Example_4.2_car_rental.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "這邊利用課本中的範例4.2來說明策略迭代的過程，程式碼主要參考的是[ShangtongZhang](https://github.com/ShangtongZhang/reinforcement-learning-an-introduction/blob/master/chapter04/car_rental.py)，再加上自己理解的註解。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Example 4.2: Jack’s Car Rental\n",
    "範例說的是，jack管理一家有兩個地點的汽車租賃公司，每天都會有人來租車。只要有車可以出租，每租一台出去就得到10元，沒車就沒搞頭。還車後的第二天，那台還回來的車就可以拿來出租。為了兩邊都有車可以用，jack會利用晚上來移動車子，每次的移動都需要2元的成本。我們假設每一個地點需求的汽車與歸還的汽車為Possion random variables([卜瓦松](https://zh.wikipedia.org/wiki/%E5%8D%9C%E7%93%A6%E6%9D%BE%E5%88%86%E5%B8%83)隨機變數)，意思就是數量$n$的機率為$\\dfrac{\\lambda^n}{n!}e^{-\\lambda}$，其中$\\lambda$是期望的數值。假設，兩地的出租需求$\\lambda$是3跟4，然後歸還的是3與2。為了稍微簡化問題，我們再假設，每個地點最多就是20台車，多的就退回總公司，因此這個問題就是最多20台，然後一個晚上最多能移動的就是5台車。我們設置discount rate $\\gamma = 0.9$，然後把這個問題弄成continuing finite MDP，time steps就是天(days)，state就是一天結束的時候每個地方擁有的汽車數量，然後actions就是晚上兩地移動的汽車的數量。Figure 4.2說明著從沒有移動任何車子的policy開始的policy iteration所找出的一系列的策略。\n",
    "\n",
    "\n",
    "![](https://i.imgur.com/d3KXJEf.png)\n",
    "Figure 4.2：Jack汽車出租問題利用策略迭代找出來的一系列的策略與最終的state-value function。前五張圖說明的是，每天結束時，每一個地點的汽車數量從一個點被移到另一個點(負值就代表從sencond location移往first location，反之，正就是從first location移到second location)。每個後續的策略(successive policy)都是從前一個策略(previous policy)嚴格改進(strict improvement)而來的，然後最後一個是optimal。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "載入需求套件"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`poisson`是關於[卜瓦松分布(Poisson distribution](https://zh.wikipedia.org/wiki/%E5%8D%9C%E7%93%A6%E6%9D%BE%E5%88%86%E5%B8%83)的function"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "更多關於Poisson distribution可以參考翁秉仁老師的[Poisson 分配、指數分配與排隊理論](http://episte.math.ntu.edu.tw/applications/ap_poisson/index.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "from scipy.stats import poisson\n",
    "\n",
    "%matplotlib inline "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "下面定義環境中需要的參數"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 每一個地點能夠放的最大汽車數，超過的就假裝\"送回總公司\"\n",
    "MAX_CARS = 20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 每個晚上能夠移動的最大汽車數\n",
    "MAX_MOVE_OF_CARS = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 甲地的出租需求數\n",
    "RENTAL_REQUEST_FIRST_LOC = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 乙地的出租需求數\n",
    "RENTAL_REQUEST_SECOND_LOC = 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 甲地的歸還汽車數\n",
    "RETURNS_FIRST_LOC = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 乙地的歸還汽車數\n",
    "RETURNS_SECOND_LOC = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# discount rate\n",
    "DISCOUNT = 0.9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 出租一台車可以賺多少\n",
    "RENTAL_CREDIT = 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 兩地之間移動汽車的成本\n",
    "MOVE_CAR_COST = 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "所有可能的action，這個案例中我們的action就是要移動幾台的車子，來幾台還是去幾台"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-5, -4, -3, -2, -1,  0,  1,  2,  3,  4,  5])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "actions = np.arange(-MAX_MOVE_OF_CARS, MAX_MOVE_OF_CARS + 1)\n",
    "actions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[scipy_poisson](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.poisson.html)：\n",
    "* 第一個參數放的是一段時間內會發生的次數\n",
    "* 第二個參數放的是一段時間內平均發生的次數，也就是$\\lambda$，在這個範例中就是3、4跟3、2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "這邊設置一個參數，最多就是計算10台的車輛，超過了，得到的機率值似乎會被截斷為0，因此作者只計算0~10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "POISSON_UPPER_BOUND = 11"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 用一個dict來記錄poisson distribution\n",
    "\n",
    "poisson_cache = dict()\n",
