{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "3e646804",
   "metadata": {},
   "source": [
    "# Cross-Entropy Methods (CEM) on MountainCarContinuous-v0\n",
    "\n",
    "> In this post, We will take a hands-on-lab of Cross-Entropy Methods (CEM for short) on openAI gym MountainCarContinuous-v0 environment. This is the coding exercise from udacity Deep Reinforcement Learning Nanodegree.\n",
    "\n",
    "- toc: true \n",
    "- badges: true\n",
    "- comments: true\n",
    "- author: Chanseok Kang\n",
    "- categories: [Python, Reinforcement_Learning, PyTorch, Udacity]\n",
    "- image: images/MountainCarContinuous-v0.gif"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6e1d4f5",
   "metadata": {},
   "source": [
    "## Cross-Entropy Methods (CEM)\n",
    "---\n",
    "In this notebook, you will implement CEM on OpenAI Gym's MountainCarContinuous-v0 environment. For summary, The **cross-entropy method** is sort of Black box optimization and it iteratively suggests a small number of neighboring policies, and uses a small percentage of the best performing policies to calculate a new estimate.\n",
    "\n",
    "### Import the Necessary Packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "063a557c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\kcsgo\\anaconda3\\lib\\site-packages\\numpy\\_distributor_init.py:32: UserWarning: loaded more than 1 DLL from .libs:\n",
      "C:\\Users\\kcsgo\\anaconda3\\lib\\site-packages\\numpy\\.libs\\libopenblas.PYQHXLVVQ7VESDPUVUADXEVJOBGHJPAY.gfortran-win_amd64.dll\n",
      "C:\\Users\\kcsgo\\anaconda3\\lib\\site-packages\\numpy\\.libs\\libopenblas.WCDJNK7YVMPZQ2ME2ZZHJJRJ3JIKNDB7.gfortran-win_amd64.dll\n",
      "  stacklevel=1)\n"
     ]
    }
   ],
   "source": [
    "import gym\n",
    "import math\n",
    "import numpy as np\n",
    "from collections import deque\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from torch.autograd import Variable\n",
    "\n",
    "import base64, io\n",
    "\n",
    "# For visualization\n",
    "from gym.wrappers.monitoring import video_recorder\n",
    "from IPython.display import HTML\n",
    "from IPython import display \n",
    "import glob"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "de727785",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "device(type='cuda', index=0)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "device"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "67faf368",
   "metadata": {},
   "source": [
    "### Instantiate the Environment and Agent\n",
    "\n",
    "MountainCar environment has two types: Discrete and Continuous. In this notebook, we used Continuous version of MountainCar. That is, we can move the car to the left (or right) precisely."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c6c02db1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "observation space: Box(-1.2000000476837158, 0.6000000238418579, (2,), float32)\n",
      "action space: Box(-1.0, 1.0, (1,), float32)\n",
      "  - low: [-1.]\n",
      "  - high: [1.]\n"
     ]
    }
   ],
   "source": [
    "env = gym.make('MountainCarContinuous-v0')\n",
    "env.seed(101)\n",
    "np.random.seed(101)\n",
    "\n",
    "print('observation space:', env.observation_space)\n",
    "print('action space:', env.action_space)\n",
    "print('  - low:', env.action_space.low)\n",
    "print('  - high:', env.action_space.high)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6ebff61",
   "metadata": {},
   "source": [
    "### Define Agent "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8a769b19",
   "metadata": {},
   "outputs": [],
   "source": [
    "class Agent(nn.Module):\n",
    "    def __init__(self, env, h_size=16):\n",
    "        super(Agent, self).__init__()\n",
    "        self.env = env\n",
    "        # state, hidden layer, action sizes\n",
    "        self.s_size = env.observation_space.shape[0]\n",
    "        self.h_size = h_size\n",
    "        self.a_size = env.action_space.shape[0]\n",
    "        # define layers (we used 2 layers)\n",
    "        self.fc1 = nn.Linear(self.s_size, self.h_size)\n",
    "        self.fc2 = nn.Linear(self.h_size, self.a_size)\n",
    "        \n",
    "    def set_weights(self, weights):\n",
    "        s_size = self.s_size\n",
    "        h_size = self.h_size\n",
    "        a_size = self.a_size\n",
    "        # separate the weights for each layer\n",
    "        fc1_end = (s_size * h_size) + h_size\n",
