regressao_ML.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "regressao_ML.ipynb",
      "provenance": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyPrgZNqYgGRsl7m3UofJEg0",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/gabifc/machine_learning_dell/blob/main/regressao_ML.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "heMtDfZDcmhW"
      },
      "source": [
        "# importar bibliotecas"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XL8h5Nk-cQqy"
      },
      "source": [
        "import numpy as np"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dnamOnCdcfIm"
      },
      "source": [
        "import pandas as pd"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nJWyW6jHchpA"
      },
      "source": [
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "q9_ReZazczRP"
      },
      "source": [
        "# dados registrados dos vendedores de sorvete"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R36do0zPc3pO"
      },
      "source": [
        "temperatura = np.array([30, 25, 36, 18, 25, 29, 30, 33, 37, 31, 26, 37, 29, 26, 30, 31, 34, 38])"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k7wvOvxOdIuj"
      },
      "source": [
        "numero_sorvetes = np.array([20, 12, 50, 10, 18, 25, 26, 32, 48, 22, 16, 52, 24, 20, 28, 29, 35, 40])"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3fM7_XdVdJay"
      },
      "source": [
        "# criar o dataframe passando em um dicionário as variáveis temperatura e numero de sorvetes como chaves dos dicionários\n",
        "# abaixo é chave : variável\n",
        "df = pd.DataFrame({'temperatura': temperatura, 'numero_sorvetes': numero_sorvetes})"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 202
        },
        "id": "EOSPwIwgd4ju",
        "outputId": "ea1b552d-9241-4da3-8a1c-b40acd1bf6bb"
      },
      "source": [
        "df.head()"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>temperatura</th>\n",
              "      <th>numero_sorvetes</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>30</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>25</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>36</td>\n",
              "      <td>50</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>18</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>25</td>\n",
              "      <td>18</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   temperatura  numero_sorvetes\n",
              "0           30               20\n",
              "1           25               12\n",
              "2           36               50\n",
              "3           18               10\n",
              "4           25               18"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "zQiPedzse9Ub",
        "outputId": "d96e7b32-4531-4e50-fc21-742872ca1617"
      },
      "source": [
        "# Gerando o gráfico\n",
        "# primeiro parametro o dataframe temperatura, o segundo o dataframe com os dados do numero de sorvetes e o terceiro é o estilo do plot, neste caso *\n",
        "plt.plot(df['temperatura'], df['numero_sorvetes'], '*')\n",
        "# título do eixo x\n",
        "plt.xlabel('Temperatura')\n",
        "# título do eixo y\n",
        "plt.ylabel('Sorvetes')\n",
        "plt.show()"
      ],