    "\n",
    "def poisson_probability(n, lam):\n",
    "    global poisson_cache\n",
    "    key = n * 10 + lam\n",
    "    if key not in poisson_cache:\n",
    "        poisson_cache[key] = poisson.pmf(n, lam)\n",
    "    return poisson_cache[key]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(21):\n",
    "    for j in [2,3,4]:\n",
    "        poisson_probability(i, j)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "下面計算expected return"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "大致流程\n",
    "* (1)初始化returns，\n",
    "* (2)計算移動車子的成本，這是由action決定，\n",
    "* (3)計算兩地還車後的車輛數，\n",
    "* (4)迴圈計算所有可能的租車排列組合，\n",
    "    * (4-1)計算機率，兩地的租車是iid，因此相乘，\n",
    "    * (4-2)確認出租車輛數，因為庫存(3)可能比(4)的租車數來的少，取小，\n",
    "    * (4-3)計算reward，\n",
    "    * (4-4)計算出租後的車輛數，\n",
    "    * (4-5)計算returns，假設固定還車數，將剩餘車數(4-4)加上還車輛就是next state，以此計算returns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "def expected_return(state, action, state_value, constant_returned_cars):\n",
    "    \"\"\"\n",
    "    state: list，[甲地的車子數, 乙地的車子數]\n",
    "    action: 甲地搬車到乙地就正值，乙地搬車到甲地就負值\n",
    "    state_value: matrix，記錄state value\n",
    "    constant_returned_cars: 如果設置為true，那每天歸還的汽車數就會是固定的，\n",
    "                            而不是經由poisson distribution計算而來，\n",
    "                            這可以降低計算時間\n",
    "    @constant_returned_cars:  if set True, model is simplified such that\n",
    "    the # of cars returned in daytime becomes constant\n",
    "    rather than a random value from poisson distribution, which will reduce calculation time\n",
    "    and leave the optimal policy/value state matrix almost the same\n",
    "    \"\"\"\n",
    "    # 初始化return\n",
    "    returns = 0.0\n",
    "\n",
    "    # 計算移動車子的成本\n",
    "    # 範例設置是2元，action有正有負，記得取abs\n",
    "    returns -= MOVE_CAR_COST * abs(action)\n",
    "\n",
    "    # 利用state、action計算移動之後兩地的汽車數\n",
    "    # 要注意的是，取min是因為範例有提過，一個地方能放的車最多就是20輛，超過的就\"自動送回總公司\"\n",
    "    NUM_OF_CARS_FIRST_LOC = min(state[0] - action, MAX_CARS)\n",
    "    NUM_OF_CARS_SECOND_LOC = min(state[1] + action, MAX_CARS)\n",
    "\n",
    "    # 下面做出所有可能的出租需求的排列組合\n",
    "    # 因為當車輛超過10台，也就是pmf(11, 3)所得到的機率變的超級小，因此作者就沒有列入計算\n",
    "    # 最多會出租POISSON_UPPER_BOUND，但這個過程包含0、1、2、3、...、POISSON_UPPER_BOUND\n",
    "    for rental_request_first_loc in range(POISSON_UPPER_BOUND):\n",
    "        for rental_request_second_loc in range(POISSON_UPPER_BOUND):\n",
    "            # 甲地、乙地兩地的出租機率是iid，因此相乘            \n",
    "            prob = poisson_probability(rental_request_first_loc, RENTAL_REQUEST_FIRST_LOC) * \\\n",
    "                poisson_probability(rental_request_second_loc, RENTAL_REQUEST_SECOND_LOC)\n",
    "\n",
    "            # 兩地的車輛數賦值\n",
    "            num_of_cars_first_loc = NUM_OF_CARS_FIRST_LOC\n",
    "            num_of_cars_second_loc = NUM_OF_CARS_SECOND_LOC\n",
    "\n",
    "            # 這邊要確認的是，你的租車需求必定小於等於你現有車輛數，所以取小\n",
    "            # 我們也會用這個實際出租車輛數來計算所得\n",
    "            valid_rental_first_loc = min(num_of_cars_first_loc, rental_request_first_loc)\n",
    "            valid_rental_second_loc = min(num_of_cars_second_loc, rental_request_second_loc)\n",
    "\n",
    "            # 計算兩地車輛出租之後的所得\n",
    "            reward = (valid_rental_first_loc + valid_rental_second_loc) * RENTAL_CREDIT\n",
    "            # 計算兩地車輛出租之後剩餘車輛\n",
    "            num_of_cars_first_loc -= valid_rental_first_loc\n",
    "            num_of_cars_second_loc -= valid_rental_second_loc\n",
    "    \n",
    "            # 判斷參數設置是否設置每日歸還數固定\n",
    "            if constant_returned_cars:\n",
