 7.2426e-03,  4.0187e-03]],\n",
        "\n",
-       "         [[[ 0.0071,  0.0224],\n",
-       "           [-0.0014, -0.0146],\n",
-       "           [ 0.0055, -0.0037]],\n",
+       "          [[-1.0519e-02, -6.4471e-05],\n",
+       "           [ 2.2689e-02,  1.2035e-02],\n",
+       "           [ 4.8530e-03, -1.4561e-03],\n",
+       "           [ 1.6576e-03,  2.0176e-02]],\n",
        "\n",
-       "          [[ 0.0101,  0.0052],\n",
-       "           [ 0.0103, -0.0091],\n",
-       "           [ 0.0132,  0.0055]],\n",
+       "          [[-8.8692e-03,  2.4320e-03],\n",
+       "           [ 4.7644e-03, -1.0054e-02],\n",
+       "           [ 1.1499e-02, -4.3900e-03],\n",
+       "           [ 8.8678e-03, -1.0401e-02]],\n",
        "\n",
-       "          [[-0.0008,  0.0241],\n",
-       "           [-0.0195,  0.0115],\n",
-       "           [-0.0165, -0.0068]]]],\n",
+       "          [[ 3.8191e-03, -1.0572e-02],\n",
+       "           [-3.9368e-03,  8.9180e-03],\n",
+       "           [-1.8331e-02, -4.8689e-03],\n",
+       "           [-1.2785e-02, -5.7243e-03]]],\n",
        "\n",
        "\n",
+       "         [[[-4.1451e-03,  5.1505e-03],\n",
+       "           [ 1.4643e-02,  8.4760e-03],\n",
+       "           [ 1.0262e-03, -5.9678e-03],\n",
+       "           [ 5.1973e-03,  1.1929e-02]],\n",
        "\n",
-       "        [[[[ 0.0028,  0.0051],\n",
-       "           [ 0.0056, -0.0025],\n",
-       "           [ 0.0075, -0.0114]],\n",
+       "          [[-6.2390e-03,  6.5752e-03],\n",
+       "           [ 9.1719e-04,  6.7939e-03],\n",
+       "           [-2.6674e-04,  2.9604e-02],\n",
+       "           [ 4.2720e-03, -1.5935e-02]],\n",
        "\n",
-       "          [[-0.0013, -0.0074],\n",
-       "           [ 0.0086, -0.0125],\n",
-       "           [-0.0140, -0.0217]],\n",
+       "          [[-5.3500e-04,  2.2230e-02],\n",
+       "           [-4.7624e-03,  1.1678e-02],\n",
+       "           [-1.1461e-03,  3.8098e-03],\n",
+       "           [ 2.2019e-02, -1.5939e-05]],\n",
        "\n",
-       "          [[-0.0125,  0.0022],\n",
-       "           [-0.0010,  0.0142],\n",
-       "           [ 0.0019, -0.0038]]],\n",
+       "          [[-1.2336e-02,  1.0585e-02],\n",
+       "           [ 1.7668e-02, -2.0162e-02],\n",
+       "           [-1.1665e-02, -5.4866e-04],\n",
+       "           [-7.0921e-03,  1.3506e-02]]],\n",
        "\n",
        "\n",
-       "         [[[ 0.0041,  0.0007],\n",
-       "           [-0.0042,  0.0151],\n",
-       "           [ 0.0067, -0.0067]],\n",
+       "         [[[ 1.5243e-03,  1.9578e-03],\n",
+       "           [ 4.2853e-03,  3.6213e-03],\n",
+       "           [-2.2286e-03, -1.7510e-02],\n",
+       "           [-6.2912e-03,  1.6289e-02]],\n",
        "\n",
-       "          [[ 0.0072, -0.0263],\n",
-       "           [-0.0084,  0.0122],\n",
-       "           [-0.0015, -0.0004]],\n",
+       "          [[-9.0679e-03, -1.2895e-02],\n",
+       "           [ 3.7751e-03, -2.3309e-02],\n",
+       "           [ 1.5525e-03,  2.4318e-03],\n",
+       "           [ 3.7684e-03, -1.7957e-02]],\n",
        "\n",
-       "          [[-0.0149,  0.0111],\n",
-       "           [ 0.0030, -0.0091],\n",
-       "           [-0.0121,  0.0127]]],\n",
+       "          [[ 2.7108e-03,  8.7287e-04],\n",
+       "           [-5.4382e-03, -1.5571e-02],\n",
+       "           [-1.1790e-02,  6.5124e-03],\n",
+       "           [ 2.2338e-02,  5.8709e-03]],\n",
        "\n",
+       "          [[ 4.9399e-04, -1.9222e-02],\n",
+       "           [-2.4067e-03,  1.6542e-02],\n",
+       "           [ 1.1662e-02,  1.1188e-02],\n",
+       "           [-6.7352e-03,  3.2604e-02]]]],\n",
        "\n",
-       "         [[[ 0.0050, -0.0017],\n",
-       "           [-0.0040,  0.0099],\n",
-       "           [-0.0021, -0.0062]],\n",
        "\n",
-       "          [[-0.0026, -0.0048],\n",
-       "           [-0.0002,  0.0143],\n",
-       "           [-0.0086,  0.0055]],\n",
        "\n",
-       "          [[ 0.0176,  0.0029],\n",
-       "           [-0.0067,  0.0101],\n",
-       "           [ 0.0047,  0.0129]]]]])"
+       "        [[[[ 7.5602e-03, -5.0242e-03],\n",
+       "           [-8.2706e-03, -1.7402e-03],\n",
+       "           [-9.3949e-03, -7.2250e-03],\n",
+       "           [ 8.7566e-03,  1.1676e-04]],\n",
+       "\n",
+       "          [[-7.5017e-03,  1.8024e-03],\n",
+       "           [ 2.5618e-03, -8.8935e-03],\n",
+       "           [-2.5026e-03, -5.5693e-03],\n",
+       "           [-9.2033e-03,  2.7998e-03]],\n",
+       "\n",
+       "          [[-1.2505e-02, -6.9063e-03],\n",
+       "           [ 1.2637e-02,  8.3032e-03],\n",
+       "           [-1.6924e-02,  1.1639e-02],\n",
+       "           [ 6.7917e-03,  2.2977e-03]],\n",
+       "\n",
+       "          [[-3.2956e-04, -5.9520e-04],\n",
+       "           [ 1.6061e-02,  9.6812e-03],\n",
+       "           [-1.1180e-02, -7.7573e-04],\n",
+       "           [ 1.3929e-03,  8.3166e-03]]],\n",
+       "\n",
+       "\n",
+       "         [[[-1.0511e-02, -9.3898e-03],\n",
+       "           [-7.4458e-03,  7.1798e-03],\n",
+       "           [ 1.8625e-02, -3.9352e-03],\n",
+       "           [ 1.1769e-02,  9.1456e-04]],\n",
+       "\n",
+       "          [[ 1.0072e-02,  7.7131e-03],\n",
+       "           [ 6.5359e-03, -9.2153e-03],\n",
+       "           [ 1.9783e-03, -2.0443e-02],\n",
+       "           [ 1.0223e-02, -8.4568e-03]],\n",
+       "\n",
+       "          [[-3.5886e-04,  3.0560e-04],\n",
+       "           [-2.9668e-03,  8.7726e-03],\n",
+       "           [ 1.6094e-02, -8.5014e-03],\n",
+       "           [ 2.4411e-02,  3.0396e-03]],\n",
+       "\n",
+       "          [[-1.3760e-02,  9.4967e-04],\n",
+       "           [-8.1352e-03,  8.6983e-03],\n",
+       "           [-1.7103e-03, -1.0511e-02],\n",
+       "           [-5.1999e-03, -6.6092e-03]]],\n",
+       "\n",
+       "\n",
+       "         [[[-4.1700e-03, -2.8946e-03],\n",
+       "           [-5.7401e-03,  4.6369e-03],\n",
+       "           [-1.2268e-02, -1.6185e-02],\n",
+       "           [-1.7004e-02, -1.8065e-03]],\n",
+       "\n",
+       "          [[ 1.4414e-03,  3.0189e-03],\n",
+       "           [ 8.3214e-04, -6.1676e-03],\n",
+       "           [ 3.0263e-03, -2.0772e-02],\n",
+       "           [ 5.5492e-03, -3.2041e-03]],\n",
+       "\n",
+       "          [[-8.9067e-03, -1.6217e-04],\n",
+       "           [-3.9358e-03,  1.3170e-03],\n",
+       "           [ 1.8509e-03,  1.9097e-02],\n",
+       "           [-6.4288e-03, -1.1963e-02]],\n",
+       "\n",
+       "          [[-5.7511e-04,  2.1690e-03],\n",
+       "           [ 6.8697e-03,  2.6131e-03],\n",
+       "           [ 9.6800e-03, -6.5693e-03],\n",
+       "           [ 7.9978e-03,  3.4935e-03]]],\n",
+       "\n",
+       "\n",
+       "         [[[-2.8822e-03, -1.7531e-02],\n",
+       "           [-9.0395e-03,  1.7119e-03],\n",
+       "           [ 4.7403e-04, -2.4928e-03],\n",
+       "           [ 2.0809e-02,  1.0375e-02]],\n",
+       "\n",
+       "          [[ 9.9379e-03,  1.1011e-03],\n",
+       "           [-2.0822e-02, -1.0136e-03],\n",
+       "           [ 5.6764e-03,  4.7663e-03],\n",
+       "           [ 5.7930e-03, -3.2342e-03]],\n",
+       "\n",
+       "          [[ 1.8071e-03, -3.8490e-03],\n",
+       "           [-5.0903e-03,  4.2883e-03],\n",
+       "           [ 6.2868e-03,  6.9501e-03],\n",
+       "           [-5.1345e-03, -4.4751e-03]],\n",
+       "\n",
+       "          [[-6.3479e-04, -1.4575e-03],\n",
+       "           [ 1.3965e-02, -4.1295e-03],\n",
+       "           [-3.1379e-03,  3.5920e-05],\n",
+       "           [ 1.0564e-02, -1.7378e-02]]]]])"
       ]
      },
-     "execution_count": 100,
+     "execution_count": 27,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -468,7 +834,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 101,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -481,7 +847,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 102,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -494,7 +860,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 103,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [
     {
@@ -514,23 +880,23 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 104,
+   "execution_count": 31,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Sample State: [ 2.7051659e+00 -1.5201922e+38 -2.6240057e-01 -7.0476830e+37]\n",
-      "Sample State: [ 5.9647828e-01  1.7090080e+38 -1.7759679e-01  1.2014864e+38]\n",
-      "Sample State: [3.5260351e+00 3.2534387e+38 2.8506801e-01 2.4451416e+38]\n",
-      "Sample State: [ 2.9073255e+00 -2.6790161e+38 -2.9842880e-01 -9.2181645e+37]\n",
-      "Sample State: [ 3.7392182e+00 -2.4017081e+38  2.4199644e-01  2.9771261e+38]\n",
-      "Sample State: [-1.6355207e+00 -2.4831722e+38  3.3985817e-01 -3.2439041e+38]\n",
-      "Sample State: [-1.1353507e+00  1.9392829e+38  3.3289015e-01  3.0479645e+38]\n",
-      "Sample State: [3.0383275e+00 2.9051530e+38 3.7706238e-01 1.9133415e+38]\n",
-      "Sample State: [-1.2916708e+00  5.8141520e+37 -1.1080918e-01 -6.8264896e+37]\n",
-      "Sample State: [ 2.1536710e+00 -3.1940789e+38 -4.0808445e-01 -8.4556874e+37]\n"
+      "Sample State: [-3.6452694e+00 -8.9159860e+37  1.4139645e-01  3.0674191e+38]\n",
+      "Sample State: [ 1.4343392e+00 -3.0131075e+38 -2.3563206e-01  1.7383354e+38]\n",
+      "Sample State: [ 3.7291312e+00  1.9267806e+38  8.8896513e-02 -1.9043992e+38]\n",
+      "Sample State: [ 5.6241733e-01 -1.1501083e+38  1.9758487e-01 -2.6513862e+38]\n",
+      "Sample State: [-3.7882154e+00 -1.8343667e+38 -4.1406271e-01  1.2239143e+38]\n",
+      "Sample State: [ 4.1043639e+00 -7.7561222e+37 -3.9738983e-01  2.0008877e+38]\n",
+      "Sample State: [ 3.2407689e+00 -3.1213367e+38 -4.0249658e-01  3.2251934e+38]\n",
+      "Sample State: [ 3.9025934e+00  1.2178617e+38 -6.5442048e-02 -1.2237320e+38]\n",
+      "Sample State: [ 3.4483566e+00 -1.4972215e+38 -1.5894611e-01 -1.3151852e+38]\n",
+      "Sample State: [ 2.0936482e+00  2.7123340e+38 -1.8713233e-01  1.7833031e+38]\n"
      ]
     }
    ],
@@ -543,7 +909,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 107,
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
    "metadata": {},
    "outputs": [
     {
@@ -551,15 +924,15 @@
      "output_type": "stream",
      "text": [
       "Episode: 0\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 1\n",
       "Episode reward: 9.0\n",
       "Episode: 2\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 3\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 4\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 5\n",
       "Episode reward: 9.0\n",
       "Episode: 6\n",
@@ -567,19 +940,19 @@
       "Episode: 7\n",
       "Episode reward: 10.0\n",
       "Episode: 8\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 9\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 10\n",
-      "Episode reward: 10.0\n",
-      "Episode: 11\n",
       "Episode reward: 9.0\n",
+      "Episode: 11\n",
+      "Episode reward: 10.0\n",
       "Episode: 12\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 71.0\n",
       "Episode: 13\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 14\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 15\n",
       "Episode reward: 9.0\n",
       "Episode: 16\n",
@@ -587,23 +960,23 @@
       "Episode: 17\n",
       "Episode reward: 9.0\n",
       "Episode: 18\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 19\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 20\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 32.0\n",
       "Episode: 21\n",
       "Episode reward: 10.0\n",
       "Episode: 22\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 23\n",
       "Episode reward: 8.0\n",
       "Episode: 24\n",
-      "Episode reward: 10.0\n",
-      "Episode: 25\n",
       "Episode reward: 9.0\n",
+      "Episode: 25\n",
+      "Episode reward: 24.0\n",
       "Episode: 26\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 27\n",
       "Episode reward: 8.0\n",
       "Episode: 28\n",
@@ -611,226 +984,1200 @@
       "Episode: 29\n",
       "Episode reward: 9.0\n",
       "Episode: 30\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 31.0\n",
       "Episode: 31\n",
-      "Episode reward: 10.0\n",
-      "Episode: 32\n",
       "Episode reward: 9.0\n",
+      "Episode: 32\n",
+      "Episode reward: 15.0\n",
       "Episode: 33\n",
       "Episode reward: 10.0\n",
       "Episode: 34\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 35\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 8.0\n",
       "Episode: 36\n",
-      "Episode reward: 10.0\n",
-      "Episode: 37\n",
       "Episode reward: 8.0\n",
-      "Episode: 38\n",
+      "Episode: 37\n",
       "Episode reward: 10.0\n",
+      "Episode: 38\n",
+      "Episode reward: 9.0\n",
       "Episode: 39\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 40\n",
       "Episode reward: 10.0\n",
       "Episode: 41\n",
       "Episode reward: 9.0\n",
       "Episode: 42\n",
-      "Episode reward: 10.0\n",
-      "Episode: 43\n",
       "Episode reward: 8.0\n",
+      "Episode: 43\n",
+      "Episode reward: 12.0\n",
       "Episode: 44\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 31.0\n",
       "Episode: 45\n",
       "Episode reward: 10.0\n",
       "Episode: 46\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 47\n",
       "Episode reward: 9.0\n",
       "Episode: 48\n",
       "Episode reward: 9.0\n",
       "Episode: 49\n",
+      "Episode reward: 15.0\n",
+      "Episode: 50\n",
+      "Episode reward: 11.0\n",
+      "Episode: 51\n",
       "Episode reward: 9.0\n",
-      "Learned Q-table:\n",
-      "tensor([[[[[ 2.5774e-04, -1.6386e-02],\n",
-      "           [-6.6059e-03, -1.2301e-02],\n",
-      "           [-1.9540e-03, -6.5953e-03]],\n",
-      "\n",
-      "          [[ 1.4049e-02,  8.9660e-03],\n",
-      "           [-1.1570e-02,  2.3123e-02],\n",
-      "           [ 3.0566e-03, -1.0416e-03]],\n",
-      "\n",
-      "          [[ 1.2075e-02,  2.0824e-02],\n",
-      "           [-2.3481e-03, -1.2590e-02],\n",
-      "           [ 1.6281e-02,  4.2098e-03]]],\n",
-      "\n",
-      "\n",
-      "         [[[ 9.7180e-03, -1.1887e-02],\n",
-      "           [-1.2298e-02,  9.9179e-03],\n",
-      "           [-1.3349e-03,  9.6772e-04]],\n",
-      "\n",
-      "          [[ 2.0868e-02,  2.6027e-03],\n",
-      "           [-1.9472e-02,  4.0829e-03],\n",
-      "           [-9.9453e-03, -7.8108e-03]],\n",
-      "\n",
-      "          [[-3.5114e-03,  8.3449e-04],\n",
-      "           [ 1.1458e-02, -1.4358e-02],\n",
-      "           [-1.6030e-02,  7.0141e-03]]],\n",
-      "\n",
-      "\n",
-      "         [[[-1.1180e-02, -2.2336e-02],\n",
-      "           [-9.6797e-03,  7.0699e-03],\n",
-      "           [ 1.2940e-02,  8.7505e-03]],\n",
-      "\n",
-      "          [[ 8.1591e-03,  1.4591e-02],\n",
-      "           [-8.2323e-04, -7.5956e-03],\n",
-      "           [ 4.1017e-03, -3.9041e-03]],\n",
-      "\n",
-      "          [[ 1.3982e-03,  1.5428e-03],\n",
-      "           [ 4.5702e-03,  2.4447e-02],\n",
-      "           [-3.8499e-03,  1.2206e-02]]]],\n",
-      "\n",
-      "\n",
-      "\n",
-      "        [[[[ 9.5825e-04,  6.5356e-04],\n",
-      "           [-1.6917e-02, -4.4886e-03],\n",
-      "           [-1.6378e-02, -4.9732e-03]],\n",
-      "\n",
-      "          [[-7.8593e-03, -8.5154e-03],\n",
-      "           [ 7.7686e-03, -7.1885e-03],\n",
-      "           [-2.0450e-02, -2.4888e-02]],\n",
-      "\n",
-      "          [[ 9.2768e-03,  1.6148e-02],\n",
-      "           [ 4.1603e-03,  1.9358e-03],\n",
-      "           [-6.2031e-03, -8.5290e-03]]],\n",
-      "\n",
-      "\n",
-      "         [[[ 3.7850e-03,  4.5296e-03],\n",
-      "           [ 7.5043e+00,  8.7035e-03],\n",
-      "           [ 1.1955e-03, -2.3893e-03]],\n",
-      "\n",
-      "          [[ 1.4752e-02,  1.2745e-02],\n",
-      "           [ 5.8729e-03,  8.4895e+00],\n",
-      "           [ 7.5059e-03, -3.2212e-03]],\n",
-      "\n",
-      "          [[-1.5970e-02,  5.5587e-03],\n",
-      "           [-1.5641e-02, -1.2862e-02],\n",
-      "           [ 5.4951e-03, -8.9383e-03]]],\n",
-      "\n",
-      "\n",
-      "         [[[ 7.0806e-03,  2.2394e-02],\n",
-      "           [-1.3594e-03, -1.4554e-02],\n",
-      "           [ 5.4938e-03, -3.7047e-03]],\n",
-      "\n",
-      "          [[ 1.0082e-02,  5.1565e-03],\n",
-      "           [ 1.0319e-02, -9.1123e-03],\n",
-      "           [ 1.3201e-02,  5.4954e-03]],\n",
-      "\n",
-      "          [[-7.8168e-04,  2.4071e-02],\n",
-      "           [-1.9452e-02,  1.1501e-02],\n",
-      "           [-1.6482e-02, -6.7994e-03]]]],\n",
-      "\n",
-      "\n",
-      "\n",
-      "        [[[[ 2.7583e-03,  5.0975e-03],\n",
-      "           [ 5.6140e-03, -2.4850e-03],\n",
-      "           [ 7.4893e-03, -1.1398e-02]],\n",
-      "\n",
-      "          [[-1.3201e-03, -7.3656e-03],\n",
-      "           [ 8.5875e-03, -1.2533e-02],\n",
-      "           [-1.3983e-02, -2.1707e-02]],\n",
-      "\n",
-      "          [[-1.2484e-02,  2.2155e-03],\n",
-      "           [-1.0199e-03,  1.4230e-02],\n",
-      "           [ 1.8682e-03, -3.7607e-03]]],\n",
-      "\n",
-      "\n",
-      "         [[[ 4.1444e-03,  6.7304e-04],\n",
-      "           [-4.1720e-03,  1.5125e-02],\n",
-      "           [ 6.7321e-03, -6.7075e-03]],\n",
-      "\n",
-      "          [[ 7.2062e-03, -2.6330e-02],\n",
-      "           [-8.4435e-03,  1.2237e-02],\n",
-      "           [-1.4670e-03, -4.3704e-04]],\n",
-      "\n",
-      "          [[-1.4862e-02,  1.1071e-02],\n",
-      "           [ 2.9910e-03, -9.0951e-03],\n",
-      "           [-1.2098e-02,  1.2734e-02]]],\n",
-      "\n",
-      "\n",
-      "         [[[ 4.9769e-03, -1.7016e-03],\n",
-      "           [-3.9668e-03,  9.9020e-03],\n",
-      "           [-2.0652e-03, -6.1593e-03]],\n",
-      "\n",
-      "          [[-2.5633e-03, -4.7999e-03],\n",
-      "           [-2.4177e-04,  1.4281e-02],\n",
-      "           [-8.6366e-03,  5.5457e-03]],\n",
-      "\n",
-      "          [[ 1.7575e-02,  2.9216e-03],\n",
-      "           [-6.7075e-03,  1.0145e-02],\n",
-      "           [ 4.7358e-03,  1.2916e-02]]]]])\n"
-     ]
-    }
-   ],
-   "source": [
-    "rewards = []  # List to store rewards for each episode\n",
-    "\n",
-    "# Training loop\n",
-    "for episode in range(num_episodes):\n",
-    "    print(\"Episode:\", episode)\n",
-    "    state, info = env.reset(seed=episode)\n",
-    "    state = discretize_state(state, num_bins)\n",
-    "    episode_reward = 0\n",
-    "\n",
-    "    while True:\n",
-    "        # Choose action using the current Q-table\n",
-    "        action = torch.argmax(q_table[state]).item()\n",
-    "\n",
-    "        # Take the chosen action and observe the next state and reward\n",
-    "        next_state, reward, terminated, truncated, info = env.step(action)\n",
-    "        next_state = discretize_state(next_state, num_bins)\n",
-    "\n",
-    "        # Update the Q-table using the Q-learning update rule\n",
-    "        q_table = update_q_table(q_table, state, action, reward, next_state, learning_rate, discount_factor)\n",
-    "\n",
-    "        episode_reward += reward\n",
-    "        state = next_state\n",
-    "\n",
-    "        if truncated or terminated:\n",
-    "            break\n",
-    "    rewards.append(episode_reward)\n",
-    "    print(\"Episode reward:\", episode_reward)\n",
-    "\n",
-    "# Print the learned Q-table\n",
-    "print(\"Learned Q-table:\")\n",
-    "print(q_table)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 108,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x158a98310>]"
-      ]
-     },
-     "execution_count": 108,
-     "metadata": {},
-     "output_type": "execute_result"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
+      "Episode: 52\n",
+      "Episode reward: 9.0\n",
+      "Episode: 53\n",
+      "Episode reward: 19.0\n",
+      "Episode: 54\n",
+      "Episode reward: 32.0\n",
+      "Episode: 55\n",
+      "Episode reward: 9.0\n",
+      "Episode: 56\n",
+      "Episode reward: 8.0\n",
+      "Episode: 57\n",
+      "Episode reward: 9.0\n",
+      "Episode: 58\n",
+      "Episode reward: 28.0\n",
+      "Episode: 59\n",
+      "Episode reward: 8.0\n",
+      "Episode: 60\n",
+      "Episode reward: 23.0\n",
+      "Episode: 61\n",
+      "Episode reward: 9.0\n",
+      "Episode: 62\n",
+      "Episode reward: 10.0\n",
+      "Episode: 63\n",
+      "Episode reward: 9.0\n",
+      "Episode: 64\n",
+      "Episode reward: 10.0\n",
+      "Episode: 65\n",
+      "Episode reward: 18.0\n",
+      "Episode: 66\n",
+      "Episode reward: 10.0\n",
+      "Episode: 67\n",
+      "Episode reward: 10.0\n",
+      "Episode: 68\n",
+      "Episode reward: 9.0\n",
+      "Episode: 69\n",
+      "Episode reward: 12.0\n",
+      "Episode: 70\n",
+      "Episode reward: 19.0\n",
+      "Episode: 71\n",
+      "Episode reward: 9.0\n",
+      "Episode: 72\n",
+      "Episode reward: 9.0\n",
+      "Episode: 73\n",
+      "Episode reward: 24.0\n",
+      "Episode: 74\n",
+      "Episode reward: 10.0\n",
+      "Episode: 75\n",
+      "Episode reward: 11.0\n",
+      "Episode: 76\n",
+      "Episode reward: 10.0\n",
+      "Episode: 77\n",
+      "Episode reward: 9.0\n",
+      "Episode: 78\n",
+      "Episode reward: 11.0\n",
+      "Episode: 79\n",
+      "Episode reward: 9.0\n",
+      "Episode: 80\n",
+      "Episode reward: 14.0\n",
+      "Episode: 81\n",
+      "Episode reward: 12.0\n",
+      "Episode: 82\n",
+      "Episode reward: 9.0\n",
+      "Episode: 83\n",
+      "Episode reward: 9.0\n",
+      "Episode: 84\n",
+      "Episode reward: 10.0\n",
+      "Episode: 85\n",
+      "Episode reward: 47.0\n",
+      "Episode: 86\n",
+      "Episode reward: 9.0\n",
+      "Episode: 87\n",
+      "Episode reward: 9.0\n",
+      "Episode: 88\n",
+      "Episode reward: 9.0\n",
+      "Episode: 89\n",
+      "Episode reward: 10.0\n",
+      "Episode: 90\n",
+      "Episode reward: 9.0\n",
+      "Episode: 91\n",
+      "Episode reward: 9.0\n",
+      "Episode: 92\n",
+      "Episode reward: 9.0\n",
+      "Episode: 93\n",
+      "Episode reward: 11.0\n",
+      "Episode: 94\n",
+      "Episode reward: 9.0\n",
+      "Episode: 95\n",
+      "Episode reward: 10.0\n",
+      "Episode: 96\n",
+      "Episode reward: 9.0\n",
+      "Episode: 97\n",
+      "Episode reward: 10.0\n",
+      "Episode: 98\n",
+      "Episode reward: 8.0\n",
+      "Episode: 99\n",
+      "Episode reward: 10.0\n",
+      "Episode: 100\n",
+      "Episode reward: 10.0\n",
+      "Episode: 101\n",
+      "Episode reward: 9.0\n",
+      "Episode: 102\n",
+      "Episode reward: 9.0\n",
+      "Episode: 103\n",
+      "Episode reward: 24.0\n",
+      "Episode: 104\n",
+      "Episode reward: 14.0\n",
+      "Episode: 105\n",
+      "Episode reward: 10.0\n",
+      "Episode: 106\n",
+      "Episode reward: 8.0\n",
+      "Episode: 107\n",
+      "Episode reward: 9.0\n",
+      "Episode: 108\n",
+      "Episode reward: 8.0\n",
+      "Episode: 109\n",
+      "Episode reward: 10.0\n",
+      "Episode: 110\n",
+      "Episode reward: 10.0\n",
+      "Episode: 111\n",
+      "Episode reward: 9.0\n",
+      "Episode: 112\n",
+      "Episode reward: 9.0\n",
+      "Episode: 113\n",
+      "Episode reward: 27.0\n",
+      "Episode: 114\n",
+      "Episode reward: 12.0\n",
+      "Episode: 115\n",
+      "Episode reward: 10.0\n",
+      "Episode: 116\n",
+      "Episode reward: 8.0\n",
+      "Episode: 117\n",
+      "Episode reward: 10.0\n",
+      "Episode: 118\n",
+      "Episode reward: 9.0\n",
+      "Episode: 119\n",
+      "Episode reward: 11.0\n",
+      "Episode: 120\n",
+      "Episode reward: 10.0\n",
+      "Episode: 121\n",
+      "Episode reward: 9.0\n",
+      "Episode: 122\n",
+      "Episode reward: 9.0\n",
+      "Episode: 123\n",
+      "Episode reward: 9.0\n",
+      "Episode: 124\n",
+      "Episode reward: 12.0\n",
+      "Episode: 125\n",
+      "Episode reward: 9.0\n",
+      "Episode: 126\n",
+      "Episode reward: 30.0\n",
+      "Episode: 127\n",
+      "Episode reward: 31.0\n",
+      "Episode: 128\n",
+      "Episode reward: 10.0\n",
+      "Episode: 129\n",
+      "Episode reward: 8.0\n",
+      "Episode: 130\n",
+      "Episode reward: 14.0\n",
+      "Episode: 131\n",
+      "Episode reward: 9.0\n",
+      "Episode: 132\n",
+      "Episode reward: 11.0\n",
+      "Episode: 133\n",
+      "Episode reward: 9.0\n",
+      "Episode: 134\n",
+      "Episode reward: 10.0\n",
+      "Episode: 135\n",
+      "Episode reward: 10.0\n",
+      "Episode: 136\n",
+      "Episode reward: 23.0\n",
+      "Episode: 137\n",
+      "Episode reward: 10.0\n",
+      "Episode: 138\n",
+      "Episode reward: 10.0\n",
+      "Episode: 139\n",
+      "Episode reward: 52.0\n",
+      "Episode: 140\n",
+      "Episode reward: 11.0\n",
+      "Episode: 141\n",
+      "Episode reward: 9.0\n",
+      "Episode: 142\n",
+      "Episode reward: 9.0\n",
+      "Episode: 143\n",
+      "Episode reward: 11.0\n",
+      "Episode: 144\n",
+      "Episode reward: 10.0\n",
+      "Episode: 145\n",
+      "Episode reward: 9.0\n",
+      "Episode: 146\n",
+      "Episode reward: 9.0\n",
+      "Episode: 147\n",
+      "Episode reward: 10.0\n",
+      "Episode: 148\n",
+      "Episode reward: 10.0\n",
+      "Episode: 149\n",
+      "Episode reward: 25.0\n",
+      "Episode: 150\n",
+      "Episode reward: 9.0\n",
+      "Episode: 151\n",
+      "Episode reward: 34.0\n",
+      "Episode: 152\n",
+      "Episode reward: 23.0\n",
+      "Episode: 153\n",
+      "Episode reward: 18.0\n",
+      "Episode: 154\n",
+      "Episode reward: 44.0\n",
+      "Episode: 155\n",
+      "Episode reward: 33.0\n",
+      "Episode: 156\n",
+      "Episode reward: 21.0\n",
+      "Episode: 157\n",
+      "Episode reward: 44.0\n",
+      "Episode: 158\n",
+      "Episode reward: 17.0\n",
+      "Episode: 159\n",
+      "Episode reward: 22.0\n",
+      "Episode: 160\n",
+      "Episode reward: 34.0\n",
+      "Episode: 161\n",
+      "Episode reward: 75.0\n",
+      "Episode: 162\n",
+      "Episode reward: 47.0\n",
+      "Episode: 163\n",
+      "Episode reward: 29.0\n",
+      "Episode: 164\n",
+      "Episode reward: 29.0\n",
+      "Episode: 165\n",
+      "Episode reward: 28.0\n",
+      "Episode: 166\n",
+      "Episode reward: 24.0\n",
+      "Episode: 167\n",
+      "Episode reward: 40.0\n",
+      "Episode: 168\n",
+      "Episode reward: 52.0\n",
+      "Episode: 169\n",
+      "Episode reward: 32.0\n",
+      "Episode: 170\n",
+      "Episode reward: 22.0\n",
+      "Episode: 171\n",
+      "Episode reward: 34.0\n",
+      "Episode: 172\n",
+      "Episode reward: 30.0\n",
+      "Episode: 173\n",
+      "Episode reward: 94.0\n",
+      "Episode: 174\n",
+      "Episode reward: 22.0\n",
+      "Episode: 175\n",
+      "Episode reward: 79.0\n",
+      "Episode: 176\n",
+      "Episode reward: 10.0\n",
+      "Episode: 177\n",
+      "Episode reward: 49.0\n",
+      "Episode: 178\n",
+      "Episode reward: 43.0\n",
+      "Episode: 179\n",
+      "Episode reward: 26.0\n",
+      "Episode: 180\n",
+      "Episode reward: 26.0\n",
+      "Episode: 181\n",
+      "Episode reward: 29.0\n",
+      "Episode: 182\n",
+      "Episode reward: 31.0\n",
+      "Episode: 183\n",
+      "Episode reward: 17.0\n",
+      "Episode: 184\n",
+      "Episode reward: 8.0\n",
+      "Episode: 185\n",
+      "Episode reward: 15.0\n",
+      "Episode: 186\n",
+      "Episode reward: 24.0\n",
+      "Episode: 187\n",
+      "Episode reward: 10.0\n",
+      "Episode: 188\n",
+      "Episode reward: 24.0\n",
+      "Episode: 189\n",
+      "Episode reward: 42.0\n",
+      "Episode: 190\n",
+      "Episode reward: 28.0\n",
+      "Episode: 191\n",
+      "Episode reward: 16.0\n",
+      "Episode: 192\n",
+      "Episode reward: 59.0\n",
+      "Episode: 193\n",
+      "Episode reward: 32.0\n",
+      "Episode: 194\n",
+      "Episode reward: 14.0\n",
+      "Episode: 195\n",
+      "Episode reward: 12.0\n",
+      "Episode: 196\n",
+      "Episode reward: 15.0\n",
+      "Episode: 197\n",
+      "Episode reward: 26.0\n",
+      "Episode: 198\n",
+      "Episode reward: 25.0\n",
+      "Episode: 199\n",
+      "Episode reward: 33.0\n",
+      "Episode: 200\n",
+      "Episode reward: 30.0\n",
+      "Episode: 201\n",
+      "Episode reward: 16.0\n",
+      "Episode: 202\n",
+      "Episode reward: 40.0\n",
+      "Episode: 203\n",
+      "Episode reward: 41.0\n",
+      "Episode: 204\n",
+      "Episode reward: 36.0\n",
+      "Episode: 205\n",
+      "Episode reward: 12.0\n",
+      "Episode: 206\n",
+      "Episode reward: 8.0\n",
+      "Episode: 207\n",
+      "Episode reward: 10.0\n",
+      "Episode: 208\n",
+      "Episode reward: 12.0\n",
+      "Episode: 209\n",
+      "Episode reward: 10.0\n",
+      "Episode: 210\n",
+      "Episode reward: 14.0\n",
+      "Episode: 211\n",
+      "Episode reward: 50.0\n",
+      "Episode: 212\n",
+      "Episode reward: 9.0\n",
+      "Episode: 213\n",
+      "Episode reward: 12.0\n",
+      "Episode: 214\n",
+      "Episode reward: 22.0\n",
+      "Episode: 215\n",
+      "Episode reward: 8.0\n",
+      "Episode: 216\n",
+      "Episode reward: 19.0\n",
+      "Episode: 217\n",
+      "Episode reward: 20.0\n",
+      "Episode: 218\n",
+      "Episode reward: 21.0\n",
+      "Episode: 219\n",
+      "Episode reward: 11.0\n",
+      "Episode: 220\n",
+      "Episode reward: 10.0\n",
+      "Episode: 221\n",
+      "Episode reward: 42.0\n",
+      "Episode: 222\n",
+      "Episode reward: 28.0\n",
+      "Episode: 223\n",
+      "Episode reward: 11.0\n",
+      "Episode: 224\n",
+      "Episode reward: 43.0\n",
+      "Episode: 225\n",
+      "Episode reward: 22.0\n",
+      "Episode: 226\n",
+      "Episode reward: 20.0\n",
+      "Episode: 227\n",
+      "Episode reward: 74.0\n",
+      "Episode: 228\n",
+      "Episode reward: 29.0\n",
+      "Episode: 229\n",
+      "Episode reward: 28.0\n",
+      "Episode: 230\n",
+      "Episode reward: 35.0\n",
+      "Episode: 231\n",
+      "Episode reward: 33.0\n",
+      "Episode: 232\n",
+      "Episode reward: 32.0\n",
+      "Episode: 233\n",
+      "Episode reward: 49.0\n",
+      "Episode: 234\n",
+      "Episode reward: 59.0\n",
+      "Episode: 235\n",
+      "Episode reward: 21.0\n",
+      "Episode: 236\n",
+      "Episode reward: 9.0\n",
+      "Episode: 237\n",
+      "Episode reward: 23.0\n",
+      "Episode: 238\n",
+      "Episode reward: 9.0\n",
+      "Episode: 239\n",
+      "Episode reward: 60.0\n",
+      "Episode: 240\n",
+      "Episode reward: 24.0\n",
+      "Episode: 241\n",
+      "Episode reward: 69.0\n",
+      "Episode: 242\n",
+      "Episode reward: 41.0\n",
+      "Episode: 243\n",
+      "Episode reward: 22.0\n",
+      "Episode: 244\n",
+      "Episode reward: 39.0\n",
+      "Episode: 245\n",
+      "Episode reward: 39.0\n",
+      "Episode: 246\n",
+      "Episode reward: 54.0\n",
+      "Episode: 247\n",
+      "Episode reward: 15.0\n",
+      "Episode: 248\n",
+      "Episode reward: 37.0\n",
+      "Episode: 249\n",
+      "Episode reward: 25.0\n",
+      "Episode: 250\n",
+      "Episode reward: 41.0\n",
+      "Episode: 251\n",
+      "Episode reward: 18.0\n",
+      "Episode: 252\n",
+      "Episode reward: 24.0\n",
+      "Episode: 253\n",