      "execution_count": 12,
      "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": "BIhts77ogJGa"
      },
      "source": [
        "# **Construindo o modelo de regressão**\n",
        "\n",
        "# Separação de varíavel independente e variável dependente.\n",
        "\n",
        "O **método to_nump** converte os valores do dataframe para float."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zA3YeTvXfSrQ"
      },
      "source": [
        "# atribuindo a variável independente X com os dados da temperatura\n",
        "\n",
        "x = df['temperatura'].to_numpy()"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EixAuzSNgsyo"
      },
      "source": [
        "# atribuindo a variável independente Y com os dados da temperatura\n",
        "y = df['numero_sorvetes'].to_numpy()"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qchlPudzinaJ"
      },
      "source": [
        "**separando os dados** para treinar (80%) e testar (20%).\n",
        "\n",
        "Uso a **biblioteca sklearn** para fazer a separação automaticamente."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NUT8bZLHhE4y"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Klc0xEGejbhZ"
      },
      "source": [
        "# a classe train_test_split recebe as variáveis x e y como entrada e retorna 4 variáveis.\n",
        "# test_size defino 20% para teste.\n",
        "\n",
        "x_treino, x_teste, y_treino, y_teste = train_test_split(x, y, test_size=0.2)"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zPjzHQ2fkgXJ"
      },
      "source": [
        "# Treinando o modelo - Regressão Linear\n",
        "\n",
        "módulo sklearn LinearRegression\n",
        "\n",
        "O **método reshape** serve para colocar os dados na forma de um vetor com n linhas e 1 coluna"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gFfuxuHvlrnS"
      },
      "source": [
        "# módulo sklearn LinearRegression\n",
        "from sklearn.linear_model import LinearRegression\n"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZtK248xSluXo"
      },
      "source": [
        "# variável modelo é criada ao instanciar o modelo de regressão - define um modelo de regressão\n",
        "modelo = LinearRegression()\n"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zV9xmYg2kRS4",
        "outputId": "0c219798-437b-4583-f8d4-20d95a6ccb2c"
      },
      "source": [
        "# treina o modelo com os dados de treino\n",
        "# uso as variáveis x_treino e y_treino como entrada\n",
        "# reshape serve para colocar os dados na forma de um vetor com n linhas e 1 coluna\n",
        "modelo.fit(x_treino.reshape(-1, 1), y_treino.reshape(-1, 1))"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KdCRfgQmmyaJ"
      },
      "source": [
        "# Realizando a previsão do número de sorvetes a serem vendidos.\n",
        "\n",
        "**Método predict** da variável modelo"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D1RiulYAmvaV"
      },
      "source": [
        "# y_previsto contém a previsão dos preços do nosso modelo\n",
        "# passo para o método predict a variável x_teste que são os dados de temperatura que não foram utilizados na etapa de treino.\n",
        "\n",