    "                # 如果歸還數固定，那每日兩地就是3、2\n",
    "                returned_cars_first_loc = RETURNS_FIRST_LOC # 3\n",
    "                returned_cars_second_loc = RETURNS_SECOND_LOC # 2\n",
    "                # 計算歸還之後車輛數，一樣的，最多最多就只能放20輛車\n",
    "                num_of_cars_first_loc = min(num_of_cars_first_loc + returned_cars_first_loc, MAX_CARS)\n",
    "                num_of_cars_second_loc = min(num_of_cars_second_loc + returned_cars_second_loc, MAX_CARS)\n",
    "                # 計算return，這一次的reward，加上next state的value\n",
    "                returns += prob * (reward + DISCOUNT * state_value[num_of_cars_first_loc, num_of_cars_second_loc])\n",
    "            else:\n",
    "                # 如果歸還數沒有固定，那就繼續迴圈把所有可能的還車數都計一次                \n",
    "                for returned_cars_first_loc in range(POISSON_UPPER_BOUND):\n",
    "                    for returned_cars_second_loc in range(POISSON_UPPER_BOUND):\n",
    "                        # 因為不固定，所以需要再考慮poisson probability\n",
    "                        prob_return = poisson_probability(\n",
    "                            returned_cars_first_loc, RETURNS_FIRST_LOC) * poisson_probability(returned_cars_second_loc, RETURNS_SECOND_LOC)\n",
    "                        # 計算剩餘車輛數，這也是next state\n",
    "                        num_of_cars_first_loc_ = min(num_of_cars_first_loc + returned_cars_first_loc, MAX_CARS)\n",
    "                        num_of_cars_second_loc_ = min(num_of_cars_second_loc + returned_cars_second_loc, MAX_CARS)\n",
    "                        # 跟固定歸還數的最大差異在於多考慮了還車的poisson probability\n",
    "                        prob_ = prob_return * prob\n",
    "                        returns += prob_ * (reward + DISCOUNT *\n",
    "                                            state_value[num_of_cars_first_loc_, num_of_cars_second_loc_])\n",
    "    return returns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "接著我們要來定義一個執行的函數。下面會說明幾個過程中會用到的函數"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`axes.flatten`，將定義的subplot做平展，有沒有做對操作索引會有差異"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`np.flipud`，[官方文件](https://numpy.org/doc/1.14/reference/generated/numpy.flipud.html)，將matrix、vector做翻轉"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0., 0.],\n",
       "       [0., 2., 0.],\n",
       "       [0., 0., 3.]])"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "A = np.diag([1.0, 2, 3])\n",
    "A"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 3.],\n",
       "       [0., 2., 0.],\n",
       "       [1., 0., 0.]])"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.flipud(A)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "下面這邊算是主要執行的部份，這分成兩塊，一個是policy evaluation，一個是policy improvement：\n",
    "* policy evaluation: 利用bellman optimal equation，用$s_t$再加上$s_{t+1}$去更新計算value function，收斂至閥值之後停止\n",
    "* policy improvement: 預設的policy都是0，我們會用policy evaluation得到的新的value function來對每個action來計算它的return，然後取最大的那一個來做為在該state下的action，因為我們的policy就是greedy\n",
    "\n",
    "接下來判斷是否穩定，如果沒有穩定就再執行下一次迭代"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "def figure_4_2(constant_returned_cars=True):\n",
    "    # 以我們的範例來看，就是設置兩個21x21的matrix\n",
    "    value = np.zeros((MAX_CARS + 1, MAX_CARS + 1))\n",
    "    policy = np.zeros(value.shape, dtype=np.int)\n",
    "    # 記錄迭代次數，也許我們會想知道收斂的進度\n",
    "    iterations = 0\n",
    "    # 設置figure、axes\n",
    "    _, axes = plt.subplots(2, 3, figsize=(40, 20))\n",
    "    # 透過flatten，就是把上面設置的2x3個sub fig平展，操作上的索引跟沒有設置flatten是不一樣的\n",
    "    axes = axes.flatten()\n",
    "    while True:\n",
    "        # 設置圖表 \n",
    "        fig = sns.heatmap(np.flipud(policy), cmap='YlGnBu', ax=axes[iterations])\n",