    "        fc1_W = torch.from_numpy(weights[:s_size*h_size].reshape(s_size, h_size))\n",
    "        fc1_b = torch.from_numpy(weights[s_size*h_size:fc1_end])\n",
    "        fc2_W = torch.from_numpy(weights[fc1_end:fc1_end+(h_size*a_size)].reshape(h_size, a_size))\n",
    "        fc2_b = torch.from_numpy(weights[fc1_end+(h_size*a_size):])\n",
    "        # set the weights for each layer\n",
    "        self.fc1.weight.data.copy_(fc1_W.view_as(self.fc1.weight.data))\n",
    "        self.fc1.bias.data.copy_(fc1_b.view_as(self.fc1.bias.data))\n",
    "        self.fc2.weight.data.copy_(fc2_W.view_as(self.fc2.weight.data))\n",
    "        self.fc2.bias.data.copy_(fc2_b.view_as(self.fc2.bias.data))\n",
    "    \n",
    "    def get_weights_dim(self):\n",
    "        return (self.s_size + 1) * self.h_size + (self.h_size + 1) * self.a_size\n",
    "        \n",
    "    def forward(self, x):\n",
    "        x = F.relu(self.fc1(x))\n",
    "        x = torch.tanh(self.fc2(x))\n",
    "        return x.cpu().data\n",
    "    \n",
    "    def act(self, state):\n",
    "        state = torch.from_numpy(state).float().to(device)\n",
    "        with torch.no_grad():\n",
    "            action = self.forward(state)\n",
    "        return action\n",
    "        \n",
    "    def evaluate(self, weights, gamma=1.0, max_t=5000):\n",
    "        self.set_weights(weights)\n",
    "        episode_return = 0.0\n",
    "        state = self.env.reset()\n",
    "        for t in range(max_t):\n",
    "            state = torch.from_numpy(state).float().to(device)\n",
    "            action = self.forward(state)\n",
    "            state, reward, done, _ = self.env.step(action)\n",
    "            episode_return += reward * math.pow(gamma, t)\n",
    "            if done:\n",
    "                break\n",
    "        return episode_return"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea18646f",
   "metadata": {},
   "source": [
    "### Cross Entropy Method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e0c930e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "def cem(agent, n_iterations=500, max_t=1000, gamma=1.0, print_every=10, pop_size=50, elite_frac=0.2, sigma=0.5):\n",
    "    \"\"\"PyTorch implementation of the cross-entropy method.\n",
    "        \n",
    "    Params\n",
    "    ======\n",
    "        Agent (object): agent instance\n",
    "        n_iterations (int): maximum number of training iterations\n",
    "        max_t (int): maximum number of timesteps per episode\n",
    "        gamma (float): discount rate\n",
    "        print_every (int): how often to print average score (over last 100 episodes)\n",
    "        pop_size (int): size of population at each iteration\n",
    "        elite_frac (float): percentage of top performers to use in update\n",
    "        sigma (float): standard deviation of additive noise\n",
    "    \"\"\"\n",
    "    n_elite=int(pop_size*elite_frac)\n",
    "\n",
    "    scores_deque = deque(maxlen=100)\n",
    "    scores = []\n",
    "    # Initialize the weight with random noise\n",
    "    best_weight = sigma * np.random.randn(agent.get_weights_dim())\n",
    "\n",
    "    for i_iteration in range(1, n_iterations+1):\n",
    "        # Define the cadidates and get the reward of each candidate\n",
    "        weights_pop = [best_weight + (sigma * np.random.randn(agent.get_weights_dim())) for i in range(pop_size)]\n",
    "        rewards = np.array([agent.evaluate(weights, gamma, max_t) for weights in weights_pop])\n",
    "        \n",
    "        # Select best candidates from collected rewards\n",
    "        elite_idxs = rewards.argsort()[-n_elite:]\n",
    "        elite_weights = [weights_pop[i] for i in elite_idxs]\n",
    "        best_weight = np.array(elite_weights).mean(axis=0)\n",
    "\n",
    "        reward = agent.evaluate(best_weight, gamma=1.0)\n",
    "        scores_deque.append(reward)\n",
    "        scores.append(reward)\n",
    "        \n",
    "        torch.save(agent.state_dict(), 'checkpoint.pth')\n",
    "        \n",
    "        if i_iteration % print_every == 0:\n",
    "            print('Episode {}\\tAverage Score: {:.2f}'.format(i_iteration, np.mean(scores_deque)))\n",
    "\n",
    "        if np.mean(scores_deque)>=90.0:\n",
    "            print('\\nEnvironment solved in {:d} iterations!\\tAverage Score: {:.2f}'.format(i_iteration-100, np.mean(scores_deque)))\n",
