le_probleme_de_monty_hall.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "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": "3.7.6"
    },
    "colab": {
      "name": "le probleme de monty hall.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Oumar199/OpenclassroomProject/blob/main/le_probleme_de_monty_hall.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sckv3USkZ9gY"
      },
      "source": [
        "#  Le probleme de monty hall\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "r4nHKUwaan1Z"
      },
      "source": [
        "## Simulation "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9iuRhogIa1eL"
      },
      "source": [
        "### Génération d'une seule partie de jeu"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_VvLCgqtcuXu"
      },
      "source": [
        "Préparons un environnement  de travail dans notebook :"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "STfy1pznY8EJ"
      },
      "source": [
        "#Pour afficher les graphiques dans la continuité du code et non dans une fenetre a part :\n",
        "%matplotlib inline\n",
        "\n",
        "#Pour utiliser la fonction randint qui permet de generer des nombres entiers aleatoires :\n",
        "from random import randint, seed\n",
        "\n",
        "#Un Enum est une structure de donnees qui consiste en un ensemble d'elements nommes, \n",
        "#Une variable de ce type peut avoir comme valeur un de ces elements\n",
        "from enum import Enum\n",
        "\n",
        "#Pour pouvoir afficher des graphiques\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a7DqE2vCc4AL"
      },
      "source": [
        "### Les stratégies du joueur"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hrbZQN5rY8ET"
      },
      "source": [
        "# Ici nous définissons une sous-classe de Enum, qui contiendra \n",
        "# les stratégies possibles\n",
        "class Strategie(Enum):\n",
        "  CHANGER = 1\n",
        "  GARDER = 2"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lbmtNpCXfSeR"
      },
      "source": [
        "# Utilise l'horloge système pour initialiser le générateur de \n",
        "# nombres pseudo-aléatoires.\n",
        "seed()\n",
        "\n",
        "def play_game(strategie):\n",
        "  '''Simule une partie du jeu Monty Hall.\n",
        "\n",
        "  Cette fonction simule le choix de la porte par le participant,\n",
        "  l'élimination d'une mauvaise porte par le présentateur, et le \n",
        "  choix final. Elle ne retourne que le résultat de la partie, parce\n",
        "  que nous n'aurons besoin que du résultat pour effectuer nos calculs.\n",
        "\n",
        "  Args:\n",
        "    strategie (Strategie): La stratégie du joueur\n",
        "\n",
        "  Returns:\n",
        "    bool: le joueur a-t-il gagné?\n",
        "  '''\n",
        "  portes = [0, 1, 2]\n",
        "  \n",
        "  bonne_porte = randint(0, 2)\n",
        "\n",
        "  # Choix du joueur\n",
        "  premier_choix = randint(0, 2)\n",
        "\n",
        "  # Il nous reste deux portes\n",
        "  portes.remove(premier_choix)\n",
        "\n",
        "  #Le présentateur élimine une porte\n",
        "  if premier_choix == bonne_porte:\n",
        "    portes.remove(portes[randint(0, 1)])\n",
        "  else:\n",
        "    portes = [bonne_porte]\n",
        "\n",
        "  deuxieme_choix = 0\n",
        "  #Le deuxieme choix depend de la strategie\n",
        "  if strategie == Strategie.CHANGER:\n",
        "    deuxieme_choix = portes[0]\n",
        "  elif strategie == Strategie.GARDER:\n",
        "    deuxieme_choix = premier_choix\n",
        "  else:\n",
        "    raise ValueError(\"Stratégie non reconnue!\")\n",
        "\n",
        "  return deuxieme_choix == bonne_porte"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BWa4KnZ2kIKq"
      },
      "source": [
        "## Simulation de plusieurs parties"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yaCVq3p1kTsC"
      },
      "source": [
        "Pour vérifier si le résultat de la fonction est aléatoire appelons la plusieurs fois"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FvoNgxC-gQWx",
        "outputId": "389bed82-36ed-4191-e50f-689e27e6c750"
      },
      "source": [
        "play_game(Strategie.CHANGER)"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "m5OCAL-0lK6R"
      },
      "source": [
        "Nous allons stocker à présent les résultats dans une liste. Donc créons une fonction nous permettant de le faire"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VpCKG78lk2hl"
      },
      "source": [
        "def play(strategie, nb_tours):\n",
        "  '''Simule une suite de tours du jeu.\n",
        "\n",
        "  Cette fonction renvoie les résultats de plusieurs parties \n",
        "  du jeu Monty Hall sous forme d'une liste de gains par le \n",
        "  joueur.\n",
        "\n",
        "  Args:\n",
        "    strategie (Strategie): La strategie du joueur\n",
        "    nb_tours (int): Nombre de tours\n",
        "\n",
        "  Returns:\n",
        "    list: Liste des gains du joueurs à chaque partie\n",
        "  '''\n",
        "\n",
        "  #Ceci est une liste en compréhension\n",
        "  return [1 if play_game(strategie) else 0 for i in range(nb_tours)]"
      ],
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9tTCwvGam2uA"
      },
      "source": [
        "## Analyse des résultats\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ebrm0uiwnEPd",
        "outputId": "3c14cbce-905b-4da8-b2be-620725b6cabb"
      },
      "source": [
        "print(\"En changeant de porte, le joueur a gagné {} sur 10000 parties.\".format(sum(play(Strategie.CHANGER, 10000))))\n",
        "print(\"En gardant son choix initial, le joueur a gagné {} sur 10000 parties.\".format(sum(play(Strategie.GARDER, 10000))))"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "En changeant de porte, le joueur a gagné 6642 sur 10000 parties.\n",