    "        fig.set_ylabel('# cars at first location', fontsize=30)\n",
    "        fig.set_yticks(list(reversed(range(MAX_CARS + 1))))\n",
    "        fig.set_xlabel('# cars at second location', fontsize=30 )\n",
    "        fig.set_title(f'policy {iterations}', fontsize=30)\n",
    "        # policy evaluation\n",
    "        while True:            \n",
    "            # 跟範例4.1一樣，我們先copy一個做舊值的記錄，\n",
    "            old_values = value.copy()\n",
    "            for i in range(MAX_CARS + 1):\n",
    "                for j in range(MAX_CARS + 1):\n",
    "                                                    # state, action, state_value, constant_returned_cars\n",
    "                    new_stata_value = expected_return([i, j], policy[i, j], value, constant_returned_cars=True)\n",
    "                    # in-place更新目前state的value\n",
    "                    value[i, j] = new_stata_value\n",
    "            # 計算最大變化        \n",
    "            max_value_change = abs(old_values - value).max()    \n",
    "            print(f'max_value_change: {max_value_change}')\n",
    "            # 判斷變化是否小於閥值，若是那就停止policy evaluation\n",
    "            if max_value_change < 1e-4:\n",
    "                print('stop policy evaluation.')\n",
    "                break\n",
    "        \n",
    "        # policy improvement\n",
    "        policy_stable = True\n",
    "        for i in range(MAX_CARS + 1):\n",
    "            for j in range(MAX_CARS + 1):\n",
    "                old_action = policy[i, j]\n",
    "                action_returns = []\n",
    "                # 兩地之間車輛的搬運是從-5~+5\n",
    "                for action in actions:\n",
    "                    if (0 <= action <= i) or (-j <= action <= 0):\n",
    "                        action_returns.append(expected_return([i, j], action, value, constant_returned_cars))\n",
    "                    else:\n",
    "                        action_returns.append(-np.inf)\n",
    "                    \n",
    "                new_action = actions[np.argmax(action_returns)]\n",
    "                policy[i, j] = new_action\n",
    "                if policy_stable and old_action != new_action:\n",
    "                    policy_stable = False\n",
    "        print(f'policy stable: {policy_stable}')\n",
    "\n",
    "        if policy_stable:\n",
    "            fig = sns.heatmap(np.flipud(value), cmap='YlGnBu', ax=axes[-1])\n",
    "            fig.set_ylabel('# cars at first location', fontsize=30)\n",
    "            fig.set_yticks(list(reversed(range(MAX_CARS + 1))))\n",
    "            fig.set_xlabel('# cars at second location', fontsize=30)\n",
    "            fig.set_title('optimal value', fontsize=30)\n",
    "            break\n",
    "            \n",
    "        iterations += 1\n",
    "    plt.savefig('./figure_4_2.png')\n",
    "    plt.show()\n",
    "    plt.close()     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max_value_change: 196.62783361783852\n",
      "max_value_change: 134.98823859766583\n",
      "max_value_change: 91.41415360228919\n",
      "max_value_change: 67.17097732555729\n",
      "max_value_change: 51.29055484635097\n",
      "max_value_change: 38.49091000659837\n",
      "max_value_change: 29.406139835126424\n",
      "max_value_change: 25.7210573245398\n",
      "max_value_change: 22.381602293031023\n",
      "max_value_change: 19.40385808254939\n",
      "max_value_change: 16.77577350573091\n",
      "max_value_change: 14.47251552455765\n",
      "max_value_change: 12.464101852186843\n",
      "max_value_change: 10.719367983418692\n",