+      "Episode reward: 32.0\n",
+      "Episode: 254\n",
+      "Episode reward: 61.0\n",
+      "Episode: 255\n",
+      "Episode reward: 14.0\n",
+      "Episode: 256\n",
+      "Episode reward: 22.0\n",
+      "Episode: 257\n",
+      "Episode reward: 20.0\n",
+      "Episode: 258\n",
+      "Episode reward: 46.0\n",
+      "Episode: 259\n",
+      "Episode reward: 50.0\n",
+      "Episode: 260\n",
+      "Episode reward: 27.0\n",
+      "Episode: 261\n",
+      "Episode reward: 48.0\n",
+      "Episode: 262\n",
+      "Episode reward: 28.0\n",
+      "Episode: 263\n",
+      "Episode reward: 23.0\n",
+      "Episode: 264\n",
+      "Episode reward: 45.0\n",
+      "Episode: 265\n",
+      "Episode reward: 12.0\n",
+      "Episode: 266\n",
+      "Episode reward: 58.0\n",
+      "Episode: 267\n",
+      "Episode reward: 43.0\n",
+      "Episode: 268\n",
+      "Episode reward: 22.0\n",
+      "Episode: 269\n",
+      "Episode reward: 82.0\n",
+      "Episode: 270\n",
+      "Episode reward: 77.0\n",
+      "Episode: 271\n",
+      "Episode reward: 22.0\n",
+      "Episode: 272\n",
+      "Episode reward: 61.0\n",
+      "Episode: 273\n",
+      "Episode reward: 50.0\n",
+      "Episode: 274\n",
+      "Episode reward: 41.0\n",
+      "Episode: 275\n",
+      "Episode reward: 48.0\n",
+      "Episode: 276\n",
+      "Episode reward: 43.0\n",
+      "Episode: 277\n",
+      "Episode reward: 9.0\n",
+      "Episode: 278\n",
+      "Episode reward: 43.0\n",
+      "Episode: 279\n",
+      "Episode reward: 45.0\n",
+      "Episode: 280\n",
+      "Episode reward: 10.0\n",
+      "Episode: 281\n",
+      "Episode reward: 70.0\n",
+      "Episode: 282\n",
+      "Episode reward: 33.0\n",
+      "Episode: 283\n",
+      "Episode reward: 29.0\n",
+      "Episode: 284\n",
+      "Episode reward: 29.0\n",
+      "Episode: 285\n",
+      "Episode reward: 59.0\n",
+      "Episode: 286\n",
+      "Episode reward: 46.0\n",
+      "Episode: 287\n",
+      "Episode reward: 47.0\n",
+      "Episode: 288\n",
+      "Episode reward: 11.0\n",
+      "Episode: 289\n",
+      "Episode reward: 48.0\n",
+      "Episode: 290\n",
+      "Episode reward: 26.0\n",
+      "Episode: 291\n",
+      "Episode reward: 69.0\n",
+      "Episode: 292\n",
+      "Episode reward: 22.0\n",
+      "Episode: 293\n",
+      "Episode reward: 51.0\n",
+      "Episode: 294\n",
+      "Episode reward: 56.0\n",
+      "Episode: 295\n",
+      "Episode reward: 47.0\n",
+      "Episode: 296\n",
+      "Episode reward: 44.0\n",
+      "Episode: 297\n",
+      "Episode reward: 25.0\n",
+      "Episode: 298\n",
+      "Episode reward: 64.0\n",
+      "Episode: 299\n",
+      "Episode reward: 89.0\n",
+      "Episode: 300\n",
+      "Episode reward: 61.0\n",
+      "Episode: 301\n",
+      "Episode reward: 23.0\n",
+      "Episode: 302\n",
+      "Episode reward: 22.0\n",
+      "Episode: 303\n",
+      "Episode reward: 66.0\n",
+      "Episode: 304\n",
+      "Episode reward: 47.0\n",
+      "Episode: 305\n",
+      "Episode reward: 20.0\n",
+      "Episode: 306\n",
+      "Episode reward: 9.0\n",
+      "Episode: 307\n",
+      "Episode reward: 23.0\n",
+      "Episode: 308\n",
+      "Episode reward: 23.0\n",
+      "Episode: 309\n",
+      "Episode reward: 50.0\n",
+      "Episode: 310\n",
+      "Episode reward: 19.0\n",
+      "Episode: 311\n",
+      "Episode reward: 114.0\n",
+      "Episode: 312\n",
+      "Episode reward: 60.0\n",
+      "Episode: 313\n",
+      "Episode reward: 26.0\n",
+      "Episode: 314\n",
+      "Episode reward: 9.0\n",
+      "Episode: 315\n",
+      "Episode reward: 14.0\n",
+      "Episode: 316\n",
+      "Episode reward: 9.0\n",
+      "Episode: 317\n",
+      "Episode reward: 8.0\n",
+      "Episode: 318\n",
+      "Episode reward: 22.0\n",
+      "Episode: 319\n",
+      "Episode reward: 32.0\n",
+      "Episode: 320\n",
+      "Episode reward: 12.0\n",
+      "Episode: 321\n",
+      "Episode reward: 19.0\n",
+      "Episode: 322\n",
+      "Episode reward: 47.0\n",
+      "Episode: 323\n",
+      "Episode reward: 37.0\n",
+      "Episode: 324\n",
+      "Episode reward: 106.0\n",
+      "Episode: 325\n",
+      "Episode reward: 31.0\n",
+      "Episode: 326\n",
+      "Episode reward: 39.0\n",
+      "Episode: 327\n",
+      "Episode reward: 20.0\n",
+      "Episode: 328\n",
+      "Episode reward: 9.0\n",
+      "Episode: 329\n",
+      "Episode reward: 27.0\n",
+      "Episode: 330\n",
+      "Episode reward: 10.0\n",
+      "Episode: 331\n",
+      "Episode reward: 33.0\n",
+      "Episode: 332\n",
+      "Episode reward: 14.0\n",
+      "Episode: 333\n",
+      "Episode reward: 22.0\n",
+      "Episode: 334\n",
+      "Episode reward: 29.0\n",
+      "Episode: 335\n",
+      "Episode reward: 38.0\n",
+      "Episode: 336\n",
+      "Episode reward: 9.0\n",
+      "Episode: 337\n",
+      "Episode reward: 45.0\n",
+      "Episode: 338\n",
+      "Episode reward: 37.0\n",
+      "Episode: 339\n",
+      "Episode reward: 26.0\n",
+      "Episode: 340\n",
+      "Episode reward: 22.0\n",
+      "Episode: 341\n",
+      "Episode reward: 33.0\n",
+      "Episode: 342\n",
+      "Episode reward: 53.0\n",
+      "Episode: 343\n",
+      "Episode reward: 24.0\n",
+      "Episode: 344\n",
+      "Episode reward: 22.0\n",
+      "Episode: 345\n",
+      "Episode reward: 21.0\n",
+      "Episode: 346\n",
+      "Episode reward: 45.0\n",
+      "Episode: 347\n",
+      "Episode reward: 37.0\n",
+      "Episode: 348\n",
+      "Episode reward: 78.0\n",
+      "Episode: 349\n",
+      "Episode reward: 38.0\n",
+      "Episode: 350\n",
+      "Episode reward: 9.0\n",
+      "Episode: 351\n",
+      "Episode reward: 58.0\n",
+      "Episode: 352\n",
+      "Episode reward: 69.0\n",
+      "Episode: 353\n",
+      "Episode reward: 26.0\n",
+      "Episode: 354\n",
+      "Episode reward: 21.0\n",
+      "Episode: 355\n",
+      "Episode reward: 161.0\n",
+      "Episode: 356\n",
+      "Episode reward: 33.0\n",
+      "Episode: 357\n",
+      "Episode reward: 101.0\n",
+      "Episode: 358\n",
+      "Episode reward: 39.0\n",
+      "Episode: 359\n",
+      "Episode reward: 36.0\n",
+      "Episode: 360\n",
+      "Episode reward: 24.0\n",
+      "Episode: 361\n",
+      "Episode reward: 174.0\n",
+      "Episode: 362\n",
+      "Episode reward: 149.0\n",
+      "Episode: 363\n",
+      "Episode reward: 15.0\n",
+      "Episode: 364\n",
+      "Episode reward: 125.0\n",
+      "Episode: 365\n",
+      "Episode reward: 37.0\n",
+      "Episode: 366\n",
+      "Episode reward: 82.0\n",
+      "Episode: 367\n",
+      "Episode reward: 23.0\n",
+      "Episode: 368\n",
+      "Episode reward: 35.0\n",
+      "Episode: 369\n",
+      "Episode reward: 31.0\n",
+      "Episode: 370\n",
+      "Episode reward: 29.0\n",
+      "Episode: 371\n",
+      "Episode reward: 74.0\n",
+      "Episode: 372\n",
+      "Episode reward: 10.0\n",
+      "Episode: 373\n",
+      "Episode reward: 20.0\n",
+      "Episode: 374\n",
+      "Episode reward: 197.0\n",
+      "Episode: 375\n",
+      "Episode reward: 46.0\n",
+      "Episode: 376\n",
+      "Episode reward: 39.0\n",
+      "Episode: 377\n",
+      "Episode reward: 25.0\n",
+      "Episode: 378\n",
+      "Episode reward: 22.0\n",
+      "Episode: 379\n",
+      "Episode reward: 31.0\n",
+      "Episode: 380\n",
+      "Episode reward: 200.0\n",
+      "Episode: 381\n",
+      "Episode reward: 22.0\n",
+      "Episode: 382\n",
+      "Episode reward: 44.0\n",
+      "Episode: 383\n",
+      "Episode reward: 77.0\n",
+      "Episode: 384\n",
+      "Episode reward: 35.0\n",
+      "Episode: 385\n",
+      "Episode reward: 179.0\n",
+      "Episode: 386\n",
+      "Episode reward: 27.0\n",
+      "Episode: 387\n",
+      "Episode reward: 28.0\n",
+      "Episode: 388\n",
+      "Episode reward: 51.0\n",
+      "Episode: 389\n",
+      "Episode reward: 36.0\n",
+      "Episode: 390\n",
+      "Episode reward: 22.0\n",
+      "Episode: 391\n",
+      "Episode reward: 19.0\n",
+      "Episode: 392\n",
+      "Episode reward: 34.0\n",
+      "Episode: 393\n",
+      "Episode reward: 28.0\n",
+      "Episode: 394\n",
+      "Episode reward: 27.0\n",
+      "Episode: 395\n",
+      "Episode reward: 23.0\n",
+      "Episode: 396\n",
+      "Episode reward: 30.0\n",
+      "Episode: 397\n",
+      "Episode reward: 42.0\n",
+      "Episode: 398\n",
+      "Episode reward: 9.0\n",
+      "Episode: 399\n",
+      "Episode reward: 17.0\n",
+      "Episode: 400\n",
+      "Episode reward: 22.0\n",
+      "Episode: 401\n",
+      "Episode reward: 43.0\n",
+      "Episode: 402\n",
+      "Episode reward: 195.0\n",
+      "Episode: 403\n",
+      "Episode reward: 59.0\n",
+      "Episode: 404\n",
+      "Episode reward: 31.0\n",
+      "Episode: 405\n",
+      "Episode reward: 14.0\n",
+      "Episode: 406\n",
+      "Episode reward: 31.0\n",
+      "Episode: 407\n",
+      "Episode reward: 23.0\n",
+      "Episode: 408\n",
+      "Episode reward: 182.0\n",
+      "Episode: 409\n",
+      "Episode reward: 24.0\n",
+      "Episode: 410\n",
+      "Episode reward: 31.0\n",
+      "Episode: 411\n",
+      "Episode reward: 25.0\n",
+      "Episode: 412\n",
+      "Episode reward: 43.0\n",
+      "Episode: 413\n",
+      "Episode reward: 45.0\n",
+      "Episode: 414\n",
+      "Episode reward: 14.0\n",
+      "Episode: 415\n",
+      "Episode reward: 25.0\n",
+      "Episode: 416\n",
+      "Episode reward: 68.0\n",
+      "Episode: 417\n",
+      "Episode reward: 9.0\n",
+      "Episode: 418\n",
+      "Episode reward: 21.0\n",
+      "Episode: 419\n",
+      "Episode reward: 12.0\n",
+      "Episode: 420\n",
+      "Episode reward: 22.0\n",
+      "Episode: 421\n",
+      "Episode reward: 131.0\n",
+      "Episode: 422\n",
+      "Episode reward: 55.0\n",
+      "Episode: 423\n",
+      "Episode reward: 22.0\n",
+      "Episode: 424\n",
+      "Episode reward: 46.0\n",
+      "Episode: 425\n",
+      "Episode reward: 24.0\n",
+      "Episode: 426\n",
+      "Episode reward: 20.0\n",
+      "Episode: 427\n",
+      "Episode reward: 21.0\n",
+      "Episode: 428\n",
+      "Episode reward: 21.0\n",
+      "Episode: 429\n",
+      "Episode reward: 26.0\n",
+      "Episode: 430\n",
+      "Episode reward: 11.0\n",
+      "Episode: 431\n",
+      "Episode reward: 58.0\n",
+      "Episode: 432\n",
+      "Episode reward: 149.0\n",
+      "Episode: 433\n",
+      "Episode reward: 25.0\n",
+      "Episode: 434\n",
+      "Episode reward: 42.0\n",
+      "Episode: 435\n",
+      "Episode reward: 55.0\n",
+      "Episode: 436\n",
+      "Episode reward: 200.0\n",
+      "Episode: 437\n",
+      "Episode reward: 73.0\n",
+      "Episode: 438\n",
+      "Episode reward: 23.0\n",
+      "Episode: 439\n",
+      "Episode reward: 25.0\n",
+      "Episode: 440\n",
+      "Episode reward: 25.0\n",
+      "Episode: 441\n",
+      "Episode reward: 104.0\n",
+      "Episode: 442\n",
+      "Episode reward: 25.0\n",
+      "Episode: 443\n",
+      "Episode reward: 15.0\n",
+      "Episode: 444\n",
+      "Episode reward: 10.0\n",
+      "Episode: 445\n",
+      "Episode reward: 56.0\n",
+      "Episode: 446\n",
+      "Episode reward: 30.0\n",
+      "Episode: 447\n",
+      "Episode reward: 31.0\n",
+      "Episode: 448\n",
+      "Episode reward: 17.0\n",
+      "Episode: 449\n",
+      "Episode reward: 9.0\n",
+      "Episode: 450\n",
+      "Episode reward: 55.0\n",
+      "Episode: 451\n",
+      "Episode reward: 24.0\n",
+      "Episode: 452\n",
+      "Episode reward: 9.0\n",
+      "Episode: 453\n",
+      "Episode reward: 24.0\n",
+      "Episode: 454\n",
+      "Episode reward: 19.0\n",
+      "Episode: 455\n",
+      "Episode reward: 29.0\n",
+      "Episode: 456\n",
+      "Episode reward: 19.0\n",
+      "Episode: 457\n",
+      "Episode reward: 42.0\n",
+      "Episode: 458\n",
+      "Episode reward: 54.0\n",
+      "Episode: 459\n",
+      "Episode reward: 30.0\n",
+      "Episode: 460\n",
+      "Episode reward: 61.0\n",
+      "Episode: 461\n",
+      "Episode reward: 35.0\n",
+      "Episode: 462\n",
+      "Episode reward: 58.0\n",
+      "Episode: 463\n",
+      "Episode reward: 13.0\n",
+      "Episode: 464\n",
+      "Episode reward: 22.0\n",
+      "Episode: 465\n",
+      "Episode reward: 10.0\n",
+      "Episode: 466\n",
+      "Episode reward: 32.0\n",
+      "Episode: 467\n",
+      "Episode reward: 103.0\n",
+      "Episode: 468\n",
+      "Episode reward: 39.0\n",
+      "Episode: 469\n",
+      "Episode reward: 68.0\n",
+      "Episode: 470\n",
+      "Episode reward: 78.0\n",
+      "Episode: 471\n",
+      "Episode reward: 82.0\n",
+      "Episode: 472\n",
+      "Episode reward: 57.0\n",
+      "Episode: 473\n",
+      "Episode reward: 43.0\n",
+      "Episode: 474\n",
+      "Episode reward: 33.0\n",
+      "Episode: 475\n",
+      "Episode reward: 46.0\n",
+      "Episode: 476\n",
+      "Episode reward: 31.0\n",
+      "Episode: 477\n",
+      "Episode reward: 45.0\n",
+      "Episode: 478\n",
+      "Episode reward: 10.0\n",
+      "Episode: 479\n",
+      "Episode reward: 22.0\n",
+      "Episode: 480\n",
+      "Episode reward: 62.0\n",
+      "Episode: 481\n",
+      "Episode reward: 35.0\n",
+      "Episode: 482\n",
+      "Episode reward: 12.0\n",
+      "Episode: 483\n",
+      "Episode reward: 27.0\n",
+      "Episode: 484\n",
+      "Episode reward: 20.0\n",
+      "Episode: 485\n",
+      "Episode reward: 27.0\n",
+      "Episode: 486\n",
+      "Episode reward: 38.0\n",
+      "Episode: 487\n",
+      "Episode reward: 52.0\n",
+      "Episode: 488\n",
+      "Episode reward: 27.0\n",
+      "Episode: 489\n",
+      "Episode reward: 10.0\n",
+      "Episode: 490\n",
+      "Episode reward: 29.0\n",
+      "Episode: 491\n",
+      "Episode reward: 34.0\n",
+      "Episode: 492\n",
+      "Episode reward: 9.0\n",
+      "Episode: 493\n",
+      "Episode reward: 25.0\n",
+      "Episode: 494\n",
+      "Episode reward: 113.0\n",
+      "Episode: 495\n",
+      "Episode reward: 56.0\n",
+      "Episode: 496\n",
+      "Episode reward: 39.0\n",
+      "Episode: 497\n",
+      "Episode reward: 67.0\n",
+      "Episode: 498\n",
+      "Episode reward: 8.0\n",
+      "Episode: 499\n",
+      "Episode reward: 9.0\n",
+      "Episode: 500\n",
+      "Episode reward: 10.0\n",
+      "Episode: 501\n",
+      "Episode reward: 32.0\n",
+      "Episode: 502\n",
+      "Episode reward: 95.0\n",
+      "Episode: 503\n",
+      "Episode reward: 25.0\n",
+      "Episode: 504\n",
+      "Episode reward: 28.0\n",
+      "Episode: 505\n",
+      "Episode reward: 135.0\n",
+      "Episode: 506\n",
+      "Episode reward: 98.0\n",
+      "Episode: 507\n",
+      "Episode reward: 127.0\n",
+      "Episode: 508\n",
+      "Episode reward: 22.0\n",
+      "Episode: 509\n",
+      "Episode reward: 62.0\n",
+      "Episode: 510\n",
+      "Episode reward: 73.0\n",
+      "Episode: 511\n",
+      "Episode reward: 49.0\n",
+      "Episode: 512\n",
+      "Episode reward: 50.0\n",
+      "Episode: 513\n",
+      "Episode reward: 58.0\n",
+      "Episode: 514\n",
+      "Episode reward: 47.0\n",
+      "Episode: 515\n",
+      "Episode reward: 36.0\n",
+      "Episode: 516\n",
+      "Episode reward: 52.0\n",
+      "Episode: 517\n",
+      "Episode reward: 28.0\n",
+      "Episode: 518\n",
+      "Episode reward: 116.0\n",
+      "Episode: 519\n",
+      "Episode reward: 152.0\n",
+      "Episode: 520\n",
+      "Episode reward: 103.0\n",
+      "Episode: 521\n",
+      "Episode reward: 29.0\n",
+      "Episode: 522\n",
+      "Episode reward: 57.0\n",
+      "Episode: 523\n",
+      "Episode reward: 89.0\n",
+      "Episode: 524\n",
+      "Episode reward: 102.0\n",
+      "Episode: 525\n",
+      "Episode reward: 19.0\n",
+      "Episode: 526\n",
+      "Episode reward: 80.0\n",
+      "Episode: 527\n",
+      "Episode reward: 92.0\n",
+      "Episode: 528\n",
+      "Episode reward: 31.0\n",
+      "Episode: 529\n",
+      "Episode reward: 44.0\n",
+      "Episode: 530\n",
+      "Episode reward: 24.0\n",
+      "Episode: 531\n",
+      "Episode reward: 11.0\n",
+      "Episode: 532\n",
+      "Episode reward: 58.0\n",
+      "Episode: 533\n",
+      "Episode reward: 27.0\n",
+      "Episode: 534\n",
+      "Episode reward: 11.0\n",
+      "Episode: 535\n",
+      "Episode reward: 138.0\n",
+      "Episode: 536\n",
+      "Episode reward: 107.0\n",
+      "Episode: 537\n",
+      "Episode reward: 55.0\n",
+      "Episode: 538\n",
+      "Episode reward: 20.0\n",
+      "Episode: 539\n",
+      "Episode reward: 49.0\n",
+      "Episode: 540\n",
+      "Episode reward: 27.0\n",
+      "Episode: 541\n",
+      "Episode reward: 35.0\n",
+      "Episode: 542\n",
+      "Episode reward: 48.0\n",
+      "Episode: 543\n",
+      "Episode reward: 33.0\n",
+      "Episode: 544\n",
+      "Episode reward: 10.0\n",
+      "Episode: 545\n",
+      "Episode reward: 26.0\n",
+      "Episode: 546\n",
+      "Episode reward: 50.0\n",
+      "Episode: 547\n",
+      "Episode reward: 32.0\n",
+      "Episode: 548\n",
+      "Episode reward: 36.0\n",
+      "Episode: 549\n",
+      "Episode reward: 44.0\n",
+      "Episode: 550\n",
+      "Episode reward: 27.0\n",
+      "Episode: 551\n",
+      "Episode reward: 17.0\n",
+      "Episode: 552\n",
+      "Episode reward: 9.0\n",
+      "Episode: 553\n",
+      "Episode reward: 98.0\n",
+      "Episode: 554\n",
+      "Episode reward: 28.0\n",
+      "Episode: 555\n",
+      "Episode reward: 27.0\n",
+      "Episode: 556\n",
+      "Episode reward: 8.0\n",
+      "Episode: 557\n",
+      "Episode reward: 54.0\n",
+      "Episode: 558\n",
+      "Episode reward: 20.0\n",
+      "Episode: 559\n",
+      "Episode reward: 61.0\n",
+      "Episode: 560\n",
+      "Episode reward: 81.0\n",
+      "Episode: 561\n",
+      "Episode reward: 42.0\n",
+      "Episode: 562\n",
+      "Episode reward: 30.0\n",
+      "Episode: 563\n",
+      "Episode reward: 33.0\n",
+      "Episode: 564\n",
+      "Episode reward: 59.0\n",
+      "Episode: 565\n",
+      "Episode reward: 44.0\n",
+      "Episode: 566\n",
+      "Episode reward: 24.0\n",
+      "Episode: 567\n",
+      "Episode reward: 37.0\n",
+      "Episode: 568\n",
+      "Episode reward: 45.0\n",
+      "Episode: 569\n",
+      "Episode reward: 48.0\n",
+      "Episode: 570\n",
+      "Episode reward: 23.0\n",
+      "Episode: 571\n",
+      "Episode reward: 50.0\n",
+      "Episode: 572\n",
+      "Episode reward: 40.0\n",
+      "Episode: 573\n",
+      "Episode reward: 34.0\n",
+      "Episode: 574\n",
+      "Episode reward: 41.0\n",
+      "Episode: 575\n",
+      "Episode reward: 9.0\n",
+      "Episode: 576\n",
+      "Episode reward: 41.0\n",
+      "Episode: 577\n",
+      "Episode reward: 54.0\n",
+      "Episode: 578\n",
+      "Episode reward: 38.0\n",
+      "Episode: 579\n",
+      "Episode reward: 22.0\n",
+      "Episode: 580\n",
+      "Episode reward: 35.0\n",
+      "Episode: 581\n",
+      "Episode reward: 54.0\n",
+      "Episode: 582\n",
+      "Episode reward: 41.0\n",
+      "Episode: 583\n",
+      "Episode reward: 23.0\n",
+      "Episode: 584\n",
+      "Episode reward: 63.0\n",
+      "Episode: 585\n"
+     ]
+    },
+    {
+     "ename": "KeyboardInterrupt",
+     "evalue": "",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
+      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 29\u001b[0m line \u001b[0;36m2\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X26sZmlsZQ%3D%3D?line=16'>17</a>\u001b[0m     action \u001b[39m=\u001b[39m env\u001b[39m.\u001b[39maction_space\u001b[39m.\u001b[39msample()\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X26sZmlsZQ%3D%3D?line=18'>19</a>\u001b[0m \u001b[39m# Take the chosen action and observe the next state and reward\u001b[39;00m\n\u001b[0;32m---> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X26sZmlsZQ%3D%3D?line=19'>20</a>\u001b[0m next_state, reward, terminated, truncated, info \u001b[39m=\u001b[39m env\u001b[39m.\u001b[39;49mstep(action)\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X26sZmlsZQ%3D%3D?line=20'>21</a>\u001b[0m next_state \u001b[39m=\u001b[39m discretize_state(next_state, num_bins)\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X26sZmlsZQ%3D%3D?line=22'>23</a>\u001b[0m \u001b[39m# Update the Q-table using the Q-learning update rule\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/gymnasium/wrappers/time_limit.py:57\u001b[0m, in \u001b[0;36mTimeLimit.step\u001b[0;34m(self, action)\u001b[0m\n\u001b[1;32m     46\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mstep\u001b[39m(\u001b[39mself\u001b[39m, action):\n\u001b[1;32m     47\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"Steps through the environment and if the number of steps elapsed exceeds ``max_episode_steps`` then truncate.\u001b[39;00m\n\u001b[1;32m     48\u001b[0m \n\u001b[1;32m     49\u001b[0m \u001b[39m    Args:\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     55\u001b[0m \n\u001b[1;32m     56\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m---> 57\u001b[0m     observation, reward, terminated, truncated, info \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49menv\u001b[39m.\u001b[39;49mstep(action)\n\u001b[1;32m     58\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_elapsed_steps \u001b[39m+\u001b[39m\u001b[39m=\u001b[39m \u001b[39m1\u001b[39m\n\u001b[1;32m     60\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_elapsed_steps \u001b[39m>\u001b[39m\u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_max_episode_steps:\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/gymnasium/wrappers/order_enforcing.py:56\u001b[0m, in \u001b[0;36mOrderEnforcing.step\u001b[0;34m(self, action)\u001b[0m\n\u001b[1;32m     54\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_has_reset:\n\u001b[1;32m     55\u001b[0m     \u001b[39mraise\u001b[39;00m ResetNeeded(\u001b[39m\"\u001b[39m\u001b[39mCannot call env.step() before calling env.reset()\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[0;32m---> 56\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49menv\u001b[39m.\u001b[39;49mstep(action)\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/gymnasium/wrappers/env_checker.py:51\u001b[0m, in \u001b[0;36mPassiveEnvChecker.step\u001b[0;34m(self, action)\u001b[0m\n\u001b[1;32m     49\u001b[0m     \u001b[39mreturn\u001b[39;00m env_step_passive_checker(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39menv, action)\n\u001b[1;32m     50\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m---> 51\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49menv\u001b[39m.\u001b[39;49mstep(action)\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/gymnasium/envs/classic_control/cartpole.py:190\u001b[0m, in \u001b[0;36mCartPoleEnv.step\u001b[0;34m(self, action)\u001b[0m\n\u001b[1;32m    187\u001b[0m     reward \u001b[39m=\u001b[39m \u001b[39m0.0\u001b[39m\n\u001b[1;32m    189\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mrender_mode \u001b[39m==\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mhuman\u001b[39m\u001b[39m\"\u001b[39m:\n\u001b[0;32m--> 190\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mrender()\n\u001b[1;32m    191\u001b[0m \u001b[39mreturn\u001b[39;00m np\u001b[39m.\u001b[39marray(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mstate, dtype\u001b[39m=\u001b[39mnp\u001b[39m.\u001b[39mfloat32), reward, terminated, \u001b[39mFalse\u001b[39;00m, {}\n",
+      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/gymnasium/envs/classic_control/cartpole.py:302\u001b[0m, in \u001b[0;36mCartPoleEnv.render\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    300\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mrender_mode \u001b[39m==\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mhuman\u001b[39m\u001b[39m\"\u001b[39m:\n\u001b[1;32m    301\u001b[0m     pygame\u001b[39m.\u001b[39mevent\u001b[39m.\u001b[39mpump()\n\u001b[0;32m--> 302\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mclock\u001b[39m.\u001b[39;49mtick(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mmetadata[\u001b[39m\"\u001b[39;49m\u001b[39mrender_fps\u001b[39;49m\u001b[39m\"\u001b[39;49m])\n\u001b[1;32m    303\u001b[0m     pygame\u001b[39m.\u001b[39mdisplay\u001b[39m.\u001b[39mflip()\n\u001b[1;32m    305\u001b[0m \u001b[39melif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mrender_mode \u001b[39m==\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mrgb_array\u001b[39m\u001b[39m\"\u001b[39m:\n",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+     ]
+    }
+   ],
+   "source": [
+    "eps = 0.1\n",
+    "num_episodes = 1000\n",
+    "rewards = []  # List to store rewards for each episode\n",
+    "\n",
+    "# Training loop\n",
+    "for episode in range(num_episodes):\n",
+    "    print(\"Episode:\", episode)\n",
+    "    state, info = env.reset(seed=episode)\n",
+    "    state = discretize_state(state, num_bins)\n",
+    "    episode_reward = 0\n",
+    "\n",
+    "    while True:\n",
+    "        # Choose action using the current Q-table or explore the environment\n",
+    "        if np.random.random() > eps:\n",
+    "            action = torch.argmax(q_table[state]).item()\n",
+    "        else:\n",
+    "            action = env.action_space.sample()\n",
+    "\n",
+    "        # Take the chosen action and observe the next state and reward\n",
+    "        next_state, reward, terminated, truncated, info = env.step(action)\n",
+    "        next_state = discretize_state(next_state, num_bins)\n",
+    "\n",
+    "        # Update the Q-table using the Q-learning update rule\n",
+    "        q_table = update_q_table(q_table, state, action, reward, next_state, learning_rate, discount_factor)\n",
+    "\n",
+    "        episode_reward += reward\n",
+    "        state = next_state\n",
+    "\n",
+    "        if truncated or terminated:\n",
+    "            break\n",
+    "    rewards.append(episode_reward)\n",
+    "    print(\"Episode reward:\", episode_reward)\n",
+    "\n",
+    "# Print the learned Q-table\n",
+    "print(\"Learned Q-table:\")\n",
+    "print(q_table)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[<matplotlib.lines.Line2D at 0x161cdf1f0>]"
+      ]
+     },
+     "execution_count": 33,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
      "metadata": {
       "image/png": {
        "height": 413,
-       "width": 556
+       "width": 555
       }
      },
      "output_type": "display_data"
@@ -840,6 +2187,361 @@
     "plt.plot(rewards)"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[<matplotlib.lines.Line2D at 0x160fc95b0>]"
+      ]
+     },
+     "execution_count": 34,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "image/png": {
+       "height": 413,
+       "width": 546
+      }
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Smooth the rewards and plot\n",
+    "def smooth_rewards(rewards, smoothing_factor=100):\n",
+    "    smoothed_rewards = []\n",
+    "    for i in range(len(rewards)):\n",
+    "        smoothed_rewards.append(np.mean(rewards[max(0, i-smoothing_factor):(i+1)]))\n",
+    "    return smoothed_rewards\n",
+    "\n",
+    "plt.plot(smooth_rewards(rewards))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Simple MLP for CartPole\n",
+    "\n",
+    "import torch.nn as nn\n",
+    "import torch.nn.functional as F\n",
+    "\n",
+    "class MLP(nn.Module):\n",
+    "    def __init__(self, input_dim, output_dim, hidden_dims=[32, 32]):\n",
+    "        super(MLP, self).__init__()\n",
+    "        self.input_dim = input_dim\n",
+    "        self.output_dim = output_dim\n",
+    "        self.hidden_dims = hidden_dims\n",
+    "\n",
+    "        self.fc1 = nn.Linear(self.input_dim, self.hidden_dims[0])\n",
+    "        self.fc2 = nn.Linear(self.hidden_dims[0], self.hidden_dims[1])\n",
+    "        self.fc3 = nn.Linear(self.hidden_dims[1], self.output_dim)\n",
+    "\n",
+    "    def forward(self, x):\n",
+    "        x = F.relu(self.fc1(x))\n",
+    "        x = F.relu(self.fc2(x))\n",
+    "        q_values = self.fc3(x)\n",
+    "        return q_values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "mlp = MLP(input_dim=4, output_dim=2, hidden_dims=[32, 32])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "MLP(\n",
+       "  (fc1): Linear(in_features=4, out_features=32, bias=True)\n",
+       "  (fc2): Linear(in_features=32, out_features=32, bias=True)\n",
+       "  (fc3): Linear(in_features=32, out_features=2, bias=True)\n",
+       ")"
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "mlp"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "tensor([-0.0455, -0.0736], grad_fn=<AddBackward0>)"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "mlp(torch.tensor([1, 2, 3, 4], dtype=torch.float32))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "tensor([ 3.7970e+35, -1.6602e+37], grad_fn=<AddBackward0>)"
+      ]
+     },
+     "execution_count": 22,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "mlp(torch.from_numpy(env.observation_space.sample()))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Episode: 0\n",
+      "Episode reward: 8.0\n",
+      "Episode: 1\n",
+      "Episode reward: 9.0\n",
+      "Episode: 2\n",
+      "Episode reward: 10.0\n",
+      "Episode: 3\n",
+      "Episode reward: 10.0\n",
+      "Episode: 4\n",
+      "Episode reward: 10.0\n",
+      "Episode: 5\n",
+      "Episode reward: 9.0\n",
+      "Episode: 6\n",
+      "Episode reward: 9.0\n",
+      "Episode: 7\n",
+      "Episode reward: 10.0\n",
+      "Episode: 8\n",
+      "Episode reward: 9.0\n",
+      "Episode: 9\n",
+      "Episode reward: 10.0\n",
+      "Episode: 10\n",
+      "Episode reward: 10.0\n",
+      "Episode: 11\n",
+      "Episode reward: 9.0\n",
+      "Episode: 12\n",
+      "Episode reward: 9.0\n",
+      "Episode: 13\n",
+      "Episode reward: 10.0\n",
+      "Episode: 14\n",
+      "Episode reward: 10.0\n",
+      "Episode: 15\n",
+      "Episode reward: 9.0\n",
+      "Episode: 16\n",
+      "Episode reward: 8.0\n",
+      "Episode: 17\n",
+      "Episode reward: 9.0\n",
+      "Episode: 18\n",
+      "Episode reward: 9.0\n",
+      "Episode: 19\n",
+      "Episode reward: 9.0\n",
+      "Episode: 20\n",
+      "Episode reward: 8.0\n",
+      "Episode: 21\n",
+      "Episode reward: 10.0\n",
+      "Episode: 22\n",
+      "Episode reward: 8.0\n",
+      "Episode: 23\n",
+      "Episode reward: 8.0\n",
+      "Episode: 24\n",
+      "Episode reward: 10.0\n",
+      "Episode: 25\n",
+      "Episode reward: 9.0\n",
+      "Episode: 26\n",
+      "Episode reward: 8.0\n",
+      "Episode: 27\n",
+      "Episode reward: 8.0\n",
+      "Episode: 28\n",
+      "Episode reward: 10.0\n",
+      "Episode: 29\n",
+      "Episode reward: 9.0\n",
+      "Episode: 30\n",
+      "Episode reward: 8.0\n",
+      "Episode: 31\n",
+      "Episode reward: 10.0\n",
+      "Episode: 32\n",
+      "Episode reward: 9.0\n",
+      "Episode: 33\n",
+      "Episode reward: 10.0\n",
+      "Episode: 34\n",
+      "Episode reward: 9.0\n",
+      "Episode: 35\n",
+      "Episode reward: 11.0\n",
+      "Episode: 36\n",
+      "Episode reward: 10.0\n",
+      "Episode: 37\n",
+      "Episode reward: 8.0\n",
+      "Episode: 38\n",
+      "Episode reward: 10.0\n",
+      "Episode: 39\n",
+      "Episode reward: 8.0\n",
+      "Episode: 40\n",
+      "Episode reward: 10.0\n",
+      "Episode: 41\n",
+      "Episode reward: 9.0\n",
+      "Episode: 42\n",
+      "Episode reward: 10.0\n",
+      "Episode: 43\n",
+      "Episode reward: 8.0\n",
+      "Episode: 44\n",