    "            break\n",
    "    return scores"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3735df07",
   "metadata": {},
   "source": [
    "### Run"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "4c662c01",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode 10\tAverage Score: -1.44\n",
      "Episode 20\tAverage Score: -3.98\n",
      "Episode 30\tAverage Score: -4.18\n",
      "Episode 40\tAverage Score: 2.57\n",
      "Episode 50\tAverage Score: 18.74\n",
      "Episode 60\tAverage Score: 29.35\n",
      "Episode 70\tAverage Score: 38.69\n",
      "Episode 80\tAverage Score: 45.65\n",
      "Episode 90\tAverage Score: 47.98\n",
      "Episode 100\tAverage Score: 52.56\n",
      "Episode 110\tAverage Score: 62.09\n",
      "Episode 120\tAverage Score: 72.28\n",
      "Episode 130\tAverage Score: 82.21\n",
      "Episode 140\tAverage Score: 89.48\n",
      "\n",
      "Environment solved in 47 iterations!\tAverage Score: 90.83\n"
     ]
    }
   ],
   "source": [
    "agent = Agent(env).to(device)\n",
    "scores = cem(agent)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51dea6d9",
   "metadata": {},
   "source": [
    "### Plot the learning progress"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "abff3eba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEGCAYAAACO8lkDAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAABBwElEQVR4nO2dd5xcdbn/38/MbE2yu9lsyWZTNqQ3SFkCEsDQISBNkSgICl6ugmJHQO8V71WvP1EUFdCIIIqiXIiXSA+9lyRASC+kbdpu6qZtmZnv74+ZMzs1OzN7Jmf25Hm/XnnNzjln5jxzMvP9nKd8n68YY1AURVGUaDxOG6AoiqLkHyoOiqIoSgIqDoqiKEoCKg6KoihKAioOiqIoSgI+pw2wg6qqKtPQ0OC0GYqiKL2KhQsX7jDGVCfb5wpxaGhoYMGCBU6boSiK0qsQkQ2p9mlYSVEURUlAxUFRFEVJQMVBURRFSSDn4iAi94lIs4gsidpWKSLzRWR1+LF/1L5bRGSNiKwUkXNybZ+iKIqSyJHwHP4EnBu37WbgeWPMKOD58HNEZDwwG5gQfs3dIuI9AjYqiqIoUeRcHIwxrwC74jZfBDwQ/vsB4OKo7X83xrQbY9YBa4DpubZRURRFicWpnEOtMWYrQPixJry9HtgUdVxTeFsCInKdiCwQkQUtLS05NVZRFOVoI98S0pJkW9Ke4saYOcaYRmNMY3V10jkciqIovY69hzp58K0NHGj3O2qHU5PgtotInTFmq4jUAc3h7U3AkKjjBgNbjrh1ii0072tjf5ufY6r7ZvXaD5v2srp5P2ePr83qPRSlN/Dcsu08srCJa08ZTn1FCZ+//x1Wbd/PE4u3cv8XjqczEOSRhU1sa23jUEeAhgF9mD68kgmDyhBJdj9tD06JwzzgauCn4cfHorb/TUTuAAYBo4B3HLFQ6RHGGK7780I27TrIG7ecTpEvVFewpnkfv5y/mqVb9vLU106lpDC23qDdH+CuF9Zw90tr8QdDTuNTH27ln9fPwOPJ3Q/hcCxu2kPT7kPMmlTnyPndzuY9h6grK478/37z4feZVF/OF2YMB8AfCCIieNP8/zfGsOdgJxWlBSkHz7bOAEU+T48HV2MMzy7bzsdGDKCsuCDj1z/41gb+87EliAhPL91Gn0IvHhG+9PER/P6VtVw+5y2adh1k54EOCn0ein0eWttCHsVl0wZz+2XH9cj+w5FzcRCRh4CZQJWINAE/ICQKD4vItcBG4DIAY8xSEXkYWAb4gRuMMYFc26jYz8urWnh/0x4Anl6yjYsm1/Powia+88gHhMd8dh/soKSwhLbOAL9+fjUbdh5kyZa9bNh5kEum1HPliUNZsW0f3/vnEua+t5nzJ9XxnUc+oMDr4dZZ46juV5RwXmMM63ceZM/BDop8XsYPKuvR52jafZDP/fEd9rf7mVRfzpDKUh5fvIU7n1vNDy+awEkjqg77+v3tfvoUenN6h9dbWbRxN796bjWvrGrhu+eO5cszR7Bq+z7mLtrMvz7YwskjqxhSWcpn/vAWLfva+e1npzJ5SAWdgSCdgSClhbHD1762Tn7zwhqeXrKNjbsO0jisPzecNpJTRlXh83ZF0P2BILN+/SpVfYv48zXTKS5IvyBy3Y4D/OTJ5dw6axzDq/rwj3c3cfPcDzl3wkDuuXJqzP/zym37+Od7m3lz7Q5G1vTjvy6aQJ+ikM17Dnbw/55ewUPvbOKMsTXcftlx/OPdTby4opkfXjSBcXVlDK8q5ea5H3LC8EruO28cxw2pAGDLnkPc++o67nt9HSccM4BPTRvcg/+F1IgblgltbGw02lspfzDGcMndb9Cyrx2fV6jpV8S9Vx/PaT9/ieFVfThnQi0/eXIFr950GkMqS1m4YRefvOdN6itKGF7Vh2tObuD0sbUABIOGS+55gy17DjF8QB/e3bCLAo+H0iIvv7x8MqeNqYk59x9fW8d/P74s8vySKfXcduEEykvSu6szxvDTp1aw+2AH1558DN99dDFrmvfT4Q/y6eMHc8t545j585do2deOR+Dy44ewZU8bK7a1Mrq2H1OH9ueMcTWMGdiPOS9/xG9eXMNxg8v5n0snMbKmX8L59rV18ujCJh5e0IQ/GOR/Lp3EtGGVPbj6cOND71FeUsB/XTThsKLUvK+NF5Y3M3NMDQPLi3t0zlSsbdnPa6t38NkThlIQNUC/uKKZax54l/6lhZQV++jwB3n1u6dz+zMr+cOrH9G3yMfImr4Mqijh8cVbqO5bxK4DHZw0sor3NuymqMDDkzeeQk1ZyG5jDNf9ZSEvrGjmlFFVTKov59GFTWzZ20ZJgZfJQyr4/gXjmDConCc/3Mr1f10EwJnjavjdldPoCATxiHQrFNf86V1eWNHMqJq+zLmqkQt/+xoeEfYe6uTXn5nChccNAmDn/nZm3v4ShzoDTKwvZ3HTHsYMLONrZ4xkyeZWHnpnI3sOdfLFk4fznXPGxIhXNHsOdlBekugBBYKGz/7hLRY37eXxG09mRJZhVxFZaIxpTLpPxUGxm5dWNvP5+9/lfy6dxL62Tn7y5ArOHFfD8yua+ddXTmbV9n188+EPeOnbM2mo6sNbH+1k9py3+NsXT+CkkYl34os27ubSu9/A5xHuuHwy4+v68fn736WmXxFzr58Rc+wld7/OgXY/N583lvc37eWuF9cwsKyYf331ZCr7FLK/3c+X/rKQmWOq+cKM4Qmhit+/vJb/eWoFPo9Ewlp3fXYqr61p4dFFm/l042AefGsjD157AnMXNTH3vc2MqO7DxPpyVm3fz8ptrQQNFBd4aOsMMnNMNe9t3MPBDj8/uWQSlzUOiTnfv/9lAc8s3c6xg8vZdaCDrXvb+OZZo7nhtJFZXfu9hzqZ8l/PEjTwrbNG89UzRgGhwfP2Z1ayYP1u/vMT4+lX7OPKP77Npl2HEIFTR1Xzm89OiYRGtu1to7asKDIo3fvqR6zavo92f5D2ziDt/gCfOG4Ql05NftcaCBrue20dP392Je3+ILMmDeTO2VMo8HrYuvcQs+58ldqyYh758km8trqFLz24iN9/bhq3zVvKuLoyZk2q49v/+wEAN507hiumD+P7jy1hyea9TG+o5LEPNnPiMQO4//PHIyLMeWUtP3lyBf9xwXiuPTkUjurwB3lu+XbeWbeLf32whQF9C3nixlO44t632bLnEP92yjH8YN5SSgu9HOwIUOjzMG1of04aMYCTRg6gorSQ+cu2s+tAB9fPHMGyLa189t63OX9SHU8t2UpxgRd/0PDkjSfzrf9dzMadB3jmG6dS06+Y7//fhzz0ziae/topjKrtx8urWvjK3xaxr82P1yOcMLyS758/vkee7ba9bZx35ytMGlzBn6/JruL/cOLgiq6sypGjwx/k0nte59tnj2Fm3F27xR9fW0d9RQmfnDqYA+1+fv7sKp5b3sxl0wYzsb6ctS37AQiEb0wC4UE4VUx56tD+/OxTxzK4oiQiHqNq+rJjf0fMcdtb23hv4x6+ffZoTh9by+lja2kc1p+r7nuH55dv57LGIby+Zgevhf89s3QbN583lqlDQxP0n122nZ8+vYLzJ9Vx24UT+POb6ykrLuD8Y+uYMKiMf7y7iQff2sisSQM5eVQVJ4+q4ieXToq529x9oIPnlm/n7XW7OHfCQM4cX8uO/e3c+NB73PrPD2mo6sPxDSHPYH+7nxdXtPD5kxq47cIJ7Gvr5Ja5H3L7Myupryjh4ilJq7gPy9sf7SRoYFJ9Ob+Yvwqf18PFUwbxmxfW8Le3N1Ja6OXiu16nb7EPAeZ8bhpLtrTy6+dX84dXPuJbZ4/hjbU7uOLet/n17Cl84rhBbNh5gB89sZz+pQWUlRRQ5PNwsCPAt//3A2rLipmRRNB//8pafvb0Ss4cV8PE+nJ+9dxqOvyLOG1sNY8ubKLdH+SuK6bSt8jHmeNqqS0r4gePLWVbaxu3zhrHBcfW8cbaHZQVF/Dlj49ARPjNZ6ZE3n9cXT9u+9cy7pi/inZ/kD++to7zJg7kmhkNkWMKfR5mTapj1qQ6Zoys4t/+vIDvPrKYd9bt4nuzxnH1SQ30KfLxwaY9DCwvZs/BDl5fs5NfzF/FL+Z3fRaPwJMfbqWkwEt9RQm/+PRxTB5SwY+fXM53zhnDyJp+/OKyYzn/16/x6d+9yU3njuVvb2/kcycOY1RtyFv8+Ohq5n/j42zYeYCJ9eWR8FJPGFhezL1XNzK0sk+P3ysZKg5KRrS2dbJkcysPvrUhqTh0BoIsWL+by48fQqHPQ6GvkAuPG8RTH27l2+eMAbpEwBKF7sQB4NNxd9xej0ReZzF/2XYAzp4wMLLt5JFVDOhTyJtrd3JZ4xDeWLODkgIvt104nh8/sZxP3vMmx1T34WB7gG2tbYyrK+P2y46ltNDHt84eE3mfhqo+XHDsIJ5aspWbzhkb2R4fhujfp5DLGofEeAhVfYu458ppXPTb1/jyg4v411dnUFdewiurWugIBDlvYsjefsUF/OryyWxvbeN7//yQ44ZUMLwq9oe/v93PbfOW4hGo6VfMF2Y0MKBvV+7ljbU7KS7w8NB1J/Lvf1nA/3t6Bf/v6RUAfHnmCP791GP40RPLeW/jbn535TRG1fbj7AkDWdu8n/teW8fnThzGDx5bijHw2Pub+cRxg3h2aei6zvvKyQypLAXgQLufi+56nRsfeo85V01je2s7tWXFTBvWn2DQ8NA7G/nYMQP4w1WNiAh9Cn38z1PLeW75djwCd3x6ciQU4vN6mH38UO58fjX9inycNb4WEeGOT09O+X246mMNPLe8md+8sAafRzh9bA0/+9SxKcNoZ42v5cxxtcx9bzMlBd7I9+lT0wYnxOx3Hejg7Y92sutgBzPH1NDc2saXH1zE6t37uXP2ZIoLvHzxlOGcMrqKMeHBf2RNP/76xRP40oOLuP6viygr9vH1M0fHvO/A8mLbw3c9DUEeDhUHJSOsAfmV1TvY3+6nb9wd0NItrRzqDETujgF+eOEEvnbGKGrD8WGvxIlD2IPIpBopmTg8u2w7DQNKGVXTFX/1eIQTRwzg9bU7MMbwxtqdHD+8ksuPH8r5xw7iicVbeOz9LVSUFvDx0dWcN6kuIdFp8aNLJnL9aSNoqMr8Tq28pIA5VzVyyV2vc9Mji/nLtScwf9l2+pcWMG1YpLUYPq+HO2dP4bw7X+XGh95j3ldmxAx4/3h3E48sbKK6XxEt+9qpqyjmihOGRfa/sXYHxzdU0rfIx4PXnsDK7ft4cUULtWVFXDKlHhHh50kqXL5x1iieWrKVy37/Jht2HmRifRmvrNpBa1snzyzdxvi6sogwAPQp8vG7K6dx8V2v88l73gSgtNDLy985jbUt+9m06xDfOmtMxPZ/O/UYZk8fwqGOAAVeD/37FMac/zPTh3LXi2uYNakurQSxxyP89rNTeOujnXxsRFVaOaXbLhzPG2t3cOnUespLUx9f2aeQ86Iq0+orSnj8xpN5+6NdESEXEcYOjA0JNTZU8q+vzuD7/1zCJ44blPAZexsqDkpGWANyhz/Iiyua+UQ4AWfx7rpQp5TGhq4Br0+RL8aNTvAcAqFHX6biEJUva23r5M21O/jCjOEJd48zRlTxxOKtvPXRLlY37+eT4TvFvkU+Lj9+KJcfPzStc5YVF1A2MPNyRYvRtf34xlmj+dETy3l++XZeWNHMmeNqE5KRgypK+PePH8PPnl7JnoOdkUEmEDQ88MZ6Gof15/efm8a0Hz0XI5DN+9pYtX1/JA9gDWDxg1gyRtb04+Ip9cxdtJnTx9Zww2kj+eQ9b/CPdzaxcONuvn7G6CSv6cvc609icdNeKvsUcN2fF/Lr51dzqDNA3yIf50R5cBDyjPqlKPccWB7KPzQMKE26PxkVpYWcOzH98uLB/Ut58dsz0y5OiKaqbxHnH9v9uerKS/jj54/P+P3zERUHJSOiB6Onl2xLFIf1uxhaWRrxEpKRIA6W55BBuadHhGCULS+tbKEzYDh7fG3CsTNGDgDgjvkrAThpxIC0z2M3V32sgb+8tYFv/ON9Wtv8nJXEXiDikUUL4Isrmtm46yA3nTsm4RoCvLl2J5D95/vW2WPo8Af57rljqa8oYVB5MXfMX4UxcPaE5HaOru3H6HBo5TPTh/LQOxvxeYWLJ9cnzGHpjsnhUs1ccrjvpRJLvrXPUPIcazAqLyngxZXNtHV2TUMxxrBgw+6YkFIyrPCRNfAF08g5xBPvOSzetIcin4cpQ/snHDu0spT6ihLeXb+bfsU+JgwqT/s8dlPo8/Ddc8fS2uanyOfh1NHJ50lYQhktgPe/sY668mLOmTAw4h1Fi8Mba3ZS1oPPV19Rwm8/O5UhlaV4PMJ5k+o41BlgaGUpYwcmluHGc+MZoyjyhaq0clV7rxw5VByUjLDKO8+ZUMvBjgAvr+pqeri25QC7DnRwfEPiAB2NFT6yBj7rPTMNK/kDXQOjP2go8nmSCoyIRO6mTzxmQEYilAvOmziQU0ZVcf5h8hu+OAH9qGU/r6/ZyZUnDqPA2/U5g1EC+da6nbZ+PmtG+NnhBHF3VPcr4qZzxzJzTHVMHkXpnag4KBlhDUYzRlZR6POwcMPuyL4F60P5huOHH95zsBLSligEs0lIi8QMjIGgOeygeFI4tORkSMlCRPjzNdO54/LJKY+xroUlgE27DwFwQvjadiX1u16z60AHgypKbLNz6tAK/vviiVz38WPSfs3VJzXwpy9M1xnhLkBzDkpGWINVkc/DwLJitu1ti+x7Z/0uKvsUckw31TyeOM8hkKXnEB1S8QcNXk/qe52zxg/kihN2R2awOk13g6cvzjOwPAhLAK2PGi2QwW4EMhsbP3fisO4PVFyJioOSEcHIIBUWh9YucfiwaS9Th1Z0O/B540ImlgeRUUI6ThxCA2Pq4/sW+fjxJZPSfn+nSV3RFfqQ3iQ5iYAxOBwxU1yEhpWUjPBHksdQW17M9rA4GGPYvOdQWrM1rYEvElbKIiHti0tI+4MmMnC6AU/KuSBx+2M8h8xCc4pyONzza1KOCF2zmT0MLCti2942jDHsPdTJwY4A9f27j3nH3/Vmk5D2SJznYAwu0oaEhHT8LPL40ByEroFXY/2KTbjo56QcCSKDlAi1ZcW0+4PsPdQZSZjWV3RfRx4fMskqIe2JnecQCLprYPTEh5WSCGh8OW/A2JtzUI5uVByUjIi+g7X6xGxrbWPLnpA4pFMtk7K3UgaDe3TXVOs93DQwJrQYSZKXCVVshf42xmBMZnkbRTkcKg5KRkSLgzXbdNveLMUhPmTizSwhnUkpa2/DuhaJnkPXT9bjSaz4ctM1UJxFq5WUjIguqawJr8S2vbWNLXvbKPJ5GJBGs7GEZGsWnoNXMitl7W1E8jJxpazRHzE67xJf6qooPcU9vybliBAIhmZdeT1CTVlIHLbtbWfz7kPUV5SkNfmpuxr+dAh5DqFwivVehytl7W1EKroCqT0Hr0hUC5LQNg0rKXbh6M9JRL4hIktFZImIPCQixSJSKSLzRWR1+FHn4ecR1oxcn0co8nmp7FPIttY2Nu85lPbs3ISBL5BdKWvInq7B000J6VShtxjPISop39W88Agaqbgax8RBROqBG4FGY8xEwAvMBm4GnjfGjAKeDz9X8gTLc7DuUGvLimkOJ6QHpVGpBFFlmPGeQyZhpSSDp5tCKukk7b2eroR0UMNKis047Yj7gBIR8QGlwBbgIuCB8P4HgIudMU1JRsRzCCdMB5YVsWn3QZr3tVNfkV4v/q67/tDzYOSuOLN5DqHXWna5SxxS5WViEtIxYaXMZ5kryuFwTByMMZuBnwMbga3AXmPMs0CtMWZr+JitQNKFikXkOhFZICILWlpakh2i5AB/VM4BQou0rG4OrQmdtucQGfiC4fc0GU2Agy6Bsexxmzgk5GWShZVEq5WU3OFkWKk/IS9hODAI6CMiV6b7emPMHGNMozGmsbq6OldmKnHEhzdqy4qxKkrrM8w5RFfaZNr2oWuGMJH3cNPAmJCXMUkS0p7EaiVtn6HYhZNhpTOBdcaYFmNMJzAXOAnYLiJ1AOHHZgdtVOKIv0MdGLWyVjqtM6Jfay3HEMzCc7CmRETnHNwUUvHEl7Im9RwSq5XclJRXnMVJcdgInCgipRKqfzwDWA7MA64OH3M18JhD9ilJiBeH2vIucRhYnl5Yqctz6AorZTqoJUvYZiow+YwvMgmO8GNyz8Hy2rrKgY+cjYq7cWwSnDHmbRF5BFgE+IH3gDlAX+BhEbmWkIBc5pSNSiJd4Y1Yz6G6XxFFvvTWDI5fqCYYzDysZE14iylldZE4WJ5DdE4ltL3rmOiwkpV70EV2FLtwdIa0MeYHwA/iNrcT8iKUPCQQV1lkiUMmK5DFL3GZTULaukN2eylrMO7zRQ/+IklakKg4KDahTqiSEfHdQStKCyj0eRichThYydZgNgnpuLbfbktIx5f7BpK04/ZK1yQ4neeg2I32VlIyIt5zEBG+cFIDU4amP5HdGr9i7vozvOP1xTWmC7otIR2XlwkEE9eriAkrabWSYjMqDkpGJFtX4JZZ4zJ6DxGJWY/Bn0VIqCsmn31oKp9J8BySrHTniWrZHdBqJcVmNKykZEQ26z0nwysSs0xopuKQLCbvprvm+GVAQ6W6scd4o9qWd1WRHTkbFXejXyUlI4JJPIds8HiieytlHitPthiOmzyHSKluoCus5PPGew5JVtNTz0GxCRUHJSP8QXsSnz6PJ2pgD2btOQRcmpCOnygYMIk5FU9Sz8E910BxFhUHJSOCJhTe6Gk9ffRdbzYJ6YR1qF2WkI6EzaxrFEj0jKIXPAqo56DYjIqDkhHZJI+TEdMXKJh5lU2kmqcHcyXyGW9cwj2ZZxTtORitVlJsRsVByYhsksfJ8HokKtkazLorazDac3DRwJg84R53jEhUy/KubYpiByoOSkZk0wcpGV6PRFaAC5jM73jj76xd5zkk7R0Vl5D2HH6lOEXpCfpVUjLCrjYVsesfZz6we+Jj8lnMss5nrI/ijxKH+I/nkcRJcOo5KHah4qBkhF3iEL3+sT8YzD4h3YNZ1vlM/ETBZJ5D8nkO7rkGirOoOCgZEUqM9vxr4/PErkWQ6Vu6vWU3xE4UTOYZeUUS1uF2k/ekOIuKg5IRgYCxZRauxyNR+YJgwl1xd0RPgstmDereQMxEwSTi5/FI4jrcLvKeFGdRcVAyImASwxvZEN1RNKuEdJTnYImM2zyH2ImCiZ5D9BrSVo8lN4XWFGdRcVAyIllJZTbEL1TjzXBMiy71dGtH0sSJgrH7Y8uBtVpJsRf9KikZkSwxmg3R4hCaWJdhWMlaEyLKc3DbXXPsRMEUXVl1PQclR6g4KBmRrKQyG7ye2FLWTPMYnqicg1srdbweT2xX1iTrOSRUK7lMIBXncFQcRKRCRB4RkRUislxEPiYilSIyX0RWhx/TX0VGyTm58ByyaZoXE1ZyrTgQNVEwSSlr9FwRl4bWFOdw2nO4E3jaGDMWOA5YDtwMPG+MGQU8H36u5Al+m9pUxDSNyyKsFL0YjlsT0tGDf7LrLjHtM9RzUOzFMXEQkTLgVOCPAMaYDmPMHuAi4IHwYQ8AFzthn5KcoLFnPoEnLp6eaUI6ehlNt941e70S0zsqoSurJzZhHdrmrmugOIeTnsMxQAtwv4i8JyL3ikgfoNYYsxUg/FiT7MUicp2ILBCRBS0tLUfO6qMcu7qy+uLi5RknpCXRc3DbXXPMJLgkLclj8jbhR5ddAsVBnBQHHzAVuMcYMwU4QAYhJGPMHGNMozGmsbq6Olc2KnHY2ZXVHxNWyuz1lpYEXJ1ziC1Vjb9GHpFIq+7IPAeXXQPFOZwUhyagyRjzdvj5I4TEYruI1AGEH5sdsk9Jgj+LVduS4YmZBJe54FjJ2UAg6NqQSkxvpWQJ6bjQHLjPe1KcwzFxMMZsAzaJyJjwpjOAZcA84OrwtquBxxwwT0lBMGjPAORLKGXNcg1pY9/SpfmGJz6slDBDOrErq9vyLopz+Bw+/1eBv4pIIfAR8AVCgvWwiFwLbAQuc9A+JQ5/MEhRQc+/Nh6P4A9ETYLLUHCsm+hg0Lh2ApjPG9+VNVEcrHCSeg6K3TgqDsaY94HGJLvOOMKmKGkSMPY0d4vuKJrNKm6RsJIxEZFx28AYXcqaPCGdWK2knoNiF07Pc1B6Gdks6ZkMrze2fUbmi/1Y9rjXc4gv903WlTUYV63ktmugOIeKg5IRgaA9d6cxk+CyWMXNexS0z/DF9Z867HoO4clwLrsEioOoOCgZYZvnEJ+QznYluOjGey4bGeMTzslKWRMS0i4LrSnOoeKgZESyqplsCJVphv7OJqwkIqH1DIx7w0rddmX1hBLSxsVzPRTnUHFQMsKu5ThDs3+DPVrFzZpI59qEtKebhLRYzQe7lgl12zVQnEPFQcmIgMk8BJQMa4nLngxq1kQ6N3sOMaWs3sRqJYjtTKvVSopdqDgoGRFaQ9q+3kqRZHKmnffoSti6NSHdXW8lT1TeJZtZ5opyOFQclIywaxDyegR/dOuLbDyHcFjJteIQt+ZFsklwQFhkNaSk2IuKg5IRAZsa71mzewM9CAl5470Pl4qDCX/GZKWs0DXXQ7VBsRMVByUj7BIHa3avtdJZVuIQLuW0Qi9uK+P0hBPSkY6rKcJKwaB93XIVxULFQckIu9ZzsNZHtsNzsBLS8Qnb3o4vnJC2PKOEhHT4aTB8HTWspNiJioOSEYEsJqwlw/IcelKf7w0373Nr0zkrIR1I4RlFJgKGq5W0UkmxExUHJSMCQZNVZVE88SGhbEtZAy7OOXgsz8HyjJKsIQ1EjnHb51ecRcVByQj7PIfQV8+awJbNXa/V0tqt4mCteRFIcY2iPYdA0H05F8VZVByUjEhWUpkN1gSujkAASLwrTus9RAgYUoZdejtWV9ZUnkNMtVIWS60qyuHQr5OSNsGgwRh7ZuFa79Hhz/6uPzR4BrsGTxcmpEOht1ATqoSV4MLPjbFv5rqiWKg4KGmT6g42G6z36Aj3ms7mrt/KW7g1IW11XbUaFMZ/Pm/0mhZBE8lBKIodqDgoaWPnamOWGHT4QyNfVmElqz+TS3MO3jjPIdUM6UC4nNdtn19xFsfFQUS8IvKeiDwefl4pIvNFZHX4sb/TNiohIvX2NrXPAOgMJA+ZpPsegWDQteJgJaQtzyEhrBRTreS+z684i+PiAHwNWB71/GbgeWPMKOD58HMlD7BzJnIkrBT2HLLtrRSdkHbb4OgJr3mRMiEdP8/BXR9fcRhHxUFEBgPnA/dGbb4IeCD89wPAxUfYLCUFQRs9B+suuN0Shyy7skbPA3CbOFhrXgRSJaQjnoN9bU0UxcJpz+FXwE1AMGpbrTFmK0D4sSbZC0XkOhFZICILWlpacm6ogq3LcVqeghVWysZziE9Iu62U1Rte6c2fQpSt/werfYbbPr/iLI6Jg4hcADQbYxZm83pjzBxjTKMxprG6utpm65RkdC2q0/OvjTcurJSNN+KxmvfZ6NHkE5G8jD+5+Fkft2ueg7s+v+IsPgfPPQO4UERmAcVAmYg8CGwXkTpjzFYRqQOaHbRRiSLVHWw2RMShhwnpzoB7E9Jd1ygQ89zCE5VzUM9BsRvHPAdjzC3GmMHGmAZgNvCCMeZKYB5wdfiwq4HHHDJRicPOpSjjq5Wya7zniXgOHsF1df5d3tXhZ0iHlkrVJUIVe3E655CMnwJnichq4KzwcyUPyInn4O+BOEhUu2oXDozW4J/Ku+rKOYTXc3DfJVAcxMmwUgRjzEvAS+G/dwJnOGmPkhw7J8FZA197D0pZrZbdwSTrK7sBTzd5GU9UbyWtVlLsJh89ByVPsTPxGz/w9WSxH3/QnmaA+Ub8XJBUCWmtVlJygYqDkjZ2lozG91bKVhwiOQcXioMnLiGdsBKcp8tz0GolxW5UHJS0yYXn0NkDz8ETNc/BjZ5DZC5IqlLWuGolFQfFTtIWBxEpEZExuTRGyW/snImckGzNMufg5oHRErz2FN6VdQ2NcW/eRXGOtMRBRD4BvA88HX4+WUTm5dAuJQ+x2jjYMRDHx9Oz78rq3oGxu4R0V1iJcM7hyNqnuJt0PYfbgOnAHgBjzPtAQy4MUvKXgLWugJ0J6Z7kHCTUW8mtCenIankpEtLWU2vNBzd6T4pzpCsOfmPM3pxaouQ9fhs9B1vmOXgEv+U5uHBgtNqURDyHFAnpYHg9Bzd6T4pzpDvPYYmIfBbwisgo4EbgjdyZpeQjQRs9BztmSHtcXsoa35wwfvD36jwHJYek6zl8FZgAtAN/A/YCX8+RTUqeYqvnELcSXDZ3vdYaywHjVs8hNvSWMAkuviurC6+B4hzdeg4i4gXmGWPOBL6Xe5OUfCXSldWG8EV3A186eCQUVgoEjOvWj4buQ2+R3krhaiU3XgPFObr1HIwxAeCgiJQfAXuUPMYfsK+U1RNfw5/tDOmge0tZrYR0ewpx6GqfgWuvgeIc6eYc2oAPRWQ+cMDaaIy5MSdWKXlJ0MZ5DlZyNVUNf1rvEVlj2Z0Do5WQTpWXsZbVCIarldRxUOwkXXF4IvxPOYqxsyurJy7nkN1iP6E1lt2ekE6Vl4lZQ9poWEmxl7TEwRjzgIgUAqPDm1YaYzpzZ5aSj9jalTWuWimrGdLhNZaDLk3GeuLmOaRcz8FotZJiP2mJg4jMBB4A1gMCDBGRq40xr+TMMiXvsLO3UvwM6exLWUO5EDfeNfviwkrxAhipVgoa1wqk4hzphpV+AZxtjFkJICKjgYeAabkyTMk/7OzKGt8aIptxzRIYfzDoyrvmyAzpVKWs8fMcXCiQinOkO8+hwBIGAGPMKqAgNyYp+UrEc7BhybHoCV5ej2S1xGd0qacbxcEa/FNVK0UmwRk0rKTYTrqewwIR+SPwl/DzK4CFuTFJyVcCuZjn4A9m/X7Rg2dZifsGRl9c+4zDVisZezw6RbFI13P4MrCUUNuMrwHLgC/15MQiMkREXhSR5SKyVES+Ft5eKSLzRWR1+LF/T86j2IflOdjZPqM9kP1df/SCQW68a45PSMeLaHRvpZDncETNU1xOul8nH3CnMeZSY8wlwK8Bbw/P7Qe+ZYwZB5wI3CAi44GbgeeNMaOA58PPlTzAVnGQnoeEPDZ4H/lM/CzyhIR0JKykjfcU+0lXHJ4HSqKelwDP9eTExpitxphF4b/3AcuBeuAiQpVRhB8v7sl5FPuwVRyi8hbZvp/1Fp0u9RyiK7qSVYhZYqDVSkouSFccio0x+60n4b9L7TJCRBqAKcDbQK0xZmv4PFuBmhSvuU5EFojIgpaWFrtMUQ5DLjyHnrzf0ZKQ7vAHkw78MYv9aLWSYjPpisMBEZlqPRGRRuCQHQaISF/gUeDrxpjWdF9njJljjGk0xjRWV1fbYYrSDX4bxcET9c3LNhzS1V7CnZU60RMFk3sOocdQWMmeyYmKYpFutdLXgf8VkS2AAQYBl/f05CJSQEgY/mqMmRvevF1E6owxW0WkDmju6XkUewhGJsH1PPMZ/R7ZTqqLXinNzeKQKqciIngE/IHkCWtF6QmH/ZWLyPEiMtAY8y4wFvgHoUTy08C6npxYQoXtfwSWG2PuiNo1D7g6/PfVwGM9OY9iH/7IJLiev1f0e2SdkLbCLgF3J6TbA8GYHE38MV0e3REzTTkK6O7r9HugI/z3x4BbgbuA3cCcHp57BvA54HQReT/8bxbwU+AsEVkNnBV+ruQBwXBb6GwmrMVj3fVCbIgpE3w2JLXzmZiKrhTXXES6Zpm78BooztFdWMlrjNkV/vtyYI4x5lHgURF5vycnNsa8RqhPUzLO6Ml7K7nBb3PS0+fx0BEIZh2m8tiQ1M5noj9TqoHfK9Kj5oWKkorufpVeEbEE5Azghah96eYrFJdgd4sGSxOyfctoW9wuDqnyMl5Plzi4MbSmOEd3A/xDwMsisoNQddKrACIyktA60spRhN3iYA1m2c9zOHrEIZVXEEpI29dKXVEsDisOxpgfi8jzQB3wrDHh5johj+OruTZOyS9sFwePJQ7ZhZXSGTx7MzGew2ES0p1WQtp9l0BxkG5DQ8aYt5JsW5Ubc5R8Jnfi0LPXgz1rTOQbMTmVlJ6D0NmDNTEUJRVa/KakjT9X4pBtV1aX5xx8aXw+T1TOQcNKip2oOChpE7S5WqnLc+hZV9aevEc+k07C3SsSacynCWnFTlQclLSx3XPQhPRhERGsj5gqpxJdreTGvIviHCoOStpYk+DswtNDz8Hj8oQ0dHlHqRLSHo9WKym5QcVBSRt/0Nia+O1pWMntCWnoEr3UpaxR1Ur6a1ZsRL9OStoEg/auGWAN7tl3Ze1+BnFvx/qMKSfBRVUrudV7UpxBxUFJG38weevobLFyBll3ZRX3ew4RAU2jWsmNeRfFOVQclLQJBO29O7UzrOTWgTEtz0GrlZQcoOKgpE0gGEyZGM2GnoaVol/n1pBKdxVdIc9BE9KK/ag4KGkTMLnxHLIVnOjX2Sla+UR3AuoR1HNQcoKKg5I2AZtzDt1V4qT7+p68R77TbVgpZob0ETNLOQrQr5OSNgGbq5V8NuYc3JqQjghoqrCSSNc8B5cKpOIMKg5K2gRsnufQ00lw0ba4Nd5uhcsO6zkEtVpJsR8VByVtcraegx2N91x61+ztxnMIVSuZmGMVxQ7yVhxE5FwRWSkia0TkZqftUewXh8hdcZbJ5Jh5Di5PSKfyHERC/y/gXu9JcYa8FAcR8QJ3AecB44HPiMh4Z61SAsberqw9TkhHfXvdGm/vrq350TDXQ3GGfF0HejqwxhjzEYCI/B24CFiWi5MZY3jyw220+wOUlxRQUVpARWkhx1T1QaJ+lGtb9vPzZ1bywaY9AIytK+MXlx1H/z6FGZ1v7qImnlm6jYUb9vDlmSO49uThPbK/eV8bl979Bv998UROG1MT87lWbd9PXUUxZcUFSV/7t7c38sSHW7h11jgmDCpn94EO3l63i+2tbQSN4aqPNUQGHX8gV4v9ZJtz8ET97c6BsbuEtNtXw1OcI1/FoR7YFPW8CTgh+gARuQ64DmDo0KE9OtmijXu44W+LErZfMqWeOz59HAC/nL+Ku15aS0mBl7PG1yLA4x9u5ZP3vMED10xnSGVpWueau6iJbz78AfUVJRR6hQfeWM81MxpiRChTFq7fTdPuQ9z0yGKe/fqplJUU8Oc31/PgWxtY23KAuvJi7r5iKlOG9o95XWtbJz99ajmtbX4u/O3rTG+oZMGGXZEYNsDImr6cMqoasL8ra49nSEeXsrpUHLpLSMeW8x4Rk5SjhHwVh2RfcxPzxJg5wByAxsZGk+T4tHli8VYKfR7mfWUG7Z1B9hzq5KWVzdz/+nqmDevPvjY/v35hDZdOqefW88dR1bcIgM+cMJQvPrCAC3/7Gv9xwXgumVJ/2EH+o5b9fP//ljB9eCV/++IJzH1vMzc9spgPmvYyeUhF1vYv29qKR2D3gQ5unruYgx0BXl29g2nD+vMfFwzjT2+s49O/f5PGYZW07G/n+IZK/vuiCTzw+npa2/z89Ysn8K8PtvDWRzu56mMNzJo0kJp+xZzxi5d5bfWOiDjkbD0HG8JKbk3GZuI5aFhJsZN8FYcmYEjU88HAllycKBg0PPnhVmaOrmbswLLI9lNGVvFRywF++K+ldAYMFx43iJ9fdlzMj/T4hkrmXn8S3/7fD/jmwx8w74Mt/OGqRgqS9E5ev+MAN/ztPQp9Hu6cPRmf18O5Ewfy/f9bwmPvb05LHNbtOMDDCzbxnbPHxNixfGsrI2v6MmtSHb96bjWFPg8/vXQSs6eHPKpPTR3MD+YtYcOug9SVF/PQOxs50O7n5VUtnDmuhhkjq5gxsirhfNOG9eeV1Tu4Jepa5ZPnEB1W8h6lCenozRpWUuwkLxPSwLvAKBEZLiKFwGxgXi5OtHDjbra1tnH+sXUx2z0e4ZeXT2ZgeTEnj6zi9suOTXr3NqK6L49+6SRunTWWl1a28NA7G2P2B4OG2+Yt5cw7Xmb9jgP88tOTqSsvAaCsuIDTx9Tw+OKtkYqTwzF3URP3vLSWzXsOxWxftqWVcXVl3HDaSL551mjmfvmkiDAAlJcW8KvZU/jn9TP4y7Un8M2zRjPvgy3sPdTJ184YnfJ8J4+qYvnWVlr2tQM5XEM668V+ot7LpQNj9+0z1HNQckNeeg7GGL+IfAV4BvAC9xljlubiXE8s3kqRz8MZ42oT9lX2KeS5b36cQq/nsOEij0f4t1OO4cUVLfzqudVcPKU+kgB+v2kPf3pjPZdMqeeWWWOp6Vcc89oLJw/i6aXbeG3NDk48pvKw51rTvB+AzXsORXIcew52sGVvG+PryijwerjxjFHdfuavnj6SAq+H1rZOJg0uT3ncqaOquf2Zlby+ZgcXT6knmG+L/RwNLbu7aWuuYSUlV+Sr54Ax5kljzGhjzAhjzI9zcY5A0PDEh1s5bUwNfYuS62SRz5tWslhEuHXWOHYd6OCel9ZGtm/d0wbAv3/8mARhADh9bOjcV9/3DmO+/zS3zP0w5TkscdgS5Tks29