      "max_value_change: 9.20806226246873\n",
      "max_value_change: 7.9019189666795455\n",
      "max_value_change: 6.775146571130392\n",
      "max_value_change: 5.8045764710083745\n",
      "max_value_change: 4.969618520007145\n",
      "max_value_change: 4.252112693842776\n",
      "max_value_change: 3.6361309524054946\n",
      "max_value_change: 3.107761240497666\n",
      "max_value_change: 2.654891834022692\n",
      "max_value_change: 2.26700589940549\n",
      "max_value_change: 1.9349911763441128\n",
      "max_value_change: 1.650966802154585\n",
      "max_value_change: 1.4081276418079938\n",
      "max_value_change: 1.2006055672075036\n",
      "max_value_change: 1.02334664187498\n",
      "max_value_change: 0.8720029351049448\n",
      "max_value_change: 0.7428376083516355\n",
      "max_value_change: 0.6326419232035505\n",
      "max_value_change: 0.5386628774742235\n",
      "max_value_change: 0.45854026040933604\n",
      "max_value_change: 0.3902520158000584\n",
      "max_value_change: 0.33206690395809346\n",
      "max_value_change: 0.28250355471067223\n",
      "max_value_change: 0.2402951004837064\n",
      "max_value_change: 0.20435866938208846\n",
      "max_value_change: 0.1737691018435612\n",
      "max_value_change: 0.14773633074884174\n",
      "max_value_change: 0.12558593365213255\n",
      "max_value_change: 0.10674242749371388\n",
      "max_value_change: 0.09071493100810812\n",
      "max_value_change: 0.07708486873008269\n",
      "max_value_change: 0.06549543334426744\n",
      "max_value_change: 0.05564256088280217\n",
      "max_value_change: 0.047267206266042194\n",
      "max_value_change: 0.040148735572074656\n",
      "max_value_change: 0.03409927655258116\n",
      "max_value_change: 0.028958890796900505\n",
      "max_value_change: 0.02459144993093787\n",
      "max_value_change: 0.02088111467702447\n",
      "max_value_change: 0.01772932984539466\n",
      "max_value_change: 0.0150522606048753\n",
      "max_value_change: 0.012778605996800252\n",
      "max_value_change: 0.010847734796413988\n",
      "max_value_change: 0.00920809667559297\n",
      "max_value_change: 0.007815868406908066\n",
      "max_value_change: 0.006633800647819044\n",
      "max_value_change: 0.005630235829983121\n",
      "max_value_change: 0.0047782719802853535\n",
      "max_value_change: 0.004055050928627679\n",
      "max_value_change: 0.003441152547054571\n",
      "max_value_change: 0.002920079317334512\n",
      "max_value_change: 0.0024778178350288727\n",
      "max_value_change: 0.0021024658457236\n",
      "max_value_change: 0.0017839150607983356\n",
      "max_value_change: 0.0015135814573454809\n",
      "max_value_change: 0.0012841759864272717\n",
      "max_value_change: 0.0010895096651211134\n",
      "max_value_change: 0.0009243279118891223\n",
      "max_value_change: 0.0007841697620847299\n",
      "max_value_change: 0.000665248240920846\n",
      "max_value_change: 0.0005643487139082026\n",
      "max_value_change: 0.00047874254232738167\n",
      "max_value_change: 0.00040611372588728045\n",
      "max_value_change: 0.0003444966009737982\n",
      "max_value_change: 0.000292222921245866\n",
      "max_value_change: 0.0002478769196159192\n",
      "max_value_change: 0.00021025715051337102\n",
      "max_value_change: 0.00017834408060934948\n",
      "max_value_change: 0.00015127258103575514\n",
      "max_value_change: 0.00012830856951495662\n",
      "max_value_change: 0.00010882917825938421\n",
      "max_value_change: 9.230592559106299e-05\n",
      "stop policy evaluation.\n",
      "policy stable: False\n",
      "max_value_change: 72.93565506480746\n",