            "En gardant son choix initial, le joueur a gagné 3382 sur 10000 parties.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hRp_dyH5oKeC"
      },
      "source": [
        "## Visualisation"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zQ89x13FoZYS"
      },
      "source": [
        "### La fonction *plot*\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "vglw9EXYndXX",
        "outputId": "71d09a94-f006-4d9a-e366-cce07138fe4f"
      },
      "source": [
        "#plot renvoie un objet, que l'on pourra manipuler plus tard pour\n",
        "#personnaliser le graphique\n",
        "plot = plt.plot(play(Strategie.CHANGER, 10))"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO2da5Ak2VXf/6feXVXZ3dOPytp59jwqUx5hQPKEEAgbAkF4hWH1AezQRoAxIVgIIwyGsEPYWHbInzAOjO2QCS8YsAEjC5mw13hBDoNsYxvJmpVAaHep6p7ZnkfvVPZruierqut9/SErq2t6qrurujIr7715fxEbO12d03Unu+4/7z33f84hxhgUCoVCIT6RoAegUCgUCm9Qgq5QKBSSoARdoVAoJEEJukKhUEiCEnSFQqGQhFhQb7y0tMRWVlaCenuFQqEQktdee22bMbY87HuBCfrKygpu374d1NsrFAqFkBDRveO+p0IuCoVCIQlK0BUKhUISlKArFAqFJChBVygUCklQgq5QKBSScKqgE9EvE9EmEX3lmO8TEf0LIlojoi8T0bu9H6ZCoVAoTmOUFfqvAnj+hO9/AECh999LAH5h8mEpFAqFYlxO9aEzxv4XEa2ccMkHAfw75tTh/RwRzRPRc4yxRx6N8Sm+sL6LPyxt+fGjx+ZrL8/jW96hBz0MLvi9rzzC1146h/xcKuihBE6l0cbv/ukjfPdfuAgiCno4gfPavV38zyIHc5YIL3zNedzIZYMeiW94kVh0AcCDga8f9l57RtCJ6CU4q3hcvnz5TG/2xXuP8S8/u3amv+sljAGLmQRe+wffFvRQAmf/oIUf/vUv4ge+8Sp++jtuBj2cwPntLz7Ex/7z67h5fhbvPD8X9HAC5+O/8yb+5MEegn62MQbc36ni5z/0rmAH4iNTzRRljL0M4GUAuHXr1pk6a/zQN13HD33TdU/HdRZ++X+/hY//zhvYrjSwlE0GPZxAWbVsAECx9/+w82dl5z6ULDv0gt7tMqxaNr7/fSv4h9/5zkDH8v2/8v9QtCqBjsFvvHC5bAC4NPD1xd5rUmPoGgCgVFYi5gp5SQk6gMPPRLEst3iMwsbeAWrNTn++BImha7izWUG70w16KL7hhaC/AuCv99wu7wWw71f8nCeMvBOHU6vSQwGznjSwV2sGPJpgYYypB9wAxd5ngxdBb3a6WN+pBT0U3xjFtvibAP4IgElED4now0T0w0T0w71LXgVwF8AagF8E8Dd9Gy1HLGeTOJeOq0kL56EWjTgB0pLkW9rTKD+pw663EY1QX8zCjPtwM/TgDyLNfG9XLfGcHcXl8uIp32cAfsSzEQkCEcHQtdALGACsWhV8w/VF/OHqNkqWjfdcXQh6SIHhfh7c+1FptJFNBlbUNHBWLRsX5megpeJBDwU3clkQOYL+7X/+uaCH4wsqU3QCzLyGUtmG80wLJ9uVBnaqTXyzmUM2GZN69TMKbvjpO7/6PIDDA+OwUrQqXKzOASAVj2JlMSP1Z1QJ+gQUdA12o41H+/WghxIYroCZuoaCng19mKFo2VjWkv1diszicRrtThd3NitcxM9dCjm5P6NK0CfA7H1Qw3ww2o+R5rMwdQ0lK9w7lpJlw9Q1XFpIIxWPhNrpsr5TQ7PT5UrQzbyG9Z0a6q1O0EPxBSXoE+BuJcNsXSxZNs6l41jOJmHoGh7XWtiqNIIeViA4nmtnRRqNEAo5LdQrdPff7h5G8oCha+h0Ge5uVYMeii8oQZ+A+XQC+mwy1AejpZ6AEVF/4q6G9H48fHyAg1YHZs/SauhK0InAVap9/zO6KefvRQn6hIR50jLGUCrb/Unibq1ljlGexKFFz7kPZj6LTbuBx9VwevNLlo2VxQxS8WjQQ+mzsphBPCqvpVQJ+oQYuobVTRudbvjixo/267AbbRR6AraUTYTam+/+u9374f4/rPejWLZR4Gh1DgCJWARXl+R1uihBnxBT11BvdfFgV97ss+NwV6Tu4bDrzQ/rIXGx7HiuXd+5GWJBr7c6WN+pcRU/d5H5M6oEfUKMfHidLqXys1mAZl7DqlUJpdOlZNlPCdhzcyloyVgoPxt3t6rodBlXDhcXU9fwYPcA1UY76KF4jhL0CXG3lGFMIClZFeizScynE/3XDF1DpdHG2yHz5rc6Xdzdqj4lYEQEIx/ObGL30JHLFXpvTGub8v1elKBPSCYZw6WFGenLcg6jZNnPrMD69TIkPXQ6jns7VTQ73b7DxcUIqTe/WLYRjxJWFjNBD+UZZM4fUYLuAUZOC52AdboMq5vPCrqRk3eynISbQFTIHbkfehZ7tRa27HB580uWjatLGSRi/EnMpYU0krGIlHOWv7stIEZew93tCloS11k+yoPdGuqtbn+14zKXjjvefAkny0kULRuRIZ5rmVeDJ1EcsnvjhWiEnDIVEv5OlKB7gKlraHUY1rflzD4bRqmf8v/spDV0DSVJEzeOY/UYz7XRL9kanpBcrdnGg92DZx72PCFr/ogSdA8wQrgK63uuh/iMTd1xuoTJm3/cinQpm8RiJhGqHYubKTzsYc8Lpq7BetLAfq0V9FA8RQm6B1xbziAaoVBN2qJVwaWFGWSG1Po28hoa7S7uh8SbX291sL5dPVbAZPY9D+NofgKP9HdOku0klaB7QCoexZXFdKgmbals9w9AjxK2EgB3tirosuO78hh6FquWjW5Idiylso1kLIJLC+mgh3Issn5GlaB7hBmi7kWtThd3tyvHrkjdMIyMMcphlE5ZkRp5DdVmBxt7B9McVmAULRsFPdtvS8gj5+dSUjZkUYLuEYau4d5OVdo6y4Osb1fR6rBjBcz15ss2WY6jZFUcz/XScM+1e59krfB3FLeEMM84ZSrka3ahBN0jzLyGLpMz++woR6sKDsOU1EUwjFLZxvXlLOLR4dOp0N/ey//Z2K+1UH5S5zp+7mLm5Uv6UoLuEa64hUHESmUb0Qjh2vLxWYCGruHuVhXNtvze/NM813MzcTw3lwrHZ2PzeDsrbxRyTkOW7Yo85Y2VoHvEymIaiWgkFAejRcvGlcX0iXWuDV1Du8vwluTe/EqjjYePD05thFzQNem298Molk/fvfFCv0yFRHNWCbpHxKIRXFvOhMK6WLIqp26pw+LNXx0h/AQApp7F2pb83vySZSObjOH8XCrooZyKjE4XJegeYoagsl691cG9neqpAuZ682WvQukm0ZxWVdDQNTTbXdzbkXvH4hRsy4KIX4eLy1I2gYVMQq3QFcMxdA0bewew63Jlnw2ytul4rk8TsFQ8ipXFtFSrn2EULRupeASXzp3suZZxe38UxhiKZZvLkrnDcJ0uMv1OlKB7yKE9Td5VemnEEANw6CKQGbeEcOQUz/WNXBZEcjtdtitNPK61hIifuxi9/BFZnC5K0D2k73SReFVatGwkohGsLJ6eBVjIabi3W8NBU15vvtM383QBSydiuHQuLfUDbpyHPS/I1pBFCbqHXDw3g5l4VOqDwFLZxrXlDGLHeK4HMfMamMTe/L1aE5t245mmFsche00XkRwuLrI1ZFGC7iGRCPXqdsgpYEDP4TJijFR2b757AD6qgJn5LNa3q2i05dyxrG7aWMgksJRNnH4xJ8jWkEUJusfIvAqz6y1s7B2MLGCuN19WQe9XFRzjASezN79YFsfh4jKXjiM/K0/SlxJ0jzF0DVt2A7tVebLPXNzD3lEFPRaN4HpOzs4wgLNN15Ix5GdH81y7wi+j84cxhpIANVyGUZDI6TKSoBPR80RUJKI1IvrokO9fJqLPEtGXiOjLRPTt3g9VDAyJ7WlunHGcOh2GnpUmPnmUomXDyGsjr0ivLvXq5kv42Xh7v45Koy2koMvUkOVUQSeiKIBPAPgAgJsAXiSim0cu+2kAn2KMvQvAhwD8K68HKgqmxHHjomVjJh7FxXMzI/8dQ9fw9n4dTyTz5jsr0vH6ZiZjUVxdykhpXew/7AXxoA8iU0OWUVbo7wGwxhi7yxhrAvgkgA8euYYBmO39eQ7A294NUSz02SRmU/LVWQbcsqjZUz3Xg/S9+ZIdFG9VGtirtWCeUsPlKKauSVlGt29ZHMHCyRsyLcJGEfQLAB4MfP2w99og/wjA9xDRQwCvAvjRYT+IiF4iottEdHtra+sMw+UfInISaiRchZ2lk7usGZLu73fcqoKGruH+bg21ZtuPYQVG0bKRn01hLh0PeihjU+g9lGUIDXp1KPoigF9ljF0E8O0Afo2InvnZjLGXGWO3GGO3lpeXPXpr/ij0nC6yZJ8BwG61iS27MbagX5ifQToRle4gcJSa8MMw81kpvfmlXpciEUknnIYsMhzejyLoGwAuDXx9sffaIB8G8CkAYIz9EYAUgCUvBigipq5h/6CFTbsR9FA8o7+lHnNFGokQCjl5XAQupbKNxUwCS9nkWH/vsNmFPPej02VYHaECJ8/I0pBlFEH/AoACEV0logScQ89XjlxzH8D7AYCI/hwcQZczpjICMpblPK1v5kkYkkyWQc4SfgKAKwtpJGJyefPv79bQaHeFaGpxHLI0ZDlV0BljbQAfAfAZAG/CcbO8TkQfJ6IXepf9JIAfJKI/AfCbAP4GkyneMCZuswOZJm3JsjGbikGfHW9FCjhx9O1KEzsVOXYsjDGsWmerKhiLRnBjOStVmeVJHva8YOadpK91wcsbx0a5iDH2KpzDzsHXPjbw5zcAvM/boYnLYjaJpWxSLkEvOyn/Z8kCPCwBUMHXjxmi4JGNvQNUm50ze67NvIbP3d3xeFTB4R4mihpDB57eVYvopXdRmaI+YehZFCVZhTHGULTsfvx3XGRzuhxWFTybgBm6hkf7dewfyOHNL1o2Li3MIJ0YaX3IJW5DFtE/o0rQfcLQNaxaNroSZJ9t2g3sH7TOvKXOaY43XwYXAXBY0/ysDzj3QSBLN6eSZQsdbgGcpC8ZGrIoQfcJM6+h1uxgY+8g6KFMzKRlUQ+9+WJPFpeSZeO5uRTmZs7muZap32qz3cXdrdNbEoqADA1ZlKD7hEylYycNMTh/15ksMpyVj5vyf5QL8zPIJKJSZM+u71TR7jIhU/6PYuhOQ5Z6S9zyxkrQfcIVPxlWYSXLxlI2icUJDjTNvIYn9TasJ2I7XTpdhtXN0WvCDyMSISf5TIIdi4hNLY7D1MVvyKIE3Se0VBzn51JShBmKvRoukyBLmOHejuNVLuQmux+yJLKULBvRCOHacibooUyMDElfStB9xMhrwjtdul3Hcz3pCkyWfqt9z/WEIYaCnsVOtYltwb35xbKNlcU0krFo0EOZGBkasihB9xFT13Bns4J2R9zss429A9SanYkFbKGXJi/6Cr1YroAIuDHpCl2SXpalMyZY8YgMDVmUoPuIoWtodrq4J3CdZS87uZv5rPBWvdKmjcsL6Yk91zKUbK23Ori3W5Mifu5iCt4TWAm6j8iwCit64HBxcZwuFaG9+SWPMgmXtSTm03GhQ3JrmxUwJnbK/1GMvIaNvQPYgjZkUYLuI9eXsyAS+yCwVLZxfi4FLTV5nWtT13DQ6uDhYzG9+Y12B29tVz15uBGR8EXLiv2Uf4kEPXdYpkJElKD7yEwiiisLabEnrVXxrIpeQXCny1vbjufaqxCD229VVG9+ybKRiEawspgOeiieIXqZCiXoPmMI7Ddud7q4s+ldnWvRq1AWPe6baeoa7EYbj/brnvy8aVO0bFzPZRGLyiMjojdkkec3wSlmXsP6Tg2NtnjZZ/d2a2h2up6tSLVUHBfmZ4QV9FWrgliEcG3Jm6qComcTO00txK2wOAw36UvUvq9K0H3G0DV0ugx3t8Srs+xHJ3dDzwq7+ilaNq4uZZCIeTNtRBZ0u97Cxt6B0E0tjsPIZfsF2ERDCbrPiDxpi5YNIudw1yuMvNMZpiWgN3/SGi5HOZdJIKclhRQP99DQPUSUCachS0PIhixK0H3m6lIGsQgJuSotWTauLKQxk/AuC9DI9bz5gnWGqTXbuO+D51pUp4tXGbM8MtiQRTSUoPtMIhbBteWMkJPWj+4trgCItirte67z3saMjV68VjRvfrFsI52I4sL8TNBD8RyRnS5K0KeAm1AjEo12B+s7Nc9XYDdyjjdftMnSDzF4/oDLot7q4sFjsbKJVzedDlaRyPgtCXknpyUxNxMX7jMKKEGfCqau4f5uDbVmO+ihjMzdrSo6HnquXVLxKFYWxduxlCwbiVgEVxa9rSo42MtSJIpl+RwuLk7SV1a4zyigBH0quAk1ItWI8LKGy1GcfqtiTZZi2caN5SyiHq9ICwIemu9UGtiuNKSq4XIUN39EtKQvJehToB83FmjSFss2YhHC1SXv61wbuob17apQnWH8qiqYTcZwYX5GqJoufoWfeELUhixK0KfA5YU0krGIUJUGS1YF15a981wPYugaugzCePOf1Ft4tF/3TcDMvCbUZ8NNupHR4eIiakMWJehTIBohFPSsYKsw7x0uLqK5CFb7Fj1/YsaGruHOVkUYb36xbGNuJo6cdvaWhLxj9MOkYnxGXZSgTwlDF6frveu59qss6spiBvEoCbP6cS2W/j3gsmh1GNa3xdixlCwbpq6BSD6Hi0u/IYsgc9ZFCfqUMHQN5Sd17Nf4r7PsHt76VRY1EYvg2lJWmAdcybKR8dFzLdL2njGGYtlGQVKHyyBmXjynixL0KdHvUCNA0Z/iFLIACwI5XRwB829Fen05iwiJ0QjFetLAk3pb6vi5i4gNWZSgTwlDoLjxqmUjGYvg8oJ/da5NXcPDxweoNvj35q9u2r525Tn05vN/xuKnnZU3RGzIogR9SpyfSyGbjAmxCitaFRR07z3Xg7gPuNVNvkVsu9LAdqXpe1VBUWq6hEnQRVqEuShBnxJu9pkIYQav+maeRD8ExfkDrl+Eyuf7YeQ1rO/w780vlm0sa0ksZBJBD8V3CjnnnECEOesykqAT0fNEVCSiNSL66DHX/DUieoOIXieif+/tMOVAhOyz/VoL5Sf+ea5dLi2kkYpHuJ8s7gPHiz6iJ2H2vPlrnO9YHDur/AeigJgNWU4VdCKKAvgEgA8AuAngRSK6eeSaAoCfAvA+xtg7Afy4D2MVHkPX8LjWwnalGfRQjsU9tPV7RRqNEG7k+HcRFK0K5tNxLPvsuRahPV+3y1CyKqEIt7iI1pBllBX6ewCsMcbuMsaaAD4J4INHrvlBAJ9gjD0GAMbYprfDlAPXGcBzskI/RjoFF4MIcePVXoKV357rlSXHm8/zwejG3gEOWh3fH/Y84TZkaQuS9DWKoF8A8GDg64e91wYxABhE9H+I6HNE9PywH0RELxHRbSK6vbW1dbYRC4wIfuNS2UY2GcP5uZTv72XqGqwnDezV+NyxMMZQtPx1uLjEoxFcX+Z7x+KuVGVsO3ccpu40ZFnfEaO8sVeHojEABQDfDOBFAL9IRPNHL2KMvcwYu8UYu7W8vOzRW4vDUjaBhUyC70nbi5FOIwvw0EXA56q0/KQOu96emoC5Zyy84i5E3MPCMCBaC8lRBH0DwKWBry/2XhvkIYBXGGMtxthbAEpwBF4xABGhkOM3JudmAU4rRmpyvmPpr0inJGBmXsPG3gHsOp/ZxCXLxoX5GWipeNBDmRpuQxZe5+xRRhH0LwAoENFVIkoA+BCAV45c85/grM5BREtwQjB3PRynNDiV9SpcOl22K008rrWmJujPzaWgcezNn7bn2l358urNdx724VmdA+I1ZDlV0BljbQAfAfAZAG8C+BRj7HUi+jgRvdC77DMAdojoDQCfBfB3GGM7fg1aZAxdg91o49F+PeihPMPqFFL+ByFyqlDyOllKVgU5LYlzU/Jc83xo3u50cXerGqr4uYso+SOAE/s+FcbYqwBePfLaxwb+zAD8RO8/xQkMNrs4z1mD3WIAWYBmXsPvfaUMxhh31fv8ampxHJfO9bz5HDbQXt+podnphsrh4mLqGv77m5uotzpIxaNBD+dEVKbolDFy/GZIliwb59JxLGWnlwXoevO3Knx1hnE81zYKuekJWCRC3Fo5w5Tyf5SCrqHTZUI0ZFGCPmXm0nHos0kut3Dugeg0V8qHJQD4WpU+eFxDvdX1ranFcRi6xu1ng8g5JAwbIjVkUYIeADyuwhhjWLUqUy+LanDab7XvcJnyitTUNWzZDTyu8uXNL1k2VhYz3Icc/ECkhixK0APA1DWsbVbQ4ajO8qP9OuxGe+oCtpR1Cj3xdhDoOk38avJxHAVOSwCEqYbLUURqyKIEPQCMvIZ6q4sHu/xkn02jqcVx8OgiKJZtXDw3g2xyJN+AZ/C4va+3Oljf8a8loQgYeU2I5jRK0AOAxxIA/aqCUzwEdDF7/VZ58ub72ST7JPKzKWipGFefjbtbVXS6bOq7FZ4wclk82OW/IYsS9ABwE0h42sIVLRv6bBJz6elnARp5DdVmBxt7fHSGaXW6uLMVTFVBIuo94Pg5JC4FuHvjBVEasihBD4BMMoZLCzNcrcKCWpECA04XTu7H+nYVrQ6busPFxcg7ThdedixFy0Y8SlhZzAQ9lMAQpSGLEvSAMHWnBAAPdLoMa5uVwGKkhb6g83E/3HEE9YAzclnsH7SwZfPhzV+1bFxbyiIRC69ciNKQJby/oYAxdA13tipotoOvs/xg1/FcB5XWPTcTR342xc3qp2jZiBBwfTm4Fbo7Dh4oWnYoU/4HEaUhixL0gDB0De0uw/pO8NlnQaT8H8UNM/BAqRys57pfhZKDB1y10caD3YOpVZzkGR7zR46iBD0gDI4mrbsyDrLOtalnscqJNz/I8wQAWMwmsZTlo26+ewgY9hU6wH9DFkAJemBcW84gGiEuJm3RsnFpYQaZKXuuBzF0Dc12F/cC3rE4nuvgqwo6JQCCP1NwH/Zh9qC78N6QBVCCHhhOneU0F4K+agV3IOpicHIwemergi4LXsAMXcOaZaMb8I6lZNlIxSO4tJAOdBw8wHtDFkAJeqCYeS1wAWu2g/NcD8JLyvuh5zrYmLGh8+HNL/YqTkYjfJU2DgK3IQtvZSoGUYIeIIWchvWdKuqtTmBjWN+pot1lgQt6OhHD5YV04KufYrmCeJRwJWDPtftA4eEBVwhpDZejuA1ZeDj3Og4l6AFi5jUwBqwFmH0WVFXBYRi9EgBBUrJsXF/OIh4NdmoUONje79WasJ40Ag8/8YSzq+Yn6esoStADhAenS8myEY0Qri0HnwVo5rN4a7saqDd/mk2yT2I2Fcf5uWC9+f0EK+Vw6cNrQxYXJegBsrKYRiIaCbSKm1PnOs1FnWvXm//WdjBOl0qjjY29A25qlhT0YM9Y+ucJHDzgeIHXhiwuStADJBaN4Hou2DrLpQCaWhxH0FUoVzlIsBrEzGtY26qg3Qlmx1KybGjJGJ6bSwXy/jxyWKaCzzi6EvSAMfRsYKsw13M9zb6ZJ9H35gf0gDvsm8nHIWDfmx9Q3fxi2TkQ5a15d5AsZRNYyPCR9DUMJegBY+gaNvYOYNdbU3/vtc0KGOOnLGoyFsXVpUxgK/RiueJ4rs/x4bkOssIfY06TbF4+G7xARFw2ZHFRgh4wZoAJNTw5XFzMAOtluCn/EU481zdyWRAFE4LaqjTwuNbi6rPBCzw2ZHFRgh4w7gooiGSF0qaNRDSClUU+VqSAk2B0f7eGg+b0vflB13A5ykwiissL6UDKLLvvqQ5En4W3hiyDKEEPmAvzM0gnooGswkplG9dzWcQC9lwPYurBePMfV5vYtPnzXDs1Xab/2ejv3lTI5Rnchz4v/QwG4Wcmh5RIhFAIqM5yyapwcwDoElQtcPf+85YVaeoa3tquotGe7o6lZNlYyCSwlE1O9X1FwO27y2McXQk6Bxi6huKUfa12vYWNvQOuQgwAcGUhjUQsMvUHHK99M428hk6X4e7WdL35Rcvm7mHPC3NpvhqyDKIEnQPMvIbtSgO71enVWS5xGiONRSO4sTz9ehlFy4aWiiE/y5fnOoh+q4wxLipw8gxPDVkGUYLOAUYAk3aV0xUp4PjAp31IXOoJGG+e66tLGcSmXDf/7f06Ko22ip+fgKlnscZJQ5ZBlKBzgJmfvqAXLRvpRBQX5mem9p6jYuQ1vL1fx5MpefNdzzWPApaIRRxv/hRDcqqpxekUdA2Ndhf3A0r6Og4l6ByQ05KYTcWmGmYoWTYKuSw3nutBTH26Vs4tu4G9WovbvplGfrre/GL/gFgJ+nHw1Pd1kJEEnYieJ6IiEa0R0UdPuO67iIgR0S3vhig/RNQvyzktiuXgm1ocx2EVyumsSvtNsjlcoQOOeNzfraHWbE/l/UplG/nZFOZm4lN5PxHhpSHLUU4VdCKKAvgEgA8AuAngRSK6OeQ6DcCPAfi814MMA0avst40ss92q01sVxpcxs8Bx5ufSUSnNll4PSB2cR9w0/Lmlzb5DD/xBC8NWY4yygr9PQDWGGN3GWNNAJ8E8MEh1/1jAD8DoO7h+EKDmdewf9DCpu1/neXDIlR8TtpIhHBjiiUASmUbS9kEFjn1XLv2wWls7ztd1+HCZ/iJJ3hoyHKUUQT9AoAHA18/7L3Wh4jeDeASY+y/nvSDiOglIrpNRLe3trbGHqzMTLPZBa+e60FMfXrJVkXOUv6PcmUxMzVv/v3dGhrtLtf3gxcMPfiGLEeZ+FCUiCIAfg7AT552LWPsZcbYLcbYreXl5UnfWiqmaV0slm3MpmLIaXyuSAHnfmxXnNCQn3S7DKucC3q0l01cnEKqOY8F23jFzAfbkGUYowj6BoBLA19f7L3mogH4KgD/g4jWAbwXwCvqYHQ83DTraa3QzTx/nutBpmXl3Ng7QLXZ4V7AzClt73ktgcAjQTdkGcYogv4FAAUiukpECQAfAvCK+03G2D5jbIkxtsIYWwHwOQAvMMZu+zJiiTHzWZR8PvhyPNf8OlxcDq2L/t6P1U03/MS3gBl5DeUndewf+OvNL1k2Li+kkU7EfH0fGQi6IcswThV0xlgbwEcAfAbAmwA+xRh7nYg+TkQv+D3AMGHoGlYtG10fs8827Qb2D1pcx88BYFlLYm4m7vvqx7VG8u65dg9G/fbm81ZCmGeCbsgyjJEew4yxVwG8euS1jx1z7TdPPqxwYuoaar06y/MhK+cAABU1SURBVJcW/KlRLkqMlIimEmYoWTbOz6Uwm+Lbcz24vb+1suDLezTbXdzdquLbbuq+/HwZMfQs3nj7SdDD6KMyRTmiMAWnC++WxUGMvNPqy09vvtM3k/970ffm+/jZeGu7inaXCfHZ4AVD13AvoIYsw1CCzhF9v7GPW7hi2cZSNomFTMK39/AKU9dg19soP/EntaHTZVjbqnAffgJ6vSx9rvBXFOhhzwtBNWQ5DiXoHKGl4rgwP+NrnLS0WeH+ANDl0Mrpz2S5t+N4iEURMFPXfD0kXrVsRCOEa8sZ395DNoJqyHIcStA5w+ko7s+kFcFzPUhf0H0KM/QTrAS5HwVdw07VP29+sWzj6lIGyVjUl58vI1cW0khEp9+Q5TiUoHOGkddwZ7OCdsf77LONvQPUBPBcu5zLJLCsJX1b/RTLFRABNzitsngUcwoPOFEebrwQi0ZwPaAWksNQgs4ZRk5Ds9PF+o73dZZFcbgMYvpY08X1XM8kxFiRGnn/zlgOmh3c262phKIzYOpZbrzoStA5w88MycNDL3EmrdGLG/vhzee9hstRlrNJnEvHfflsrG1WwJg44SeemHZDlpNQgs4ZN3JZEPkj6KuWjQvzM9A491wPYuazOGh18PDxgac/t9HuYH27KpSAEVG/zLLX9O2sAjh+eGPaDVlOQgk6Z6TiUawsZnxaoVeEWp0DA958j+9H33MtmIC5JVu99uaXLBuJWARXfEpok5lpN2Q5CSXoHFLIed/1vt3p4s4m/zVcjlLI+dMZ5vA8QawHnJHXYDfaeLTvrTe/aNm4vpxFLKokYVwuzM8gPcWGLCehfnscYuY1rO/UUG95l322vlNDsyOO59rF9eZ7/YArWTZiEcK1JbEE3fRpx1Iq26qpxRmJRAiFKTZkOXEsQQ9A8SyGrqHTZbi75V2dZRGaWhyHH/1Wi+UKri45jSNEwt1ReOmqeFJv4e39unDhJ56YZkOWkxDr0xwSXNF1S7t6QcmyhfJcD2LoGu5uVdHy0Ju/KmjfzPl0Avps0tOD0VXOe6qKwLQaspyGEnQOWVnMIB4lT8MMJcvGymIGqbgYnutBDD2LZqeLezve7FhqzTbu79aEFTDD4+29SAXbeGWaHcdOQgk6hyRiEVxd8tbpUizb/QNG0fDaReB6rkU7EHUxdA2rmzY6Hnnzi2Ub6UQUF+ZnPPl5YaS/q55Cm8CTUILOKYbuXWW9equD9Z2akPFzwAkTRci7g0ARM2YHMXUN9VYXD3a9ySYuWU4J4UiE35aEvJObUkOW01CCzimmruHB7gGqjfbEP+vuVhUdgetc9735HoWg+p7rRTGrCnpd4c+p4SLmboUXptWQ5TSUoHOKO2m9qLN82DdTTEEHenFjjw6JS1YFhVwWUUFXpG7ozIvMxJ1KA9uVprAPe56YRkOW01CCzile+o2LZRvxKGFF0BUp4MS717ernnjzRa8qmEnGcPHcjCdlll23jMgPe14wfG7IMgpK0Dnl0kIayVjEky1cybKF9FwPYuQ1dBlwZ2syEds/aOHRfl2ItnMn4dX2XjlcvMPvhiyjIO4Ml5xohFDQs96s0AWrKjgM0yNb2Go/wUrsmLGR13Bnq4JmezJvftGyMTcTR05LejSy8OJ3Q5ZRUILOMV74jauNNh7sHggdYgCAlSXXmz/Z6keWvpmmrqHdZVif0JvvpPxrIBLzPIEnFnxuyDIKStA5xtQ1WE8a2K+dvc6ye6gqYlbkIPFoBNeXsxMfBK5aFWQk8Fx7kcjCGEPJsvuNMxST42dDllFQgs4xrghP4u5wVwuir9ABp5TupKufYtlJ+Rd9RXptOYMITba9t5408KTeluKzwQsFPetbQ5ZRUILOMYcZkmeftKWyjWQsgksS1Lk29SwePj5AZQJvfsmyYeTEF7BUPIqVpcxEDzj374p+QMwTpq750pBlVJSgc8z5uRSyydhEW7iiZaOgi+u5HsR9wJ017LJdaWCn2hQ+/ORiTti9qCR4xiyPeJ30NS5K0DnGaTk2WbOLkgQOF5dJ+626AiZLiMHQNazvnN2bX7RsLGtJLGQSHo8svPjVkGVUlKBzjlsL/CzZZ/u1FqwnDWkE7NK5NFLxyJlXpYd9M+U4BDTzGhg7ezbxquAJVjziV0OWUVGCzjmGruFxrYXtSnPsv+sepsoSYohECIXc2V0ERauCc+k4lrNyeK77zS7OcD+6XYaSJV5LQhEwAmx2oQSdcyaxp4leVXAYhq6defXjVhUU3eHicmUxg0Q0cqZ47cPHBzhodYQtIcwzRt77hiyjMpKgE9HzRFQkojUi+uiQ7/8EEb1BRF8mot8noiveDzWcTOJ0KVk2sskYzs+lvB5WYJj5LDbtBh5Xx9uxMMb6STSyEI9GcG35bFUo+wlWkuzeeMLUNU8bsozDqYJORFEAnwDwAQA3AbxIRDePXPYlALcYY18N4NMA/onXAw0rS9kEFjKJM6/QDT0rzYoUOPuO5dF+HXajLZ2AOWcs48fQ3fsnatMTnvG6Ics4jLJCfw+ANcbYXcZYE8AnAXxw8ALG2GcZY261/c8BuOjtMMOL63QZV8DcLEDZquj1nS5jHgSWJEqwGsTQNWzsHcCuj5dNXLJsXJifgZaK+zSy8OJ1Q5ZxGEXQLwB4MPD1w95rx/FhAL877BtE9BIR3Sai21tbW6OPMuS4fuNxnC7blSYe11pSxc8BID+bgpaMjR1mOKwqKNeKtO/NH/MBVyzL97DnhVQ8iiuLGU/q1Y+Lp4eiRPQ9AG4B+Nlh32eMvcwYu8UYu7W8vOzlW0tNQddQabTx9v7odZZlLYtKRDDy45cAKJYryGlJzKfl8lz3q1CO8YBrdbq4u1VFQbKHG08YHlVKHZdRBH0DwKWBry/2XnsKIvpWAH8fwAuMsYY3w1MAA2GGMSatjA4XF7cK5Tg7FhnDTwBw8dwMZuLRscTj3k4VzU5XuvATT5i65llDlnEYRdC/AKBARFeJKAHgQwBeGbyAiN4F4F/DEfNN74cZbtzaI+NM2pJlYyGTwFJWrhUp4NR02au1sGWPtm7odhlWN+XJmB0kEhn/jMU9rJPxfvCCVw1ZxuVUQWeMtQF8BMBnALwJ4FOMsdeJ6ONE9ELvsp8FkAXwW0T0x0T0yjE/TnEG5tJx5GdTY01aJ+VfLoeLS78K5YjujgePa6i35F2RGmPWdClZNiLkHN4p/MGL8sZnITbKRYyxVwG8euS1jw38+Vs9HpfiCEZ+9AxJx+FSwXe9+6Sza3Hp28IsG99YWDr1+n74ScKQC+Dcj9967SF2q82R6rKULBsrixmk4tEpjC6crCw6DVmm3Y5OZYoKgpFz6ix3Rqiz/PZ+HZVGW9qyqEvZJBYziZHPFGT3XBtjFi1zK3Aq/CMRi+DaUnbq7eiUoAuCkdfQaHdxf7d26rX9qoKSrkiBXgmAkQWsgovnZpBJjrQhFY5x+q3WWx2sb1elDT/xxFncWJOiBF0Qxpm0fcuiBI0cjsPMa1gd0ekie1VBfTaJ2dRodfPvblXRZfKGn3jCi4Ys46IEXRDcLfIoW7iiZSM/m8JcWt4sQEPXUG12sLF3cmeYVqeLO1sVqQWMiJwSACOkmsuaMcsjbshzmglGStAFIZ2I4fJCeqQtnNP4V+4JO2rp2PXtKlodJr2Auf1WT9uxFC0b8ShhZSkzpZGFF7Mv6NM7GFWCLhCj+I07XYZVqwJD0gNAl8KIBZAO+2bKfT9MXcP+QQubp3jzS2Ub15ayiEfV1PebSwtOQ5ZpxtHVb1UgDN2ps9xsH19n+f5uDY12V/oV+txMHM/Nne7NL5Udz/X1ZbkFfdQyy8UQ7N54ITphQ5azoARdIMy8hnaXYf2EOsthipG6JQBOomRVsLIkv+d6lBBUtdHGw8cHMCXfrfDEJA1ZzoISdIEYZRXmHprKHmIAek6XzZO9+SXJHS4ui9kklrLJEwXdrcioUv6nh6E7DVn2auO3kDwLStAF4tpyBtEInThpi5aNywtppBNyeq4HKeSyaLaP7wxTb3WwvlMNjYA5Ff6OP1MIQ34Cb4xbpmJSlKALRDIWxcpi+uQVeq+GSxgwT8mQXNusOJ7r0Ai6483vHrNjKVo2UvEILp1LT3lk4cUcKFMxDZSgC4Z5Qk2XZtupcx0WAbuRy4LoeKdL/zwhH54HXO0Eb37JslHIaYhE5CvYxivPzZ2tIctZUYIuGIau4d5ubWid5fWdKtpdFpottevNL20Onywlq4JENIIri+HwXJ9W4c/ZvYXjs8ELZ23IclaUoAuGqWtgzAknHEXmphbHYejasaufkmXj2nImNJ5rN9Q2TDz2ak1YTxqh2a3whJs/Mk5DlrMSjk+6RBROcLqULBvRCOHacjhWpIAzWd7arqLRfnbHUiyHa0WqpeI4P5ca+oBzD+VkrcDJM4auOQ1ZKv43clOCLhgri2kkopGh2+pi2cbKYhrJmNye60EM3fHmv7X9tNPFrrewsXcQmvCTi7O9H7J7C1F+Am8c9n313+miBF0wYtEIrueGN6CVtW/mSbj/3qM7lrB6rk1dw53NCtqdp7OJS2UbWjKG5+ZSAY0svLjWxWnE0ZWgC4ipZ58p+FNvdXBvtxY6Abu2lEUsQs/cj9WQrkgNXUOz08W9I3Xz3YJtMrYk5J1xG7JMghJ0ATHyGjb2DmDXW/3X1jYrYCx8ApaIRXB1KfPM6qdYrmAmHsXFczMBjSwY+t78AfFwWhKG6zyBNwr68F211yhBFxC3ccVg9lmxn/Ifvkk7rKZLqddmLWye6+vLPW/+wP3YqjTwuNYKTcIZj5j66A1ZJkEJuoAMy5AsWTYS0QhWFsOXBWjoGu7v1lBrHnaGKYZ0RTqTiOLKQvrpz0bvMC5suzeeMPKjNWSZFCXoAnJhfgbpRPSpg8CiZeN6LotYSDzXg5j57FPe/MfVJrbsRmgF7GiFP3e1rsrmBsc4LSQnIXyzXwIiEUJB17A6kCG5alVCWxb1MEOy0vt/uAXMzGtY36n1vfmrlo3FTAJL2WTAIwsvozZkmRQl6IJi6tn+h8P1XIdVwK4sZpCIHXrzw1QTfhiGrqHTZbi75Xjzwxp+4om5mTjys6c3ZJkUJeiCYugatisN7FQa/ZWpe1gaNqIRwo3lbD/MULRsaKkY9NlwrkgHa7owxlAqh6cCJ88YJxTW8wol6IIyGGY4rCoYTkEHnq5CWSpXYOrh9VxfXcogFiEUyzY29g5QbXZCu3vjCVPPntqQZVKUoAvKoNOlWLaRTkRxYT5cnutBDF3Do/069g9aoe+bmYhFcG05g5Jlhz78xBOGrp3YkMULlKALSk5LYm4mjpJlY3XTRkEPd51rt4rg/13bxv5BK/QC5njzK6ooF0ec1pDFC5SgCwoRwewl1BTL4XW4uLghqP/y5bef+jqsmD1v/h/f38NzcynMzcSDHlLouZHrlTf20emiBF1gCnoWX364j+1KI/QCdmF+BplEFL//5iYAhP4Q0F2Rf7a4qVbnnHBaQxYvUIIuMGZeQ6PtVNULu6ATOd78RruLpWwCiyH3XLvb+0a7G/rdG0+c1JDFC0YSdCJ6noiKRLRGRB8d8v0kEf2H3vc/T0QrXg9U8SyDIh5mh4uLGzcP+8MNAC4vpJGMOdNb3Q9+MPPHN2TxglMFnYiiAD4B4AMAbgJ4kYhuHrnswwAeM8ZuAPhnAH7G64EqnsWdqHMzceS0cK9IgcPMUCVgjje/0FuZq4c9PxzXkMUrRlmhvwfAGmPsLmOsCeCTAD545JoPAvi3vT9/GsD7Kawm4CmykElgWUuG2nM9iLtCVwLmYOgaiA4P4xTB4y42hrWQ9ILYCNdcAPBg4OuHAL7uuGsYY20i2gewCGB78CIiegnASwBw+fLlMw5ZMchHn38HFrKJoIfBBbdWzuEH/+JV/OV35oMeChd839ev4OZzs0gnRpnmimlwbTmD978jh/m0P3OWTqvPS0TfDeB5xtgP9L7+XgBfxxj7yMA1X+ld87D39Z3eNdvDfiYA3Lp1i92+fduDf4JCoVCEByJ6jTF2a9j3Rgm5bAC4NPD1xd5rQ68hohiAOQA74w9VoVAoFGdlFEH/AoACEV0logSADwF45cg1rwD4vt6fvxvAHzC/W3MoFAqF4ilODa71YuIfAfAZAFEAv8wYe52IPg7gNmPsFQD/BsCvEdEagF04oq9QKBSKKTLSaQlj7FUArx557WMDf64D+KveDk2hUCgU46AyRRUKhUISlKArFAqFJChBVygUCklQgq5QKBSScGpikW9vTLQF4N4Z//oSjmShhhx1P55G3Y9D1L14GhnuxxXG2PKwbwQm6JNARLePy5QKI+p+PI26H4eoe/E0st8PFXJRKBQKSVCCrlAoFJIgqqC/HPQAOEPdj6dR9+MQdS+eRur7IWQMXaFQKBTPIuoKXaFQKBRHUIKuUCgUkiCcoJ/WsDosENElIvosEb1BRK8T0Y8FPSYeIKIoEX2JiH4n6LEEDRHNE9GniejPiOhNIvr6oMcUFET0t3vz5CtE9JtElAp6TH4glKCP2LA6LLQB/CRj7CaA9wL4kRDfi0F+DMCbQQ+CE/45gN9jjL0DwNcgpPeFiC4A+FsAbjHGvgpOGXApS3wLJegYrWF1KGCMPWKMfbH3ZxvOZL0Q7KiChYguAvgrAH4p6LEEDRHNAfhLcHoVgDHWZIztBTuqQIkBmOl1VEsDeDvg8fiCaII+rGF1qEUMAIhoBcC7AHw+2JEEzs8D+LsAukEPhAOuAtgC8Cu9ENQvEVEm6EEFAWNsA8A/BXAfwCMA+4yx/xbsqPxBNEFXHIGIsgD+I4AfZ4w9CXo8QUFE3wFgkzH2WtBj4YQYgHcD+AXG2LsAVAGE8syJiM7B2clfBXAeQIaIvifYUfmDaII+SsPq0EBEcThi/huMsd8OejwB8z4ALxDROpxQ3LcQ0a8HO6RAeQjgIWPM3bV9Go7Ah5FvBfAWY2yLMdYC8NsAviHgMfmCaII+SsPqUEBEBCc++iZj7OeCHk/QMMZ+ijF2kTG2Audz8QeMMSlXYaPAGCsDeEBEZu+l9wN4I8AhBcl9AO8lonRv3rwfkh4Qj9RTlBeOa1gd8LCC4n0AvhfAnxLRH/de+3u9/q8KBQD8KIDf6C1+7gL4/oDHEwiMsc8T0acBfBGOO+xLkLQEgEr9VygUCkkQLeSiUCgUimNQgq5QKBSSoARdoVAoJEEJukKhUEiCEnSFQqGQBCXoCoVCIQlK0BUKhUIS/j8vH5ZD+KLjXQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uXlafOOHpEC8"