+      "Episode reward: 9.0\n",
+      "Episode: 45\n",
+      "Episode reward: 10.0\n",
+      "Episode: 46\n",
+      "Episode reward: 9.0\n",
+      "Episode: 47\n",
+      "Episode reward: 9.0\n",
+      "Episode: 48\n",
+      "Episode reward: 9.0\n",
+      "Episode: 49\n",
+      "Episode reward: 9.0\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Train the MLP\n",
+    "\n",
+    "# Hyperparameters\n",
+    "learning_rate = 0.1\n",
+    "discount_factor = 0.9\n",
+    "num_episodes = 50\n",
+    "\n",
+    "# Initialize the MLP\n",
+    "mlp = MLP(input_dim=4, output_dim=2, hidden_dims=[32, 32])\n",
+    "\n",
+    "# Define the loss function\n",
+    "loss_fn = nn.MSELoss()\n",
+    "\n",
+    "# Define the optimizer\n",
+    "optimizer = torch.optim.Adam(mlp.parameters(), lr=learning_rate)\n",
+    "\n",
+    "# List to store rewards for each episode\n",
+    "rewards = []\n",
+    "\n",
+    "# Training loop\n",
+    "for episode in range(num_episodes):\n",
+    "    print(\"Episode:\", episode)\n",
+    "    state, info = env.reset(seed=episode)\n",
+    "    state = torch.from_numpy(state).float()\n",
+    "    episode_reward = 0\n",
+    "\n",
+    "    while True:\n",
+    "        # Choose action using the current Q-table\n",
+    "        q_values = mlp(state)\n",
+    "        action = torch.argmax(q_values).item()\n",
+    "\n",
+    "        # Take the chosen action and observe the next state and reward\n",
+    "        next_state, reward, terminated, truncated, info = env.step(action)\n",
+    "        next_state = torch.from_numpy(next_state).float()\n",
+    "\n",
+    "        # Update the Q-table using the Q-learning update rule\n",
+    "        q_values_next = mlp(next_state)\n",
+    "        q_values_target = q_values.clone()\n",
+    "        q_values_target[action] = reward + discount_factor * torch.max(q_values_next)\n",
+    "        loss = loss_fn(q_values, q_values_target)\n",
+    "\n",
+    "        optimizer.zero_grad()\n",
+    "        loss.backward()\n",
+    "        optimizer.step()\n",
+    "\n",
+    "        episode_reward += reward\n",
+    "        state = next_state\n",
+    "\n",
+    "        if terminated or truncated:\n",
+    "            break\n",
+    "    rewards.append(episode_reward)\n",
+    "    print(\"Episode reward:\", episode_reward)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "MLP(\n",
+       "  (fc1): Linear(in_features=4, out_features=32, bias=True)\n",
+       "  (fc2): Linear(in_features=32, out_features=32, bias=True)\n",
+       "  (fc3): Linear(in_features=32, out_features=2, bias=True)\n",
+       ")"
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "mlp"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "tensor([-inf, nan], grad_fn=<AddBackward0>)"
+      ]
+     },
+     "execution_count": 25,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "mlp(torch.from_numpy(env.observation_space.sample()).float())"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,

From 22664559fdac256249888c9a26a035089b24473f Mon Sep 17 00:00:00 2001
From: Nipun Batra <nipunbatra0@gmail.com>
Date: Tue, 12 Dec 2023 11:21:25 +0530
Subject: [PATCH 4/5] added some details of the cartpole reference.

---
 posts/2023-Dec-11-gym.ipynb | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/posts/2023-Dec-11-gym.ipynb b/posts/2023-Dec-11-gym.ipynb
index 5c6a90a..c34d638 100644
--- a/posts/2023-Dec-11-gym.ipynb
+++ b/posts/2023-Dec-11-gym.ipynb
@@ -19,6 +19,15 @@
     "\n"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Reference\n",
+    "\n",
+    "1. [Detailed Explanation and Python Implementation of Q-Learning Algorithm in OpenAI Gym (Cart-Pole)](https://www.youtube.com/watch?v=KMjQmG5Uzis)\n"
+   ]
+  },
   {
    "attachments": {},
    "cell_type": "markdown",

From d74212ca9b2dd6dc2e769f144285400138a84064 Mon Sep 17 00:00:00 2001
From: Nipun Batra <nipunbatra0@gmail.com>
Date: Tue, 12 Dec 2023 17:32:43 +0530
Subject: [PATCH 5/5] added decaying epsilon

---
 posts/2023-Dec-11-gym.ipynb | 5055 ++++++++++++++++++++++++++++-------
 1 file changed, 4108 insertions(+), 947 deletions(-)

diff --git a/posts/2023-Dec-11-gym.ipynb b/posts/2023-Dec-11-gym.ipynb
index c34d638..0ab5b99 100644
--- a/posts/2023-Dec-11-gym.ipynb
+++ b/posts/2023-Dec-11-gym.ipynb
@@ -39,7 +39,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 3,
    "id": "f41ca63d",
    "metadata": {},
    "outputs": [],
@@ -55,7 +55,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -114,7 +114,7 @@
        " 'GymV26Environment-v0']"
       ]
      },
-     "execution_count": 2,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -126,7 +126,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
@@ -139,13 +139,13 @@
     }
    ],
    "source": [
-    "env = gym.make(\"CartPole-v0\", render_mode=\"human\")\n",
+    "env = gym.make(\"CartPole-v0\")\n",
     "observation, info = env.reset(seed=42)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -154,7 +154,7 @@
        "Discrete(2)"
       ]
      },
-     "execution_count": 4,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -165,16 +165,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "0"
+       "1"
       ]
      },
-     "execution_count": 5,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -185,16 +185,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "0"
+       "1"
       ]
      },
-     "execution_count": 6,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -205,7 +205,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
@@ -222,7 +222,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
@@ -231,7 +231,7 @@
        "array([ 0.0273956 , -0.00611216,  0.03585979,  0.0197368 ], dtype=float32)"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -242,14 +242,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "1\n"
+      "0\n"
      ]
     }
    ],
@@ -260,7 +260,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -276,7 +276,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "[ 0.02727336  0.18847767  0.03625453 -0.26141977] 1.0 False False {}\n"
+      "[ 0.02727336 -0.20172954  0.03625453  0.32351476] 1.0 False False {}\n"
      ]
     }
    ],
@@ -293,7 +293,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "[ 0.03104291  0.38306385  0.03102613 -0.5424507 ] 1.0 False False {}\n"
+      "[ 0.02323877 -0.39734846  0.04272482  0.62740684] 1.0 False False {}\n"
      ]
     }
    ],
@@ -368,9 +368,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 17,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/nipun/miniconda3/lib/python3.9/site-packages/gymnasium/envs/classic_control/cartpole.py:180: UserWarning: \u001b[33mWARN: You are calling 'step()' even though this environment has already returned terminated = True. You should always call 'reset()' once you receive 'terminated = True' -- any further steps are undefined behavior.\u001b[0m\n",
+      "  logger.warn(\n"
+     ]
+    }
+   ],
    "source": [
     "for _ in range(100):\n",
     "   action = env.action_space.sample()  # this is where you would insert your policy\n",
@@ -383,7 +392,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
@@ -417,7 +426,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 41,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -428,7 +437,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 59,
+   "execution_count": 42,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -438,16 +447,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 63,
+   "execution_count": 43,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "array([7])"
+       "array([1])"
       ]
      },
-     "execution_count": 63,
+     "execution_count": 43,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -458,7 +467,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 44,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -487,352 +496,352 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 45,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "tensor([[[[[-1.1406e-02, -1.0509e-02],\n",
-       "           [ 1.3085e-03,  1.3717e-03],\n",
-       "           [-6.0147e-03,  1.3680e-02],\n",
-       "           [-1.2355e-02, -1.0242e-02]],\n",
-       "\n",
-       "          [[-7.4378e-03,  1.3998e-03],\n",
-       "           [ 8.4903e-03,  2.0094e-03],\n",
-       "           [ 3.5405e-03, -2.5198e-03],\n",
-       "           [ 6.9709e-03,  1.0338e-02]],\n",
-       "\n",
-       "          [[-6.8366e-03,  9.4885e-03],\n",
-       "           [ 7.2773e-03, -6.8727e-03],\n",
-       "           [ 1.7096e-02,  1.0054e-02],\n",
-       "           [-9.2244e-03,  2.8881e-05]],\n",
-       "\n",
-       "          [[ 5.0180e-03, -1.0984e-02],\n",
-       "           [ 6.4403e-03,  1.6484e-02],\n",
-       "           [ 6.2945e-03, -1.0362e-02],\n",
-       "           [ 1.0791e-03, -1.1073e-02]]],\n",
-       "\n",
-       "\n",
-       "         [[[ 1.0317e-02, -3.5725e-03],\n",
-       "           [-3.7283e-03,  5.4095e-03],\n",
-       "           [ 1.6884e-02, -2.2501e-03],\n",
-       "           [-4.3716e-03, -8.7110e-03]],\n",
-       "\n",
-       "          [[-4.2364e-03,  1.4712e-02],\n",
-       "           [ 1.1050e-03,  1.4572e-02],\n",
-       "           [-3.4920e-03, -1.4701e-02],\n",
-       "           [ 1.1567e-02,  2.2020e-02]],\n",
-       "\n",
-       "          [[-8.6948e-04, -2.7715e-02],\n",
-       "           [-2.0818e-03,  9.2057e-03],\n",
-       "           [-2.1844e-03, -5.6509e-03],\n",
-       "           [-1.4649e-02, -2.0757e-02]],\n",
-       "\n",
-       "          [[ 1.4552e-02, -1.1368e-03],\n",
-       "           [ 1.7741e-03,  1.2114e-03],\n",
-       "           [ 3.1583e-03,  7.2406e-03],\n",
-       "           [-4.3346e-04, -1.6101e-02]]],\n",
-       "\n",
-       "\n",
-       "         [[[-2.5550e-04,  1.7958e-02],\n",
-       "           [-5.8764e-03, -7.2153e-04],\n",
-       "           [-4.4413e-03,  2.5568e-03],\n",
-       "           [-8.1304e-03,  4.7102e-03]],\n",
-       "\n",
-       "          [[-1.2885e-03, -5.7140e-03],\n",
-       "           [ 4.2125e-03, -6.7077e-03],\n",
-       "           [ 8.2713e-04,  1.6447e-02],\n",
-       "           [ 4.4602e-03, -6.7215e-03]],\n",
-       "\n",
-       "          [[ 2.2286e-04, -5.4333e-03],\n",
-       "           [-5.5148e-03,  7.2770e-03],\n",
-       "           [-7.1460e-03,  2.2344e-02],\n",
-       "           [ 9.4010e-03, -1.0193e-04]],\n",
-       "\n",
-       "          [[-1.2127e-02, -1.0535e-02],\n",
-       "           [-1.7398e-02,  7.0977e-03],\n",
-       "           [-5.5518e-03,  1.7752e-02],\n",
-       "           [ 4.2644e-03,  6.5037e-04]]],\n",
-       "\n",
-       "\n",
-       "         [[[-1.7868e-03, -4.7429e-03],\n",
-       "           [ 3.5122e-03, -3.5208e-03],\n",
-       "           [-7.4866e-03, -1.0636e-02],\n",
-       "           [ 1.0040e-02, -1.1361e-02]],\n",
-       "\n",
-       "          [[-2.1231e-03, -5.4856e-04],\n",
-       "           [-2.5714e-04,  1.5240e-03],\n",
-       "           [ 1.1099e-02,  7.3871e-03],\n",
-       "           [-1.9287e-02,  1.2078e-02]],\n",
-       "\n",
-       "          [[-7.6529e-03, -8.6546e-03],\n",
-       "           [ 8.2139e-03,  1.1296e-02],\n",
-       "           [ 1.9106e-02, -2.2947e-03],\n",
-       "           [ 2.8244e-03,  1.3176e-02]],\n",
-       "\n",
-       "          [[-4.5813e-03,  7.6768e-04],\n",
-       "           [ 3.7834e-03, -3.0948e-03],\n",
-       "           [-1.2757e-02,  8.3703e-03],\n",
-       "           [ 5.4267e-04,  4.7438e-04]]]],\n",
-       "\n",
-       "\n",
-       "\n",
-       "        [[[[ 2.9412e-03,  8.9573e-03],\n",
-       "           [ 1.7503e-02, -7.0822e-03],\n",
-       "           [-1.3228e-02,  8.8431e-03],\n",
-       "           [-1.8440e-02,  3.7296e-03]],\n",
-       "\n",
-       "          [[ 7.3911e-03,  8.4983e-03],\n",
-       "           [-4.7013e-03, -2.8349e-02],\n",
-       "           [ 4.0318e-03,  2.3796e-02],\n",
-       "           [-1.3484e-02,  1.8661e-02]],\n",
-       "\n",
-       "          [[ 1.1991e-02,  2.6918e-05],\n",
-       "           [ 1.3168e-02, -1.1984e-03],\n",
-       "           [-2.1404e-02, -2.9017e-03],\n",
-       "           [ 7.4044e-03,  5.5205e-03]],\n",
-       "\n",
-       "          [[-5.1570e-03, -7.7488e-03],\n",
-       "           [-1.0394e-02,  5.5803e-03],\n",
-       "           [-4.9909e-04,  2.5824e-02],\n",
-       "           [-9.5470e-03,  1.2008e-03]]],\n",
-       "\n",
-       "\n",
-       "         [[[-5.0071e-03,  3.1824e-03],\n",
-       "           [ 6.2639e-03, -1.1144e-02],\n",
-       "           [ 7.6656e-04,  1.6723e-02],\n",
-       "           [-6.5249e-03, -7.0401e-03]],\n",
-       "\n",
-       "          [[-6.7663e-03, -1.2750e-03],\n",
-       "           [ 2.2510e-03,  4.3926e-03],\n",
-       "           [-1.6809e-03,  8.1863e-03],\n",
-       "           [-1.5351e-02, -1.1513e-03]],\n",
-       "\n",
-       "          [[ 2.3877e-03, -2.0612e-03],\n",
-       "           [ 1.1877e-02,  8.4327e-03],\n",
-       "           [-1.1223e-03, -4.2275e-03],\n",
-       "           [-1.5361e-03, -1.3486e-03]],\n",
-       "\n",
-       "          [[-3.0515e-03, -8.0040e-03],\n",
-       "           [ 1.0299e-03, -1.3834e-04],\n",
-       "           [-1.9414e-02, -7.4649e-03],\n",
-       "           [ 9.8406e-03,  1.7082e-02]]],\n",
-       "\n",
-       "\n",
-       "         [[[-7.0127e-03, -1.6560e-03],\n",
-       "           [-2.8100e-03, -5.4193e-03],\n",
-       "           [-1.6859e-02,  1.1542e-02],\n",
-       "           [-1.1415e-02,  1.4802e-02]],\n",
-       "\n",
-       "          [[ 6.0110e-03,  1.5143e-02],\n",
-       "           [ 4.0649e-03, -8.9268e-03],\n",
-       "           [-1.8577e-03, -2.3410e-02],\n",
-       "           [-3.9312e-03,  1.2311e-03]],\n",
-       "\n",
-       "          [[ 1.8359e-03, -3.4620e-03],\n",
-       "           [ 1.3212e-02, -3.8299e-03],\n",
-       "           [-1.0617e-02, -9.8063e-03],\n",
-       "           [ 7.7875e-03, -1.1448e-02]],\n",
-       "\n",
-       "          [[-9.5362e-03,  1.5079e-02],\n",
-       "           [-6.8809e-03,  8.7390e-04],\n",
-       "           [ 8.9715e-03,  1.6310e-03],\n",
-       "           [-1.3025e-02,  6.7466e-03]]],\n",
-       "\n",
-       "\n",
-       "         [[[ 2.5851e-03,  6.6429e-03],\n",
-       "           [ 1.2671e-02, -8.8480e-03],\n",
-       "           [-3.8258e-03,  1.5023e-03],\n",
-       "           [ 9.2233e-03,  6.4037e-03]],\n",
-       "\n",
-       "          [[ 5.9047e-03, -6.3933e-03],\n",
-       "           [-8.4737e-03,  8.7749e-03],\n",
-       "           [ 7.2186e-03, -1.0333e-02],\n",
-       "           [-7.0906e-03,  2.4680e-02]],\n",
+       "tensor([[[[[ 7.5282e-03, -5.0158e-03],\n",
+       "           [ 1.8066e-04, -1.1957e-02],\n",
+       "           [ 1.7809e-02,  1.9935e-02],\n",
+       "           [ 1.2077e-02,  1.2135e-03]],\n",
+       "\n",
+       "          [[ 1.1068e-02,  1.3167e-02],\n",
+       "           [-3.6065e-03, -2.3091e-02],\n",
+       "           [-5.3917e-03, -3.6806e-03],\n",
+       "           [-2.0194e-02,  1.4977e-02]],\n",
+       "\n",
+       "          [[-3.9683e-03, -1.1291e-03],\n",
+       "           [ 1.6093e-03,  1.7743e-02],\n",
+       "           [ 3.1133e-03, -1.3254e-02],\n",
+       "           [-2.2300e-03,  1.5660e-02]],\n",
+       "\n",
+       "          [[-9.9277e-03, -3.0461e-03],\n",
+       "           [ 1.5977e-02, -9.5863e-03],\n",
+       "           [ 9.4414e-03, -4.6137e-03],\n",
+       "           [-6.0294e-03,  3.8514e-03]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 3.2934e-03, -2.7038e-03],\n",
+       "           [ 1.1472e-03, -7.2562e-03],\n",
+       "           [ 7.2273e-03, -7.4928e-03],\n",
+       "           [-3.5465e-03, -1.3511e-02]],\n",
+       "\n",
+       "          [[-1.1883e-02, -3.6573e-03],\n",
+       "           [-9.0871e-03, -1.4479e-02],\n",
+       "           [ 1.2498e-04,  2.4612e-03],\n",
+       "           [-1.4339e-02, -3.3635e-03]],\n",
+       "\n",
+       "          [[ 1.4458e-02,  2.2707e-02],\n",
+       "           [ 1.1106e-03, -1.9436e-02],\n",
+       "           [ 1.7882e-02,  5.9812e-03],\n",
+       "           [ 2.0743e-02, -8.5244e-03]],\n",
+       "\n",
+       "          [[-3.0426e-03, -1.1320e-03],\n",
+       "           [-1.7067e-02,  5.2065e-03],\n",
+       "           [ 5.5506e-03, -2.2826e-03],\n",
+       "           [-3.7662e-03, -9.6092e-03]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 4.2712e-03, -1.7112e-02],\n",
+       "           [-7.6057e-04,  1.5980e-02],\n",
+       "           [ 2.6001e-03,  4.0400e-03],\n",
+       "           [-1.3805e-03, -1.1307e-02]],\n",
+       "\n",
+       "          [[ 5.0641e-03, -1.3241e-02],\n",
+       "           [ 1.7783e-03,  8.7516e-03],\n",
+       "           [-9.8789e-03, -9.4022e-03],\n",
+       "           [ 1.1799e-02, -8.4989e-03]],\n",
+       "\n",
+       "          [[-8.7781e-03, -1.9099e-02],\n",
+       "           [-2.1311e-03,  1.3072e-02],\n",
+       "           [-9.6554e-03, -9.8139e-03],\n",
+       "           [ 1.0881e-02, -7.0734e-03]],\n",
+       "\n",
+       "          [[-1.2324e-02, -1.8049e-02],\n",
+       "           [ 2.6147e-03,  7.3541e-03],\n",
+       "           [-1.2201e-02, -2.3215e-02],\n",
+       "           [ 1.8518e-02,  1.3268e-02]]],\n",
+       "\n",
+       "\n",
+       "         [[[-3.7732e-04,  6.0749e-03],\n",
+       "           [ 9.5228e-03,  1.7512e-03],\n",
+       "           [-6.6485e-03, -6.8077e-04],\n",
+       "           [-1.4016e-02,  4.3504e-03]],\n",
+       "\n",
+       "          [[-1.1826e-02,  3.7322e-03],\n",
+       "           [-1.4521e-02, -5.1530e-03],\n",
+       "           [-8.8820e-04,  5.4113e-03],\n",
+       "           [ 1.0400e-02, -1.4762e-03]],\n",
+       "\n",
+       "          [[ 1.0607e-02,  6.7451e-03],\n",
+       "           [ 1.0964e-02,  2.8386e-03],\n",
+       "           [ 5.6247e-03,  1.0890e-02],\n",
+       "           [ 4.6335e-03, -3.7274e-03]],\n",
+       "\n",
+       "          [[-9.0740e-03,  6.4783e-03],\n",
+       "           [-1.2299e-03, -5.6999e-03],\n",
+       "           [-1.6512e-02,  1.0905e-02],\n",
+       "           [ 2.4370e-03, -1.5498e-02]]]],\n",
+       "\n",
+       "\n",
+       "\n",
+       "        [[[[ 8.5643e-03,  1.1855e-02],\n",
+       "           [-4.0238e-03, -1.9470e-02],\n",
+       "           [ 8.6876e-03, -1.3815e-03],\n",
+       "           [ 1.3968e-02, -1.1944e-02]],\n",
+       "\n",
+       "          [[-4.9274e-03, -3.5420e-03],\n",
+       "           [ 1.1639e-02, -3.9007e-03],\n",
+       "           [-6.9517e-03, -9.7298e-03],\n",
+       "           [-9.1168e-03,  1.6739e-03]],\n",
+       "\n",
+       "          [[ 9.1662e-03,  3.5723e-04],\n",
+       "           [-7.4430e-04, -2.2149e-03],\n",
+       "           [-3.0224e-03,  1.0551e-02],\n",
+       "           [-5.0447e-03, -1.4064e-02]],\n",
+       "\n",
+       "          [[-3.5565e-03, -7.7704e-04],\n",
+       "           [-7.0354e-03,  8.5815e-03],\n",
+       "           [-8.7700e-03, -1.2051e-02],\n",
+       "           [ 8.8658e-03, -2.5521e-02]]],\n",
+       "\n",
+       "\n",
+       "         [[[-1.2974e-03, -8.8698e-03],\n",
+       "           [-3.6456e-03,  1.5029e-02],\n",
+       "           [-8.8230e-03,  9.4439e-03],\n",
+       "           [ 2.5990e-03,  1.5128e-02]],\n",
+       "\n",
+       "          [[ 5.6906e-03,  4.9925e-03],\n",
+       "           [ 1.2224e-04, -1.1197e-02],\n",
+       "           [-1.9241e-02,  2.9079e-03],\n",
+       "           [ 7.9246e-03, -1.4221e-02]],\n",
+       "\n",
+       "          [[ 1.8949e-04, -9.8628e-03],\n",
+       "           [-1.2586e-02,  2.9915e-02],\n",
+       "           [ 1.6699e-02, -3.6897e-03],\n",
+       "           [-6.5158e-03,  1.5208e-03]],\n",
+       "\n",
+       "          [[ 2.0201e-04, -2.1695e-02],\n",
+       "           [-2.6849e-04, -1.5498e-02],\n",
+       "           [-1.1011e-02,  1.1365e-03],\n",
+       "           [-7.1765e-03,  1.2924e-02]]],\n",
        "\n",
-       "          [[ 2.5566e-02, -2.3371e-03],\n",
-       "           [-5.3729e-03, -4.7599e-03],\n",
-       "           [ 1.9879e-02,  1.8122e-02],\n",
-       "           [-3.3057e-03, -1.3735e-02]],\n",
-       "\n",
-       "          [[ 4.4050e-03, -1.1497e-02],\n",
-       "           [-1.1812e-02,  9.4849e-03],\n",
-       "           [ 1.0993e-02, -1.5682e-02],\n",
-       "           [ 6.7096e-03,  2.5279e-02]]]],\n",
        "\n",
+       "         [[[ 9.7294e-03,  2.4206e-02],\n",
+       "           [-1.2685e-02, -1.0336e-03],\n",
+       "           [ 8.7326e-03, -2.7275e-02],\n",
+       "           [-2.1335e-02,  4.4139e-03]],\n",
        "\n",
+       "          [[-1.5349e-02, -5.3750e-03],\n",
+       "           [ 5.4290e-03,  3.5145e-03],\n",
+       "           [ 9.8669e-03,  5.5941e-03],\n",
+       "           [-2.8471e-03,  4.4822e-03]],\n",
        "\n",
-       "        [[[[ 5.1784e-03, -1.4195e-03],\n",
-       "           [-5.3990e-03, -1.0834e-02],\n",
-       "           [ 1.0641e-02,  6.1596e-03],\n",
-       "           [ 8.7506e-03,  2.9205e-03]],\n",
+       "          [[-3.5157e-03,  5.6956e-03],\n",
+       "           [ 8.4872e-03, -6.3729e-03],\n",
+       "           [-5.0063e-03, -1.1296e-02],\n",
+       "           [-1.5232e-02, -1.2576e-02]],\n",
        "\n",
-       "          [[-1.6737e-02,  5.6834e-03],\n",
-       "           [ 6.5168e-03, -3.1135e-03],\n",
-       "           [ 2.9460e-03, -1.2015e-03],\n",
-       "           [-1.1608e-02,  7.6579e-03]],\n",
+       "          [[-1.5377e-02,  4.6587e-03],\n",
+       "           [-1.2515e-02, -8.6466e-03],\n",
+       "           [-3.5109e-03,  1.8327e-02],\n",
+       "           [ 5.5513e-03, -7.4427e-03]]],\n",
        "\n",
-       "          [[ 1.0303e-02, -1.0884e-02],\n",
-       "           [ 9.5722e-03, -4.6254e-03],\n",
-       "           [ 4.6554e-03,  1.0371e-02],\n",
-       "           [-2.4556e-03,  4.5003e-03]],\n",
-       "\n",
-       "          [[-5.7525e-03, -6.7471e-03],\n",
-       "           [ 4.3139e-03, -5.5204e-03],\n",
-       "           [ 9.1538e-03,  4.8076e-03],\n",
-       "           [ 9.0984e-03, -2.3925e-03]]],\n",
-       "\n",
-       "\n",
-       "         [[[ 3.8985e-03, -7.3933e-03],\n",
-       "           [-5.1886e-03,  1.1477e-03],\n",
-       "           [ 6.4889e-03, -3.6765e-03],\n",
-       "           [ 7.2426e-03,  4.0187e-03]],\n",
-       "\n",
-       "          [[-1.0519e-02, -6.4471e-05],\n",
-       "           [ 2.2689e-02,  1.2035e-02],\n",
-       "           [ 4.8530e-03, -1.4561e-03],\n",
-       "           [ 1.6576e-03,  2.0176e-02]],\n",
-       "\n",
-       "          [[-8.8692e-03,  2.4320e-03],\n",
-       "           [ 4.7644e-03, -1.0054e-02],\n",
-       "           [ 1.1499e-02, -4.3900e-03],\n",
-       "           [ 8.8678e-03, -1.0401e-02]],\n",
-       "\n",
-       "          [[ 3.8191e-03, -1.0572e-02],\n",
-       "           [-3.9368e-03,  8.9180e-03],\n",
-       "           [-1.8331e-02, -4.8689e-03],\n",
-       "           [-1.2785e-02, -5.7243e-03]]],\n",
-       "\n",
-       "\n",
-       "         [[[-4.1451e-03,  5.1505e-03],\n",
-       "           [ 1.4643e-02,  8.4760e-03],\n",
-       "           [ 1.0262e-03, -5.9678e-03],\n",
-       "           [ 5.1973e-03,  1.1929e-02]],\n",
-       "\n",
-       "          [[-6.2390e-03,  6.5752e-03],\n",
-       "           [ 9.1719e-04,  6.7939e-03],\n",
-       "           [-2.6674e-04,  2.9604e-02],\n",
-       "           [ 4.2720e-03, -1.5935e-02]],\n",
-       "\n",
-       "          [[-5.3500e-04,  2.2230e-02],\n",
-       "           [-4.7624e-03,  1.1678e-02],\n",
-       "           [-1.1461e-03,  3.8098e-03],\n",
-       "           [ 2.2019e-02, -1.5939e-05]],\n",
-       "\n",
-       "          [[-1.2336e-02,  1.0585e-02],\n",
-       "           [ 1.7668e-02, -2.0162e-02],\n",
-       "           [-1.1665e-02, -5.4866e-04],\n",
-       "           [-7.0921e-03,  1.3506e-02]]],\n",
-       "\n",
-       "\n",
-       "         [[[ 1.5243e-03,  1.9578e-03],\n",
-       "           [ 4.2853e-03,  3.6213e-03],\n",
-       "           [-2.2286e-03, -1.7510e-02],\n",
-       "           [-6.2912e-03,  1.6289e-02]],\n",
-       "\n",
-       "          [[-9.0679e-03, -1.2895e-02],\n",
-       "           [ 3.7751e-03, -2.3309e-02],\n",
-       "           [ 1.5525e-03,  2.4318e-03],\n",
-       "           [ 3.7684e-03, -1.7957e-02]],\n",
-       "\n",
-       "          [[ 2.7108e-03,  8.7287e-04],\n",
-       "           [-5.4382e-03, -1.5571e-02],\n",
-       "           [-1.1790e-02,  6.5124e-03],\n",
-       "           [ 2.2338e-02,  5.8709e-03]],\n",
-       "\n",
-       "          [[ 4.9399e-04, -1.9222e-02],\n",
-       "           [-2.4067e-03,  1.6542e-02],\n",
-       "           [ 1.1662e-02,  1.1188e-02],\n",
-       "           [-6.7352e-03,  3.2604e-02]]]],\n",
-       "\n",
-       "\n",
-       "\n",
-       "        [[[[ 7.5602e-03, -5.0242e-03],\n",
-       "           [-8.2706e-03, -1.7402e-03],\n",
-       "           [-9.3949e-03, -7.2250e-03],\n",
-       "           [ 8.7566e-03,  1.1676e-04]],\n",
-       "\n",
-       "          [[-7.5017e-03,  1.8024e-03],\n",
-       "           [ 2.5618e-03, -8.8935e-03],\n",
-       "           [-2.5026e-03, -5.5693e-03],\n",
-       "           [-9.2033e-03,  2.7998e-03]],\n",
-       "\n",
-       "          [[-1.2505e-02, -6.9063e-03],\n",
-       "           [ 1.2637e-02,  8.3032e-03],\n",
-       "           [-1.6924e-02,  1.1639e-02],\n",
-       "           [ 6.7917e-03,  2.2977e-03]],\n",
-       "\n",
-       "          [[-3.2956e-04, -5.9520e-04],\n",
-       "           [ 1.6061e-02,  9.6812e-03],\n",
-       "           [-1.1180e-02, -7.7573e-04],\n",
-       "           [ 1.3929e-03,  8.3166e-03]]],\n",
-       "\n",
-       "\n",
-       "         [[[-1.0511e-02, -9.3898e-03],\n",
-       "           [-7.4458e-03,  7.1798e-03],\n",
-       "           [ 1.8625e-02, -3.9352e-03],\n",
-       "           [ 1.1769e-02,  9.1456e-04]],\n",
-       "\n",
-       "          [[ 1.0072e-02,  7.7131e-03],\n",
-       "           [ 6.5359e-03, -9.2153e-03],\n",
-       "           [ 1.9783e-03, -2.0443e-02],\n",
-       "           [ 1.0223e-02, -8.4568e-03]],\n",
-       "\n",
-       "          [[-3.5886e-04,  3.0560e-04],\n",
-       "           [-2.9668e-03,  8.7726e-03],\n",
-       "           [ 1.6094e-02, -8.5014e-03],\n",
-       "           [ 2.4411e-02,  3.0396e-03]],\n",
-       "\n",
-       "          [[-1.3760e-02,  9.4967e-04],\n",
-       "           [-8.1352e-03,  8.6983e-03],\n",
-       "           [-1.7103e-03, -1.0511e-02],\n",
-       "           [-5.1999e-03, -6.6092e-03]]],\n",
-       "\n",
-       "\n",
-       "         [[[-4.1700e-03, -2.8946e-03],\n",
-       "           [-5.7401e-03,  4.6369e-03],\n",
-       "           [-1.2268e-02, -1.6185e-02],\n",
-       "           [-1.7004e-02, -1.8065e-03]],\n",
-       "\n",
-       "          [[ 1.4414e-03,  3.0189e-03],\n",
-       "           [ 8.3214e-04, -6.1676e-03],\n",
-       "           [ 3.0263e-03, -2.0772e-02],\n",
-       "           [ 5.5492e-03, -3.2041e-03]],\n",
-       "\n",
-       "          [[-8.9067e-03, -1.6217e-04],\n",
-       "           [-3.9358e-03,  1.3170e-03],\n",
-       "           [ 1.8509e-03,  1.9097e-02],\n",
-       "           [-6.4288e-03, -1.1963e-02]],\n",
-       "\n",
-       "          [[-5.7511e-04,  2.1690e-03],\n",
-       "           [ 6.8697e-03,  2.6131e-03],\n",
-       "           [ 9.6800e-03, -6.5693e-03],\n",
-       "           [ 7.9978e-03,  3.4935e-03]]],\n",
-       "\n",
-       "\n",
-       "         [[[-2.8822e-03, -1.7531e-02],\n",
-       "           [-9.0395e-03,  1.7119e-03],\n",
-       "           [ 4.7403e-04, -2.4928e-03],\n",
-       "           [ 2.0809e-02,  1.0375e-02]],\n",
-       "\n",
-       "          [[ 9.9379e-03,  1.1011e-03],\n",
-       "           [-2.0822e-02, -1.0136e-03],\n",
-       "           [ 5.6764e-03,  4.7663e-03],\n",
-       "           [ 5.7930e-03, -3.2342e-03]],\n",
-       "\n",
-       "          [[ 1.8071e-03, -3.8490e-03],\n",
-       "           [-5.0903e-03,  4.2883e-03],\n",
-       "           [ 6.2868e-03,  6.9501e-03],\n",
-       "           [-5.1345e-03, -4.4751e-03]],\n",
-       "\n",
-       "          [[-6.3479e-04, -1.4575e-03],\n",
-       "           [ 1.3965e-02, -4.1295e-03],\n",
-       "           [-3.1379e-03,  3.5920e-05],\n",
-       "           [ 1.0564e-02, -1.7378e-02]]]]])"
+       "\n",
+       "         [[[-1.2366e-02, -1.8556e-02],\n",
+       "           [-7.6673e-03, -6.9909e-03],\n",
+       "           [-1.7524e-02, -1.2405e-02],\n",
+       "           [-2.0454e-04, -1.0710e-02]],\n",
+       "\n",
+       "          [[-7.5223e-04,  8.1430e-03],\n",
+       "           [-4.2421e-03, -2.4979e-03],\n",
+       "           [-2.6400e-03, -7.0397e-03],\n",
+       "           [ 7.2088e-03,  1.2598e-02]],\n",
+       "\n",
+       "          [[ 5.4684e-03, -2.1672e-03],\n",
+       "           [ 7.8071e-04,  8.7978e-04],\n",
+       "           [-4.5486e-03, -1.1562e-02],\n",
+       "           [-1.2760e-02, -4.7683e-03]],\n",
+       "\n",
+       "          [[-5.0534e-03,  2.6844e-02],\n",
+       "           [ 6.5477e-03, -2.0174e-03],\n",
+       "           [-2.3228e-03, -1.9718e-03],\n",
+       "           [-1.0325e-02,  1.4051e-02]]]],\n",
+       "\n",
+       "\n",
+       "\n",
+       "        [[[[ 1.2235e-02,  3.8194e-03],\n",
+       "           [-1.2756e-02,  2.0048e-03],\n",
+       "           [-1.0588e-02, -6.2585e-03],\n",
+       "           [-1.1622e-02,  7.8146e-03]],\n",
+       "\n",
+       "          [[-1.3627e-02,  4.8715e-03],\n",
+       "           [-2.6081e-03, -1.2547e-02],\n",
+       "           [ 2.5712e-02,  6.9268e-04],\n",
+       "           [ 5.7297e-03,  1.6908e-02]],\n",
+       "\n",
+       "          [[-6.0016e-03,  3.8388e-03],\n",
+       "           [-4.2425e-03,  5.6236e-03],\n",
+       "           [-2.2077e-03,  8.3500e-03],\n",
+       "           [-6.5338e-03, -5.6204e-03]],\n",
+       "\n",
+       "          [[ 3.6226e-03,  6.5813e-03],\n",
+       "           [ 8.4147e-03, -4.6316e-03],\n",
+       "           [ 1.1701e-03, -3.8150e-03],\n",
+       "           [ 3.4847e-03, -1.6027e-02]]],\n",
+       "\n",
+       "\n",
+       "         [[[-1.2393e-03, -8.3570e-03],\n",
+       "           [ 4.4231e-03, -6.3459e-03],\n",
+       "           [-6.1139e-04,  5.3770e-03],\n",
+       "           [ 5.4646e-03,  3.0140e-03]],\n",
+       "\n",
+       "          [[-2.6489e-04,  3.1112e-04],\n",
+       "           [ 4.5194e-03,  3.8941e-03],\n",
+       "           [-1.0284e-02, -3.8570e-04],\n",
+       "           [-7.3893e-03,  1.7081e-04]],\n",
+       "\n",
+       "          [[ 1.2995e-02,  4.9129e-03],\n",
+       "           [-3.2257e-03,  4.4844e-03],\n",
+       "           [-1.6129e-03, -3.0462e-03],\n",
+       "           [-4.6639e-03,  9.7619e-03]],\n",
+       "\n",
+       "          [[-3.7702e-03, -1.7440e-02],\n",
+       "           [ 2.7172e-03, -7.0420e-04],\n",
+       "           [ 1.0267e-02,  6.5922e-03],\n",
+       "           [-4.3092e-03, -6.1909e-03]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 8.2225e-03,  2.0020e-02],\n",
+       "           [ 1.3149e-02, -1.5706e-02],\n",
+       "           [ 1.4529e-02,  2.0051e-02],\n",