      "max_value_change: 5.771584637253568\n",
      "max_value_change: 2.1472970104344995\n",
      "max_value_change: 1.070365975080108\n",
      "max_value_change: 0.8619106467957636\n",
      "max_value_change: 0.7181428891676092\n",
      "max_value_change: 0.611364010490604\n",
      "max_value_change: 0.5169059906119742\n",
      "max_value_change: 0.4358272831748309\n",
      "max_value_change: 0.3670218562992318\n",
      "max_value_change: 0.30890785349942007\n",
      "max_value_change: 0.259927010978231\n",
      "max_value_change: 0.21868429274547907\n",
      "max_value_change: 0.18397356667821896\n",
      "max_value_change: 0.15476712387498992\n",
      "max_value_change: 0.1301950284682789\n",
      "max_value_change: 0.10952318723241206\n",
      "max_value_change: 0.09213308201026393\n",
      "max_value_change: 0.07750397308279844\n",
      "max_value_change: 0.065197614125168\n",
      "max_value_change: 0.054845259444334715\n",
      "max_value_change: 0.046136675532011395\n",
      "max_value_change: 0.038810871455780216\n",
      "max_value_change: 0.03264828952961807\n",
      "max_value_change: 0.02746423075132043\n",
      "max_value_change: 0.02310332179803254\n",
      "max_value_change: 0.019434859485443212\n",
      "max_value_change: 0.016348893962117472\n",
      "max_value_change: 0.013752933586260951\n",
      "max_value_change: 0.011569172929284832\n",
      "max_value_change: 0.009732160858675343\n",
      "max_value_change: 0.008186838889457704\n",
      "max_value_change: 0.006886890992063854\n",
      "max_value_change: 0.005793355427499591\n",
      "max_value_change: 0.004873456997415815\n",
      "max_value_change: 0.004099624717810002\n",
      "max_value_change: 0.003448665460666689\n",
      "max_value_change: 0.0029010688248831684\n",
      "max_value_change: 0.0024404223665897007\n",
      "max_value_change: 0.002052919695017863\n",
      "max_value_change: 0.0017269466670768452\n",
      "max_value_change: 0.0014527332954799022\n",
      "max_value_change: 0.0012220609166320173\n",
      "max_value_change: 0.001028015870474519\n",
      "max_value_change: 0.0008647822832017482\n",
      "max_value_change: 0.000727467756007627\n",
      "max_value_change: 0.0006119567266296144\n",
      "max_value_change: 0.0005147871261783621\n",
      "max_value_change: 0.0004330466088617868\n",
      "max_value_change: 0.00036428526584586507\n",
      "max_value_change: 0.0003064422000988998\n",
      "max_value_change: 0.0002577837491344326\n",
      "max_value_change: 0.00021685153535599966\n",
      "max_value_change: 0.00018241874482782805\n",
      "max_value_change: 0.00015345336976224644\n",
      "max_value_change: 0.00012908726307614415\n",
      "max_value_change: 0.00010859013127628714\n",
      "max_value_change: 9.134763803331225e-05\n",
      "stop policy evaluation.\n",
      "policy stable: False\n",
      "max_value_change: 4.7865793901779625\n",
      "max_value_change: 3.2947349349497017\n",
      "max_value_change: 2.2411823866665372\n",
      "max_value_change: 1.616931343950455\n",
      "max_value_change: 1.1197864003121367\n",
      "max_value_change: 0.7204544260453076\n",
      "max_value_change: 0.443826224180043\n",
      "max_value_change: 0.270089591177225\n",
      "max_value_change: 0.16639579119885184\n",
      "max_value_change: 0.1097569388878128\n",
      "max_value_change: 0.09306955083684443\n",
      "max_value_change: 0.07883243113371918\n",
      "max_value_change: 0.06673516197616891\n",
      "max_value_change: 0.05647744756430484\n",
      "max_value_change: 0.04778890580797679\n",