      },
      "source": [
        "Essayons de créer un graphique avec 10000 points."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "000OgcOEnjKA",
        "outputId": "dd143b6e-044f-47b1-bb6e-5a3a2f761b96"
      },
      "source": [
        "plot = plt.plot(play(Strategie.CHANGER, 10000))"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2SrqMmclpiTh"
      },
      "source": [
        "Le carré que vous voyez est dû au fait que  plot  relie les points que nous lui demandons de placer sur le graphe par des lignes. Ce sont ces lignes qui recouvrent toute la surface du graphe, et non pas les points que nous voudrions en fait voir."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3IwyT88_pp_p"
      },
      "source": [
        "De manière générale, ces lignes peuvent être trompeuses, parce qu'elles peuvent aussi donner l'impression que nous avons des données entre deux points, alors que ce n'est pas le cas. Pour cette raison, nous allons placer les points sur le graphe sans les relier entre eux."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CQV1wphyp24g"
      },
      "source": [
        "#### La fonction scatter"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "c74KS-azpW6A",
        "outputId": "a100a760-5594-476e-869f-a1794ca0242b"
      },
      "source": [
        "plot = plt.scatter(range(10000), play(Strategie.CHANGER, 10000))"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAPdklEQVR4nO3dfYxcV33G8e+DjQPlLTHeomC72KQG1UKIhFWUyFUblTcnQjZSX7AFItA0ltqmgoKoHFGFkv5ToKKA6gIBUl4ECSFF1EpNrRZSVYpImk0DIU4wrB3ANtAsBIJUCiHtr3/MNUzWu96xfe3xHn8/0sj3nnv2zu/MWT+eufeOb6oKSdLi97hxFyBJ6oeBLkmNMNAlqREGuiQ1wkCXpEYsHdcTr1ixotasWTOup5ekRemuu+76XlVNzLVtbIG+Zs0apqamxvX0krQoJfnmfNs85CJJjTDQJakRBrokNcJAl6RGGOiS1IgFr3JJcj3wcuDBqnreHNsDvAe4DPgx8Nqq+s++CwV4/lv/mR/99H9Pxq4l6ZTbcN5yPnHlxb3tb5R36B8BNh5l+6XAuu6xDXjfiZd1JMNcUmtu2/cQr/rgF3vb34KBXlX/Djx0lC6bgY/VwO3A2UnO7avAwwxzSS26bd/R4vXY9HEMfSVwYGj9YNd2hCTbkkwlmZqZmenhqSVJh53Sk6JVdV1VTVbV5MTEnN9clSQdpz4C/RCwemh9VdfWq6eetaTvXUrS2G04b3lv++oj0HcCr8nARcDDVfWdHvb7GPe8baOhLqkpfV/lMsplizcAlwArkhwE3go8HqCq3g/sYnDJ4jSDyxZf11t1s9zztqNdbCNJZ7YFA72qti6wvYA/7q0iSdJx8ZuiktQIA12SGmGgS1IjDHRJaoSBLkmNMNAlqREGuiQ1wkCXpEYY6JLUCANdkhphoEtSIwx0SWqEgS5JjTDQJakRBrokNcJAl6RGGOiS1AgDXZIaYaBLUiMMdElqhIEuSY0w0CWpEQa6JDXCQJekRhjoktQIA12SGmGgS1IjDHRJaoSBLkmNMNAlqREGuiQ1YqRAT7Ixyd4k00m2z7H9V5LcmuTuJPckuaz/UiVJR7NgoCdZAuwALgXWA1uTrJ/V7c+Bm6rqfGAL8Hd9FypJOrpR3qFfCExX1f6qegS4Edg8q08BT+2WnwZ8u78SJUmjGCXQVwIHhtYPdm3D/gJ4dZKDwC7gT+baUZJtSaaSTM3MzBxHuZKk+fR1UnQr8JGqWgVcBnw8yRH7rqrrqmqyqiYnJiZ6empJEowW6IeA1UPrq7q2YVcANwFU1ReBJwAr+ihQkjSaUQL9TmBdkrVJljE46blzVp9vAS8CSPJrDALdYyqSdAotGOhV9ShwFbAbuJ/B1Sx7klybZFPX7U3AlUm+DNwAvLaq6mQVLUk60tJROlXVLgYnO4fbrhlavg/Y0G9pkqRj4TdFJakRBrokNcJAl6RGGOiS1AgDXZIaYaBLUiMMdElqhIEuSY0w0CWpEQa6JDXCQJekRhjoktQIA12SGmGgS1IjDHRJaoSBLkmNMNAlqREGuiQ1wkCXpEYY6JLUCANdkhphoEtSIwx0SWqEgS5JjTDQJakRBrokNcJAl6RGGOiS1AgDXZIaYaBLUiNGCvQkG5PsTTKdZPs8fX4vyX1J9iT5ZL9lSpIWsnShDkmWADuAlwAHgTuT7Kyq+4b6rAOuBjZU1Q+S/PLJKliSNLdR3qFfCExX1f6qegS4Edg8q8+VwI6q+gFAVT3Yb5mSpIWMEugrgQND6we7tmHPAZ6T5LYktyfZONeOkmxLMpVkamZm5vgqliTNqa+TokuBdcAlwFbgg0nOnt2pqq6rqsmqmpyYmOjpqSVJMFqgHwJWD62v6tqGHQR2VtXPquoB4GsMAl6SdIqMEuh3AuuSrE2yDNgC7JzV57MM3p2TZAWDQzD7e6xTkrSABQO9qh4FrgJ2A/cDN1XVniTXJtnUddsNfD/JfcCtwJur6vsnq2hJ0pFSVWN54snJyZqamhrLc0vSYpXkrqqanGub3xSVpEYY6JLUCANdkhphoEtSIwx0SWqEgS5JjTDQJakRBrokNcJAl6RGGOiS1AgDXZIaYaBLUiMMdElqhIEuSY0w0CWpEQa6JDXCQJekRhjoktQIA12SGmGgS1IjDHRJaoSBLkmNMNAlqREGuiQ1wkCXpEYY6JLUCANdkhphoEtSIwx0SWqEgS5JjTDQJakRIwV6ko1J9iaZTrL9KP1+O0klmeyvREnSKBYM9CRLgB3ApcB6YGuS9XP0ewrweuCOvouUJC1slHfoFwLTVbW/qh4BbgQ2z9HvL4G3Az/psT5J0ohGCfSVwIGh9YNd288luQBYXVX/dLQdJdmWZCrJ1MzMzDEXK0ma3wmfFE3yOOBdwJsW6ltV11XVZFVNTkxMnOhTS5KGjBLoh4DVQ+ururbDngI8D/i3JN8ALgJ2emJUkk6tUQL9TmBdkrVJlgFbgJ2HN1bVw1W1oqrWVNUa4HZgU1VNnZSKJUlzWjDQq+pR4CpgN3A/cFNV7UlybZJNJ7tASdJolo7Sqap2AbtmtV0zT99LTrwsSdKx8puiktQIA12SGmGgS1IjDHRJaoSBLkmNMNAlqREGuiQ1wkCXpEYY6JLUCANdkhphoEtSIwx0SWqEgS5JjTDQJakRBrokNcJAl6RGGOiS1AgDXZIaYaBLUiMMdElqhIEuSY0w0CWpEQa6JDXCQJekRhjoktQIA12SGmGgS1IjDHRJaoSBLkmNMNAlqREjBXqSjUn2JplOsn2O7W9Mcl+Se5J8Psmz+i9VknQ0CwZ6kiXADuBSYD2wNcn6Wd3uBiar6vnAzcA7+i5UknR0o7xDvxCYrqr9VfUIcCOwebhDVd1aVT/uVm8HVvVbpiRpIaME+krgwND6wa5tPlcAn5trQ5JtSaaSTM3MzIxepSRpQb2eFE3yamASeOdc26vquqqarKrJiYmJPp9aks54S0focwhYPbS+qmt7jCQvBt4C/GZV/bSf8iRJoxrlHfqdwLoka5MsA7YAO4c7JDkf+ACwqaoe7L9MSdJCFgz0qnoUuArYDdwP3FRVe5Jcm2RT1+2dwJOBTyf5UpKd8+xOknSSjHLIharaBeya1XbN0PKLe65LknSM/KaoJDXCQJekRhjoktQIA12SGmGgS1IjDHRJaoSBLkmNMNAlqREGuiQ1wkCXpEYY6JLUCANdkhphoEtSIwx0SWqEgS5JjTDQJakRBrokNcJAl6RGGOiS1AgDXZIaYaBLUiMMdElqhIEuSY0w0CWpEQa6JDXCQJekRhjoktQIA12SGmGgS1IjDHRJaoSBLkmNWDpKpyQbgfcAS4APVdVfzdp+FvAx4IXA94FXVtU3+i0VPnv3Id7wqS/1vVtJGosN5y3nE1de3Nv+FnyHnmQJsAO4FFgPbE2yfla3K4AfVNWvAn8DvL23CjuGuaTW3LbvIV71wS/2tr9RDrlcCExX1f6qegS4Edg8q89m4KPd8s3Ai5KktyqBd+7e2+fuJOm0cNu+h3rb1yiBvhI4MLR+sGubs09VPQo8DDx99o6SbEsylWRqZmbmmAr99g//55j6S9KZ5pSeFK2q66pqsqomJyYmjulnn3n2E09SVZLUhlEC/RCwemh9Vdc2Z58kS4GnMTg52ps3v+y5fe5Okk4LG85b3tu+Rgn0O4F1SdYmWQZsAXbO6rMTuLxb/h3gC1VVvVUJvOL8lbz7lS/oc5eSNFZ9X+Wy4GWLVfVokquA3QwuW7y+qvYkuRaYqqqdwIeBjyeZBh5iEPq9e8X5K3nF+bMP30uSYMTr0KtqF7BrVts1Q8s/AX6339IkScfCb4pKUiMMdElqhIEuSY0w0CWpEen56sLRnziZAb55nD++Avhej+UsBo75zOCYzwwnMuZnVdWc38wcW6CfiCRTVTU57jpOJcd8ZnDMZ4aTNWYPuUhSIwx0SWrEYg3068ZdwBg45jODYz4znJQxL8pj6JKkIy3Wd+iSpFkMdElqxKIL9CQbk+xNMp1k+7jrOV5JVie5Ncl9SfYkeX3XvjzJvyT5evfnOV17kry3G/c9SS4Y2tflXf+vJ7l8vuc8XSRZkuTuJLd062uT3NGN7VPdf9NMkrO69elu+5qhfVzdte9N8rLxjGQ0Sc5OcnOSrya5P8nFrc9zkj/tfq/vTXJDkie0Ns9Jrk/yYJJ7h9p6m9ckL0zyle5n3puMcFvPqlo0Dwb/fe8+4NnAMuDLwPpx13WcYzkXuKBbfgrwNQY34X4HsL1r3w68vVu+DPgcEOAi4I6ufTmwv/vznG75nHGPb4GxvxH4JHBLt34TsKVbfj/wh93yHwHv75a3AJ/qltd3c38WsLb7nVgy7nEdZbwfBf6gW14GnN3yPDO4JeUDwBOH5ve1rc0z8BvABcC9Q229zSvwH13fdD976YI1jftFOcYX8GJg99D61cDV466rp7H9I/ASYC9wbtd2LrC3W/4AsHWo/95u+1bgA0Ptj+l3uj0Y3PHq88BvAbd0v6zfA5bOnmMG/wf/xd3y0q5fZs/7cL/T7cHg7l0P0F2AMHv+WpxnfnGP4eXdvN0CvKzFeQbWzAr0Xua12/bVofbH9JvvsdgOuYxyw+pFp/uIeT5wB/CMqvpOt+m7wDO65fnGvthek3cDfwb8X7f+dOCHNbi5ODy2/vluPr6YxrwWmAH+vjvM9KEkT6Lhea6qQ8BfA98CvsNg3u6i7Xk+rK95Xdktz24/qsUW6M1J8mTgH4A3VNWPhrfV4J/mZq4rTfJy4MGqumvctZxCSxl8LH9fVZ0P/DeDj+I/1+A8nwNsZvCP2TOBJwEbx1rUGIxjXhdboI9yw+pFI8njGYT5J6rqM13zfyU5t9t+LvBg1z7f2BfTa7IB2JTkG8CNDA67vAc4O4Obi8Nj65/v5uOLacwHgYNVdUe3fjODgG95nl8MPFBVM1X1M+AzDOa+5Xk+rK95PdQtz24/qsUW6KPcsHpR6M5Yfxi4v6reNbRp+IbblzM4tn64/TXd2fKLgIe7j3a7gZcmOad7Z/TSru20U1VXV9WqqlrDYO6+UFWvAm5lcHNxOHLMc918fCewpbs6Yi2wjsEJpNNOVX0XOJDkuV3Ti4D7aHieGRxquSjJL3W/54fH3Ow8D+llXrttP0pyUfcavmZoX/Mb90mF4zgJcRmDK0L2AW8Zdz0nMI5fZ/Bx7B7gS93jMgbHDj8PfB34V2B51z/Ajm7cXwEmh/b1+8B093jduMc24vgv4RdXuTybwV/UaeDTwFld+xO69elu+7OHfv4t3WuxlxHO/o95rC8Aprq5/iyDqxmanmfgbcBXgXuBjzO4UqWpeQZuYHCO4GcMPold0ee8ApPd67cP+FtmnVif6+FX/yWpEYvtkIskaR4GuiQ1wkCXpEYY6JLUCANdkhphoEtSIwx0SWrE/wNCW+JzvLgMXgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a7atqoYdqQPT"
      },
      "source": [
        "La fonction  scatter  se comporte comme  plot, mais ne relie pas les points entre eux. Ici nous voyons bien que nos points sont bien situés là où ils doivent être."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cKpTfc4Lqdv3"
      },
      "source": [
        "Une compétence essentielle pour tout data scientist est de savoir présenter les résultats dans le bon format. Dans notre cas, nous voulons montrer la différence entre les gains des deux stratégies."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KCnQ7fRUqjNc"
      },
      "source": [
        "#### La fonction bar"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 267
        },
        "id": "mauYvGsoqHG3",
        "outputId": "061987d3-d2e9-40e9-a70d-6f98d6e371e5"
      },
      "source": [
        "plot = plt.bar([1, 2], [sum(play(Strategie.CHANGER, 10000)), sum(play(Strategie.GARDER, 10000))], tick_label = [\"Changer\", \"Garder\"])"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD6CAYAAABNu5eFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAATI0lEQVR4nO3df5Bd5X3f8fcnkjGx3RoRNiojaSoSy3hwWtt4C7hOUtvY4oczEZ5igqcTVIYZdRqasdvOxLjTVi2YKfzROiaO6ahBjcg4xoSGoDhM6AZMnWTMjxXGBIGp1tiMpAJaW4I0YWwX8u0f91n7Gu+yd9HqSuZ5v2Z27jnf85xznwNnP/fsc597lapCktSHHzvaHZAkjY+hL0kdMfQlqSOGviR1xNCXpI4Y+pLUkUVDP8mpSR4c+vnLJB9JcmKSqSR72uOq1j5Jrksyk+ShJKcPHWtza78nyeYjeWKSpB+WpczTT7IC2A+cCVwOHKyqa5JcAayqqo8mOR/4VeD81u6TVXVmkhOBaWASKGAX8PaqOrTQ85100km1fv36l3dmktSpXbt2fbOqJubbtnKJxzob+FpVPZFkE/CuVt8B3A18FNgE3FiDV5N7kpyQ5OTWdqqqDgIkmQLOBT670JOtX7+e6enpJXZRkvqW5ImFti11TP9ivh/Sq6vqybb8FLC6La8B9g7ts6/VFqq/uLNbkkwnmZ6dnV1i9yRJL2Xk0E9yHPCLwO+9eFu7q1+W73Ooqm1VNVlVkxMT8/51Ikl6mZZyp38e8EBVPd3Wn27DNrTHA62+H1g3tN/aVluoLkkak6WE/of4wfH3ncDcDJzNwG1D9UvaLJ6zgGfbMNAdwMYkq9pMn42tJkkak5HeyE3yWuB9wD8bKl8D3JzkMuAJ4KJWv53BzJ0Z4DngUoCqOpjkKuD+1u7KuTd1JUnjsaQpm+M2OTlZzt6RpKVJsquqJufb5idyJakjhr4kdcTQl6SOLPUTuT9S1l/xR0e7CzpGfeOa9x/tLkhHhXf6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6MlLoJzkhyS1Jvprk0STvSHJikqkke9rjqtY2Sa5LMpPkoSSnDx1nc2u/J8nmI3VSkqT5jXqn/0ngj6vqTcBbgEeBK4A7q2oDcGdbBzgP2NB+tgDXAyQ5EdgKnAmcAWyde6GQJI3HoqGf5PXAzwM3AFTVd6vqGWATsKM12wFc0JY3ATfWwD3ACUlOBs4BpqrqYFUdAqaAc5f1bCRJL2mUO/1TgFngvyf5cpLfSvJaYHVVPdnaPAWsbstrgL1D++9rtYXqPyDJliTTSaZnZ2eXdjaSpJc0SuivBE4Hrq+qtwF/zfeHcgCoqgJqOTpUVduqarKqJicmJpbjkJKkZpTQ3wfsq6p72/otDF4Enm7DNrTHA237fmDd0P5rW22huiRpTBYN/ap6Ctib5NRWOht4BNgJzM3A2Qzc1pZ3Ape0WTxnAc+2YaA7gI1JVrU3cDe2miRpTFaO2O5Xgc8kOQ54HLiUwQvGzUkuA54ALmptbwfOB2aA51pbqupgkquA+1u7K6vq4LKchSRpJCOFflU9CEzOs+nsedoWcPkCx9kObF9KByVJy8dP5EpSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR0x9CWpI4a+JHXE0Jekjhj6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI6MFPpJvpHkL5I8mGS61U5MMpVkT3tc1epJcl2SmSQPJTl96DibW/s9STYfmVOSJC1kKXf6766qt1bVZFu/ArizqjYAd7Z1gPOADe1nC3A9DF4kgK3AmcAZwNa5FwpJ0ngczvDOJmBHW94BXDBUv7EG7gFOSHIycA4wVVUHq+oQMAWcexjPL0laolFDv4D/mWRXki2ttrqqnmzLTwGr2/IaYO/QvvtabaH6D0iyJcl0kunZ2dkRuydJGsXKEdv9bFXtT/KTwFSSrw5vrKpKUsvRoaraBmwDmJycXJZjSpIGRrrTr6r97fEAcCuDMfmn27AN7fFAa74fWDe0+9pWW6guSRqTRUM/yWuT/K25ZWAj8DCwE5ibgbMZuK0t7wQuabN4zgKebcNAdwAbk6xqb+BubDVJ0piMMryzGrg1yVz7362qP05yP3BzksuAJ4CLWvvbgfOBGeA54FKAqjqY5Crg/tbuyqo6uGxnIkla1KKhX1WPA2+Zp/4t4Ox56gVcvsCxtgPbl95NSdJy8BO5ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR0x9CWpIyOHfpIVSb6c5PNt/ZQk9yaZSfK5JMe1+qvb+kzbvn7oGB9r9ceSnLPcJyNJemlLudP/MPDo0Pq1wCeq6g3AIeCyVr8MONTqn2jtSHIacDHwZuBc4NNJVhxe9yVJSzFS6CdZC7wf+K22HuA9wC2tyQ7ggra8qa3Ttp/d2m8Cbqqq71TV14EZ4IzlOAlJ0mhGvdP/deDXgL9p6z8BPFNVz7f1fcCatrwG2AvQtj/b2n+vPs8+35NkS5LpJNOzs7NLOBVJ0mIWDf0kvwAcqKpdY+gPVbWtqiaranJiYmIcTylJ3Vg5Qpt3Ar+Y5HzgeOBvA58ETkiyst3NrwX2t/b7gXXAviQrgdcD3xqqzxneR5I0Bove6VfVx6pqbVWtZ/BG7F1V9U+ALwAXtmabgdva8s62Ttt+V1VVq1/cZvecAmwA7lu2M5EkLWqUO/2FfBS4KcnHgS8DN7T6DcDvJJkBDjJ4oaCqdie5GXgEeB64vKpeOIznlyQt0ZJCv6ruBu5uy48zz+ybqvo28MEF9r8auHqpnZQkLQ8/kStJHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR0x9CWpI4a+JHXE0Jekjhj6ktQRQ1+SOmLoS1JHDH1J6sjh/MtZkg7T+iv+6Gh3Qceob1zz/iNyXO/0Jakjhr4kdcTQl6SOGPqS1JFFQz/J8UnuS/KVJLuT/MdWPyXJvUlmknwuyXGt/uq2PtO2rx861sda/bEk5xypk5IkzW+UO/3vAO+pqrcAbwXOTXIWcC3wiap6A3AIuKy1vww41OqfaO1IchpwMfBm4Fzg00lWLOfJSJJe2qKhXwN/1VZf1X4KeA9wS6vvAC5oy5vaOm372UnS6jdV1Xeq6uvADHDGspyFJGkkI43pJ1mR5EHgADAFfA14pqqeb032AWva8hpgL0Db/izwE8P1efYZfq4tSaaTTM/Ozi79jCRJCxop9Kvqhap6K7CWwd35m45Uh6pqW1VNVtXkxMTEkXoaSerSkmbvVNUzwBeAdwAnJJn7RO9aYH9b3g+sA2jbXw98a7g+zz6SpDEYZfbORJIT2vKPA+8DHmUQ/he2ZpuB29ryzrZO235XVVWrX9xm95wCbADuW64TkSQtbpTv3jkZ2NFm2vwYcHNVfT7JI8BNST4OfBm4obW/AfidJDPAQQYzdqiq3UluBh4Bngcur6oXlvd0JEkvZdHQr6qHgLfNU3+ceWbfVNW3gQ8ucKyrgauX3k1J0nLwE7mS1BFDX5I6YuhLUkcMfUnqiKEvSR0x9CWpI4a+JHXE0Jekjhj6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakji4Z+knVJvpDkkSS7k3y41U9MMpVkT3tc1epJcl2SmSQPJTl96FibW/s9STYfudOSJM1nlDv954F/XVWnAWcBlyc5DbgCuLOqNgB3tnWA84AN7WcLcD0MXiSArcCZwBnA1rkXCknSeCwa+lX1ZFU90Jb/L/AosAbYBOxozXYAF7TlTcCNNXAPcEKSk4FzgKmqOlhVh4Ap4NxlPRtJ0kta0ph+kvXA24B7gdVV9WTb9BSwui2vAfYO7bav1Raqv/g5tiSZTjI9Ozu7lO5JkhYxcugneR3wP4CPVNVfDm+rqgJqOTpUVduqarKqJicmJpbjkJKkZqTQT/IqBoH/mar6/VZ+ug3b0B4PtPp+YN3Q7mtbbaG6JGlMRpm9E+AG4NGq+i9Dm3YCczNwNgO3DdUvabN4zgKebcNAdwAbk6xqb+BubDVJ0pisHKHNO4FfBv4iyYOt9m+Aa4Cbk1wGPAFc1LbdDpwPzADPAZcCVNXBJFcB97d2V1bVwWU5C0nSSBYN/ar6MyALbD57nvYFXL7AsbYD25fSQUnS8vETuZLUEUNfkjpi6EtSRwx9SeqIoS9JHTH0Jakjhr4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR0x9CWpI4a+JHXE0Jekjhj6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOLhn6S7UkOJHl4qHZikqkke9rjqlZPkuuSzCR5KMnpQ/tsbu33JNl8ZE5HkvRSRrnT/23g3BfVrgDurKoNwJ1tHeA8YEP72QJcD4MXCWArcCZwBrB17oVCkjQ+i4Z+VX0ROPii8iZgR1veAVwwVL+xBu4BTkhyMnAOMFVVB6vqEDDFD7+QSJKOsJc7pr+6qp5sy08Bq9vyGmDvULt9rbZQ/Yck2ZJkOsn07Ozsy+yeJGk+h/1GblUVUMvQl7njbauqyaqanJiYWK7DSpJ4+aH/dBu2oT0eaPX9wLqhdmtbbaG6JGmMXm7o7wTmZuBsBm4bql/SZvGcBTzbhoHuADYmWdXewN3YapKkMVq5WIMknwXeBZyUZB+DWTjXADcnuQx4ArioNb8dOB+YAZ4DLgWoqoNJrgLub+2urKoXvzksSTrCFg39qvrQApvOnqdtAZcvcJztwPYl9U6StKz8RK4kdcTQl6SOGPqS1BFDX5I6YuhLUkcMfUnqiKEvSR0x9CWpI4a+JHXE0Jekjhj6ktQRQ1+SOmLoS1JHDH1J6oihL0kdMfQlqSOGviR1xNCXpI4Y+pLUEUNfkjpi6EtSRwx9SerI2EM/yblJHksyk+SKcT+/JPVsrKGfZAXwm8B5wGnAh5KcNs4+SFLPxn2nfwYwU1WPV9V3gZuATWPugyR1a+WYn28NsHdofR9w5nCDJFuALW31r5I8Nqa+vdKdBHzzaHfiWJFrj3YPNA+v0SGHeY3+3YU2jDv0F1VV24BtR7sfrzRJpqtq8mj3Q1qI1+h4jHt4Zz+wbmh9batJksZg3KF/P7AhySlJjgMuBnaOuQ+S1K2xDu9U1fNJ/gVwB7AC2F5Vu8fZh445ZKZjndfoGKSqjnYfJElj4idyJakjhr4kdcTQP0Yl+TtJbkrytSS7ktyeZEuSzx/tvkkLSbI6ye8mebxdt19K8oHDON5vJ7lwOfvYO0P/GJQkwK3A3VX101X1duBjwOqj27Pva1+pIX1Pu27/APhiVf1Uu24vZjA1e5T9D3tiyXIc45XO0D82vRv4f1X1X+cKVfUV4E+B1yW5JclXk3ym/aKR5N8nuT/Jw0m2DdXvTnJtkvuS/O8kP9fqr0lyc5JHktya5N4kk23bxnaH9kCS30vyulb/RjvWA8AHx/ufRD8C3gN890XX7RNV9RtJ1if503ZNPZDkHwIkeVer7wQeycCn2pcy/gnwk3PHSvL2JP+r/QVxR5KTW/3uJL+eZBr48HhP+UePoX9s+hlg1wLb3gZ8hMEX1v0U8M5W/1RV/YOq+hngx4FfGNpnZVWd0fbb2mq/AhyqqtOAfwe8HSDJScC/Bd5bVacD08C/GjrWt6rq9Kq66TDPUa88bwYeWGDbAeB97Zr6JeC6oW2nAx+uqjcCHwBOZXB9XwLMvTi8CvgN4ML2F8R24OqhYxxXVZNV9Z+X8XxekfxT6EfPfVW1DyDJg8B64M+Adyf5NeA1wInAbuAP2z6/3x53tfYAPwt8EqCqHk7yUKufxeAX7s/bHwvHAV8aev7PLfsZ6RUpyW8yuM6+C7wX+FSStwIvAG8canpfVX29Lf888NmqegH4P0nuavVTGdwMTbXrcgXw5NAxvC5HZOgfm3YDC7159Z2h5ReAlUmOBz4NTFbV3iT/ATh+nn1eYPH/5wGmqupDC2z/60X2V792A/94bqWqLm9/OU4D/xJ4GngLgxGGbw/tN8o1FWB3Vb1jge1elyNyeOfYdBfw6vaNowAk+fvAzy3Qfi7gv9nG30eZ7fDnwEXt2KcBf6/V7wHemeQNbdtrk7xx/kNIP+Au4Pgk/3yo9pr2+Hrgyar6G+CXGdypz+eLwC8lWdHG7N/d6o8BE0neAYPhniRvXvYz6IChfwyqwcekPwC8t03Z3A38J+CpBdo/A/w34GEGX3Fx/whP82kGv0SPAB9ncJf2bFXNAv8U+Gwb8vkS8KbDOyP1oF23FwD/KMnXk9wH7AA+yuB625zkKwyup4XuzG8F9gCPADfShhbbv79xIXBtO8aDtPF+LY1fw9CpNuXyVVX17SQ/DfwJcGr75ZL0CuWYfr9eA3yhzYoI8CsGvvTK552+JHXEMX1J6oihL0kdMfQlqSOGviR1xNCXpI78f7qsSAAOhA24AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ebSxNygRrPpt"
      },
      "source": [
        "### Visualiser une liste de nombres"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "k1TnDmnNrhp8"
      },
      "source": [
        "Nous pouvons vérifier que notre programme se comporte de manière raisonnable si nous détectons une relation linéaire entre les gains des joueurs et le nombre de parties qu'ils ont joué."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t5sbIP6Uq8k2"
      },
      "source": [
        "gains_changer = []\n",
        "gains_garder = []\n",
        "samples = [1000, 10000, 20000, 50000, 80000, 100000]\n",
        "for tours in samples:\n",
        "  gains_changer.append(play(Strategie.CHANGER, tours))\n",
        "  gains_garder.append(play(Strategie.GARDER, tours))"
      ],
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YgX3N5GhsB_q"
      },
      "source": [
        "Nous avons maintenant deux listes. L'une contenant les gains d'un joueur qui change de porte systématiquement après 1000, 10000, 20000, 50000, 80000 et 100000 parties. L'autre contient la même chose, mais pour un joueur qui ne change jamais de porte. Nous allons maintenant afficher les deux courbes correspondant à ces listes sur le même graphique. Remarquez que nous donnons deux arguments à  scatter. Le premier est la liste des abscisses."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "ypFKVUspr--g",
        "outputId": "ae743df3-7f77-4b04-c7e9-c372e0a2178b"
      },
      "source": [
        "figure = plt.figure()\n",
        "plot = plt.scatter(samples, [sum(x)/samples[i] for i, x in enumerate(gains_changer)])\n",
        "plot = plt.scatter(samples, [sum(x)/samples[i] for i, x in enumerate(gains_garder)])"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD4CAYAAADrRI2NAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAUpUlEQVR4nO3dcayd9X3f8fcnNiQsyzCM24rYTmwig0bTKJAjRMYaRUsBL+sANVXkdFJhW4qyFSVbJzZQp3Rz/miTTFkWzSqhGVU0lTosi9hNqsmirKm0SlAfCwaxEwfjpPV16bgBQqTNBQzf/XEek+Pba99zr4/P8b2/90s68nl+z+85z+/n38Pn+Z3neQ5OVSFJascbpt0ASdJkGfyS1BiDX5IaY/BLUmMMfklqzPppN2ChSy65pLZs2TLtZkjSqrJv374fVNXMKHXPueDfsmUL/X5/2s2QpFUlyZ+OWtdLPZLUGINfkhpj8EtSYwx+SWqMwS9JjTnnnupZDR587Cif3XOQP//hMd664QLuvPEKbrlq47SbJUkjMfiX6cHHjnL3157k2CuvAnD0h8e4+2tPAkwk/D3pSDpTBv8yfXbPwddD/4Rjr7zKZ/ccPOsBPO2TjjQJTm7OPq/xL9Of//DYssrH6XQnHWktODG5OfrDYxQ/ntw8+NjRaTdtTXHGv0xv3XABRxcJ+bduuOCs73uaJ51WOfucrGl+o56mSR9nzviX6c4br+CC89adVHbBeeu488Yrzvq+T3VymcRJp0XOPievxcnNNI4zg3+ZbrlqI7/x8z/Nxg0XEGDjhgv4jZ//6YnMRqZ50mmRl9Ymr8XJzTSOMy/1rMAtV22cytfOE/v00sNktDj7nLY7b7zipAcYYO1PbqZxnBn8q8y0Tjotmub9nFa1OLmZxnFm8Eun0OLs81zQ2uRmGseZwS+dQouzT03eNI6zVNVZ+/CV6PV6tZJ/iMXH7iS1LMm+quqNUnekp3qSbE9yMMmhJHedos6HkxxIsj/J/UPlryZ5vHvNjtaF5fGxO0ka3ZKXepKsA3YB1wNzwN4ks1V1YKjONuBu4LqqeiHJTwx9xLGqeveY232SVn/0IUkrMcqM/xrgUFUdrqqXgd3AzQvq/DKwq6peAKiqZ8fbzNPzsTtJGt0owb8RODK0PNeVDbscuDzJHyd5JMn2oXVvStLvym9ZbAdJbu/q9Ofn55fVAWjzRx+StFLj+uXuemAb8H7gI8BvJ9nQrXt7d8PhF4HPJ3nHwo2r6t6q6lVVb2ZmZtk79xetkjS6UYL/KLB5aHlTVzZsDpitqleq6nvAdxmcCKiqo92fh4FvAledYZv/imn+bxQkabUZ5Tn+vcC2JFsZBP4OBrP3YQ8ymOn/TpJLGFz6OZzkIuD/VdVLXfl1wGfG1vohrf3oQ5JWasngr6rjSe4A9gDrgPuqan+SnUC/qma7dTckOQC8CtxZVc8l+dvAF5O8xuDbxW8OPw0kSZq8NfMDLklq2dh/wCVJWjsMfklqjMEvSY0x+CWpMQa/JDXG4Jekxhj8ktQYg1+SGmPwS1JjDH5JaozBL0mNMfglqTEGvyQ1xuCXpMYY/JLUGINfkhpj8EtSYwx+SWqMwS9JjTH4JakxIwV/ku1JDiY5lOSuU9T5cJIDSfYnuX+o/NYkT3WvW8fVcEnSyqxfqkKSdcAu4HpgDtibZLaqDgzV2QbcDVxXVS8k+Ymu/GLg14EeUMC+btsXxt8VSdIoRpnxXwMcqqrDVfUysBu4eUGdXwZ2nQj0qnq2K78ReKiqnu/WPQRsH0/TJUkrMUrwbwSODC3PdWXDLgcuT/LHSR5Jsn0Z25Lk9iT9JP35+fnRWy9JWrZx3dxdD2wD3g98BPjtJBtG3biq7q2qXlX1ZmZmxtQkSdJiRgn+o8DmoeVNXdmwOWC2ql6pqu8B32VwIhhlW0nSBI0S/HuBbUm2Jjkf2AHMLqjzIIPZPkkuYXDp5zCwB7ghyUVJLgJu6MokSVOy5FM9VXU8yR0MAnsdcF9V7U+yE+hX1Sw/DvgDwKvAnVX1HECSTzE4eQDsrKrnz0ZHJEmjSVVNuw0n6fV61e/3p90MSVpVkuyrqt4odf3lriQ1xuCXpMYY/JLUGINfkhpj8EtSYwx+SWqMwS9JjTH4JakxBr8kNcbgl6TGGPyS1BiDX5IaY/BLUmMMfklqjMEvSY0x+CWpMQa/JDXG4Jekxhj8ktQYg1+SGjNS8CfZnuRgkkNJ7lpk/W1J5pM83r0+OrTu1aHy2XE2XpK0fOuXqpBkHbALuB6YA/Ymma2qAwuqfqWq7ljkI45V1bvPvKmSpHEYZcZ/DXCoqg5X1cvAbuDms9ssSdLZMkrwbwSODC3PdWULfSjJE0m+mmTzUPmbkvSTPJLklsV2kOT2rk5/fn5+9NZLkpZtXDd3vw5sqap3AQ8BXx5a9/aq6gG/CHw+yTsWblxV91ZVr6p6MzMzY2qSJGkxowT/UWB4Br+pK3tdVT1XVS91i18C3jO07mj352Hgm8BVZ9BeSdIZGiX49wLbkmxNcj6wAzjp6Zwklw4t3gR8uyu/KMkbu/eXANcBC28KS5ImaMmneqrqeJI7gD3AOuC+qtqfZCfQr6pZ4ONJbgKOA88Dt3Wb/y3gi0leY3CS+c1FngaSJE1QqmrabThJr9erfr8/7WZI0qqSZF93P3VJ/nJXkhpj8EtSYwx+SWqMwS9JjTH4JakxBr8kNcbgl6TGGPyS1BiDX5IaY/BLUmMMfklqjMEvSY0x+CWpMQa/JDXG4Jekxhj8ktQYg1+SGmPwS1JjDH5JaozBL0mNGSn4k2xPcjDJoSR3LbL+tiTzSR7vXh8dWndrkqe6163jbLwkafnWL1UhyTpgF3A9MAfsTTJbVQcWVP1KVd2xYNuLgV8HekAB+7ptXxhL6yVJyzbKjP8a4FBVHa6ql4HdwM0jfv6NwENV9XwX9g8B21fWVEnSOIwS/BuBI0PLc13ZQh9K8kSSrybZvJxtk9yepJ+kPz8/P2LTJUkrMa6bu18HtlTVuxjM6r+8nI2r6t6q6lVVb2ZmZkxNkiQtZpTgPwpsHlre1JW9rqqeq6qXusUvAe8ZdVtJ0mSNEvx7gW1JtiY5H9gBzA5XSHLp0OJNwLe793uAG5JclOQi4IauTJI0JUs+1VNVx5PcwSCw1wH3VdX+JDuBflXNAh9PchNwHHgeuK3b9vkkn2Jw8gDYWVXPn4V+SJJGlKqadhtO0uv1qt/vT7sZkrSqJNlXVb1R6vrLXUlqjMEvSY0x+CWpMQa/JDXG4Jekxhj8ktQYg1+SGmPwS1JjDH5JaozBL0mNMfglqTEGvyQ1xuCXpMYY/JLUGINfkhpj8EtSYwx+SWqMwS9JjTH4JakxBr8kNWak4E+yPcnBJIeS3HWaeh9KUkl63fKWJMeSPN697hlXwyVJK7N+qQpJ1gG7gOuBOWBvktmqOrCg3luATwCPLviIp6vq3WNqryTpDI0y478GOFRVh6vqZWA3cPMi9T4FfBr4yzG2T5I0ZqME/0bgyNDyXFf2uiRXA5ur6vcX2X5rkseS/FGSn1l5UyVJ47DkpZ6lJHkD8DngtkVWPwO8raqeS/Ie4MEkP1VVP1rwGbcDtwO87W1vO9MmSZJOY5QZ/1Fg89Dypq7shLcA7wS+meT7wLXAbJJeVb1UVc8BVNU+4Gng8oU7qKp7q6pXVb2ZmZmV9USSNJJRgn8vsC3J1iTnAzuA2RMrq+rFqrqkqrZU1RbgEeCmquonmeluDpPkMmAbcHjsvZAkjWzJSz1VdTzJHcAeYB1wX1XtT7IT6FfV7Gk2fx+wM8krwGvAx6rq+XE0XJK0MqmqabfhJL1er/r9/rSbIUmrSpJ9VdUbpa6/3JWkxhj8ktQYg1+SGmPwS1JjDH5JaozBL0mNMfglqTEGvyQ1xuCXpMYY/JLUGINfkhpj8EtSYwx+SWqMwS9JjTH4JakxBr8kNcbgl6TGGPyS1BiDX5IaY/BLUmNGCv4k25McTHIoyV2nqfehJJWkN1R2d7fdwSQ3jqPRkqSVW79UhSTrgF3A9cAcsDfJbFUdWFDvLcAngEeHyq4EdgA/BbwV+IMkl1fVq+PrgiRpOUaZ8V8DHKqqw1X1MrAbuHmRep8CPg385VDZzcDuqnqpqr4HHOo+T5I0JaME/0bgyNDyXFf2uiRXA5ur6veXu60kabLO+OZukjcAnwP+5Rl8xu1J+kn68/PzZ9okSdJpjBL8R4HNQ8uburIT3gK8E/hmku8D1wKz3Q3epbYFoKrurapeVfVmZmaW1wNJ0rKMEvx7gW1JtiY5n8HN2tkTK6vqxaq6pKq2VNUW4BHgpqrqd/V2JHljkq3ANuBPxt4LSdLIlnyqp6qOJ7kD2AOsA+6rqv1JdgL9qpo9zbb7kzwAHACOA7/iEz2SNF2pqmm34SS9Xq/6/f60myFJq0qSfVXVW7qmv9yVpOYY/JLUGINfkhpj8EtSYwx+SWqMwS9JjTH4JakxBr8kNcbgl6TGGPyS1BiDX5IaY/BLUmMMfklqjMEvSY0x+CWpMQa/JDXG4Jekxhj8ktQYg1+SGmPwS1JjDH5JasxIwZ9ke5KDSQ4luWuR9R9L8mSSx5P8ryRXduVbkhzryh9Pcs+4OyBJWp71S1VIsg7YBVwPzAF7k8xW1YGhavdX1T1d/ZuAzwHbu3VPV9W7x9tsSdJKjTLjvwY4VFWHq+plYDdw83CFqvrR0OKbgRpfEyVJ4zRK8G8Ejgwtz3VlJ0nyK0meBj4DfHxo1dYkjyX5oyQ/s9gOktyepJ+kPz8/v4zmS5KWa2w3d6tqV1W9A/jXwL/pip8B3lZVVwG/Ctyf5G8ssu29VdWrqt7MzMy4miRJWsQowX8U2Dy0vKkrO5XdwC0AVfVSVT3Xvd8HPA1cvrKmSpLGYZTg3wtsS7I1yfnADmB2uEKSbUOLfx94qiuf6W4Ok+QyYBtweBwNlyStzJJP9VTV8SR3AHuAdcB9VbU/yU6gX1WzwB1JfhZ4BXgBuLXb/H3AziSvAK8BH6uq589GR3jiAXh4J7w4Bxdugg98Et714bOyK0lazVJ1bj2A0+v1qt/vL2+jJx6Ar38cXjn247LzLoB/8AXDX1ITkuyrqt4oddfGL3cf3nly6MNg+eGd02mP1o4nHoD/8E74txsGfz7xwLRbpLVowsfZkpd6VoUX55ZXfqa8rNSGhd8kXzwyWAbHW+MzheNsbcz4L9y0vPIzcWKQXjwC1I8HaVIzQWegk+M3yelo7RifwnG2NoL/A58cXNMfdt4Fg/Jxm2YYTPuk05pJf5NUm8f4FI6ztRH87/rw4EbuhZuBDP48Wzd2pxkGzkAna5LfJDXQ4jE+heNsbVzjh0HIT+K664WbutnIIuVnmzPQyfrAJxd/WuxsfJPUQIvH+BSOs7Ux45+kSV5WWsgZ6GRN8pukBlo8xqdwnK2dGf+knBiMaTzV4wx08ib1TVIDrR7jEz7ODP6VmFYYTPOkI02Cx/hErI1f7kpS49r75a4kaWQGvyQ1xuCXpMYY/JLUGINfkhpzzj3Vk2Qe+NNlbnYJ8IOz0JxzWYt9hjb73WKfoc1+n0mf315VI/2j5edc8K9Ekv6ojzGtFS32Gdrsd4t9hjb7Pak+e6lHkhpj8EtSY9ZK8N877QZMQYt9hjb73WKfoc1+T6TPa+IavyRpdGtlxi9JGpHBL0mNWdXBn2R7koNJDiW5a9rtWa4km5P8YZIDSfYn+URXfnGSh5I81f15UVeeJF/o+vtEkquHPuvWrv5TSW4dKn9Pkie7bb6QJJPv6V+VZF2Sx5J8o1vemuTRrp1fSXJ+V/7GbvlQt37L0Gfc3ZUfTHLjUPk5eVwk2ZDkq0m+k+TbSd7byFj/i+74/laS30vyprU23knuS/Jskm8NlZ31sT3VPpZUVavyBawDngYuA84H/jdw5bTbtcw+XApc3b1/C/Bd4ErgM8BdXfldwKe79x8E/gcQ4Frg0a78YuBw9+dF3fuLunV/0tVNt+3fm3a/u3b9KnA/8I1u+QFgR/f+HuCfdu//GXBP934H8JXu/ZXdmL8R2NodC+vO5eMC+DLw0e79+cCGtT7WwEbge8AFQ+N821obb+B9wNXAt4bKzvrYnmofS7Z32gfGGfxFvxfYM7R8N3D3tNt1hn3678D1wEHg0q7sUuBg9/6LwEeG6h/s1n8E+OJQ+Re7skuB7wyVn1Rviv3cBDwM/F3gG93B/ANg/cKxBfYA7+3er+/qZeF4n6h3rh4XwIVdAGZB+Vof643AkS7M1nfjfeNaHG9gCycH/1kf21PtY6nXar7Uc+KAOmGuK1uVuq+0VwGPAj9ZVc90q/4C+Mnu/an6fLryuUXKp+3zwL8CXuuW/ybww6o63i0Pt/P1vnXrX+zqL/fvYtq2AvPA73SXuL6U5M2s8bGuqqPAvwf+DHiGwfjtY+2PN0xmbE+1j9NazcG/ZiT568B/A/55Vf1oeF0NTuVr5pnbJD8HPFtV+6bdlglbz+BSwG9V1VXA/2Xw1fx1a22sAbprzjczOPG9FXgzsH2qjZqCSYztcvaxmoP/KLB5aHlTV7aqJDmPQej/blV9rSv+P0ku7dZfCjzblZ+qz6cr37RI+TRdB9yU5PvAbgaXe/4jsCHJiX8Deridr/etW38h8BzL/7uYtjlgrqoe7Za/yuBEsJbHGuBnge9V1XxVvQJ8jcExsNbHGyYztqfax2mt5uDfC2zrng44n8GNoNkpt2lZujvz/xn4dlV9bmjVLHDijv6tDK79nyj/pe6pgGuBF7uveXuAG5Jc1M2wbmBw3fMZ4EdJru329UtDnzUVVXV3VW2qqi0Mxux/VtU/BP4Q+IWu2sI+n/i7+IWufnXlO7qnQLYC2xjcADsnj4uq+gvgSJIruqIPAAdYw2Pd+TPg2iR/rWvXiX6v6fHuTGJsT7WP05vWTZ8x3Uz5IIMnYZ4Gfm3a7VlB+/8Og69mTwCPd68PMrim+TDwFPAHwMVd/QC7uv4+CfSGPusfA4e61z8aKu8B3+q2+U8suLk45f6/nx8/1XMZg/+QDwH/FXhjV/6mbvlQt/6yoe1/revXQYaeYDlXjwvg3UC/G+8HGTy5sebHGvh3wHe6tv0XBk/mrKnxBn6PwT2MVxh8u/snkxjbU+1jqZf/ywZJasxqvtQjSVoBg1+SGmPwS1JjDH5JaozBL0mNMfglqTEGvyQ15v8DEh4zAP2nsnwAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "vzGVRsTbsmuB",
        "outputId": "fb24742f-cf99-4061-bc54-7621e616759e"
      },
      "source": [
        "plot = plt.plot(samples, [sum(x) for x in gains_changer])\n",
        "plot = plt.plot(samples, [sum(x) for x in gains_garder])"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-9lwYIBLkpFG"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}