oKwPhBZUlfk8rWL88cwXfPHXvY4yYMKqN/aQGvrG4B7PccelrKejQt9nO4rqyRv13qPSnOkLficCR4d/0uWva1c8Fxdd0fnAaTBpdzyZR67nttHXsOdgCwdW9oIB9YligMAMUFXu6+YirfOWcMYwf2Y+GG3UmP6wwEWbfjABAnDltC4jCuLn1xSBePR5gxsorXVu/AGGN7tZKvm3h6d4hIJObu1rvm7hb78WpYSckRR7U4TB3an/u/cDynj63p/uA0uWjyINr9wchd/vbWNop8HspLks8zADh1dDU3nDaSGSOraNp9CGMS8w8bdh7EH85LbI7zHGr6FUUqqOzm1FHVNO9rZ3Xz/pDnYOckuG4GvnToaWgq38kkIe3WvIviDEe1OBT6PJw2pobSQvtSL4P7h5LN1gC+rbWdgeXFaYWm6itKONQZYPfBzoR9ltgU+TxsDoeqAJZv3ZcTr8Hi2CGhnMSq7fvC7TPs+8pYRV09Gdg9PSyHzXe6C71FC6tLL4HiEEe1OOSCQRUhcWjaHRKH7XvbqE0RUorHEpam3QcT9q1p3gfA9OGVkbBShz/ImuZ9GeUbMqU+6vOExMG+97ZKUXuSTO5p2+98p7vQm4aVlFyh4mAzpYU+KvsURsRhW2sbdeXpiUO95XXsPpSwb03zfgaVFzOqph+bw6Gn1c376AyYnHoO/YpD7USadh+03XPoaUIaet72O9/pNueg1UpKjlBxyAGD+5eweU9oAN/W2pYyGZ3wuopQeWr8PAaANS37GVnbj/r+odDTnoOdLG7aC8DEHHoOEPo8ufAcrPfqSUdVt+ccPN3lHLRaSckRKg45oL6ihM27D7L7YCcd/mDaYaWyEh/9inwRr+PFlc387OkVBIOGNc37GVndl/qK0Htt3nOI9zfuobykgOFVfXL2WSAkWk27D4W6stqacwi9V08GtUgLDpeKQ3dhs5iEtEuvgeIMeTkJrrdTX1HCiyubu8pY0wwriQj14bt0gPteW8erq3cA0NYZZGRN30hOY/OeQ7y/aQ+Th1T0qGlfOgzuH/o8xtib+LUjId1dS+veTneht5icg0uvgeIM6jnkgMH9S2jrDLJ8ayiJnK7nYL22afdBgkHDB5v2IAJ3hyfVjazpG0kQr96+j1XN+3rUsC8Tm9r9oaUo7VxUx467freHlboLvcWElfTXrNiIfp1yQH3/UO5g4YZdAGknpCEcktpziHU7D9Da5ucbZ46OzJEYVdOXyj6FFPk8PL10G8bA5KEVttsfz+D+Xe3I7Yxre2y46/e4PKzUXegtpm25eg6KjWhYKQdYd/cL1u9GBKr7pT9BbXD/Uva1+XllVahlxbkTBzJ5SAUvrGiOLCpUX1HCks2hmdGTB1fYa3wymypLIn/b2cPIjjJU647aveIQetTeSsqRRsUhB1glqaub91PdryhpC+/uXvv44q30LfIxorovo2v7cero6phjPtpxgIYBpRmvQpcNltiBvT2M7ChDdX9COuw5pFxDWj0HJTdoWCkHlJcU0K84pLvplrFaWBPhFm7YzbGDy5MOeoPCLb+PRL4BuuY6gL2egx0Dux2hqXzGGvBTew7Rf7vzGijOoOKQI6y77XQrleJfBzAlRT7Bqlg6UuIAXXbZ6Tl0t1ZBWu9hCYxrF/uxHtPJORwJi5SjBRWHHGF5AJl6DpV9Cikp8AIweUj/pMcMGxBKEE8dlnx/LrA+Ty7Wc+jJe7q9lLW7hHSkeaGQ85Jm5ehCcw45wqrwydRzsOY6rGnen9IzOG/SQMpKGjn2CCSjLazPY+88By1l7Y5uE9Iuz7kozqHikCOsMEwmcxwshlWW0u4PpKxyKvJ5OX1s4sp1ucTyHHKxTGhPQlWu760kh79GHhtCc4qSDBWHHJFtWAngPy4Yz4EOv90m9YiI55CDhLQtXVldOjh6u+lca0fzQkVJhiM5BxG5XURWiMhiEfmniFRE7btFRNaIyEoROccJ++xg5pgavn32aKYPr8z4tQ1VfZgwKPXazk5g9W9KtZxqNthx1+sVQcTNy4SGHlN3ZbUe3fn5FedwKiE9H5hojDkWWAXcAiAi44HZwATgXOBuEfE6ZGOPKCn08pXTR1Hoc0fOf2RNX/55/UnMHFPd/cFpElmroAeVRh6Pe70GSKMra8RzOGImKUcJjoxcxphnjTFW3OQtYHD474uAvxtj2o0x64A1wHQnbFQSmTK0Pz4be3bbUsrqEVffNXcXNnN7Ql5xjny4rb0GeCr8dz2wKWpfU3hbAiJynYgsEJEFLS0tOTZRyQV29EXyejyuHhg93SWktVpJyRE5S0iLyHPAwCS7vmeMeSx8zPcAP/BX62VJjjfJ3t8YMweYA9DY2Jj0GCW/6W4JzHTwirvDSt3NBdFqJSVX5EwcjDFnHm6/iFwNXACcYYyxBvcmYEjUYYOBLbmxUHGakTV9GVpZGpnxnQ1ej7h2djR0iULKhLR6DkqOcKpa6Vzgu8CFxpiDUbvmAbNFpEhEhgOjgHecsFHJPaNq+/HKTadR2YPmgR4RV3sO3SWkI9VMLr4GijM4Nc/ht0ARMD885f8tY8yXjDFLReRhYBmhcNMNxpiAQzYqvQCf190JaW838xi6chJHzCTlKMERcTDGjDzMvh8DPz6C5ii9GI+4XBy6KfeNJKTVc1BsRu83lF6N1yOuDql011jQjhYkipIMbZ+h9GrOmziQUTV9nTYjZ9SVl1Do86Rc1Mnt61kozqHioPRqzp1Y57QJOeXEYyp5/z/PorQw+U9Vq5WUXKFhJUXJY0QkpTCAVispuUPFQVF6MaLVSkqO0K+UovRivFqtpOQIFQdF6cVotZKSK1QcFKUXo9VKSq5QcVCUXkx3y4gqSraoOChKL8bSBPUcFLtRcVCUXoxHF/tRcoSKg6L0YjQhreQKFQdF6cXoGtJKrlBxUJReTHeN+RQlW1QcFKUXY3kMGlZS7EbFQVF6Mbqeg5IrVBwUpRfj1WolJUeoOChKL0arlZRc4ag4iMi3RcSISFXUtltEZI2IrBSRc5y0T1Hyna6wksOGKK7DscV+RGQIcBawMWrbeGA2MAEYBDwnIqONMQFnrFSU/CaSkNacg2IzTnoOvwRuAkzUtouAvxtj2o0x64A1wHQnjFOU3oCGlZRc4Yg4iMiFwGZjzAdxu+qBTVHPm8Lbkr3HdSKyQEQWtLS05MhSRclvtFpJyRU5CyuJyHPAwCS7vgfcCpyd7GVJtpkk2zDGzAHmADQ2NiY9RlHcjnoOSq7ImTgYY85Mtl1EJgHDgQ/CSxwOBhaJyHRCnsKQqMMHA1tyZaOi9Ha6SlkdNkRxHUf8K2WM+dAYU2OMaTDGNBAShKnGmG3APGC2iBSJyHBgFPDOkbZRUXoLVjRJw0qK3ThWrZQMY8xSEXkYWAb4gRu0UklRUqOL/Si5wnFxCHsP0c9/DPzYGWsUpXehjfeUXKGRSkXpxXg0Ia3kCBUHRenFRMJK6jkoNqPioCi9mMg8B/0lKzajXylF6cV4wr9gzTkodqPioCi9GK1WUnKF49VKiqJkj8/r4dZZYzl9bI3TpiguQ8VBUXo51506wmkTFBeiYSVFURQlARUHRVEUJQEVB0VRFCUBFQdFURQlARUHRVEUJQEVB0VRFCUBFQdFURQlARUHRVEUJQExpvcvvywiLcCGDF9WBezIgTl2o3baS2+wszfYCGqnnThl4zBjTHWyHa4Qh2wQkQXGmEan7egOtdNeeoOdvcFGUDvtJB9t1LCSoiiKkoCKg6IoipLA0SwOc5w2IE3UTnvpDXb2BhtB7bSTvLPxqM05KIqiKKk5mj0HRVEUJQUqDoqiKEoCR6U4iMi5IrJSRNaIyM1O22MhIkNE5EURWS4iS0Xka+HtlSIyX0RWhx/754GtXhF5T0Qez2MbK0TkERFZEb6mH8s3O0XkG+H/6yUi8pCIFOeDjSJyn4g0i8iSqG0p7RKRW8K/p5Uico7Ddt4e/j9fLCL/FJGKfLQzat+3RcSISJXTdkZz1ImDiHiBu4DzgPHAZ0RkvLNWRfAD3zLGjANOBG4I23Yz8LwxZhTwfPi503wNWB71PB9tvBN42hgzFjiOkL15Y6eI1AM3Ao3GmImAF5idJzb+CTg3bltSu8Lf0dnAhPBr7g7/zpyycz4w0RhzLLAKuCVP7UREhgBnARujtjlpZ4SjThyA6cAaY8xHxpgO4O/ARQ7bBIAxZqsxZlH4732EBrN6QvY9ED7sAeBiRwwMIyKDgfOBe6M255uNZcCpwB8BjDEdxpg95JmdhJbqLRERH1AKbCEPbDTGvALsitucyq6LgL8bY9qNMeuANYR+Z47YaYx51hjjDz99Cxicj3aG+SVwExBdGeSYndEcjeJQD2yKet4U3pZXiEgDMAV4G6g1xmyFkIAATq8m/ytCX+hg1LZ8s/EYoAW4Pxz+uldE+pBHdhpjNgM/J3TXuBXYa