      "max_value_change: 0.04043363544485601\n",
      "max_value_change: 0.03420889623009771\n",
      "max_value_change: 0.02894175601244342\n",
      "max_value_change: 0.0244852782967655\n",
      "max_value_change: 0.020714866829109724\n",
      "max_value_change: 0.01752498189301832\n",
      "max_value_change: 0.014826276628639334\n",
      "max_value_change: 0.01254313570910881\n",
      "max_value_change: 0.010611575530163009\n",
      "max_value_change: 0.008977459770164842\n",
      "max_value_change: 0.00759498609147613\n",
      "max_value_change: 0.006425404193464601\n",
      "max_value_change: 0.005435930515432119\n",
      "max_value_change: 0.004598829696703888\n",
      "max_value_change: 0.00389063733302919\n",
      "max_value_change: 0.003291502340118768\n",
      "max_value_change: 0.0027846305547996053\n",
      "max_value_change: 0.0023558140037494013\n",
      "max_value_change: 0.0019930326484427496\n",
      "max_value_change: 0.0016861174636346732\n",
      "max_value_change: 0.0014264653930240456\n",
      "max_value_change: 0.001206798198722936\n",
      "max_value_change: 0.0010209584465883381\n",
      "max_value_change: 0.0008637369118673632\n",
      "max_value_change: 0.0007307265586291578\n",
      "max_value_change: 0.0006181990110007973\n",
      "max_value_change: 0.0005230000359119913\n",
      "max_value_change: 0.00044246113805002096\n",
      "max_value_change: 0.00037432475153309497\n",
      "max_value_change: 0.00031668096374914967\n",
      "max_value_change: 0.0002679139766428307\n",
      "max_value_change: 0.00022665681694888917\n",
      "max_value_change: 0.00019175301355289776\n",
      "max_value_change: 0.00016222418821598694\n",
      "max_value_change: 0.00013724262737468962\n",
      "max_value_change: 0.00011610807808892787\n",
      "max_value_change: 9.822812404536307e-05\n",
      "stop policy evaluation.\n",
      "policy stable: False\n",
      "max_value_change: 0.5643315459673204\n",
      "max_value_change: 0.19760617142037518\n",
      "max_value_change: 0.10013580858492332\n",
      "max_value_change: 0.06076229858263105\n",
      "max_value_change: 0.04080851176706801\n",
      "max_value_change: 0.02724975517776329\n",
      "max_value_change: 0.01637959485265128\n",
      "max_value_change: 0.00917172069227945\n",
      "max_value_change: 0.0049277609952014245\n",
      "max_value_change: 0.0025834353657501197\n",
      "max_value_change: 0.0013420404746966597\n",
      "max_value_change: 0.0007016294298409775\n",
      "max_value_change: 0.00037558255417025066\n",
      "max_value_change: 0.00020989058543818828\n",
      "max_value_change: 0.00013043237390775175\n",
      "max_value_change: 0.00011051700198549952\n",
      "max_value_change: 9.361574132071837e-05\n",
      "stop policy evaluation.\n",
      "policy stable: False\n",
      "max_value_change: 0.04079438312567163\n",
      "max_value_change: 0.010408227162770345\n",
      "max_value_change: 0.005110707129347247\n",
      "max_value_change: 0.0032318390198042835\n",
      "max_value_change: 0.0021719229242762594\n",
      "max_value_change: 0.0013911772695109903\n",
      "max_value_change: 0.0008154469392138708\n",
      "max_value_change: 0.0004459807777266178\n",
      "max_value_change: 0.0002340408432246477\n",
      "max_value_change: 0.00012037610895276885\n",
      "max_value_change: 6.173777182993945e-05\n",
      "stop policy evaluation.\n",
      "policy stable: True\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2880x1440 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure_4_2()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}