+       "           [-3.3207e-03,  2.9005e-03]],\n",
+       "\n",
+       "          [[ 1.6814e-02, -1.1611e-02],\n",
+       "           [ 1.1958e-02, -1.1217e-02],\n",
+       "           [-7.3994e-03,  3.3478e-03],\n",
+       "           [-1.9186e-03, -3.3058e-03]],\n",
+       "\n",
+       "          [[ 1.8627e-02, -4.3313e-03],\n",
+       "           [-2.5333e-04,  9.4670e-03],\n",
+       "           [ 1.6627e-02, -3.7056e-03],\n",
+       "           [-4.0002e-03, -4.0428e-03]],\n",
+       "\n",
+       "          [[-5.8534e-03, -8.5589e-03],\n",
+       "           [ 1.8699e-02,  1.4041e-02],\n",
+       "           [-4.5634e-03, -1.2667e-02],\n",
+       "           [ 2.9807e-03, -1.0735e-02]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 2.3185e-03, -1.1426e-02],\n",
+       "           [ 1.0432e-02,  8.7466e-03],\n",
+       "           [-2.0517e-02, -6.4246e-03],\n",
+       "           [ 1.6325e-02, -2.2658e-03]],\n",
+       "\n",
+       "          [[-1.3443e-03, -7.4122e-03],\n",
+       "           [ 2.6692e-03,  1.6455e-03],\n",
+       "           [-1.2263e-02, -3.9807e-04],\n",
+       "           [-1.1106e-02, -1.2117e-03]],\n",
+       "\n",
+       "          [[ 1.2348e-02, -4.9474e-03],\n",
+       "           [-5.5995e-03, -2.2941e-03],\n",
+       "           [-8.0381e-03,  8.6414e-03],\n",
+       "           [-1.2672e-02, -8.6333e-03]],\n",
+       "\n",
+       "          [[ 2.2708e-03,  1.5464e-02],\n",
+       "           [-6.9604e-07,  9.0496e-03],\n",
+       "           [ 4.9665e-03, -1.3779e-04],\n",
+       "           [ 2.5279e-03, -9.1125e-03]]]],\n",
+       "\n",
+       "\n",
+       "\n",
+       "        [[[[-4.1377e-03, -5.1594e-04],\n",
+       "           [-3.2999e-03, -3.2629e-03],\n",
+       "           [ 1.5395e-02, -1.9373e-03],\n",
+       "           [-2.1155e-02, -2.2015e-03]],\n",
+       "\n",
+       "          [[-4.7272e-03,  4.2632e-03],\n",
+       "           [-7.7089e-03, -4.4246e-03],\n",
+       "           [ 1.1186e-02,  1.2331e-02],\n",
+       "           [-8.5431e-03,  9.3936e-03]],\n",
+       "\n",
+       "          [[ 2.0678e-02, -1.5143e-03],\n",
+       "           [ 1.1107e-03, -1.5056e-03],\n",
+       "           [-1.5841e-02, -1.0837e-03],\n",
+       "           [ 1.2758e-03, -1.2761e-03]],\n",
+       "\n",
+       "          [[-1.0277e-02, -2.0925e-02],\n",
+       "           [ 2.1294e-04, -7.8277e-03],\n",
+       "           [-8.0540e-04,  6.9005e-03],\n",
+       "           [ 8.6584e-03, -3.0194e-03]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 1.7248e-02, -6.2072e-03],\n",
+       "           [ 5.1232e-03, -8.9823e-03],\n",
+       "           [-1.4095e-02, -1.8089e-02],\n",
+       "           [ 8.5848e-03, -7.7452e-03]],\n",
+       "\n",
+       "          [[ 6.3832e-03, -9.4521e-04],\n",
+       "           [-6.0866e-03, -1.7911e-02],\n",
+       "           [ 7.1645e-03,  1.8826e-03],\n",
+       "           [-1.1272e-03, -1.6942e-03]],\n",
+       "\n",
+       "          [[-1.9407e-04,  2.6855e-02],\n",
+       "           [ 2.4922e-03,  4.0395e-03],\n",
+       "           [-9.0179e-03,  1.1456e-02],\n",
+       "           [ 6.2417e-03,  1.6225e-03]],\n",
+       "\n",
+       "          [[-1.3302e-02, -7.9396e-03],\n",
+       "           [ 1.6342e-02, -8.3640e-04],\n",
+       "           [ 8.2515e-03, -1.9994e-02],\n",
+       "           [-1.5314e-02,  1.1206e-03]]],\n",
+       "\n",
+       "\n",
+       "         [[[-1.6801e-02,  2.5082e-03],\n",
+       "           [-1.2035e-02,  1.0600e-02],\n",
+       "           [-3.4271e-03,  8.3387e-03],\n",
+       "           [ 1.1338e-03, -1.3025e-03]],\n",
+       "\n",
+       "          [[-4.3297e-04,  6.9727e-03],\n",
+       "           [-9.4044e-03, -1.8042e-02],\n",
+       "           [-1.6318e-02, -1.8721e-03],\n",
+       "           [ 9.9107e-03, -1.0611e-02]],\n",
+       "\n",
+       "          [[ 1.3804e-02, -5.9115e-03],\n",
+       "           [-8.4717e-03,  1.2661e-02],\n",
+       "           [ 5.9259e-03,  6.3610e-03],\n",
+       "           [-1.8119e-03, -1.0934e-02]],\n",
+       "\n",
+       "          [[-6.4295e-03,  1.4941e-03],\n",
+       "           [-1.3924e-02,  1.0059e-02],\n",
+       "           [ 3.7694e-03,  1.3317e-02],\n",
+       "           [ 7.0841e-03,  8.2705e-03]]],\n",
+       "\n",
+       "\n",
+       "         [[[ 5.0314e-03, -1.1748e-03],\n",
+       "           [-6.9476e-03, -6.9823e-05],\n",
+       "           [ 3.5111e-03, -3.0028e-03],\n",
+       "           [-6.3763e-03,  1.4920e-03]],\n",
+       "\n",
+       "          [[-9.0027e-04, -3.8365e-03],\n",
+       "           [ 1.1627e-04, -4.9486e-03],\n",
+       "           [ 1.3806e-02,  5.5603e-03],\n",
+       "           [-1.1555e-02, -3.8007e-03]],\n",
+       "\n",
+       "          [[-3.6884e-04,  1.0951e-02],\n",
+       "           [ 2.6923e-03,  6.7217e-04],\n",
+       "           [ 1.1842e-02, -1.7159e-02],\n",
+       "           [ 1.3960e-04,  8.7688e-03]],\n",
+       "\n",
+       "          [[-6.5861e-03,  1.1069e-03],\n",
+       "           [-9.9205e-03,  1.4599e-02],\n",
+       "           [-4.2275e-03, -3.8720e-03],\n",
+       "           [ 1.4915e-02, -2.9827e-03]]]]])"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 45,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -843,7 +852,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 46,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -856,7 +865,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 47,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -869,7 +878,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 48,
    "metadata": {},
    "outputs": [
     {
@@ -889,23 +898,23 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 49,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Sample State: [-3.6452694e+00 -8.9159860e+37  1.4139645e-01  3.0674191e+38]\n",
-      "Sample State: [ 1.4343392e+00 -3.0131075e+38 -2.3563206e-01  1.7383354e+38]\n",
-      "Sample State: [ 3.7291312e+00  1.9267806e+38  8.8896513e-02 -1.9043992e+38]\n",
-      "Sample State: [ 5.6241733e-01 -1.1501083e+38  1.9758487e-01 -2.6513862e+38]\n",
-      "Sample State: [-3.7882154e+00 -1.8343667e+38 -4.1406271e-01  1.2239143e+38]\n",
-      "Sample State: [ 4.1043639e+00 -7.7561222e+37 -3.9738983e-01  2.0008877e+38]\n",
-      "Sample State: [ 3.2407689e+00 -3.1213367e+38 -4.0249658e-01  3.2251934e+38]\n",
-      "Sample State: [ 3.9025934e+00  1.2178617e+38 -6.5442048e-02 -1.2237320e+38]\n",
-      "Sample State: [ 3.4483566e+00 -1.4972215e+38 -1.5894611e-01 -1.3151852e+38]\n",
-      "Sample State: [ 2.0936482e+00  2.7123340e+38 -1.8713233e-01  1.7833031e+38]\n"
+      "Sample State: [-2.1946154e+00 -2.5122178e+38  8.9282773e-02 -1.3416754e+38]\n",
+      "Sample State: [ 2.7149630e-01  1.2149416e+37 -9.9354312e-02 -1.2340108e+38]\n",
+      "Sample State: [ 4.0275431e+00 -2.6682660e+38 -2.4785740e-02  2.4654679e+38]\n",
+      "Sample State: [-2.3548093e-01 -6.2527667e+37  3.5589758e-01 -4.8273950e+37]\n",
+      "Sample State: [ 1.6777289e+00 -6.2703186e+37  3.0708086e-01  1.9350419e+38]\n",
+      "Sample State: [-2.8664367e+00 -9.9032201e+37 -4.1302589e-01  3.0742409e+38]\n",
+      "Sample State: [-2.4482067e+00 -6.4236516e+37 -1.9836776e-01  2.5115714e+38]\n",
+      "Sample State: [-4.2023139e+00 -1.8168436e+37 -6.5094754e-02  1.7452279e+38]\n",
+      "Sample State: [-2.8398631e+00 -3.0210992e+38 -1.6737616e-01 -2.1272714e+38]\n",
+      "Sample State: [-2.1986934e-04  9.8844897e+36 -7.7146210e-02 -7.3564386e+37]\n"
      ]
     }
    ],
@@ -925,7 +934,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 50,
    "metadata": {},
    "outputs": [
     {
@@ -933,1202 +942,4354 @@
      "output_type": "stream",
      "text": [
       "Episode: 0\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 1\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 2\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 15.0\n",
       "Episode: 3\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 15.0\n",
       "Episode: 4\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 5\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 6\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 34.0\n",
       "Episode: 7\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 15.0\n",
       "Episode: 8\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 9\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 10\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 11\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 21.0\n",
       "Episode: 12\n",
-      "Episode reward: 71.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 13\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 14\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 15\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 16\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 17\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 18\n",
-      "Episode reward: 30.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 19\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 15.0\n",
       "Episode: 20\n",
-      "Episode reward: 32.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 21\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 22\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 23\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 24\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 25\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 21.0\n",
       "Episode: 26\n",
       "Episode reward: 10.0\n",
       "Episode: 27\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 28\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 37.0\n",
       "Episode: 29\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 30\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 31\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 32\n",
-      "Episode reward: 15.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 33\n",
       "Episode reward: 10.0\n",
       "Episode: 34\n",
-      "Episode reward: 19.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 35\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 36\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 37\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 38\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 39\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 40\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 41\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 42\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 43\n",
       "Episode reward: 12.0\n",
       "Episode: 44\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 45\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 46\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 15.0\n",
       "Episode: 47\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 32.0\n",
       "Episode: 48\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 32.0\n",
       "Episode: 49\n",
-      "Episode reward: 15.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 50\n",
       "Episode reward: 11.0\n",
       "Episode: 51\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 52\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 53\n",
-      "Episode reward: 19.0\n",
+      "Episode reward: 27.0\n",
       "Episode: 54\n",
-      "Episode reward: 32.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 55\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 56\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 57\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 58\n",
-      "Episode reward: 28.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 59\n",
-      "Episode reward: 8.0\n",
-      "Episode: 60\n",
       "Episode reward: 23.0\n",
+      "Episode: 60\n",
+      "Episode reward: 16.0\n",
       "Episode: 61\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 36.0\n",
       "Episode: 62\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 63\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 64\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 65\n",
-      "Episode reward: 18.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 66\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 67\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 68\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 69\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 70\n",
-      "Episode reward: 19.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 71\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 72\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 73\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 74\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 75\n",
-      "Episode reward: 11.0\n",
-      "Episode: 76\n",
       "Episode reward: 10.0\n",
+      "Episode: 76\n",
+      "Episode reward: 59.0\n",
       "Episode: 77\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 53.0\n",
       "Episode: 78\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 79\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 80\n",
-      "Episode reward: 14.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 81\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 48.0\n",
       "Episode: 82\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 83\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 43.0\n",
       "Episode: 84\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 85\n",
-      "Episode reward: 47.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 86\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 87\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 88\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 89\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 90\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 91\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 39.0\n",
       "Episode: 92\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 93\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 94\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 95\n",
       "Episode reward: 10.0\n",
       "Episode: 96\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 47.0\n",
       "Episode: 97\n",
       "Episode reward: 10.0\n",
       "Episode: 98\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 99\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 100\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 101\n",
       "Episode reward: 9.0\n",
       "Episode: 102\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 103\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 104\n",
-      "Episode reward: 14.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 105\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 106\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 107\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 108\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 44.0\n",
       "Episode: 109\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 110\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 60.0\n",
       "Episode: 111\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 112\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 113\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 114\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 115\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 116\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 117\n",
-      "Episode reward: 10.0\n",
-      "Episode: 118\n",
       "Episode reward: 9.0\n",
+      "Episode: 118\n",
+      "Episode reward: 62.0\n",
       "Episode: 119\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 120\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 121\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 122\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 32.0\n",
       "Episode: 123\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 31.0\n",
       "Episode: 124\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 27.0\n",
       "Episode: 125\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 126\n",
-      "Episode reward: 30.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 127\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 128\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 129\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 130\n",
-      "Episode reward: 14.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 131\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 33.0\n",
       "Episode: 132\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 133\n",
-      "Episode reward: 9.0\n",
-      "Episode: 134\n",
       "Episode reward: 10.0\n",
+      "Episode: 134\n",
+      "Episode reward: 16.0\n",
       "Episode: 135\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 136\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 137\n",
       "Episode reward: 10.0\n",
       "Episode: 138\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 139\n",
-      "Episode reward: 52.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 140\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 141\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 85.0\n",
       "Episode: 142\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 51.0\n",
       "Episode: 143\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 46.0\n",
       "Episode: 144\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 145\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 55.0\n",
       "Episode: 146\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 147\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 50.0\n",
       "Episode: 148\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 149\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 150\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 151\n",
-      "Episode reward: 34.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 152\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 21.0\n",
       "Episode: 153\n",
-      "Episode reward: 18.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 154\n",
-      "Episode reward: 44.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 155\n",
-      "Episode reward: 33.0\n",
+      "Episode reward: 15.0\n",
       "Episode: 156\n",
-      "Episode reward: 21.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 157\n",
-      "Episode reward: 44.0\n",
+      "Episode reward: 63.0\n",
       "Episode: 158\n",
-      "Episode reward: 17.0\n",
+      "Episode reward: 21.0\n",
       "Episode: 159\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 160\n",
-      "Episode reward: 34.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 161\n",
-      "Episode reward: 75.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 162\n",
-      "Episode reward: 47.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 163\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 164\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 165\n",
-      "Episode reward: 28.0\n",
+      "Episode reward: 45.0\n",
       "Episode: 166\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 167\n",
-      "Episode reward: 40.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 168\n",
-      "Episode reward: 52.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 169\n",
-      "Episode reward: 32.0\n",
+      "Episode reward: 65.0\n",
       "Episode: 170\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 171\n",
-      "Episode reward: 34.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 172\n",
-      "Episode reward: 30.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 173\n",
-      "Episode reward: 94.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 174\n",
       "Episode reward: 22.0\n",
       "Episode: 175\n",
-      "Episode reward: 79.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 176\n",
       "Episode reward: 10.0\n",
       "Episode: 177\n",
-      "Episode reward: 49.0\n",
+      "Episode reward: 75.0\n",
       "Episode: 178\n",
-      "Episode reward: 43.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 179\n",
       "Episode reward: 26.0\n",
       "Episode: 180\n",
-      "Episode reward: 26.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 181\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 182\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 183\n",
       "Episode reward: 17.0\n",
       "Episode: 184\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 45.0\n",
       "Episode: 185\n",
-      "Episode reward: 15.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 186\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 187\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 188\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 50.0\n",
       "Episode: 189\n",
-      "Episode reward: 42.0\n",
+      "Episode reward: 29.0\n",
       "Episode: 190\n",
-      "Episode reward: 28.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 191\n",
-      "Episode reward: 16.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 192\n",
-      "Episode reward: 59.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 193\n",
       "Episode reward: 32.0\n",
       "Episode: 194\n",
       "Episode reward: 14.0\n",
       "Episode: 195\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 196\n",
-      "Episode reward: 15.0\n",
+      "Episode reward: 21.0\n",
       "Episode: 197\n",
-      "Episode reward: 26.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 198\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 199\n",
-      "Episode reward: 33.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 200\n",
       "Episode reward: 30.0\n",
       "Episode: 201\n",
-      "Episode reward: 16.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 202\n",
-      "Episode reward: 40.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 203\n",
-      "Episode reward: 41.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 204\n",
-      "Episode reward: 36.0\n",
+      "Episode reward: 47.0\n",
       "Episode: 205\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 206\n",
-      "Episode reward: 8.0\n",
-      "Episode: 207\n",
       "Episode reward: 10.0\n",
+      "Episode: 207\n",
+      "Episode reward: 37.0\n",
       "Episode: 208\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 15.0\n",
       "Episode: 209\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 31.0\n",
       "Episode: 210\n",
-      "Episode reward: 14.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 211\n",
-      "Episode reward: 50.0\n",
+      "Episode reward: 31.0\n",
       "Episode: 212\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 213\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 34.0\n",
       "Episode: 214\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 215\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 216\n",
-      "Episode reward: 19.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 217\n",
       "Episode reward: 20.0\n",
       "Episode: 218\n",
-      "Episode reward: 21.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 219\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 8.0\n",
       "Episode: 220\n",
       "Episode reward: 10.0\n",
       "Episode: 221\n",
-      "Episode reward: 42.0\n",
+      "Episode reward: 8.0\n",
       "Episode: 222\n",
-      "Episode reward: 28.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 223\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 224\n",
-      "Episode reward: 43.0\n",
+      "Episode reward: 45.0\n",
       "Episode: 225\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 226\n",
-      "Episode reward: 20.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 227\n",
-      "Episode reward: 74.0\n",
+      "Episode reward: 21.0\n",
       "Episode: 228\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 229\n",
-      "Episode reward: 28.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 230\n",
-      "Episode reward: 35.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 231\n",
-      "Episode reward: 33.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 232\n",
-      "Episode reward: 32.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 233\n",
-      "Episode reward: 49.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 234\n",
-      "Episode reward: 59.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 235\n",
-      "Episode reward: 21.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 236\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 237\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 38.0\n",
       "Episode: 238\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 41.0\n",
       "Episode: 239\n",
-      "Episode reward: 60.0\n",
+      "Episode reward: 33.0\n",
       "Episode: 240\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 241\n",
-      "Episode reward: 69.0\n",
+      "Episode reward: 46.0\n",
       "Episode: 242\n",
-      "Episode reward: 41.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 243\n",
       "Episode reward: 22.0\n",
       "Episode: 244\n",
-      "Episode reward: 39.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 245\n",
-      "Episode reward: 39.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 246\n",
-      "Episode reward: 54.0\n",
+      "Episode reward: 50.0\n",
       "Episode: 247\n",
-      "Episode reward: 15.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 248\n",
-      "Episode reward: 37.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 249\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 250\n",
-      "Episode reward: 41.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 251\n",
-      "Episode reward: 18.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 252\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 253\n",
-      "Episode reward: 32.0\n",
+      "Episode reward: 57.0\n",
       "Episode: 254\n",
-      "Episode reward: 61.0\n",
+      "Episode reward: 57.0\n",
       "Episode: 255\n",
-      "Episode reward: 14.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 256\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 257\n",
-      "Episode reward: 20.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 258\n",
-      "Episode reward: 46.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 259\n",
-      "Episode reward: 50.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 260\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 261\n",
-      "Episode reward: 48.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 262\n",
-      "Episode reward: 28.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 263\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 264\n",
-      "Episode reward: 45.0\n",
+      "Episode reward: 44.0\n",
       "Episode: 265\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 51.0\n",
       "Episode: 266\n",
-      "Episode reward: 58.0\n",
+      "Episode reward: 41.0\n",
       "Episode: 267\n",
-      "Episode reward: 43.0\n",
+      "Episode reward: 54.0\n",
       "Episode: 268\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 48.0\n",
       "Episode: 269\n",
-      "Episode reward: 82.0\n",
+      "Episode reward: 39.0\n",
       "Episode: 270\n",
       "Episode reward: 77.0\n",
       "Episode: 271\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 58.0\n",
       "Episode: 272\n",
-      "Episode reward: 61.0\n",
+      "Episode reward: 53.0\n",
       "Episode: 273\n",
-      "Episode reward: 50.0\n",
+      "Episode reward: 52.0\n",
       "Episode: 274\n",
-      "Episode reward: 41.0\n",
+      "Episode reward: 71.0\n",
       "Episode: 275\n",
-      "Episode reward: 48.0\n",
-      "Episode: 276\n",
       "Episode reward: 43.0\n",
+      "Episode: 276\n",
+      "Episode reward: 48.0\n",
       "Episode: 277\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 41.0\n",
       "Episode: 278\n",
-      "Episode reward: 43.0\n",
+      "Episode reward: 48.0\n",
       "Episode: 279\n",
-      "Episode reward: 45.0\n",
+      "Episode reward: 46.0\n",
       "Episode: 280\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 58.0\n",
       "Episode: 281\n",
-      "Episode reward: 70.0\n",
+      "Episode reward: 62.0\n",
       "Episode: 282\n",
-      "Episode reward: 33.0\n",
+      "Episode reward: 49.0\n",
       "Episode: 283\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 27.0\n",
       "Episode: 284\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 43.0\n",
       "Episode: 285\n",
-      "Episode reward: 59.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 286\n",
-      "Episode reward: 46.0\n",
+      "Episode reward: 44.0\n",
       "Episode: 287\n",
-      "Episode reward: 47.0\n",
+      "Episode reward: 59.0\n",
       "Episode: 288\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 289\n",
-      "Episode reward: 48.0\n",
+      "Episode reward: 60.0\n",
       "Episode: 290\n",
-      "Episode reward: 26.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 291\n",
-      "Episode reward: 69.0\n",
+      "Episode reward: 47.0\n",
       "Episode: 292\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 54.0\n",
       "Episode: 293\n",
-      "Episode reward: 51.0\n",
+      "Episode reward: 65.0\n",
       "Episode: 294\n",
       "Episode reward: 56.0\n",
       "Episode: 295\n",
-      "Episode reward: 47.0\n",
+      "Episode reward: 37.0\n",
       "Episode: 296\n",
-      "Episode reward: 44.0\n",
+      "Episode reward: 84.0\n",
       "Episode: 297\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 27.0\n",
       "Episode: 298\n",
-      "Episode reward: 64.0\n",
+      "Episode reward: 34.0\n",
       "Episode: 299\n",
-      "Episode reward: 89.0\n",
+      "Episode reward: 41.0\n",
       "Episode: 300\n",
-      "Episode reward: 61.0\n",
+      "Episode reward: 54.0\n",
       "Episode: 301\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 39.0\n",
       "Episode: 302\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 47.0\n",
       "Episode: 303\n",
-      "Episode reward: 66.0\n",
+      "Episode reward: 42.0\n",
       "Episode: 304\n",
-      "Episode reward: 47.0\n",
+      "Episode reward: 94.0\n",
       "Episode: 305\n",
-      "Episode reward: 20.0\n",
+      "Episode reward: 89.0\n",
       "Episode: 306\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 307\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 47.0\n",
       "Episode: 308\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 39.0\n",
       "Episode: 309\n",
-      "Episode reward: 50.0\n",
+      "Episode reward: 38.0\n",
       "Episode: 310\n",
-      "Episode reward: 19.0\n",
+      "Episode reward: 46.0\n",
       "Episode: 311\n",
-      "Episode reward: 114.0\n",
+      "Episode reward: 43.0\n",
       "Episode: 312\n",
-      "Episode reward: 60.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 313\n",
-      "Episode reward: 26.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 314\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 47.0\n",
       "Episode: 315\n",
-      "Episode reward: 14.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 316\n",
       "Episode reward: 9.0\n",
       "Episode: 317\n",
       "Episode reward: 8.0\n",
       "Episode: 318\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 319\n",
-      "Episode reward: 32.0\n",
+      "Episode reward: 27.0\n",