4x5Np9sjCOVXfn8m7oGeCr8d17ZKSIXApuNMR/E7coLO49GcZAk2/KqnldE+gKPAl83xrQ6bU80InIB0GyMWei0Ld3gA6YC9xhjpgAHyI9QV4RwzP4iYDgwCOgjIlc6a1VW5OVvSkS+RyhU+1drU5LDHLFTREqB7wH/mWx3km1H3M6jURyagCFRzwcTcuXzAhEpICQMfzXGzA1v3i4ideH9dUCzU/YBM4ALRWQ9oZDc6SLyIPllI4T+n5uMMW+Hnz9CSCzyyc4zgXXGmBZjTCcwFzgpz2yMJpVdefebEpGrgQuAK0zXZK58snMEoZuCD8K/pcHAIhEZSJ7YeTSKw7vAKBEZLiKFhBI/8xy2CQAREUIx8uXGmDuids0Drg7/fTXw2JG2zcIYc4sxZrAxpoHQtXvBGHMleWQjgDFmG7BJRMaEN50BLCO/7NwInCgipeH/+zMI5ZnyycZoUtk1D5gtIkUiMhwYBbzjgH1AqBoR+C5woTHmYNSuvLHTGPOhMabGGNMQ/i01AVPD39v8sNMYc9T9A2YRqmJYC3zPaXui7DqZkPu4GHg//G8WMIBQdcjq8GOl07aG7Z0JPB7+O+9sBCYDC8LX8/+A/vlmJ/BDYAWwBPgLUJQPNgIPEcqDdBIauK49nF2EQiRrgZXAeQ7buYZQzN76Df0uH+2M278eqHLazuh/2j5DURRFSeBoDCspiqIo3aDioCiKoiSg4qAoiqIkoOKgKIqiJKDioCiKoiSg4qAc1YhIQETej/p32FnUIvIlEbnKhvOuj+7CmcHrzhGR20Skv4g82VM7FCUVPqcNUBSHOWSMmZzuwcaY3+XQlnQ4BXiRUFPB1x22RXExKg6KkoRwS4N/AKeFN33WGLNGRG4D9htjfi4iNwJfItS/Z5kxZraIVAL3EWr8dxC4zhizWEQGEJoIVU1otqtEnetKQq27Cwk1WrzeGBOIs+dyQq2njyHUj6kWaBWRE4wxF+biGihHNxpWUo52SuLCSpdH7Ws1xkwHfkuoE208NwNTTGjdgC+Ft/0QeC+87Vbgz+HtPwBeM6EmgPOAoQAiMg64HJgR9mACwBXxJzLG/INQb6glxphJhGZUT1FhUHKFeg7K0c7hwkoPRT3+Msn+xcBfReT/CLXngFALlE8CGGNeEJEBIlJOKAx0aXj7EyKyO3z8GcA04N1QeyVKSN1obxShlgoApSa05oei5AQVB0VJjUnxt8X5hAb9C4H/EJEJHL7dcrL3EOABY8wthzNERBYAVYBPRJYBdSLyPvBVY8yrh/0UipIFGlZSlNRcHvX4ZvQOEfEAQ4wxLxJa+KgC6Au8QjgsJCIzgR0mtCZH9PbzCDUBhFADu0+JSE14X6WIDIs3xBjTCDxBKN/wM0INIyerMCi5Qj0H5WinJHwHbvG0McYqZy0SkbcJ3UR9Ju51XuDBcMhIgF8aY/aEE9b3i8hiQglpq8X1D4GHRGQR8DLhNYONMctE5PvAs2HB6QRuADYksXUqocT19cAdSfYrim1oV1ZFSUK4WqnRGLPDaVsUxQk0rKQoiqIkoJ6DoiiKkoB6DoqiKEoCKg6KoihKAioOiqIoSgIqDoqiKEoCKg6KoihKAv8fBXQOPQFdcLMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the scores\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "plt.plot(np.arange(1, len(scores)+1), scores)\n",
    "plt.ylabel('Score')\n",
    "plt.xlabel('Episode #')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75403f40",
   "metadata": {},
   "source": [
    "### Animate it with Video"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7fae8a12",
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_video(env_name):\n",
    "    mp4list = glob.glob('video/*.mp4')\n",
    "    if len(mp4list) > 0:\n",
    "        mp4 = 'video/{}.mp4'.format(env_name)\n",
    "        video = io.open(mp4, 'r+b').read()\n",
    "        encoded = base64.b64encode(video)\n",
    "        display.display(HTML(data='''<video alt=\"test\" autoplay \n",
    "                loop controls style=\"height: 400px;\">\n",
    "                <source src=\"data:video/mp4;base64,{0}\" type=\"video/mp4\" />\n",
    "             </video>'''.format(encoded.decode('ascii'))))\n",
    "    else:\n",
    "        print(\"Could not find video\")\n",
    "        \n",
    "def show_video_of_model(agent, env_name):\n",
    "    env = gym.make(env_name)\n",
    "    vid = video_recorder.VideoRecorder(env, path=\"video/{}.mp4\".format(env_name))\n",
    "    agent.load_state_dict(torch.load('checkpoint.pth'))\n",
    "    state = env.reset()\n",
    "    done = False\n",
    "    while not done:\n",
    "        vid.capture_frame()\n",
    "        \n",
    "        action = agent.act(state)\n",
    "        next_state, reward, done, _ = env.step(action)\n",
    "        state = next_state\n",
    "        if done:\n",
    "            break        \n",
    "    env.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "c8a347e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "agent = Agent(env).to(device)\n",
    "show_video_of_model(agent, 'MountainCarContinuous-v0')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "7fd10379",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<video alt=\"test\" autoplay \n",
       "                loop controls style=\"height: 400px;\">\n",
       "                <source src=\"data:video/mp4;base64,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\" type=\"video/mp4\" />\n",
       "             </video>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_video('MountainCarContinuous-v0')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ab1adc92-90a4-47a1-9ab8-65d5e947bec7",
   "metadata": {},
   "source": [
    "> Note: While I tried to execute the model with VideoRecorder in Linux, it doesn't show correctly. But in windows, it works! Perhaps, It may be bug in VideoRecoder."
   ]
  }
 ],
 "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": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}