       "Episode: 320\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 70.0\n",
       "Episode: 321\n",
-      "Episode reward: 19.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 322\n",
-      "Episode reward: 47.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 323\n",
-      "Episode reward: 37.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 324\n",
-      "Episode reward: 106.0\n",
+      "Episode reward: 36.0\n",
       "Episode: 325\n",
       "Episode reward: 31.0\n",
       "Episode: 326\n",
-      "Episode reward: 39.0\n",
+      "Episode reward: 52.0\n",
       "Episode: 327\n",
-      "Episode reward: 20.0\n",
+      "Episode reward: 60.0\n",
       "Episode: 328\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 15.0\n",
       "Episode: 329\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 32.0\n",
       "Episode: 330\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 51.0\n",
       "Episode: 331\n",
-      "Episode reward: 33.0\n",
+      "Episode reward: 31.0\n",
       "Episode: 332\n",
       "Episode reward: 14.0\n",
       "Episode: 333\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 64.0\n",
       "Episode: 334\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 335\n",
-      "Episode reward: 38.0\n",
+      "Episode reward: 9.0\n",
       "Episode: 336\n",
       "Episode reward: 9.0\n",
       "Episode: 337\n",
-      "Episode reward: 45.0\n",
+      "Episode reward: 29.0\n",
       "Episode: 338\n",
-      "Episode reward: 37.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 339\n",
-      "Episode reward: 26.0\n",
+      "Episode reward: 21.0\n",
       "Episode: 340\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 21.0\n",
       "Episode: 341\n",
-      "Episode reward: 33.0\n",
+      "Episode reward: 31.0\n",
       "Episode: 342\n",
-      "Episode reward: 53.0\n",
+      "Episode reward: 46.0\n",
       "Episode: 343\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 34.0\n",
       "Episode: 344\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 345\n",
       "Episode reward: 21.0\n",
       "Episode: 346\n",
-      "Episode reward: 45.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 347\n",
-      "Episode reward: 37.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 348\n",
-      "Episode reward: 78.0\n",
-      "Episode: 349\n",
       "Episode reward: 38.0\n",
+      "Episode: 349\n",
+      "Episode reward: 71.0\n",
       "Episode: 350\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 17.0\n",
       "Episode: 351\n",
-      "Episode reward: 58.0\n",
+      "Episode reward: 49.0\n",
       "Episode: 352\n",
-      "Episode reward: 69.0\n",
+      "Episode reward: 36.0\n",
       "Episode: 353\n",
-      "Episode reward: 26.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 354\n",
-      "Episode reward: 21.0\n",
+      "Episode reward: 16.0\n",
       "Episode: 355\n",
-      "Episode reward: 161.0\n",
+      "Episode reward: 46.0\n",
       "Episode: 356\n",
-      "Episode reward: 33.0\n",
+      "Episode reward: 48.0\n",
       "Episode: 357\n",
-      "Episode reward: 101.0\n",
+      "Episode reward: 29.0\n",
       "Episode: 358\n",
-      "Episode reward: 39.0\n",
+      "Episode reward: 46.0\n",
       "Episode: 359\n",
-      "Episode reward: 36.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 360\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 361\n",
-      "Episode reward: 174.0\n",
+      "Episode reward: 39.0\n",
       "Episode: 362\n",
-      "Episode reward: 149.0\n",
+      "Episode reward: 61.0\n",
       "Episode: 363\n",
       "Episode reward: 15.0\n",
       "Episode: 364\n",
-      "Episode reward: 125.0\n",
+      "Episode reward: 39.0\n",
       "Episode: 365\n",
-      "Episode reward: 37.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 366\n",
-      "Episode reward: 82.0\n",
+      "Episode reward: 41.0\n",
       "Episode: 367\n",
       "Episode reward: 23.0\n",
       "Episode: 368\n",
-      "Episode reward: 35.0\n",
+      "Episode reward: 52.0\n",
       "Episode: 369\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 370\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 371\n",
-      "Episode reward: 74.0\n",
+      "Episode reward: 90.0\n",
       "Episode: 372\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 81.0\n",
       "Episode: 373\n",
       "Episode reward: 20.0\n",
       "Episode: 374\n",
-      "Episode reward: 197.0\n",
+      "Episode reward: 109.0\n",
       "Episode: 375\n",
-      "Episode reward: 46.0\n",
+      "Episode reward: 14.0\n",
       "Episode: 376\n",
       "Episode reward: 39.0\n",
       "Episode: 377\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 31.0\n",
       "Episode: 378\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 55.0\n",
       "Episode: 379\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 68.0\n",
       "Episode: 380\n",
-      "Episode reward: 200.0\n",
+      "Episode reward: 59.0\n",
       "Episode: 381\n",
       "Episode reward: 22.0\n",
       "Episode: 382\n",
-      "Episode reward: 44.0\n",
+      "Episode reward: 75.0\n",
       "Episode: 383\n",
-      "Episode reward: 77.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 384\n",
       "Episode reward: 35.0\n",
       "Episode: 385\n",
-      "Episode reward: 179.0\n",
+      "Episode reward: 109.0\n",
       "Episode: 386\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 73.0\n",
       "Episode: 387\n",
-      "Episode reward: 28.0\n",
+      "Episode reward: 52.0\n",
       "Episode: 388\n",
-      "Episode reward: 51.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 389\n",
-      "Episode reward: 36.0\n",
+      "Episode reward: 107.0\n",
       "Episode: 390\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 391\n",
-      "Episode reward: 19.0\n",
+      "Episode reward: 62.0\n",
       "Episode: 392\n",
-      "Episode reward: 34.0\n",
+      "Episode reward: 32.0\n",
       "Episode: 393\n",
-      "Episode reward: 28.0\n",
+      "Episode reward: 46.0\n",
       "Episode: 394\n",
       "Episode reward: 27.0\n",
       "Episode: 395\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 54.0\n",
       "Episode: 396\n",
-      "Episode reward: 30.0\n",
+      "Episode reward: 44.0\n",
       "Episode: 397\n",
-      "Episode reward: 42.0\n",
+      "Episode reward: 57.0\n",
       "Episode: 398\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 399\n",
-      "Episode reward: 17.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 400\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 39.0\n",
       "Episode: 401\n",
-      "Episode reward: 43.0\n",
+      "Episode reward: 47.0\n",
       "Episode: 402\n",
-      "Episode reward: 195.0\n",
+      "Episode reward: 63.0\n",
       "Episode: 403\n",
-      "Episode reward: 59.0\n",
+      "Episode reward: 98.0\n",
       "Episode: 404\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 39.0\n",
       "Episode: 405\n",
-      "Episode reward: 14.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 406\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 67.0\n",
       "Episode: 407\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 49.0\n",
       "Episode: 408\n",
-      "Episode reward: 182.0\n",
+      "Episode reward: 93.0\n",
       "Episode: 409\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 410\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 42.0\n",
       "Episode: 411\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 412\n",
-      "Episode reward: 43.0\n",
+      "Episode reward: 59.0\n",
       "Episode: 413\n",
-      "Episode reward: 45.0\n",
+      "Episode reward: 8.0\n",
       "Episode: 414\n",
-      "Episode reward: 14.0\n",
+      "Episode reward: 64.0\n",
       "Episode: 415\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 416\n",
-      "Episode reward: 68.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 417\n",
       "Episode reward: 9.0\n",
       "Episode: 418\n",
-      "Episode reward: 21.0\n",
+      "Episode reward: 44.0\n",
       "Episode: 419\n",
       "Episode reward: 12.0\n",
       "Episode: 420\n",
       "Episode reward: 22.0\n",
       "Episode: 421\n",
-      "Episode reward: 131.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 422\n",
-      "Episode reward: 55.0\n",
+      "Episode reward: 38.0\n",
       "Episode: 423\n",
       "Episode reward: 22.0\n",
       "Episode: 424\n",
-      "Episode reward: 46.0\n",
+      "Episode reward: 61.0\n",
       "Episode: 425\n",
       "Episode reward: 24.0\n",
       "Episode: 426\n",
-      "Episode reward: 20.0\n",
+      "Episode reward: 33.0\n",
       "Episode: 427\n",
       "Episode reward: 21.0\n",
       "Episode: 428\n",
-      "Episode reward: 21.0\n",
+      "Episode reward: 33.0\n",
       "Episode: 429\n",
-      "Episode reward: 26.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 430\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 431\n",
-      "Episode reward: 58.0\n",
+      "Episode reward: 50.0\n",
       "Episode: 432\n",
-      "Episode reward: 149.0\n",
+      "Episode reward: 71.0\n",
       "Episode: 433\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 109.0\n",
       "Episode: 434\n",
-      "Episode reward: 42.0\n",
-      "Episode: 435\n",
       "Episode reward: 55.0\n",
+      "Episode: 435\n",
+      "Episode reward: 41.0\n",
       "Episode: 436\n",
-      "Episode reward: 200.0\n",
+      "Episode reward: 33.0\n",
       "Episode: 437\n",
-      "Episode reward: 73.0\n",
+      "Episode reward: 77.0\n",
       "Episode: 438\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 439\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 440\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 441\n",
-      "Episode reward: 104.0\n",
+      "Episode reward: 71.0\n",
       "Episode: 442\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 49.0\n",
       "Episode: 443\n",
-      "Episode reward: 15.0\n",
+      "Episode reward: 52.0\n",
       "Episode: 444\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 445\n",
-      "Episode reward: 56.0\n",
+      "Episode reward: 43.0\n",
       "Episode: 446\n",
-      "Episode reward: 30.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 447\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 448\n",
-      "Episode reward: 17.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 449\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 450\n",
-      "Episode reward: 55.0\n",
+      "Episode reward: 46.0\n",
       "Episode: 451\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 38.0\n",
       "Episode: 452\n",
       "Episode reward: 9.0\n",
       "Episode: 453\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 49.0\n",
       "Episode: 454\n",
-      "Episode reward: 19.0\n",
+      "Episode reward: 56.0\n",
       "Episode: 455\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 126.0\n",
       "Episode: 456\n",
-      "Episode reward: 19.0\n",
+      "Episode reward: 47.0\n",
       "Episode: 457\n",
-      "Episode reward: 42.0\n",
+      "Episode reward: 61.0\n",
       "Episode: 458\n",
-      "Episode reward: 54.0\n",
+      "Episode reward: 61.0\n",
       "Episode: 459\n",
-      "Episode reward: 30.0\n",
+      "Episode reward: 24.0\n",
       "Episode: 460\n",
-      "Episode reward: 61.0\n",
+      "Episode reward: 104.0\n",
       "Episode: 461\n",
-      "Episode reward: 35.0\n",
+      "Episode reward: 56.0\n",
       "Episode: 462\n",
-      "Episode reward: 58.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 463\n",
       "Episode reward: 13.0\n",
       "Episode: 464\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 465\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 466\n",
-      "Episode reward: 32.0\n",
+      "Episode reward: 43.0\n",
       "Episode: 467\n",
-      "Episode reward: 103.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 468\n",
-      "Episode reward: 39.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 469\n",
-      "Episode reward: 68.0\n",
+      "Episode reward: 75.0\n",
       "Episode: 470\n",
-      "Episode reward: 78.0\n",
+      "Episode reward: 30.0\n",
       "Episode: 471\n",
-      "Episode reward: 82.0\n",
+      "Episode reward: 29.0\n",
       "Episode: 472\n",
-      "Episode reward: 57.0\n",
+      "Episode reward: 33.0\n",
       "Episode: 473\n",
-      "Episode reward: 43.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 474\n",
-      "Episode reward: 33.0\n",
+      "Episode reward: 27.0\n",
       "Episode: 475\n",
-      "Episode reward: 46.0\n",
+      "Episode reward: 88.0\n",
       "Episode: 476\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 49.0\n",
       "Episode: 477\n",
-      "Episode reward: 45.0\n",
+      "Episode reward: 18.0\n",
       "Episode: 478\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 479\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 64.0\n",
       "Episode: 480\n",
-      "Episode reward: 62.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 481\n",
-      "Episode reward: 35.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 482\n",
-      "Episode reward: 12.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 483\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 36.0\n",
       "Episode: 484\n",
-      "Episode reward: 20.0\n",
+      "Episode reward: 51.0\n",
       "Episode: 485\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 39.0\n",
       "Episode: 486\n",
-      "Episode reward: 38.0\n",
+      "Episode reward: 37.0\n",
       "Episode: 487\n",
-      "Episode reward: 52.0\n",
+      "Episode reward: 29.0\n",
       "Episode: 488\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 69.0\n",
       "Episode: 489\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 34.0\n",
       "Episode: 490\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 48.0\n",
       "Episode: 491\n",
-      "Episode reward: 34.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 492\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 29.0\n",
       "Episode: 493\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 20.0\n",
       "Episode: 494\n",
-      "Episode reward: 113.0\n",
+      "Episode reward: 19.0\n",
       "Episode: 495\n",
-      "Episode reward: 56.0\n",
+      "Episode reward: 37.0\n",
       "Episode: 496\n",
-      "Episode reward: 39.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 497\n",
-      "Episode reward: 67.0\n",
+      "Episode reward: 21.0\n",
       "Episode: 498\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 11.0\n",
       "Episode: 499\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 29.0\n",
       "Episode: 500\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 34.0\n",
       "Episode: 501\n",
-      "Episode reward: 32.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 502\n",
-      "Episode reward: 95.0\n",
+      "Episode reward: 80.0\n",
       "Episode: 503\n",
-      "Episode reward: 25.0\n",
+      "Episode reward: 75.0\n",
       "Episode: 504\n",
-      "Episode reward: 28.0\n",
+      "Episode reward: 31.0\n",
       "Episode: 505\n",
-      "Episode reward: 135.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 506\n",
-      "Episode reward: 98.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 507\n",
-      "Episode reward: 127.0\n",
+      "Episode reward: 46.0\n",
       "Episode: 508\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 12.0\n",
       "Episode: 509\n",
-      "Episode reward: 62.0\n",
+      "Episode reward: 32.0\n",
       "Episode: 510\n",
-      "Episode reward: 73.0\n",
+      "Episode reward: 66.0\n",
       "Episode: 511\n",
-      "Episode reward: 49.0\n",
+      "Episode reward: 13.0\n",
       "Episode: 512\n",
-      "Episode reward: 50.0\n",
+      "Episode reward: 42.0\n",
       "Episode: 513\n",
-      "Episode reward: 58.0\n",
+      "Episode reward: 99.0\n",
       "Episode: 514\n",
-      "Episode reward: 47.0\n",
+      "Episode reward: 48.0\n",
       "Episode: 515\n",
-      "Episode reward: 36.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 516\n",
-      "Episode reward: 52.0\n",
+      "Episode reward: 8.0\n",
       "Episode: 517\n",
-      "Episode reward: 28.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 518\n",
-      "Episode reward: 116.0\n",
+      "Episode reward: 43.0\n",
       "Episode: 519\n",
-      "Episode reward: 152.0\n",
+      "Episode reward: 36.0\n",
       "Episode: 520\n",
-      "Episode reward: 103.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 521\n",
-      "Episode reward: 29.0\n",
+      "Episode reward: 27.0\n",
       "Episode: 522\n",
-      "Episode reward: 57.0\n",
+      "Episode reward: 59.0\n",
       "Episode: 523\n",
-      "Episode reward: 89.0\n",
+      "Episode reward: 82.0\n",
       "Episode: 524\n",
-      "Episode reward: 102.0\n",
+      "Episode reward: 89.0\n",
       "Episode: 525\n",
-      "Episode reward: 19.0\n",
+      "Episode reward: 53.0\n",
       "Episode: 526\n",
-      "Episode reward: 80.0\n",
+      "Episode reward: 27.0\n",
       "Episode: 527\n",
-      "Episode reward: 92.0\n",
+      "Episode reward: 86.0\n",
       "Episode: 528\n",
-      "Episode reward: 31.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 529\n",
-      "Episode reward: 44.0\n",
+      "Episode reward: 71.0\n",
       "Episode: 530\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 37.0\n",
       "Episode: 531\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 27.0\n",
       "Episode: 532\n",
-      "Episode reward: 58.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 533\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 534\n",
-      "Episode reward: 11.0\n",
+      "Episode reward: 125.0\n",
       "Episode: 535\n",
-      "Episode reward: 138.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 536\n",
-      "Episode reward: 107.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 537\n",
-      "Episode reward: 55.0\n",
+      "Episode reward: 117.0\n",
       "Episode: 538\n",
-      "Episode reward: 20.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 539\n",
-      "Episode reward: 49.0\n",
+      "Episode reward: 51.0\n",
       "Episode: 540\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 29.0\n",
       "Episode: 541\n",
       "Episode reward: 35.0\n",
       "Episode: 542\n",
-      "Episode reward: 48.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 543\n",
-      "Episode reward: 33.0\n",
+      "Episode reward: 81.0\n",
       "Episode: 544\n",
-      "Episode reward: 10.0\n",
+      "Episode reward: 104.0\n",
       "Episode: 545\n",
-      "Episode reward: 26.0\n",
-      "Episode: 546\n",
       "Episode reward: 50.0\n",
+      "Episode: 546\n",
+      "Episode reward: 27.0\n",
       "Episode: 547\n",
-      "Episode reward: 32.0\n",
+      "Episode reward: 98.0\n",
       "Episode: 548\n",
-      "Episode reward: 36.0\n",
+      "Episode reward: 87.0\n",
       "Episode: 549\n",
-      "Episode reward: 44.0\n",
+      "Episode reward: 61.0\n",
       "Episode: 550\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 551\n",
-      "Episode reward: 17.0\n",
+      "Episode reward: 10.0\n",
       "Episode: 552\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 36.0\n",
       "Episode: 553\n",
-      "Episode reward: 98.0\n",
+      "Episode reward: 37.0\n",
       "Episode: 554\n",
       "Episode reward: 28.0\n",
       "Episode: 555\n",
-      "Episode reward: 27.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 556\n",
-      "Episode reward: 8.0\n",
+      "Episode reward: 76.0\n",
       "Episode: 557\n",
-      "Episode reward: 54.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 558\n",
-      "Episode reward: 20.0\n",
+      "Episode reward: 35.0\n",
       "Episode: 559\n",
-      "Episode reward: 61.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 560\n",
-      "Episode reward: 81.0\n",
+      "Episode reward: 68.0\n",
       "Episode: 561\n",
-      "Episode reward: 42.0\n",
+      "Episode reward: 47.0\n",
       "Episode: 562\n",
-      "Episode reward: 30.0\n",
+      "Episode reward: 25.0\n",
       "Episode: 563\n",
-      "Episode reward: 33.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 564\n",
-      "Episode reward: 59.0\n",
+      "Episode reward: 34.0\n",
       "Episode: 565\n",
-      "Episode reward: 44.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 566\n",
-      "Episode reward: 24.0\n",
+      "Episode reward: 29.0\n",
       "Episode: 567\n",
-      "Episode reward: 37.0\n",
+      "Episode reward: 51.0\n",
       "Episode: 568\n",
-      "Episode reward: 45.0\n",
+      "Episode reward: 40.0\n",
       "Episode: 569\n",
-      "Episode reward: 48.0\n",
+      "Episode reward: 26.0\n",
       "Episode: 570\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 51.0\n",
       "Episode: 571\n",
-      "Episode reward: 50.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 572\n",
-      "Episode reward: 40.0\n",
+      "Episode reward: 36.0\n",
       "Episode: 573\n",
       "Episode reward: 34.0\n",
       "Episode: 574\n",
-      "Episode reward: 41.0\n",
+      "Episode reward: 74.0\n",
       "Episode: 575\n",
-      "Episode reward: 9.0\n",
+      "Episode reward: 42.0\n",
       "Episode: 576\n",
-      "Episode reward: 41.0\n",
+      "Episode reward: 22.0\n",
       "Episode: 577\n",
-      "Episode reward: 54.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 578\n",
-      "Episode reward: 38.0\n",
+      "Episode reward: 32.0\n",
       "Episode: 579\n",
-      "Episode reward: 22.0\n",
+      "Episode reward: 62.0\n",
       "Episode: 580\n",
-      "Episode reward: 35.0\n",
+      "Episode reward: 23.0\n",
       "Episode: 581\n",
-      "Episode reward: 54.0\n",
+      "Episode reward: 53.0\n",
       "Episode: 582\n",
-      "Episode reward: 41.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 583\n",
-      "Episode reward: 23.0\n",
+      "Episode reward: 28.0\n",
       "Episode: 584\n",
+      "Episode reward: 17.0\n",
+      "Episode: 585\n",
+      "Episode reward: 57.0\n",
+      "Episode: 586\n",
+      "Episode reward: 49.0\n",
+      "Episode: 587\n",
       "Episode reward: 63.0\n",
-      "Episode: 585\n"
-     ]
-    },
-    {
-     "ename": "KeyboardInterrupt",
-     "evalue": "",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
-      "\u001b[1;32m/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb Cell 29\u001b[0m line \u001b[0;36m2\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X26sZmlsZQ%3D%3D?line=16'>17</a>\u001b[0m     action \u001b[39m=\u001b[39m env\u001b[39m.\u001b[39maction_space\u001b[39m.\u001b[39msample()\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X26sZmlsZQ%3D%3D?line=18'>19</a>\u001b[0m \u001b[39m# Take the chosen action and observe the next state and reward\u001b[39;00m\n\u001b[0;32m---> <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X26sZmlsZQ%3D%3D?line=19'>20</a>\u001b[0m next_state, reward, terminated, truncated, info \u001b[39m=\u001b[39m env\u001b[39m.\u001b[39;49mstep(action)\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X26sZmlsZQ%3D%3D?line=20'>21</a>\u001b[0m next_state \u001b[39m=\u001b[39m discretize_state(next_state, num_bins)\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/nipun/git/blog/posts/2023-Dec-11-gym.ipynb#X26sZmlsZQ%3D%3D?line=22'>23</a>\u001b[0m \u001b[39m# Update the Q-table using the Q-learning update rule\u001b[39;00m\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/gymnasium/wrappers/time_limit.py:57\u001b[0m, in \u001b[0;36mTimeLimit.step\u001b[0;34m(self, action)\u001b[0m\n\u001b[1;32m     46\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mstep\u001b[39m(\u001b[39mself\u001b[39m, action):\n\u001b[1;32m     47\u001b[0m \u001b[39m    \u001b[39m\u001b[39m\"\"\"Steps through the environment and if the number of steps elapsed exceeds ``max_episode_steps`` then truncate.\u001b[39;00m\n\u001b[1;32m     48\u001b[0m \n\u001b[1;32m     49\u001b[0m \u001b[39m    Args:\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     55\u001b[0m \n\u001b[1;32m     56\u001b[0m \u001b[39m    \"\"\"\u001b[39;00m\n\u001b[0;32m---> 57\u001b[0m     observation, reward, terminated, truncated, info \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49menv\u001b[39m.\u001b[39;49mstep(action)\n\u001b[1;32m     58\u001b[0m     \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_elapsed_steps \u001b[39m+\u001b[39m\u001b[39m=\u001b[39m \u001b[39m1\u001b[39m\n\u001b[1;32m     60\u001b[0m     \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_elapsed_steps \u001b[39m>\u001b[39m\u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_max_episode_steps:\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/gymnasium/wrappers/order_enforcing.py:56\u001b[0m, in \u001b[0;36mOrderEnforcing.step\u001b[0;34m(self, action)\u001b[0m\n\u001b[1;32m     54\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_has_reset:\n\u001b[1;32m     55\u001b[0m     \u001b[39mraise\u001b[39;00m ResetNeeded(\u001b[39m\"\u001b[39m\u001b[39mCannot call env.step() before calling env.reset()\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[0;32m---> 56\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49menv\u001b[39m.\u001b[39;49mstep(action)\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/gymnasium/wrappers/env_checker.py:51\u001b[0m, in \u001b[0;36mPassiveEnvChecker.step\u001b[0;34m(self, action)\u001b[0m\n\u001b[1;32m     49\u001b[0m     \u001b[39mreturn\u001b[39;00m env_step_passive_checker(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39menv, action)\n\u001b[1;32m     50\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m---> 51\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49menv\u001b[39m.\u001b[39;49mstep(action)\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/gymnasium/envs/classic_control/cartpole.py:190\u001b[0m, in \u001b[0;36mCartPoleEnv.step\u001b[0;34m(self, action)\u001b[0m\n\u001b[1;32m    187\u001b[0m     reward \u001b[39m=\u001b[39m \u001b[39m0.0\u001b[39m\n\u001b[1;32m    189\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mrender_mode \u001b[39m==\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mhuman\u001b[39m\u001b[39m\"\u001b[39m:\n\u001b[0;32m--> 190\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mrender()\n\u001b[1;32m    191\u001b[0m \u001b[39mreturn\u001b[39;00m np\u001b[39m.\u001b[39marray(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mstate, dtype\u001b[39m=\u001b[39mnp\u001b[39m.\u001b[39mfloat32), reward, terminated, \u001b[39mFalse\u001b[39;00m, {}\n",
-      "File \u001b[0;32m~/miniconda3/lib/python3.9/site-packages/gymnasium/envs/classic_control/cartpole.py:302\u001b[0m, in \u001b[0;36mCartPoleEnv.render\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    300\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mrender_mode \u001b[39m==\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mhuman\u001b[39m\u001b[39m\"\u001b[39m:\n\u001b[1;32m    301\u001b[0m     pygame\u001b[39m.\u001b[39mevent\u001b[39m.\u001b[39mpump()\n\u001b[0;32m--> 302\u001b[0m     \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mclock\u001b[39m.\u001b[39;49mtick(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mmetadata[\u001b[39m\"\u001b[39;49m\u001b[39mrender_fps\u001b[39;49m\u001b[39m\"\u001b[39;49m])\n\u001b[1;32m    303\u001b[0m     pygame\u001b[39m.\u001b[39mdisplay\u001b[39m.\u001b[39mflip()\n\u001b[1;32m    305\u001b[0m \u001b[39melif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mrender_mode \u001b[39m==\u001b[39m \u001b[39m\"\u001b[39m\u001b[39mrgb_array\u001b[39m\u001b[39m\"\u001b[39m:\n",
-      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+      "Episode: 588\n",
+      "Episode reward: 17.0\n",
+      "Episode: 589\n",
+      "Episode reward: 39.0\n",
+      "Episode: 590\n",
+      "Episode reward: 39.0\n",
+      "Episode: 591\n",
+      "Episode reward: 35.0\n",
+      "Episode: 592\n",
+      "Episode reward: 27.0\n",
+      "Episode: 593\n",
+      "Episode reward: 22.0\n",
+      "Episode: 594\n",
+      "Episode reward: 59.0\n",
+      "Episode: 595\n",
+      "Episode reward: 22.0\n",
+      "Episode: 596\n",
+      "Episode reward: 9.0\n",
+      "Episode: 597\n",
+      "Episode reward: 42.0\n",
+      "Episode: 598\n",
+      "Episode reward: 28.0\n",
+      "Episode: 599\n",
+      "Episode reward: 19.0\n",
+      "Episode: 600\n",
+      "Episode reward: 62.0\n",
+      "Episode: 601\n",
+      "Episode reward: 27.0\n",
+      "Episode: 602\n",
+      "Episode reward: 37.0\n",
+      "Episode: 603\n",
+      "Episode reward: 27.0\n",
+      "Episode: 604\n",
+      "Episode reward: 20.0\n",
+      "Episode: 605\n",
+      "Episode reward: 100.0\n",
+      "Episode: 606\n",
+      "Episode reward: 56.0\n",
+      "Episode: 607\n",
+      "Episode reward: 23.0\n",
+      "Episode: 608\n",
+      "Episode reward: 120.0\n",
+      "Episode: 609\n",
+      "Episode reward: 39.0\n",
+      "Episode: 610\n",
+      "Episode reward: 26.0\n",
+      "Episode: 611\n",
+      "Episode reward: 36.0\n",
+      "Episode: 612\n",
+      "Episode reward: 49.0\n",
+      "Episode: 613\n",
+      "Episode reward: 20.0\n",
+      "Episode: 614\n",
+      "Episode reward: 37.0\n",
+      "Episode: 615\n",
+      "Episode reward: 12.0\n",
+      "Episode: 616\n",
+      "Episode reward: 117.0\n",
+      "Episode: 617\n",
+      "Episode reward: 32.0\n",
+      "Episode: 618\n",
+      "Episode reward: 81.0\n",
+      "Episode: 619\n",
+      "Episode reward: 59.0\n",
+      "Episode: 620\n",
+      "Episode reward: 12.0\n",
+      "Episode: 621\n",
+      "Episode reward: 28.0\n",
+      "Episode: 622\n",
+      "Episode reward: 30.0\n",
+      "Episode: 623\n",
+      "Episode reward: 40.0\n",
+      "Episode: 624\n",
+      "Episode reward: 24.0\n",
+      "Episode: 625\n",
+      "Episode reward: 107.0\n",
+      "Episode: 626\n",
+      "Episode reward: 79.0\n",
+      "Episode: 627\n",
+      "Episode reward: 68.0\n",
+      "Episode: 628\n",
+      "Episode reward: 29.0\n",
+      "Episode: 629\n",
+      "Episode reward: 27.0\n",
+      "Episode: 630\n",
+      "Episode reward: 45.0\n",
+      "Episode: 631\n",
+      "Episode reward: 42.0\n",
+      "Episode: 632\n",
+      "Episode reward: 87.0\n",
+      "Episode: 633\n",
+      "Episode reward: 56.0\n",
+      "Episode: 634\n",
+      "Episode reward: 29.0\n",
+      "Episode: 635\n",
+      "Episode reward: 26.0\n",
+      "Episode: 636\n",
+      "Episode reward: 26.0\n",
+      "Episode: 637\n",
+      "Episode reward: 53.0\n",
+      "Episode: 638\n",
+      "Episode reward: 35.0\n",
+      "Episode: 639\n",
+      "Episode reward: 28.0\n",
+      "Episode: 640\n",
+      "Episode reward: 42.0\n",
+      "Episode: 641\n",
+      "Episode reward: 62.0\n",
+      "Episode: 642\n",
+      "Episode reward: 53.0\n",
+      "Episode: 643\n",
+      "Episode reward: 34.0\n",
+      "Episode: 644\n",
+      "Episode reward: 23.0\n",
+      "Episode: 645\n",
+      "Episode reward: 52.0\n",
+      "Episode: 646\n",
+      "Episode reward: 27.0\n",
+      "Episode: 647\n",
+      "Episode reward: 49.0\n",
+      "Episode: 648\n",
+      "Episode reward: 39.0\n",
+      "Episode: 649\n",
+      "Episode reward: 36.0\n",
+      "Episode: 650\n",
+      "Episode reward: 21.0\n",
+      "Episode: 651\n",
+      "Episode reward: 89.0\n",
+      "Episode: 652\n",
+      "Episode reward: 38.0\n",
+      "Episode: 653\n",
+      "Episode reward: 45.0\n",
+      "Episode: 654\n",
+      "Episode reward: 39.0\n",
+      "Episode: 655\n",
+      "Episode reward: 41.0\n",
+      "Episode: 656\n",
+      "Episode reward: 75.0\n",
+      "Episode: 657\n",
+      "Episode reward: 19.0\n",
+      "Episode: 658\n",
+      "Episode reward: 34.0\n",
+      "Episode: 659\n",
+      "Episode reward: 32.0\n",
+      "Episode: 660\n",
+      "Episode reward: 63.0\n",
+      "Episode: 661\n",
+      "Episode reward: 61.0\n",
+      "Episode: 662\n",
+      "Episode reward: 29.0\n",
+      "Episode: 663\n",
+      "Episode reward: 36.0\n",
+      "Episode: 664\n",
+      "Episode reward: 40.0\n",
+      "Episode: 665\n",
+      "Episode reward: 51.0\n",
+      "Episode: 666\n",
+      "Episode reward: 17.0\n",
+      "Episode: 667\n",
+      "Episode reward: 105.0\n",
+      "Episode: 668\n",
+      "Episode reward: 38.0\n",
+      "Episode: 669\n",
+      "Episode reward: 74.0\n",
+      "Episode: 670\n",
+      "Episode reward: 38.0\n",
+      "Episode: 671\n",
+      "Episode reward: 22.0\n",
+      "Episode: 672\n",
+      "Episode reward: 45.0\n",
+      "Episode: 673\n",
+      "Episode reward: 19.0\n",
+      "Episode: 674\n",
+      "Episode reward: 64.0\n",
+      "Episode: 675\n",
+      "Episode reward: 115.0\n",
+      "Episode: 676\n",
+      "Episode reward: 43.0\n",
+      "Episode: 677\n",
+      "Episode reward: 61.0\n",
+      "Episode: 678\n",
+      "Episode reward: 53.0\n",
+      "Episode: 679\n",
+      "Episode reward: 23.0\n",
+      "Episode: 680\n",
+      "Episode reward: 32.0\n",
+      "Episode: 681\n",
+      "Episode reward: 36.0\n",
+      "Episode: 682\n",
+      "Episode reward: 89.0\n",
+      "Episode: 683\n",
+      "Episode reward: 19.0\n",
+      "Episode: 684\n",
+      "Episode reward: 23.0\n",
+      "Episode: 685\n",
+      "Episode reward: 38.0\n",
+      "Episode: 686\n",
+      "Episode reward: 49.0\n",
+      "Episode: 687\n",
+      "Episode reward: 83.0\n",
+      "Episode: 688\n",
+      "Episode reward: 53.0\n",
+      "Episode: 689\n",
+      "Episode reward: 62.0\n",
+      "Episode: 690\n",
+      "Episode reward: 29.0\n",
+      "Episode: 691\n",
+      "Episode reward: 25.0\n",
+      "Episode: 692\n",
+      "Episode reward: 24.0\n",
+      "Episode: 693\n",
+      "Episode reward: 27.0\n",
+      "Episode: 694\n",
+      "Episode reward: 9.0\n",
+      "Episode: 695\n",
+      "Episode reward: 39.0\n",
+      "Episode: 696\n",
+      "Episode reward: 78.0\n",
+      "Episode: 697\n",
+      "Episode reward: 115.0\n",
+      "Episode: 698\n",
+      "Episode reward: 38.0\n",
+      "Episode: 699\n",
+      "Episode reward: 23.0\n",
+      "Episode: 700\n",
+      "Episode reward: 41.0\n",
+      "Episode: 701\n",
+      "Episode reward: 18.0\n",
+      "Episode: 702\n",
+      "Episode reward: 37.0\n",
+      "Episode: 703\n",
+      "Episode reward: 37.0\n",
+      "Episode: 704\n",
+      "Episode reward: 23.0\n",
+      "Episode: 705\n",
+      "Episode reward: 119.0\n",
+      "Episode: 706\n",
+      "Episode reward: 41.0\n",
+      "Episode: 707\n",
+      "Episode reward: 37.0\n",
+      "Episode: 708\n",
+      "Episode reward: 23.0\n",
+      "Episode: 709\n",
+      "Episode reward: 36.0\n",
+      "Episode: 710\n",
+      "Episode reward: 66.0\n",
+      "Episode: 711\n",
+      "Episode reward: 33.0\n",
+      "Episode: 712\n",
+      "Episode reward: 33.0\n",
+      "Episode: 713\n",
+      "Episode reward: 61.0\n",
+      "Episode: 714\n",
+      "Episode reward: 66.0\n",
+      "Episode: 715\n",
+      "Episode reward: 44.0\n",
+      "Episode: 716\n",
+      "Episode reward: 19.0\n",
+      "Episode: 717\n",
+      "Episode reward: 72.0\n",
+      "Episode: 718\n",
+      "Episode reward: 25.0\n",
+      "Episode: 719\n",
+      "Episode reward: 26.0\n",
+      "Episode: 720\n",
+      "Episode reward: 29.0\n",
+      "Episode: 721\n",
+      "Episode reward: 40.0\n",
+      "Episode: 722\n",
+      "Episode reward: 41.0\n",
+      "Episode: 723\n",
+      "Episode reward: 37.0\n",
+      "Episode: 724\n",
+      "Episode reward: 70.0\n",
+      "Episode: 725\n",
+      "Episode reward: 56.0\n",
+      "Episode: 726\n",
+      "Episode reward: 37.0\n",
+      "Episode: 727\n",
+      "Episode reward: 85.0\n",
+      "Episode: 728\n",
+      "Episode reward: 22.0\n",
+      "Episode: 729\n",
+      "Episode reward: 23.0\n",
+      "Episode: 730\n",
+      "Episode reward: 50.0\n",
+      "Episode: 731\n",
+      "Episode reward: 23.0\n",
+      "Episode: 732\n",
+      "Episode reward: 24.0\n",
+      "Episode: 733\n",
+      "Episode reward: 96.0\n",
+      "Episode: 734\n",
+      "Episode reward: 18.0\n",
+      "Episode: 735\n",
+      "Episode reward: 23.0\n",
+      "Episode: 736\n",
+      "Episode reward: 23.0\n",
+      "Episode: 737\n",
+      "Episode reward: 42.0\n",
+      "Episode: 738\n",
+      "Episode reward: 22.0\n",
+      "Episode: 739\n",
+      "Episode reward: 27.0\n",
+      "Episode: 740\n",
+      "Episode reward: 72.0\n",
+      "Episode: 741\n",
+      "Episode reward: 39.0\n",
+      "Episode: 742\n",
+      "Episode reward: 18.0\n",
+      "Episode: 743\n",
+      "Episode reward: 22.0\n",
+      "Episode: 744\n",
+      "Episode reward: 23.0\n",
+      "Episode: 745\n",
+      "Episode reward: 22.0\n",
+      "Episode: 746\n",
+      "Episode reward: 26.0\n",
+      "Episode: 747\n",
+      "Episode reward: 23.0\n",
+      "Episode: 748\n",
+      "Episode reward: 20.0\n",
+      "Episode: 749\n",
+      "Episode reward: 32.0\n",
+      "Episode: 750\n",
+      "Episode reward: 17.0\n",
+      "Episode: 751\n",
+      "Episode reward: 33.0\n",
+      "Episode: 752\n",
+      "Episode reward: 56.0\n",
+      "Episode: 753\n",
+      "Episode reward: 22.0\n",
+      "Episode: 754\n",
+      "Episode reward: 43.0\n",
+      "Episode: 755\n",
+      "Episode reward: 91.0\n",
+      "Episode: 756\n",
+      "Episode reward: 39.0\n",
+      "Episode: 757\n",
+      "Episode reward: 43.0\n",
+      "Episode: 758\n",
+      "Episode reward: 49.0\n",
+      "Episode: 759\n",
+      "Episode reward: 38.0\n",
+      "Episode: 760\n",
+      "Episode reward: 66.0\n",
+      "Episode: 761\n",
+      "Episode reward: 147.0\n",
+      "Episode: 762\n",
+      "Episode reward: 172.0\n",
+      "Episode: 763\n",
+      "Episode reward: 146.0\n",
+      "Episode: 764\n",
+      "Episode reward: 187.0\n",
+      "Episode: 765\n",
+      "Episode reward: 135.0\n",
+      "Episode: 766\n",
+      "Episode reward: 153.0\n",
+      "Episode: 767\n",
+      "Episode reward: 145.0\n",
+      "Episode: 768\n",
+      "Episode reward: 153.0\n",
+      "Episode: 769\n",
+      "Episode reward: 191.0\n",
+      "Episode: 770\n",
+      "Episode reward: 114.0\n",
+      "Episode: 771\n",
+      "Episode reward: 160.0\n",
+      "Episode: 772\n",
+      "Episode reward: 195.0\n",
+      "Episode: 773\n",
+      "Episode reward: 200.0\n",
+      "Episode: 774\n",
+      "Episode reward: 151.0\n",
+      "Episode: 775\n",
+      "Episode reward: 196.0\n",
+      "Episode: 776\n",
+      "Episode reward: 113.0\n",
+      "Episode: 777\n",
+      "Episode reward: 158.0\n",
+      "Episode: 778\n",
+      "Episode reward: 141.0\n",
+      "Episode: 779\n",
+      "Episode reward: 160.0\n",
+      "Episode: 780\n",
+      "Episode reward: 200.0\n",
+      "Episode: 781\n",
+      "Episode reward: 160.0\n",
+      "Episode: 782\n",
+      "Episode reward: 117.0\n",
+      "Episode: 783\n",
+      "Episode reward: 193.0\n",
+      "Episode: 784\n",
+      "Episode reward: 119.0\n",
+      "Episode: 785\n",
+      "Episode reward: 195.0\n",
+      "Episode: 786\n",
+      "Episode reward: 118.0\n",
+      "Episode: 787\n",
+      "Episode reward: 156.0\n",
+      "Episode: 788\n",
+      "Episode reward: 195.0\n",
+      "Episode: 789\n",
+      "Episode reward: 139.0\n",
+      "Episode: 790\n",
+      "Episode reward: 188.0\n",
+      "Episode: 791\n",
+      "Episode reward: 155.0\n",
+      "Episode: 792\n",
+      "Episode reward: 113.0\n",
+      "Episode: 793\n",
+      "Episode reward: 187.0\n",
+      "Episode: 794\n",
+      "Episode reward: 126.0\n",
+      "Episode: 795\n",
+      "Episode reward: 157.0\n",
+      "Episode: 796\n",
+      "Episode reward: 181.0\n",
+      "Episode: 797\n",
+      "Episode reward: 110.0\n",
+      "Episode: 798\n",
+      "Episode reward: 200.0\n",
+      "Episode: 799\n",
+      "Episode reward: 141.0\n",
+      "Episode: 800\n",
+      "Episode reward: 170.0\n",
+      "Episode: 801\n",
+      "Episode reward: 136.0\n",
+      "Episode: 802\n",
+      "Episode reward: 153.0\n",
+      "Episode: 803\n",
+      "Episode reward: 200.0\n",
+      "Episode: 804\n",
+      "Episode reward: 115.0\n",
+      "Episode: 805\n",
+      "Episode reward: 168.0\n",
+      "Episode: 806\n",
+      "Episode reward: 165.0\n",
+      "Episode: 807\n",
+      "Episode reward: 198.0\n",
+      "Episode: 808\n",
+      "Episode reward: 146.0\n",
+      "Episode: 809\n",
+      "Episode reward: 194.0\n",
+      "Episode: 810\n",
+      "Episode reward: 152.0\n",
+      "Episode: 811\n",
+      "Episode reward: 200.0\n",
+      "Episode: 812\n",
+      "Episode reward: 162.0\n",
+      "Episode: 813\n",
+      "Episode reward: 150.0\n",
+      "Episode: 814\n",
+      "Episode reward: 140.0\n",
+      "Episode: 815\n",
+      "Episode reward: 155.0\n",
+      "Episode: 816\n",
+      "Episode reward: 200.0\n",
+      "Episode: 817\n",
+      "Episode reward: 145.0\n",
+      "Episode: 818\n",
+      "Episode reward: 181.0\n",
+      "Episode: 819\n",
+      "Episode reward: 200.0\n",
+      "Episode: 820\n",
+      "Episode reward: 104.0\n",
+      "Episode: 821\n",
+      "Episode reward: 142.0\n",
+      "Episode: 822\n",
+      "Episode reward: 107.0\n",
+      "Episode: 823\n",
+      "Episode reward: 200.0\n",
+      "Episode: 824\n",
+      "Episode reward: 147.0\n",
+      "Episode: 825\n",
+      "Episode reward: 144.0\n",
+      "Episode: 826\n",
+      "Episode reward: 167.0\n",
+      "Episode: 827\n",
+      "Episode reward: 122.0\n",
+      "Episode: 828\n",
+      "Episode reward: 127.0\n",
+      "Episode: 829\n",
+      "Episode reward: 143.0\n",
+      "Episode: 830\n",
+      "Episode reward: 113.0\n",
+      "Episode: 831\n",
+      "Episode reward: 131.0\n",
+      "Episode: 832\n",
+      "Episode reward: 120.0\n",
+      "Episode: 833\n",
+      "Episode reward: 159.0\n",
+      "Episode: 834\n",
+      "Episode reward: 200.0\n",
+      "Episode: 835\n",
+      "Episode reward: 200.0\n",
+      "Episode: 836\n",
+      "Episode reward: 193.0\n",
+      "Episode: 837\n",
+      "Episode reward: 194.0\n",
+      "Episode: 838\n",
+      "Episode reward: 119.0\n",
+      "Episode: 839\n",
+      "Episode reward: 143.0\n",
+      "Episode: 840\n",
+      "Episode reward: 141.0\n",
+      "Episode: 841\n",
+      "Episode reward: 165.0\n",
+      "Episode: 842\n",
+      "Episode reward: 84.0\n",
+      "Episode: 843\n",
+      "Episode reward: 139.0\n",
+      "Episode: 844\n",
+      "Episode reward: 130.0\n",
+      "Episode: 845\n",
+      "Episode reward: 70.0\n",
+      "Episode: 846\n",
+      "Episode reward: 200.0\n",
+      "Episode: 847\n",
+      "Episode reward: 161.0\n",
+      "Episode: 848\n",
+      "Episode reward: 142.0\n",
+      "Episode: 849\n",
+      "Episode reward: 177.0\n",
+      "Episode: 850\n",
+      "Episode reward: 133.0\n",
+      "Episode: 851\n",
+      "Episode reward: 200.0\n",
+      "Episode: 852\n",
+      "Episode reward: 134.0\n",
+      "Episode: 853\n",
+      "Episode reward: 119.0\n",
+      "Episode: 854\n",
+      "Episode reward: 200.0\n",
+      "Episode: 855\n",
+      "Episode reward: 121.0\n",
+      "Episode: 856\n",
+      "Episode reward: 183.0\n",
+      "Episode: 857\n",
+      "Episode reward: 140.0\n",
+      "Episode: 858\n",
+      "Episode reward: 196.0\n",
+      "Episode: 859\n",
+      "Episode reward: 200.0\n",
+      "Episode: 860\n",
+      "Episode reward: 110.0\n",
+      "Episode: 861\n",
+      "Episode reward: 138.0\n",
+      "Episode: 862\n",
+      "Episode reward: 200.0\n",
+      "Episode: 863\n",
+      "Episode reward: 153.0\n",
+      "Episode: 864\n",
+      "Episode reward: 161.0\n",
+      "Episode: 865\n",
+      "Episode reward: 135.0\n",
+      "Episode: 866\n",
+      "Episode reward: 169.0\n",
+      "Episode: 867\n",
+      "Episode reward: 181.0\n",
+      "Episode: 868\n",
+      "Episode reward: 112.0\n",
+      "Episode: 869\n",
+      "Episode reward: 185.0\n",
+      "Episode: 870\n",
+      "Episode reward: 147.0\n",
+      "Episode: 871\n",
+      "Episode reward: 121.0\n",
+      "Episode: 872\n",
+      "Episode reward: 118.0\n",
+      "Episode: 873\n",
+      "Episode reward: 167.0\n",
+      "Episode: 874\n",
+      "Episode reward: 184.0\n",
+      "Episode: 875\n",
+      "Episode reward: 200.0\n",
+      "Episode: 876\n",
+      "Episode reward: 146.0\n",
+      "Episode: 877\n",
+      "Episode reward: 150.0\n",
+      "Episode: 878\n",
+      "Episode reward: 149.0\n",
+      "Episode: 879\n",
+      "Episode reward: 107.0\n",
+      "Episode: 880\n",
+      "Episode reward: 151.0\n",
+      "Episode: 881\n",
+      "Episode reward: 160.0\n",
+      "Episode: 882\n",
+      "Episode reward: 153.0\n",
+      "Episode: 883\n",
+      "Episode reward: 99.0\n",
+      "Episode: 884\n",
+      "Episode reward: 119.0\n",
+      "Episode: 885\n",
+      "Episode reward: 200.0\n",
+      "Episode: 886\n",
+      "Episode reward: 158.0\n",
+      "Episode: 887\n",
+      "Episode reward: 155.0\n",
+      "Episode: 888\n",
+      "Episode reward: 143.0\n",
+      "Episode: 889\n",
+      "Episode reward: 200.0\n",
+      "Episode: 890\n",
+      "Episode reward: 188.0\n",
+      "Episode: 891\n",
+      "Episode reward: 147.0\n",
+      "Episode: 892\n",
+      "Episode reward: 155.0\n",
+      "Episode: 893\n",
+      "Episode reward: 118.0\n",
+      "Episode: 894\n",
+      "Episode reward: 113.0\n",
+      "Episode: 895\n",
+      "Episode reward: 134.0\n",
+      "Episode: 896\n",
+      "Episode reward: 118.0\n",
+      "Episode: 897\n",
+      "Episode reward: 153.0\n",
+      "Episode: 898\n",
+      "Episode reward: 112.0\n",
+      "Episode: 899\n",
+      "Episode reward: 116.0\n",
+      "Episode: 900\n",
+      "Episode reward: 120.0\n",
+      "Episode: 901\n",
+      "Episode reward: 182.0\n",
+      "Episode: 902\n",
+      "Episode reward: 149.0\n",
+      "Episode: 903\n",
+      "Episode reward: 200.0\n",
+      "Episode: 904\n",
+      "Episode reward: 200.0\n",
+      "Episode: 905\n",
+      "Episode reward: 199.0\n",
+      "Episode: 906\n",
+      "Episode reward: 143.0\n",
+      "Episode: 907\n",
+      "Episode reward: 133.0\n",
+      "Episode: 908\n",
+      "Episode reward: 126.0\n",
+      "Episode: 909\n",
+      "Episode reward: 158.0\n",
+      "Episode: 910\n",
+      "Episode reward: 144.0\n",
+      "Episode: 911\n",
+      "Episode reward: 149.0\n",
+      "Episode: 912\n",
+      "Episode reward: 173.0\n",
+      "Episode: 913\n",
+      "Episode reward: 138.0\n",
+      "Episode: 914\n",
+      "Episode reward: 159.0\n",
+      "Episode: 915\n",
+      "Episode reward: 137.0\n",
+      "Episode: 916\n",
+      "Episode reward: 169.0\n",
+      "Episode: 917\n",
+      "Episode reward: 200.0\n",
+      "Episode: 918\n",
+      "Episode reward: 134.0\n",
+      "Episode: 919\n",
+      "Episode reward: 200.0\n",
+      "Episode: 920\n",
+      "Episode reward: 165.0\n",
+      "Episode: 921\n",
+      "Episode reward: 160.0\n",
+      "Episode: 922\n",
+      "Episode reward: 142.0\n",
+      "Episode: 923\n",
+      "Episode reward: 200.0\n",
+      "Episode: 924\n",
+      "Episode reward: 159.0\n",
+      "Episode: 925\n",
+      "Episode reward: 117.0\n",
+      "Episode: 926\n",
+      "Episode reward: 145.0\n",
+      "Episode: 927\n",
+      "Episode reward: 136.0\n",
+      "Episode: 928\n",
+      "Episode reward: 149.0\n",
+      "Episode: 929\n",
+      "Episode reward: 128.0\n",
+      "Episode: 930\n",
+      "Episode reward: 200.0\n",
+      "Episode: 931\n",
+      "Episode reward: 166.0\n",
+      "Episode: 932\n",
+      "Episode reward: 153.0\n",
+      "Episode: 933\n",
+      "Episode reward: 175.0\n",
+      "Episode: 934\n",
+      "Episode reward: 111.0\n",
+      "Episode: 935\n",
+      "Episode reward: 149.0\n",
+      "Episode: 936\n",
+      "Episode reward: 118.0\n",
+      "Episode: 937\n",
+      "Episode reward: 135.0\n",
+      "Episode: 938\n",
+      "Episode reward: 117.0\n",
+      "Episode: 939\n",
+      "Episode reward: 147.0\n",
+      "Episode: 940\n",
+      "Episode reward: 134.0\n",
+      "Episode: 941\n",
+      "Episode reward: 132.0\n",
+      "Episode: 942\n",
+      "Episode reward: 151.0\n",
+      "Episode: 943\n",
+      "Episode reward: 143.0\n",
+      "Episode: 944\n",
+      "Episode reward: 138.0\n",
+      "Episode: 945\n",
+      "Episode reward: 156.0\n",
+      "Episode: 946\n",
+      "Episode reward: 142.0\n",
+      "Episode: 947\n",
+      "Episode reward: 132.0\n",
+      "Episode: 948\n",
+      "Episode reward: 175.0\n",
+      "Episode: 949\n",
+      "Episode reward: 136.0\n",
+      "Episode: 950\n",
+      "Episode reward: 145.0\n",
+      "Episode: 951\n",
+      "Episode reward: 121.0\n",
+      "Episode: 952\n",
+      "Episode reward: 149.0\n",
+      "Episode: 953\n",
+      "Episode reward: 124.0\n",
+      "Episode: 954\n",
+      "Episode reward: 132.0\n",
+      "Episode: 955\n",
+      "Episode reward: 200.0\n",
+      "Episode: 956\n",
+      "Episode reward: 200.0\n",
+      "Episode: 957\n",
+      "Episode reward: 123.0\n",
+      "Episode: 958\n",
+      "Episode reward: 200.0\n",
+      "Episode: 959\n",
+      "Episode reward: 200.0\n",
+      "Episode: 960\n",
+      "Episode reward: 136.0\n",
+      "Episode: 961\n",
+      "Episode reward: 169.0\n",
+      "Episode: 962\n",
+      "Episode reward: 151.0\n",
+      "Episode: 963\n",
+      "Episode reward: 113.0\n",
+      "Episode: 964\n",
+      "Episode reward: 148.0\n",
+      "Episode: 965\n",
+      "Episode reward: 138.0\n",
+      "Episode: 966\n",
+      "Episode reward: 200.0\n",
+      "Episode: 967\n",
+      "Episode reward: 167.0\n",
+      "Episode: 968\n",
+      "Episode reward: 200.0\n",
+      "Episode: 969\n",
+      "Episode reward: 160.0\n",
+      "Episode: 970\n",
+      "Episode reward: 157.0\n",
+      "Episode: 971\n",
+      "Episode reward: 160.0\n",
+      "Episode: 972\n",
+      "Episode reward: 143.0\n",
+      "Episode: 973\n",
+      "Episode reward: 163.0\n",
+      "Episode: 974\n",
+      "Episode reward: 110.0\n",
+      "Episode: 975\n",
+      "Episode reward: 200.0\n",
+      "Episode: 976\n",
+      "Episode reward: 113.0\n",
+      "Episode: 977\n",
+      "Episode reward: 200.0\n",
+      "Episode: 978\n",
+      "Episode reward: 135.0\n",
+      "Episode: 979\n",
+      "Episode reward: 159.0\n",
+      "Episode: 980\n",
+      "Episode reward: 118.0\n",
+      "Episode: 981\n",
+      "Episode reward: 200.0\n",
+      "Episode: 982\n",
+      "Episode reward: 122.0\n",
+      "Episode: 983\n",
+      "Episode reward: 191.0\n",
+      "Episode: 984\n",
+      "Episode reward: 149.0\n",
+      "Episode: 985\n",
+      "Episode reward: 156.0\n",
+      "Episode: 986\n",
+      "Episode reward: 119.0\n",
+      "Episode: 987\n",
+      "Episode reward: 164.0\n",
+      "Episode: 988\n",
+      "Episode reward: 112.0\n",
+      "Episode: 989\n",
+      "Episode reward: 155.0\n",
+      "Episode: 990\n",
+      "Episode reward: 148.0\n",
+      "Episode: 991\n",
+      "Episode reward: 194.0\n",
+      "Episode: 992\n",
+      "Episode reward: 112.0\n",
+      "Episode: 993\n",
+      "Episode reward: 162.0\n",
+      "Episode: 994\n",
+      "Episode reward: 160.0\n",
+      "Episode: 995\n",
+      "Episode reward: 190.0\n",
+      "Episode: 996\n",
+      "Episode reward: 178.0\n",
+      "Episode: 997\n",
+      "Episode reward: 109.0\n",
+      "Episode: 998\n",
+      "Episode reward: 145.0\n",
+      "Episode: 999\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1000\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1001\n",
+      "Episode reward: 160.0\n",
+      "Episode: 1002\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1003\n",
+      "Episode reward: 167.0\n",
+      "Episode: 1004\n",
+      "Episode reward: 166.0\n",
+      "Episode: 1005\n",
+      "Episode reward: 183.0\n",
+      "Episode: 1006\n",
+      "Episode reward: 193.0\n",
+      "Episode: 1007\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1008\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1009\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1010\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1011\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1012\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1013\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1014\n",
+      "Episode reward: 120.0\n",
+      "Episode: 1015\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1016\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1017\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1018\n",
+      "Episode reward: 116.0\n",
+      "Episode: 1019\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1020\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1021\n",
+      "Episode reward: 90.0\n",
+      "Episode: 1022\n",
+      "Episode reward: 170.0\n",
+      "Episode: 1023\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1024\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1025\n",
+      "Episode reward: 172.0\n",
+      "Episode: 1026\n",
+      "Episode reward: 167.0\n",
+      "Episode: 1027\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1028\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1029\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1030\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1031\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1032\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1033\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1034\n",
+      "Episode reward: 141.0\n",
+      "Episode: 1035\n",
+      "Episode reward: 132.0\n",
+      "Episode: 1036\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1037\n",
+      "Episode reward: 177.0\n",
+      "Episode: 1038\n",
+      "Episode reward: 145.0\n",
+      "Episode: 1039\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1040\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1041\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1042\n",
+      "Episode reward: 160.0\n",
+      "Episode: 1043\n",
+      "Episode reward: 159.0\n",
+      "Episode: 1044\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1045\n",
+      "Episode reward: 132.0\n",
+      "Episode: 1046\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1047\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1048\n",
+      "Episode reward: 87.0\n",
+      "Episode: 1049\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1050\n",
+      "Episode reward: 82.0\n",
+      "Episode: 1051\n",
+      "Episode reward: 194.0\n",
+      "Episode: 1052\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1053\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1054\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1055\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1056\n",
+      "Episode reward: 139.0\n",
+      "Episode: 1057\n",
+      "Episode reward: 190.0\n",
+      "Episode: 1058\n",
+      "Episode reward: 159.0\n",
+      "Episode: 1059\n",
+      "Episode reward: 139.0\n",
+      "Episode: 1060\n",
+      "Episode reward: 120.0\n",
+      "Episode: 1061\n",
+      "Episode reward: 189.0\n",
+      "Episode: 1062\n",
+      "Episode reward: 145.0\n",
+      "Episode: 1063\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1064\n",
+      "Episode reward: 159.0\n",
+      "Episode: 1065\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1066\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1067\n",
+      "Episode reward: 152.0\n",
+      "Episode: 1068\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1069\n",
+      "Episode reward: 178.0\n",
+      "Episode: 1070\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1071\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1072\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1073\n",
+      "Episode reward: 178.0\n",
+      "Episode: 1074\n",
+      "Episode reward: 124.0\n",
+      "Episode: 1075\n",
+      "Episode reward: 174.0\n",
+      "Episode: 1076\n",
+      "Episode reward: 106.0\n",
+      "Episode: 1077\n",
+      "Episode reward: 180.0\n",
+      "Episode: 1078\n",
+      "Episode reward: 167.0\n",
+      "Episode: 1079\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1080\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1081\n",
+      "Episode reward: 190.0\n",
+      "Episode: 1082\n",
+      "Episode reward: 152.0\n",
+      "Episode: 1083\n",
+      "Episode reward: 145.0\n",
+      "Episode: 1084\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1085\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1086\n",
+      "Episode reward: 144.0\n",
+      "Episode: 1087\n",
+      "Episode reward: 128.0\n",
+      "Episode: 1088\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1089\n",
+      "Episode reward: 142.0\n",
+      "Episode: 1090\n",
+      "Episode reward: 145.0\n",
+      "Episode: 1091\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1092\n",
+      "Episode reward: 188.0\n",
+      "Episode: 1093\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1094\n",
+      "Episode reward: 165.0\n",
+      "Episode: 1095\n",
+      "Episode reward: 145.0\n",
+      "Episode: 1096\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1097\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1098\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1099\n",
+      "Episode reward: 164.0\n",
+      "Episode: 1100\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1101\n",
+      "Episode reward: 105.0\n",
+      "Episode: 1102\n",
+      "Episode reward: 114.0\n",
+      "Episode: 1103\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1104\n",
+      "Episode reward: 156.0\n",
+      "Episode: 1105\n",
+      "Episode reward: 189.0\n",
+      "Episode: 1106\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1107\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1108\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1109\n",
+      "Episode reward: 144.0\n",
+      "Episode: 1110\n",
+      "Episode reward: 185.0\n",
+      "Episode: 1111\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1112\n",
+      "Episode reward: 109.0\n",
+      "Episode: 1113\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1114\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1115\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1116\n",
+      "Episode reward: 195.0\n",
+      "Episode: 1117\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1118\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1119\n",
+      "Episode reward: 108.0\n",
+      "Episode: 1120\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1121\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1122\n",
+      "Episode reward: 118.0\n",
+      "Episode: 1123\n",
+      "Episode reward: 196.0\n",
+      "Episode: 1124\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1125\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1126\n",
+      "Episode reward: 199.0\n",
+      "Episode: 1127\n",
+      "Episode reward: 187.0\n",
+      "Episode: 1128\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1129\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1130\n",
+      "Episode reward: 166.0\n",
+      "Episode: 1131\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1132\n",
+      "Episode reward: 188.0\n",
+      "Episode: 1133\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1134\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1135\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1136\n",
+      "Episode reward: 108.0\n",
+      "Episode: 1137\n",
+      "Episode reward: 177.0\n",
+      "Episode: 1138\n",
+      "Episode reward: 102.0\n",
+      "Episode: 1139\n",
+      "Episode reward: 92.0\n",
+      "Episode: 1140\n",
+      "Episode reward: 174.0\n",
+      "Episode: 1141\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1142\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1143\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1144\n",
+      "Episode reward: 114.0\n",
+      "Episode: 1145\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1146\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1147\n",
+      "Episode reward: 158.0\n",
+      "Episode: 1148\n",
+      "Episode reward: 193.0\n",
+      "Episode: 1149\n",
+      "Episode reward: 103.0\n",
+      "Episode: 1150\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1151\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1152\n",
+      "Episode reward: 187.0\n",
+      "Episode: 1153\n",
+      "Episode reward: 108.0\n",
+      "Episode: 1154\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1155\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1156\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1157\n",
+      "Episode reward: 181.0\n",
+      "Episode: 1158\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1159\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1160\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1161\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1162\n",
+      "Episode reward: 135.0\n",
+      "Episode: 1163\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1164\n",
+      "Episode reward: 85.0\n",
+      "Episode: 1165\n",
+      "Episode reward: 101.0\n",
+      "Episode: 1166\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1167\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1168\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1169\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1170\n",
+      "Episode reward: 183.0\n",
+      "Episode: 1171\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1172\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1173\n",
+      "Episode reward: 126.0\n",
+      "Episode: 1174\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1175\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1176\n",
+      "Episode reward: 159.0\n",
+      "Episode: 1177\n",
+      "Episode reward: 184.0\n",
+      "Episode: 1178\n",
+      "Episode reward: 124.0\n",
+      "Episode: 1179\n",
+      "Episode reward: 174.0\n",
+      "Episode: 1180\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1181\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1182\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1183\n",
+      "Episode reward: 132.0\n",
+      "Episode: 1184\n",
+      "Episode reward: 190.0\n",
+      "Episode: 1185\n",
+      "Episode reward: 167.0\n",
+      "Episode: 1186\n",
+      "Episode reward: 126.0\n",
+      "Episode: 1187\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1188\n",
+      "Episode reward: 189.0\n",
+      "Episode: 1189\n",
+      "Episode reward: 169.0\n",
+      "Episode: 1190\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1191\n",
+      "Episode reward: 194.0\n",
+      "Episode: 1192\n",
+      "Episode reward: 125.0\n",
+      "Episode: 1193\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1194\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1195\n",
+      "Episode reward: 128.0\n",
+      "Episode: 1196\n",
+      "Episode reward: 165.0\n",
+      "Episode: 1197\n",
+      "Episode reward: 173.0\n",
+      "Episode: 1198\n",
+      "Episode reward: 197.0\n",
+      "Episode: 1199\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1200\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1201\n",
+      "Episode reward: 135.0\n",
+      "Episode: 1202\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1203\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1204\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1205\n",
+      "Episode reward: 122.0\n",
+      "Episode: 1206\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1207\n",
+      "Episode reward: 125.0\n",
+      "Episode: 1208\n",
+      "Episode reward: 132.0\n",
+      "Episode: 1209\n",
+      "Episode reward: 122.0\n",
+      "Episode: 1210\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1211\n",
+      "Episode reward: 159.0\n",
+      "Episode: 1212\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1213\n",
+      "Episode reward: 126.0\n",
+      "Episode: 1214\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1215\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1216\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1217\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1218\n",
+      "Episode reward: 158.0\n",
+      "Episode: 1219\n",
+      "Episode reward: 142.0\n",
+      "Episode: 1220\n",
+      "Episode reward: 165.0\n",
+      "Episode: 1221\n",
+      "Episode reward: 147.0\n",
+      "Episode: 1222\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1223\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1224\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1225\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1226\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1227\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1228\n",
+      "Episode reward: 148.0\n",
+      "Episode: 1229\n",
+      "Episode reward: 189.0\n",
+      "Episode: 1230\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1231\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1232\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1233\n",
+      "Episode reward: 179.0\n",
+      "Episode: 1234\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1235\n",
+      "Episode reward: 179.0\n",
+      "Episode: 1236\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1237\n",
+      "Episode reward: 188.0\n",
+      "Episode: 1238\n",
+      "Episode reward: 137.0\n",
+      "Episode: 1239\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1240\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1241\n",
+      "Episode reward: 159.0\n",
+      "Episode: 1242\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1243\n",
+      "Episode reward: 171.0\n",
+      "Episode: 1244\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1245\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1246\n",
+      "Episode reward: 122.0\n",
+      "Episode: 1247\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1248\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1249\n",
+      "Episode reward: 147.0\n",
+      "Episode: 1250\n",
+      "Episode reward: 173.0\n",
+      "Episode: 1251\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1252\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1253\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1254\n",
+      "Episode reward: 120.0\n",
+      "Episode: 1255\n",
+      "Episode reward: 160.0\n",
+      "Episode: 1256\n",
+      "Episode reward: 169.0\n",
+      "Episode: 1257\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1258\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1259\n",
+      "Episode reward: 195.0\n",
+      "Episode: 1260\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1261\n",
+      "Episode reward: 118.0\n",
+      "Episode: 1262\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1263\n",
+      "Episode reward: 127.0\n",
+      "Episode: 1264\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1265\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1266\n",
+      "Episode reward: 80.0\n",
+      "Episode: 1267\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1268\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1269\n",
+      "Episode reward: 186.0\n",
+      "Episode: 1270\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1271\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1272\n",
+      "Episode reward: 106.0\n",
+      "Episode: 1273\n",
+      "Episode reward: 193.0\n",
+      "Episode: 1274\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1275\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1276\n",
+      "Episode reward: 139.0\n",
+      "Episode: 1277\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1278\n",
+      "Episode reward: 184.0\n",
+      "Episode: 1279\n",
+      "Episode reward: 85.0\n",
+      "Episode: 1280\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1281\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1282\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1283\n",
+      "Episode reward: 142.0\n",
+      "Episode: 1284\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1285\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1286\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1287\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1288\n",
+      "Episode reward: 178.0\n",
+      "Episode: 1289\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1290\n",
+      "Episode reward: 137.0\n",
+      "Episode: 1291\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1292\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1293\n",
+      "Episode reward: 120.0\n",
+      "Episode: 1294\n",
+      "Episode reward: 185.0\n",
+      "Episode: 1295\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1296\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1297\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1298\n",
+      "Episode reward: 148.0\n",
+      "Episode: 1299\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1300\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1301\n",
+      "Episode reward: 177.0\n",
+      "Episode: 1302\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1303\n",
+      "Episode reward: 161.0\n",
+      "Episode: 1304\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1305\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1306\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1307\n",
+      "Episode reward: 167.0\n",
+      "Episode: 1308\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1309\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1310\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1311\n",
+      "Episode reward: 174.0\n",
+      "Episode: 1312\n",
+      "Episode reward: 111.0\n",
+      "Episode: 1313\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1314\n",
+      "Episode reward: 92.0\n",
+      "Episode: 1315\n",
+      "Episode reward: 104.0\n",
+      "Episode: 1316\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1317\n",
+      "Episode reward: 125.0\n",
+      "Episode: 1318\n",
+      "Episode reward: 173.0\n",
+      "Episode: 1319\n",
+      "Episode reward: 128.0\n",
+      "Episode: 1320\n",
+      "Episode reward: 156.0\n",
+      "Episode: 1321\n",
+      "Episode reward: 175.0\n",
+      "Episode: 1322\n",
+      "Episode reward: 106.0\n",
+      "Episode: 1323\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1324\n",
+      "Episode reward: 147.0\n",
+      "Episode: 1325\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1326\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1327\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1328\n",
+      "Episode reward: 169.0\n",
+      "Episode: 1329\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1330\n",
+      "Episode reward: 166.0\n",
+      "Episode: 1331\n",
+      "Episode reward: 122.0\n",
+      "Episode: 1332\n",
+      "Episode reward: 148.0\n",
+      "Episode: 1333\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1334\n",
+      "Episode reward: 170.0\n",
+      "Episode: 1335\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1336\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1337\n",
+      "Episode reward: 122.0\n",
+      "Episode: 1338\n",
+      "Episode reward: 142.0\n",
+      "Episode: 1339\n",
+      "Episode reward: 68.0\n",
+      "Episode: 1340\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1341\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1342\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1343\n",
+      "Episode reward: 109.0\n",
+      "Episode: 1344\n",
+      "Episode reward: 133.0\n",
+      "Episode: 1345\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1346\n",
+      "Episode reward: 147.0\n",
+      "Episode: 1347\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1348\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1349\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1350\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1351\n",
+      "Episode reward: 178.0\n",
+      "Episode: 1352\n",
+      "Episode reward: 122.0\n",
+      "Episode: 1353\n",
+      "Episode reward: 152.0\n",
+      "Episode: 1354\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1355\n",
+      "Episode reward: 105.0\n",
+      "Episode: 1356\n",
+      "Episode reward: 126.0\n",
+      "Episode: 1357\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1358\n",
+      "Episode reward: 116.0\n",
+      "Episode: 1359\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1360\n",
+      "Episode reward: 191.0\n",
+      "Episode: 1361\n",
+      "Episode reward: 144.0\n",
+      "Episode: 1362\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1363\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1364\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1365\n",
+      "Episode reward: 181.0\n",
+      "Episode: 1366\n",
+      "Episode reward: 158.0\n",
+      "Episode: 1367\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1368\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1369\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1370\n",
+      "Episode reward: 196.0\n",
+      "Episode: 1371\n",
+      "Episode reward: 156.0\n",
+      "Episode: 1372\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1373\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1374\n",
+      "Episode reward: 199.0\n",
+      "Episode: 1375\n",
+      "Episode reward: 134.0\n",
+      "Episode: 1376\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1377\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1378\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1379\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1380\n",
+      "Episode reward: 165.0\n",
+      "Episode: 1381\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1382\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1383\n",
+      "Episode reward: 128.0\n",
+      "Episode: 1384\n",
+      "Episode reward: 187.0\n",
+      "Episode: 1385\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1386\n",
+      "Episode reward: 108.0\n",
+      "Episode: 1387\n",
+      "Episode reward: 148.0\n",
+      "Episode: 1388\n",
+      "Episode reward: 147.0\n",
+      "Episode: 1389\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1390\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1391\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1392\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1393\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1394\n",
+      "Episode reward: 118.0\n",
+      "Episode: 1395\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1396\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1397\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1398\n",
+      "Episode reward: 147.0\n",
+      "Episode: 1399\n",
+      "Episode reward: 190.0\n",
+      "Episode: 1400\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1401\n",
+      "Episode reward: 183.0\n",
+      "Episode: 1402\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1403\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1404\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1405\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1406\n",
+      "Episode reward: 173.0\n",
+      "Episode: 1407\n",
+      "Episode reward: 108.0\n",
+      "Episode: 1408\n",
+      "Episode reward: 180.0\n",
+      "Episode: 1409\n",
+      "Episode reward: 189.0\n",
+      "Episode: 1410\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1411\n",
+      "Episode reward: 120.0\n",
+      "Episode: 1412\n",
+      "Episode reward: 167.0\n",
+      "Episode: 1413\n",
+      "Episode reward: 187.0\n",
+      "Episode: 1414\n",
+      "Episode reward: 110.0\n",
+      "Episode: 1415\n",
+      "Episode reward: 194.0\n",
+      "Episode: 1416\n",
+      "Episode reward: 126.0\n",
+      "Episode: 1417\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1418\n",
+      "Episode reward: 137.0\n",
+      "Episode: 1419\n",
+      "Episode reward: 111.0\n",
+      "Episode: 1420\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1421\n",
+      "Episode reward: 178.0\n",
+      "Episode: 1422\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1423\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1424\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1425\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1426\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1427\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1428\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1429\n",
+      "Episode reward: 148.0\n",
+      "Episode: 1430\n",
+      "Episode reward: 133.0\n",
+      "Episode: 1431\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1432\n",
+      "Episode reward: 166.0\n",
+      "Episode: 1433\n",
+      "Episode reward: 137.0\n",
+      "Episode: 1434\n",
+      "Episode reward: 167.0\n",
+      "Episode: 1435\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1436\n",
+      "Episode reward: 183.0\n",
+      "Episode: 1437\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1438\n",
+      "Episode reward: 192.0\n",
+      "Episode: 1439\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1440\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1441\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1442\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1443\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1444\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1445\n",
+      "Episode reward: 187.0\n",
+      "Episode: 1446\n",
+      "Episode reward: 170.0\n",
+      "Episode: 1447\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1448\n",
+      "Episode reward: 198.0\n",
+      "Episode: 1449\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1450\n",
+      "Episode reward: 175.0\n",
+      "Episode: 1451\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1452\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1453\n",
+      "Episode reward: 108.0\n",
+      "Episode: 1454\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1455\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1456\n",
+      "Episode reward: 192.0\n",
+      "Episode: 1457\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1458\n",
+      "Episode reward: 159.0\n",
+      "Episode: 1459\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1460\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1461\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1462\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1463\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1464\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1465\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1466\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1467\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1468\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1469\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1470\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1471\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1472\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1473\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1474\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1475\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1476\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1477\n",
+      "Episode reward: 125.0\n",
+      "Episode: 1478\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1479\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1480\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1481\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1482\n",
+      "Episode reward: 170.0\n",
+      "Episode: 1483\n",
+      "Episode reward: 133.0\n",
+      "Episode: 1484\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1485\n",
+      "Episode reward: 141.0\n",
+      "Episode: 1486\n",
+      "Episode reward: 183.0\n",
+      "Episode: 1487\n",
+      "Episode reward: 192.0\n",
+      "Episode: 1488\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1489\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1490\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1491\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1492\n",
+      "Episode reward: 137.0\n",
+      "Episode: 1493\n",
+      "Episode reward: 190.0\n",
+      "Episode: 1494\n",
+      "Episode reward: 188.0\n",
+      "Episode: 1495\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1496\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1497\n",
+      "Episode reward: 180.0\n",
+      "Episode: 1498\n",
+      "Episode reward: 137.0\n",
+      "Episode: 1499\n",
+      "Episode reward: 148.0\n",
+      "Episode: 1500\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1501\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1502\n",
+      "Episode reward: 139.0\n",
+      "Episode: 1503\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1504\n",
+      "Episode reward: 191.0\n",
+      "Episode: 1505\n",
+      "Episode reward: 144.0\n",
+      "Episode: 1506\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1507\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1508\n",
+      "Episode reward: 165.0\n",
+      "Episode: 1509\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1510\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1511\n",
+      "Episode reward: 169.0\n",
+      "Episode: 1512\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1513\n",
+      "Episode reward: 158.0\n",
+      "Episode: 1514\n",
+      "Episode reward: 125.0\n",
+      "Episode: 1515\n",
+      "Episode reward: 195.0\n",
+      "Episode: 1516\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1517\n",
+      "Episode reward: 174.0\n",
+      "Episode: 1518\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1519\n",
+      "Episode reward: 160.0\n",
+      "Episode: 1520\n",
+      "Episode reward: 144.0\n",
+      "Episode: 1521\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1522\n",
+      "Episode reward: 118.0\n",
+      "Episode: 1523\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1524\n",
+      "Episode reward: 169.0\n",
+      "Episode: 1525\n",
+      "Episode reward: 63.0\n",
+      "Episode: 1526\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1527\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1528\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1529\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1530\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1531\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1532\n",
+      "Episode reward: 105.0\n",
+      "Episode: 1533\n",
+      "Episode reward: 145.0\n",
+      "Episode: 1534\n",
+      "Episode reward: 197.0\n",
+      "Episode: 1535\n",
+      "Episode reward: 156.0\n",
+      "Episode: 1536\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1537\n",
+      "Episode reward: 125.0\n",
+      "Episode: 1538\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1539\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1540\n",
+      "Episode reward: 193.0\n",
+      "Episode: 1541\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1542\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1543\n",
+      "Episode reward: 135.0\n",
+      "Episode: 1544\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1545\n",
+      "Episode reward: 124.0\n",
+      "Episode: 1546\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1547\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1548\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1549\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1550\n",
+      "Episode reward: 159.0\n",
+      "Episode: 1551\n",
+      "Episode reward: 169.0\n",
+      "Episode: 1552\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1553\n",
+      "Episode reward: 196.0\n",
+      "Episode: 1554\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1555\n",
+      "Episode reward: 148.0\n",
+      "Episode: 1556\n",
+      "Episode reward: 191.0\n",
+      "Episode: 1557\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1558\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1559\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1560\n",
+      "Episode reward: 174.0\n",
+      "Episode: 1561\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1562\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1563\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1564\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1565\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1566\n",
+      "Episode reward: 196.0\n",
+      "Episode: 1567\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1568\n",
+      "Episode reward: 142.0\n",
+      "Episode: 1569\n",
+      "Episode reward: 145.0\n",
+      "Episode: 1570\n",
+      "Episode reward: 145.0\n",
+      "Episode: 1571\n",
+      "Episode reward: 147.0\n",
+      "Episode: 1572\n",
+      "Episode reward: 120.0\n",
+      "Episode: 1573\n",
+      "Episode reward: 156.0\n",
+      "Episode: 1574\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1575\n",
+      "Episode reward: 135.0\n",
+      "Episode: 1576\n",
+      "Episode reward: 182.0\n",
+      "Episode: 1577\n",
+      "Episode reward: 127.0\n",
+      "Episode: 1578\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1579\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1580\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1581\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1582\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1583\n",
+      "Episode reward: 125.0\n",
+      "Episode: 1584\n",
+      "Episode reward: 199.0\n",
+      "Episode: 1585\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1586\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1587\n",
+      "Episode reward: 95.0\n",
+      "Episode: 1588\n",
+      "Episode reward: 134.0\n",
+      "Episode: 1589\n",
+      "Episode reward: 120.0\n",
+      "Episode: 1590\n",
+      "Episode reward: 184.0\n",
+      "Episode: 1591\n",
+      "Episode reward: 137.0\n",
+      "Episode: 1592\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1593\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1594\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1595\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1596\n",
+      "Episode reward: 101.0\n",
+      "Episode: 1597\n",
+      "Episode reward: 134.0\n",
+      "Episode: 1598\n",
+      "Episode reward: 134.0\n",
+      "Episode: 1599\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1600\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1601\n",
+      "Episode reward: 139.0\n",
+      "Episode: 1602\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1603\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1604\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1605\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1606\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1607\n",
+      "Episode reward: 185.0\n",
+      "Episode: 1608\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1609\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1610\n",
+      "Episode reward: 198.0\n",
+      "Episode: 1611\n",
+      "Episode reward: 107.0\n",
+      "Episode: 1612\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1613\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1614\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1615\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1616\n",
+      "Episode reward: 135.0\n",
+      "Episode: 1617\n",
+      "Episode reward: 126.0\n",
+      "Episode: 1618\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1619\n",
+      "Episode reward: 174.0\n",
+      "Episode: 1620\n",
+      "Episode reward: 173.0\n",
+      "Episode: 1621\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1622\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1623\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1624\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1625\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1626\n",
+      "Episode reward: 116.0\n",
+      "Episode: 1627\n",
+      "Episode reward: 164.0\n",
+      "Episode: 1628\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1629\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1630\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1631\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1632\n",
+      "Episode reward: 116.0\n",
+      "Episode: 1633\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1634\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1635\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1636\n",
+      "Episode reward: 175.0\n",
+      "Episode: 1637\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1638\n",
+      "Episode reward: 105.0\n",
+      "Episode: 1639\n",
+      "Episode reward: 180.0\n",
+      "Episode: 1640\n",
+      "Episode reward: 183.0\n",
+      "Episode: 1641\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1642\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1643\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1644\n",
+      "Episode reward: 118.0\n",
+      "Episode: 1645\n",
+      "Episode reward: 187.0\n",
+      "Episode: 1646\n",
+      "Episode reward: 114.0\n",
+      "Episode: 1647\n",
+      "Episode reward: 197.0\n",
+      "Episode: 1648\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1649\n",
+      "Episode reward: 156.0\n",
+      "Episode: 1650\n",
+      "Episode reward: 172.0\n",
+      "Episode: 1651\n",
+      "Episode reward: 114.0\n",
+      "Episode: 1652\n",
+      "Episode reward: 191.0\n",
+      "Episode: 1653\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1654\n",
+      "Episode reward: 111.0\n",
+      "Episode: 1655\n",
+      "Episode reward: 182.0\n",
+      "Episode: 1656\n",
+      "Episode reward: 152.0\n",
+      "Episode: 1657\n",
+      "Episode reward: 191.0\n",
+      "Episode: 1658\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1659\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1660\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1661\n",
+      "Episode reward: 189.0\n",
+      "Episode: 1662\n",
+      "Episode reward: 166.0\n",
+      "Episode: 1663\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1664\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1665\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1666\n",
+      "Episode reward: 125.0\n",
+      "Episode: 1667\n",
+      "Episode reward: 152.0\n",
+      "Episode: 1668\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1669\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1670\n",
+      "Episode reward: 183.0\n",
+      "Episode: 1671\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1672\n",
+      "Episode reward: 198.0\n",
+      "Episode: 1673\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1674\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1675\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1676\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1677\n",
+      "Episode reward: 184.0\n",
+      "Episode: 1678\n",
+      "Episode reward: 167.0\n",
+      "Episode: 1679\n",
+      "Episode reward: 175.0\n",
+      "Episode: 1680\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1681\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1682\n",
+      "Episode reward: 160.0\n",
+      "Episode: 1683\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1684\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1685\n",
+      "Episode reward: 158.0\n",
+      "Episode: 1686\n",
+      "Episode reward: 114.0\n",
+      "Episode: 1687\n",
+      "Episode reward: 160.0\n",
+      "Episode: 1688\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1689\n",
+      "Episode reward: 175.0\n",
+      "Episode: 1690\n",
+      "Episode reward: 178.0\n",
+      "Episode: 1691\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1692\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1693\n",
+      "Episode reward: 116.0\n",
+      "Episode: 1694\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1695\n",
+      "Episode reward: 118.0\n",
+      "Episode: 1696\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1697\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1698\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1699\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1700\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1701\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1702\n",
+      "Episode reward: 159.0\n",
+      "Episode: 1703\n",
+      "Episode reward: 122.0\n",
+      "Episode: 1704\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1705\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1706\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1707\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1708\n",
+      "Episode reward: 189.0\n",
+      "Episode: 1709\n",
+      "Episode reward: 108.0\n",
+      "Episode: 1710\n",
+      "Episode reward: 118.0\n",
+      "Episode: 1711\n",
+      "Episode reward: 110.0\n",
+      "Episode: 1712\n",
+      "Episode reward: 111.0\n",
+      "Episode: 1713\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1714\n",
+      "Episode reward: 139.0\n",
+      "Episode: 1715\n",
+      "Episode reward: 185.0\n",
+      "Episode: 1716\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1717\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1718\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1719\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1720\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1721\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1722\n",
+      "Episode reward: 118.0\n",
+      "Episode: 1723\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1724\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1725\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1726\n",
+      "Episode reward: 161.0\n",
+      "Episode: 1727\n",
+      "Episode reward: 148.0\n",
+      "Episode: 1728\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1729\n",
+      "Episode reward: 142.0\n",
+      "Episode: 1730\n",
+      "Episode reward: 125.0\n",
+      "Episode: 1731\n",
+      "Episode reward: 99.0\n",
+      "Episode: 1732\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1733\n",
+      "Episode reward: 166.0\n",
+      "Episode: 1734\n",
+      "Episode reward: 187.0\n",
+      "Episode: 1735\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1736\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1737\n",
+      "Episode reward: 169.0\n",
+      "Episode: 1738\n",
+      "Episode reward: 142.0\n",
+      "Episode: 1739\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1740\n",
+      "Episode reward: 141.0\n",
+      "Episode: 1741\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1742\n",
+      "Episode reward: 184.0\n",
+      "Episode: 1743\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1744\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1745\n",
+      "Episode reward: 126.0\n",
+      "Episode: 1746\n",
+      "Episode reward: 165.0\n",
+      "Episode: 1747\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1748\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1749\n",
+      "Episode reward: 192.0\n",
+      "Episode: 1750\n",
+      "Episode reward: 162.0\n",
+      "Episode: 1751\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1752\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1753\n",
+      "Episode reward: 152.0\n",
+      "Episode: 1754\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1755\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1756\n",
+      "Episode reward: 195.0\n",
+      "Episode: 1757\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1758\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1759\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1760\n",
+      "Episode reward: 158.0\n",
+      "Episode: 1761\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1762\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1763\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1764\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1765\n",
+      "Episode reward: 158.0\n",
+      "Episode: 1766\n",
+      "Episode reward: 124.0\n",
+      "Episode: 1767\n",
+      "Episode reward: 158.0\n",
+      "Episode: 1768\n",
+      "Episode reward: 132.0\n",
+      "Episode: 1769\n",
+      "Episode reward: 199.0\n",
+      "Episode: 1770\n",
+      "Episode reward: 195.0\n",
+      "Episode: 1771\n",
+      "Episode reward: 111.0\n",
+      "Episode: 1772\n",
+      "Episode reward: 177.0\n",
+      "Episode: 1773\n",
+      "Episode reward: 114.0\n",
+      "Episode: 1774\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1775\n",
+      "Episode reward: 144.0\n",
+      "Episode: 1776\n",
+      "Episode reward: 109.0\n",
+      "Episode: 1777\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1778\n",
+      "Episode reward: 127.0\n",
+      "Episode: 1779\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1780\n",
+      "Episode reward: 158.0\n",
+      "Episode: 1781\n",
+      "Episode reward: 114.0\n",
+      "Episode: 1782\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1783\n",
+      "Episode reward: 160.0\n",
+      "Episode: 1784\n",
+      "Episode reward: 124.0\n",
+      "Episode: 1785\n",
+      "Episode reward: 175.0\n",
+      "Episode: 1786\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1787\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1788\n",
+      "Episode reward: 175.0\n",
+      "Episode: 1789\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1790\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1791\n",
+      "Episode reward: 111.0\n",
+      "Episode: 1792\n",
+      "Episode reward: 137.0\n",
+      "Episode: 1793\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1794\n",
+      "Episode reward: 174.0\n",
+      "Episode: 1795\n",
+      "Episode reward: 108.0\n",
+      "Episode: 1796\n",
+      "Episode reward: 158.0\n",
+      "Episode: 1797\n",
+      "Episode reward: 145.0\n",
+      "Episode: 1798\n",
+      "Episode reward: 106.0\n",
+      "Episode: 1799\n",
+      "Episode reward: 125.0\n",
+      "Episode: 1800\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1801\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1802\n",
+      "Episode reward: 181.0\n",
+      "Episode: 1803\n",
+      "Episode reward: 161.0\n",
+      "Episode: 1804\n",
+      "Episode reward: 165.0\n",
+      "Episode: 1805\n",
+      "Episode reward: 173.0\n",
+      "Episode: 1806\n",
+      "Episode reward: 116.0\n",
+      "Episode: 1807\n",
+      "Episode reward: 175.0\n",
+      "Episode: 1808\n",
+      "Episode reward: 128.0\n",
+      "Episode: 1809\n",
+      "Episode reward: 132.0\n",
+      "Episode: 1810\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1811\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1812\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1813\n",
+      "Episode reward: 111.0\n",
+      "Episode: 1814\n",
+      "Episode reward: 128.0\n",
+      "Episode: 1815\n",
+      "Episode reward: 126.0\n",
+      "Episode: 1816\n",
+      "Episode reward: 161.0\n",
+      "Episode: 1817\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1818\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1819\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1820\n",
+      "Episode reward: 120.0\n",
+      "Episode: 1821\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1822\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1823\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1824\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1825\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1826\n",
+      "Episode reward: 108.0\n",
+      "Episode: 1827\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1828\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1829\n",
+      "Episode reward: 134.0\n",
+      "Episode: 1830\n",
+      "Episode reward: 127.0\n",
+      "Episode: 1831\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1832\n",
+      "Episode reward: 191.0\n",
+      "Episode: 1833\n",
+      "Episode reward: 142.0\n",
+      "Episode: 1834\n",
+      "Episode reward: 174.0\n",
+      "Episode: 1835\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1836\n",
+      "Episode reward: 129.0\n",
+      "Episode: 1837\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1838\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1839\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1840\n",
+      "Episode reward: 172.0\n",
+      "Episode: 1841\n",
+      "Episode reward: 172.0\n",
+      "Episode: 1842\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1843\n",
+      "Episode reward: 192.0\n",
+      "Episode: 1844\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1845\n",
+      "Episode reward: 134.0\n",
+      "Episode: 1846\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1847\n",
+      "Episode reward: 111.0\n",
+      "Episode: 1848\n",
+      "Episode reward: 126.0\n",
+      "Episode: 1849\n",
+      "Episode reward: 160.0\n",
+      "Episode: 1850\n",
+      "Episode reward: 118.0\n",
+      "Episode: 1851\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1852\n",
+      "Episode reward: 182.0\n",
+      "Episode: 1853\n",
+      "Episode reward: 111.0\n",
+      "Episode: 1854\n",
+      "Episode reward: 173.0\n",
+      "Episode: 1855\n",
+      "Episode reward: 144.0\n",
+      "Episode: 1856\n",
+      "Episode reward: 120.0\n",
+      "Episode: 1857\n",
+      "Episode reward: 169.0\n",
+      "Episode: 1858\n",
+      "Episode reward: 111.0\n",
+      "Episode: 1859\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1860\n",
+      "Episode reward: 83.0\n",
+      "Episode: 1861\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1862\n",
+      "Episode reward: 142.0\n",
+      "Episode: 1863\n",
+      "Episode reward: 108.0\n",
+      "Episode: 1864\n",
+      "Episode reward: 114.0\n",
+      "Episode: 1865\n",
+      "Episode reward: 140.0\n",
+      "Episode: 1866\n",
+      "Episode reward: 187.0\n",
+      "Episode: 1867\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1868\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1869\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1870\n",
+      "Episode reward: 134.0\n",
+      "Episode: 1871\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1872\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1873\n",
+      "Episode reward: 199.0\n",
+      "Episode: 1874\n",
+      "Episode reward: 105.0\n",
+      "Episode: 1875\n",
+      "Episode reward: 147.0\n",
+      "Episode: 1876\n",
+      "Episode reward: 152.0\n",
+      "Episode: 1877\n",
+      "Episode reward: 117.0\n",
+      "Episode: 1878\n",
+      "Episode reward: 89.0\n",
+      "Episode: 1879\n",
+      "Episode reward: 138.0\n",
+      "Episode: 1880\n",
+      "Episode reward: 144.0\n",
+      "Episode: 1881\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1882\n",
+      "Episode reward: 128.0\n",
+      "Episode: 1883\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1884\n",
+      "Episode reward: 167.0\n",
+      "Episode: 1885\n",
+      "Episode reward: 121.0\n",
+      "Episode: 1886\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1887\n",
+      "Episode reward: 176.0\n",
+      "Episode: 1888\n",
+      "Episode reward: 175.0\n",
+      "Episode: 1889\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1890\n",
+      "Episode reward: 122.0\n",
+      "Episode: 1891\n",
+      "Episode reward: 116.0\n",
+      "Episode: 1892\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1893\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1894\n",
+      "Episode reward: 192.0\n",
+      "Episode: 1895\n",
+      "Episode reward: 190.0\n",
+      "Episode: 1896\n",
+      "Episode reward: 193.0\n",
+      "Episode: 1897\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1898\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1899\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1900\n",
+      "Episode reward: 184.0\n",
+      "Episode: 1901\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1902\n",
+      "Episode reward: 156.0\n",
+      "Episode: 1903\n",
+      "Episode reward: 131.0\n",
+      "Episode: 1904\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1905\n",
+      "Episode reward: 144.0\n",
+      "Episode: 1906\n",
+      "Episode reward: 173.0\n",
+      "Episode: 1907\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1908\n",
+      "Episode reward: 189.0\n",
+      "Episode: 1909\n",
+      "Episode reward: 136.0\n",
+      "Episode: 1910\n",
+      "Episode reward: 157.0\n",
+      "Episode: 1911\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1912\n",
+      "Episode reward: 112.0\n",
+      "Episode: 1913\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1914\n",
+      "Episode reward: 156.0\n",
+      "Episode: 1915\n",
+      "Episode reward: 180.0\n",
+      "Episode: 1916\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1917\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1918\n",
+      "Episode reward: 139.0\n",
+      "Episode: 1919\n",
+      "Episode reward: 144.0\n",
+      "Episode: 1920\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1921\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1922\n",
+      "Episode reward: 152.0\n",
+      "Episode: 1923\n",
+      "Episode reward: 166.0\n",
+      "Episode: 1924\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1925\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1926\n",
+      "Episode reward: 181.0\n",
+      "Episode: 1927\n",
+      "Episode reward: 106.0\n",
+      "Episode: 1928\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1929\n",
+      "Episode reward: 116.0\n",
+      "Episode: 1930\n",
+      "Episode reward: 191.0\n",
+      "Episode: 1931\n",
+      "Episode reward: 95.0\n",
+      "Episode: 1932\n",
+      "Episode reward: 105.0\n",
+      "Episode: 1933\n",
+      "Episode reward: 156.0\n",
+      "Episode: 1934\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1935\n",
+      "Episode reward: 155.0\n",
+      "Episode: 1936\n",
+      "Episode reward: 182.0\n",
+      "Episode: 1937\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1938\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1939\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1940\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1941\n",
+      "Episode reward: 66.0\n",
+      "Episode: 1942\n",
+      "Episode reward: 115.0\n",
+      "Episode: 1943\n",
+      "Episode reward: 192.0\n",
+      "Episode: 1944\n",
+      "Episode reward: 146.0\n",
+      "Episode: 1945\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1946\n",
+      "Episode reward: 135.0\n",
+      "Episode: 1947\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1948\n",
+      "Episode reward: 126.0\n",
+      "Episode: 1949\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1950\n",
+      "Episode reward: 191.0\n",
+      "Episode: 1951\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1952\n",
+      "Episode reward: 154.0\n",
+      "Episode: 1953\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1954\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1955\n",
+      "Episode reward: 118.0\n",
+      "Episode: 1956\n",
+      "Episode reward: 160.0\n",
+      "Episode: 1957\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1958\n",
+      "Episode reward: 168.0\n",
+      "Episode: 1959\n",
+      "Episode reward: 148.0\n",
+      "Episode: 1960\n",
+      "Episode reward: 130.0\n",
+      "Episode: 1961\n",
+      "Episode reward: 152.0\n",
+      "Episode: 1962\n",
+      "Episode reward: 141.0\n",
+      "Episode: 1963\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1964\n",
+      "Episode reward: 119.0\n",
+      "Episode: 1965\n",
+      "Episode reward: 107.0\n",
+      "Episode: 1966\n",
+      "Episode reward: 156.0\n",
+      "Episode: 1967\n",
+      "Episode reward: 193.0\n",
+      "Episode: 1968\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1969\n",
+      "Episode reward: 164.0\n",
+      "Episode: 1970\n",
+      "Episode reward: 151.0\n",
+      "Episode: 1971\n",
+      "Episode reward: 109.0\n",
+      "Episode: 1972\n",
+      "Episode reward: 110.0\n",
+      "Episode: 1973\n",
+      "Episode reward: 198.0\n",
+      "Episode: 1974\n",
+      "Episode reward: 145.0\n",
+      "Episode: 1975\n",
+      "Episode reward: 139.0\n",
+      "Episode: 1976\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1977\n",
+      "Episode reward: 141.0\n",
+      "Episode: 1978\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1979\n",
+      "Episode reward: 143.0\n",
+      "Episode: 1980\n",
+      "Episode reward: 153.0\n",
+      "Episode: 1981\n",
+      "Episode reward: 124.0\n",
+      "Episode: 1982\n",
+      "Episode reward: 193.0\n",
+      "Episode: 1983\n",
+      "Episode reward: 148.0\n",
+      "Episode: 1984\n",
+      "Episode reward: 123.0\n",
+      "Episode: 1985\n",
+      "Episode reward: 150.0\n",
+      "Episode: 1986\n",
+      "Episode reward: 180.0\n",
+      "Episode: 1987\n",
+      "Episode reward: 196.0\n",
+      "Episode: 1988\n",
+      "Episode reward: 113.0\n",
+      "Episode: 1989\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1990\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1991\n",
+      "Episode reward: 183.0\n",
+      "Episode: 1992\n",
+      "Episode reward: 179.0\n",
+      "Episode: 1993\n",
+      "Episode reward: 141.0\n",
+      "Episode: 1994\n",
+      "Episode reward: 149.0\n",
+      "Episode: 1995\n",
+      "Episode reward: 163.0\n",
+      "Episode: 1996\n",
+      "Episode reward: 120.0\n",
+      "Episode: 1997\n",
+      "Episode reward: 200.0\n",
+      "Episode: 1998\n",
+      "Episode reward: 103.0\n",
+      "Episode: 1999\n",
+      "Episode reward: 197.0\n",
+      "Learned Q-table:\n",
+      "tensor([[[[[ 7.5282e-03, -5.0158e-03],\n",
+      "           [ 1.8066e-04, -1.1957e-02],\n",
+      "           [ 1.7809e-02,  1.9935e-02],\n",
+      "           [ 1.2077e-02,  1.2135e-03]],\n",
+      "\n",
+      "          [[ 1.1068e-02,  1.3167e-02],\n",
+      "           [-3.6065e-03, -2.3091e-02],\n",
+      "           [-5.3917e-03, -3.6806e-03],\n",
+      "           [-2.0194e-02,  1.4977e-02]],\n",
+      "\n",
+      "          [[-3.9683e-03, -1.1291e-03],\n",
+      "           [ 1.6093e-03,  1.7743e-02],\n",
+      "           [ 3.1133e-03, -1.3254e-02],\n",
+      "           [-2.2300e-03,  1.5660e-02]],\n",
+      "\n",
+      "          [[-9.9277e-03, -3.0461e-03],\n",
+      "           [ 1.5977e-02, -9.5863e-03],\n",
+      "           [ 9.4414e-03, -4.6137e-03],\n",
+      "           [-6.0294e-03,  3.8514e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 3.2934e-03, -2.7038e-03],\n",
+      "           [ 1.1472e-03, -7.2562e-03],\n",
+      "           [ 7.2273e-03, -7.4928e-03],\n",
+      "           [-3.5465e-03, -1.3511e-02]],\n",
+      "\n",
+      "          [[-1.1883e-02, -3.6573e-03],\n",
+      "           [-9.0871e-03, -1.4479e-02],\n",
+      "           [ 1.2498e-04,  2.4612e-03],\n",
+      "           [-1.4339e-02, -3.3635e-03]],\n",
+      "\n",
+      "          [[ 1.4458e-02,  2.2707e-02],\n",
+      "           [ 1.1106e-03, -1.9436e-02],\n",
+      "           [ 1.7882e-02,  5.9812e-03],\n",
+      "           [ 2.0743e-02, -8.5244e-03]],\n",
+      "\n",
+      "          [[-3.0426e-03, -1.1320e-03],\n",
+      "           [-1.7067e-02,  5.2065e-03],\n",
+      "           [ 5.5506e-03, -2.2826e-03],\n",
+      "           [-3.7662e-03, -9.6092e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 4.2712e-03, -1.7112e-02],\n",
+      "           [-7.6057e-04,  1.5980e-02],\n",
+      "           [ 2.6001e-03,  4.0400e-03],\n",
+      "           [-1.3805e-03, -1.1307e-02]],\n",
+      "\n",
+      "          [[ 5.0641e-03, -1.3241e-02],\n",
+      "           [ 1.7783e-03,  8.7516e-03],\n",
+      "           [-9.8789e-03, -9.4022e-03],\n",
+      "           [ 1.1799e-02, -8.4989e-03]],\n",
+      "\n",
+      "          [[-8.7781e-03, -1.9099e-02],\n",
+      "           [-2.1311e-03,  1.3072e-02],\n",
+      "           [-9.6554e-03, -9.8139e-03],\n",
+      "           [ 1.0881e-02, -7.0734e-03]],\n",
+      "\n",
+      "          [[-1.2324e-02, -1.8049e-02],\n",
+      "           [ 2.6147e-03,  7.3541e-03],\n",
+      "           [-1.2201e-02, -2.3215e-02],\n",
+      "           [ 1.8518e-02,  1.3268e-02]]],\n",
+      "\n",
+      "\n",
+      "         [[[-3.7732e-04,  6.0749e-03],\n",
+      "           [ 9.5228e-03,  1.7512e-03],\n",
+      "           [-6.6485e-03, -6.8077e-04],\n",
+      "           [-1.4016e-02,  4.3504e-03]],\n",
+      "\n",
+      "          [[-1.1826e-02,  3.7322e-03],\n",
+      "           [-1.4521e-02, -5.1530e-03],\n",
+      "           [-8.8820e-04,  5.4113e-03],\n",
+      "           [ 1.0400e-02, -1.4762e-03]],\n",
+      "\n",
+      "          [[ 1.0607e-02,  6.7451e-03],\n",
+      "           [ 1.0964e-02,  2.8386e-03],\n",
+      "           [ 5.6247e-03,  1.0890e-02],\n",
+      "           [ 4.6335e-03, -3.7274e-03]],\n",
+      "\n",
+      "          [[-9.0740e-03,  6.4783e-03],\n",
+      "           [-1.2299e-03, -5.6999e-03],\n",
+      "           [-1.6512e-02,  1.0905e-02],\n",
+      "           [ 2.4370e-03, -1.5498e-02]]]],\n",
+      "\n",
+      "\n",
+      "\n",
+      "        [[[[ 8.5643e-03,  1.1855e-02],\n",
+      "           [-4.0238e-03, -1.9470e-02],\n",
+      "           [ 8.6876e-03, -1.3815e-03],\n",
+      "           [ 1.3968e-02, -1.1944e-02]],\n",
+      "\n",
+      "          [[-4.9274e-03, -3.5420e-03],\n",
+      "           [ 1.1639e-02, -3.9007e-03],\n",
+      "           [-6.9517e-03, -9.7298e-03],\n",
+      "           [-9.1168e-03,  1.6739e-03]],\n",
+      "\n",
+      "          [[ 9.1662e-03,  3.5723e-04],\n",
+      "           [-7.4430e-04, -2.2149e-03],\n",
+      "           [-3.0224e-03,  1.0551e-02],\n",
+      "           [-5.0447e-03, -1.4064e-02]],\n",
+      "\n",
+      "          [[-3.5565e-03, -7.7704e-04],\n",
+      "           [-7.0354e-03,  8.5815e-03],\n",
+      "           [-8.7700e-03, -1.2051e-02],\n",
+      "           [ 8.8658e-03, -2.5521e-02]]],\n",
+      "\n",
+      "\n",
+      "         [[[-1.2974e-03, -8.8698e-03],\n",
+      "           [-3.6456e-03,  1.5029e-02],\n",
+      "           [-8.8230e-03,  9.4439e-03],\n",
+      "           [ 2.5990e-03,  1.5128e-02]],\n",
+      "\n",
+      "          [[ 5.6906e-03,  4.9925e-03],\n",
+      "           [ 8.7974e+00,  6.8648e+00],\n",
+      "           [ 7.6102e+00,  9.6593e+00],\n",
+      "           [ 7.9246e-03, -1.4221e-02]],\n",
+      "\n",
+      "          [[ 1.8949e-04, -9.8628e-03],\n",
+      "           [ 9.7750e+00,  2.2845e+00],\n",
+      "           [ 4.7479e+00,  9.7569e+00],\n",
+      "           [-6.5158e-03,  1.5208e-03]],\n",
+      "\n",
+      "          [[ 2.0201e-04, -2.1695e-02],\n",
+      "           [-2.6849e-04, -1.5498e-02],\n",
+      "           [-1.1011e-02,  1.1365e-03],\n",
+      "           [-7.1765e-03,  1.2924e-02]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 9.7294e-03,  2.4206e-02],\n",
+      "           [-1.2685e-02, -1.0336e-03],\n",
+      "           [ 8.7326e-03, -2.7275e-02],\n",
+      "           [-2.1335e-02,  4.4139e-03]],\n",
+      "\n",
+      "          [[-1.5349e-02, -5.3750e-03],\n",
+      "           [ 9.6216e+00,  6.4663e+00],\n",
+      "           [ 9.2758e+00,  1.1250e+00],\n",
+      "           [-2.8471e-03,  4.4822e-03]],\n",
+      "\n",
+      "          [[-3.5157e-03,  5.6956e-03],\n",
+      "           [ 8.0273e+00,  9.5288e+00],\n",
+      "           [ 6.7466e+00,  9.7377e+00],\n",
+      "           [-1.5232e-02, -1.2576e-02]],\n",
+      "\n",
+      "          [[-1.5377e-02,  4.6587e-03],\n",
+      "           [-1.2515e-02, -8.6466e-03],\n",
+      "           [-3.5109e-03,  1.8327e-02],\n",
+      "           [ 5.5513e-03, -7.4427e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[-1.2366e-02, -1.8556e-02],\n",
+      "           [-7.6673e-03, -6.9909e-03],\n",
+      "           [-1.7524e-02, -1.2405e-02],\n",
+      "           [-2.0454e-04, -1.0710e-02]],\n",
+      "\n",
+      "          [[-7.5223e-04,  8.1430e-03],\n",
+      "           [-4.2421e-03, -2.4979e-03],\n",
+      "           [-2.6400e-03, -7.0397e-03],\n",
+      "           [ 7.2088e-03,  1.2598e-02]],\n",
+      "\n",
+      "          [[ 5.4684e-03, -2.1672e-03],\n",
+      "           [ 7.8071e-04,  8.7978e-04],\n",
+      "           [-4.5486e-03, -1.1562e-02],\n",
+      "           [-1.2760e-02, -4.7683e-03]],\n",
+      "\n",
+      "          [[-5.0534e-03,  2.6844e-02],\n",
+      "           [ 6.5477e-03, -2.0174e-03],\n",
+      "           [-2.3228e-03, -1.9718e-03],\n",
+      "           [-1.0325e-02,  1.4051e-02]]]],\n",
+      "\n",
+      "\n",
+      "\n",
+      "        [[[[ 1.2235e-02,  3.8194e-03],\n",
+      "           [-1.2756e-02,  2.0048e-03],\n",
+      "           [-1.0588e-02, -6.2585e-03],\n",
+      "           [-1.1622e-02,  7.8146e-03]],\n",
+      "\n",
+      "          [[-1.3627e-02,  4.8715e-03],\n",
+      "           [-2.6081e-03, -1.2547e-02],\n",
+      "           [ 2.5712e-02,  6.9268e-04],\n",
+      "           [ 5.7297e-03,  1.6908e-02]],\n",
+      "\n",
+      "          [[-6.0016e-03,  3.8388e-03],\n",
+      "           [-4.2425e-03,  5.6236e-03],\n",
+      "           [-2.2077e-03,  8.3500e-03],\n",
+      "           [-6.5338e-03, -5.6204e-03]],\n",
+      "\n",
+      "          [[ 3.6226e-03,  6.5813e-03],\n",
+      "           [ 8.4147e-03, -4.6316e-03],\n",
+      "           [ 1.1701e-03, -3.8150e-03],\n",
+      "           [ 3.4847e-03, -1.6027e-02]]],\n",
+      "\n",
+      "\n",
+      "         [[[-1.2393e-03, -8.3570e-03],\n",
+      "           [ 4.4231e-03, -6.3459e-03],\n",
+      "           [-6.1139e-04,  5.3770e-03],\n",
+      "           [ 5.4646e-03,  3.0140e-03]],\n",
+      "\n",
+      "          [[-2.6489e-04,  3.1112e-04],\n",
+      "           [ 9.8092e+00,  7.6335e+00],\n",
+      "           [ 8.1881e+00,  9.7766e+00],\n",
+      "           [-7.3893e-03,  1.7081e-04]],\n",
+      "\n",
+      "          [[ 1.2995e-02,  4.9129e-03],\n",
+      "           [ 9.8921e+00,  2.2475e+00],\n",
+      "           [ 6.9135e+00,  9.8888e+00],\n",
+      "           [-4.6639e-03,  9.7619e-03]],\n",
+      "\n",
+      "          [[-3.7702e-03, -1.7440e-02],\n",
+      "           [ 2.7172e-03, -7.0420e-04],\n",
+      "           [ 1.0267e-02,  6.5922e-03],\n",
+      "           [-4.3092e-03, -6.1909e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 8.2225e-03,  2.0020e-02],\n",
+      "           [ 1.3149e-02, -1.5706e-02],\n",
+      "           [ 1.4529e-02,  2.0051e-02],\n",
+      "           [-3.3207e-03,  2.9005e-03]],\n",
+      "\n",
+      "          [[ 1.6814e-02, -1.1611e-02],\n",
+      "           [ 9.8325e+00,  5.7229e+00],\n",
+      "           [ 9.8201e+00,  2.8158e+00],\n",
+      "           [-1.9186e-03, -3.3058e-03]],\n",
+      "\n",
+      "          [[ 1.8627e-02, -4.3313e-03],\n",
+      "           [ 7.2040e+00,  9.8860e+00],\n",
+      "           [ 1.7174e+00,  9.8843e+00],\n",
+      "           [-4.0002e-03, -4.0428e-03]],\n",
+      "\n",
+      "          [[-5.8534e-03, -8.5589e-03],\n",
+      "           [ 1.8699e-02,  1.4041e-02],\n",
+      "           [-4.5634e-03, -1.2667e-02],\n",
+      "           [ 2.9807e-03, -1.0735e-02]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 2.3185e-03, -1.1426e-02],\n",
+      "           [ 1.0432e-02,  8.7466e-03],\n",
+      "           [-2.0517e-02, -6.4246e-03],\n",
+      "           [ 1.6325e-02, -2.2658e-03]],\n",
+      "\n",
+      "          [[-1.3443e-03, -7.4122e-03],\n",
+      "           [ 2.6692e-03,  1.6455e-03],\n",
+      "           [-1.2263e-02, -3.9807e-04],\n",
+      "           [-1.1106e-02, -1.2117e-03]],\n",
+      "\n",
+      "          [[ 1.2348e-02, -4.9474e-03],\n",
+      "           [-5.5995e-03, -2.2941e-03],\n",
+      "           [-8.0381e-03,  8.6414e-03],\n",
+      "           [-1.2672e-02, -8.6333e-03]],\n",
+      "\n",
+      "          [[ 2.2708e-03,  1.5464e-02],\n",
+      "           [-6.9604e-07,  9.0496e-03],\n",
+      "           [ 4.9665e-03, -1.3779e-04],\n",
+      "           [ 2.5279e-03, -9.1125e-03]]]],\n",
+      "\n",
+      "\n",
+      "\n",
+      "        [[[[-4.1377e-03, -5.1594e-04],\n",
+      "           [-3.2999e-03, -3.2629e-03],\n",
+      "           [ 1.5395e-02, -1.9373e-03],\n",
+      "           [-2.1155e-02, -2.2015e-03]],\n",
+      "\n",
+      "          [[-4.7272e-03,  4.2632e-03],\n",
+      "           [-7.7089e-03, -4.4246e-03],\n",
+      "           [ 1.1186e-02,  1.2331e-02],\n",
+      "           [-8.5431e-03,  9.3936e-03]],\n",
+      "\n",
+      "          [[ 2.0678e-02, -1.5143e-03],\n",
+      "           [ 1.1107e-03, -1.5056e-03],\n",
+      "           [-1.5841e-02, -1.0837e-03],\n",
+      "           [ 1.2758e-03, -1.2761e-03]],\n",
+      "\n",
+      "          [[-1.0277e-02, -2.0925e-02],\n",
+      "           [ 2.1294e-04, -7.8277e-03],\n",
+      "           [-8.0540e-04,  6.9005e-03],\n",
+      "           [ 8.6584e-03, -3.0194e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 1.7248e-02, -6.2072e-03],\n",
+      "           [ 5.1232e-03, -8.9823e-03],\n",
+      "           [-1.4095e-02, -1.8089e-02],\n",
+      "           [ 8.5848e-03, -7.7452e-03]],\n",
+      "\n",
+      "          [[ 6.3832e-03, -9.4521e-04],\n",
+      "           [-6.0866e-03, -1.7911e-02],\n",
+      "           [ 7.1645e-03,  1.8826e-03],\n",
+      "           [-1.1272e-03, -1.6942e-03]],\n",
+      "\n",
+      "          [[-1.9407e-04,  2.6855e-02],\n",
+      "           [ 2.4922e-03,  4.0395e-03],\n",
+      "           [-9.0179e-03,  1.1456e-02],\n",
+      "           [ 6.2417e-03,  1.6225e-03]],\n",
+      "\n",
+      "          [[-1.3302e-02, -7.9396e-03],\n",
+      "           [ 1.6342e-02, -8.3640e-04],\n",
+      "           [ 8.2515e-03, -1.9994e-02],\n",
+      "           [-1.5314e-02,  1.1206e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[-1.6801e-02,  2.5082e-03],\n",
+      "           [-1.2035e-02,  1.0600e-02],\n",
+      "           [-3.4271e-03,  8.3387e-03],\n",
+      "           [ 1.1338e-03, -1.3025e-03]],\n",
+      "\n",
+      "          [[-4.3297e-04,  6.9727e-03],\n",
+      "           [-9.4044e-03, -1.8042e-02],\n",
+      "           [-1.6318e-02, -1.8721e-03],\n",
+      "           [ 9.9107e-03, -1.0611e-02]],\n",
+      "\n",
+      "          [[ 1.3804e-02, -5.9115e-03],\n",
+      "           [-8.4717e-03,  1.2661e-02],\n",
+      "           [ 5.9259e-03,  6.3610e-03],\n",
+      "           [-1.8119e-03, -1.0934e-02]],\n",
+      "\n",
+      "          [[-6.4295e-03,  1.4941e-03],\n",
+      "           [-1.3924e-02,  1.0059e-02],\n",
+      "           [ 3.7694e-03,  1.3317e-02],\n",
+      "           [ 7.0841e-03,  8.2705e-03]]],\n",
+      "\n",
+      "\n",
+      "         [[[ 5.0314e-03, -1.1748e-03],\n",
+      "           [-6.9476e-03, -6.9823e-05],\n",
+      "           [ 3.5111e-03, -3.0028e-03],\n",
+      "           [-6.3763e-03,  1.4920e-03]],\n",
+      "\n",
+      "          [[-9.0027e-04, -3.8365e-03],\n",
+      "           [ 1.1627e-04, -4.9486e-03],\n",
+      "           [ 1.3806e-02,  5.5603e-03],\n",
+      "           [-1.1555e-02, -3.8007e-03]],\n",
+      "\n",
+      "          [[-3.6884e-04,  1.0951e-02],\n",
+      "           [ 2.6923e-03,  6.7217e-04],\n",
+      "           [ 1.1842e-02, -1.7159e-02],\n",
+      "           [ 1.3960e-04,  8.7688e-03]],\n",
+      "\n",
+      "          [[-6.5861e-03,  1.1069e-03],\n",
+      "           [-9.9205e-03,  1.4599e-02],\n",
+      "           [-4.2275e-03, -3.8720e-03],\n",
+      "           [ 1.4915e-02, -2.9827e-03]]]]])\n"
      ]
     }
    ],
    "source": [
-    "eps = 0.1\n",
-    "num_episodes = 1000\n",
+    "eps = 1.0\n",
+    "num_episodes = 2000\n",
     "rewards = []  # List to store rewards for each episode\n",
     "\n",
     "# Training loop\n",
     "for episode in range(num_episodes):\n",
+    "    eps = eps * 0.99\n",
     "    print(\"Episode:\", episode)\n",
     "    state, info = env.reset(seed=episode)\n",
     "    state = discretize_state(state, num_bins)\n",
@@ -2163,22 +5324,22 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 51,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x161cdf1f0>]"
+       "[<matplotlib.lines.Line2D at 0x138355700>]"
       ]
      },
-     "execution_count": 33,
+     "execution_count": 51,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -2186,7 +5347,7 @@
      "metadata": {
       "image/png": {
        "height": 413,
-       "width": 555
+       "width": 552
       }
      },
      "output_type": "display_data"
@@ -2198,22 +5359,22 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
+   "execution_count": 52,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x160fc95b0>]"
+       "[<matplotlib.lines.Line2D at 0x138444fd0>]"
       ]
      },
-     "execution_count": 34,
+     "execution_count": 52,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -2221,7 +5382,7 @@
      "metadata": {
       "image/png": {
        "height": 413,
-       "width": 546
+       "width": 552
       }
      },
      "output_type": "display_data"