        "y_previsto = modelo.predict(x_teste.reshape(-1,1))"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "MQDdmjyfplDB",
        "outputId": "f8549e2b-d831-46fc-c88f-a4eeeb116c72"
      },
      "source": [
        "# visualização\n",
        "plt.plot(range(y_previsto.shape[0]), y_previsto, 'r--')\n",
        "plt.plot(range(y_teste.shape[0]), y_teste, 'g--')\n",
        "plt.legend(['Sorvetes previstos', 'Sorvetes vendidos'])\n",
        "plt.xlabel('índice')\n",
        "plt.ylabel('Sorvetes')\n",
        "plt.show()"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEGCAYAAABiq/5QAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd3wU1RbA8d9NgQSItIQmvdcQOghCaGIBUaRIE2woFqxUpYuC+EQQVFAfIKDAQ0GpIr1I76H3GggtIQnpue+PWWpCWMjuzpbz/XzyYXd2yplsODt7595zldYaIYQQnsPL7ACEEEI4liR+IYTwMJL4hRDCw0jiF0IIDyOJXwghPIyP2QFYIzAwUBcvXtzsMIQQwqVs27btktY66O7lLpH4ixcvztatW80OQwghXIpS6mR6y6WpRwghPIwkfiGE8DCS+IUQwsO4RBt/epKSkjhz5gzx8fFmhyIczM/Pj8KFC+Pr62t2KEK4JJdN/GfOnCEgIIDixYujlDI7HOEgWmsuX77MmTNnKFGihNnhCOGS7NrUo5Q6oZTao5TaqZTaalmWRyn1j1LqsOXf3A+z7/j4ePLmzStJ38MopcibN6980xMiExzRxt9Yax2ita5ped4PWK61LgMstzx/KJL0PZO870Jkjhk3d1sDUy2PpwLPmRCDEOI2W89tZenRpWaHIRzE3olfA0uVUtuUUj0sy/JrrcMtj88D+dPbUCnVQym1VSm19eLFi3YO8+GMGDGCSpUqERwcTEhICJs2bbLr8aZMmcK5c+fseozMOHfuHG3btr3n65GRkXz33XcOjEhYIyE5gRdmv8BzM5/jXLTz/n0J27F34m+gta4OPAW8rZRqePuL2pgFJt2ZYLTWk7TWNbXWNYOC0ow4Nt2GDRtYsGAB27dvZ/fu3SxbtowiRYpYvX1KSsoDH9ORiV9rTWpq6gNtU6hQIebMmXPP1yXxO6eZYTM5FXWK+OR4hq4aanY4wgHsmvi11mct/0YAc4HawAWlVEEAy78R9ozBXsLDwwkMDCRr1qwABAYGUqhQIQCWL19OtWrVqFKlCq+88goJCQmAUXqib9++VK9endGjR1O7du2b+ztx4gRVqlQBYNu2bTRq1IgaNWrQokULwsPDmTNnDlu3bqVz586EhIQQFxeX7noA48aNo2LFigQHB/Piiy+miX3KlCm0bt2a0NBQypQpw9ChQ2/GUK5cOV566SUqV67M6dOnGT16NLVq1SI4OJjBgwcD0K9fPyZMmHBzf0OGDOGrr77ixIkTVK5cGYC9e/dSu3ZtQkJCCA4O5vDhw/Tr14+jR48SEhJC79690VrTu3dvKleuTJUqVZg1a9bN323Dhg0JCQmhcuXKrF271nZvnEijc3Bn5necz9u13ubnHT9z9MpRs0MS9qa1tssPkB0IuO3xv8CTwGign2V5P+DL++2rRo0a+m779u27c0GjRml/JkwwXouNTf/1yZON1y9eTPvafURHR+uqVavqMmXK6J49e+pVq1ZprbWOi4vThQsX1gcPHtRaa921a1c9ZswYrbXWxYoV06NGjbq5j6pVq+pjx45prbUeOXKkHj58uE5MTNT16tXTERERWmutZ86cqV9++WXLKTbSW7Zs0VrrDNcrWLCgjo+P11prffXq1TSxT548WRcoUEBfunRJX79+XVeqVElv2bJFHz9+XCul9IYNG7TWWv/999/69ddf16mpqTolJUU/88wzevXq1Xr79u26YcOGN/dXoUIFferUKX38+HFdqVIlrbXW77zzjp4+fbrWWuuEhAR9/fr1O17XWus5c+boZs2a6eTkZH3+/HldpEgRfe7cOf3VV1/pzz77TGutdXJysr527Vqac0jz/ouHkpqaevPxhZgLelbYLJ2SmmJiRMKWgK06nZxqzyv+/MA6pdQuYDOwUGu9BBgJNFdKHQaaWZ67nBw5crBt2zYmTZpEUFAQHTp0YMqUKRw8eJASJUpQtmxZALp168aaNWtubtehQ4ebj9u3b3/zKnfWrFl06NCBgwcPEhYWRvPmzQkJCeGzzz7jzJkzaY6f0XrBwcF07tyZ6dOn4+OT/lCN5s2bkzdvXvz9/WnTpg3r1q0DoFixYtStWxeApUuXsnTpUqpVq0b16tU5cOAAhw8fplq1akRERHDu3Dl27dpF7ty50zRz1atXj88//5xRo0Zx8uRJ/P3908Swbt06OnbsiLe3N/nz56dRo0Zs2bKFWrVqMXnyZIYMGcKePXsICAiw+n0R1otNjKXK91X4fd/vAOTLno/2ldrjpWRAv7uz2wAurfUxoGo6yy8DTW1+wFWr7v1atmwZvx4YmPHr9+Dt7U1oaCihoaFUqVKFqVOnUq1atQy3yZ49+83HHTp0oF27drRp0walFGXKlGHPnj1UqlSJDRs2ZLgfrfU911u4cCFr1qxh/vz5jBgxgj179qT5ALi7S+SN57fHp7Wmf//+vPHGG2mO0a5dO+bMmcP58+fv+DC7oVOnTtSpU4eFCxfy9NNPM3HiREqWLJnhOd3QsGFD1qxZw8KFC+nevTsffvghL730klXbCuuN3zyevRf3UiigUJrlK0+s5Pf2v5sUmbA3+Wh/SAcPHuTw4cM3n+/cuZNixYpRrlw5Tpw4wZEjRwCYNm0ajRo1SncfpUqVwtvbm+HDh99MnuXKlePixYs3E3pSUhJ79+4FICAggOjo6AzXS01N5fTp0zRu3JhRo0YRFRVFTExMmmP/888/XLlyhbi4OObNm0f9+vXTrNOiRQv++9//3tz+7NmzREQYt2Q6dOjAzJkzmTNnDu3atUuz7bFjxyhZsiS9evWidevW7N69+474AR5//HFmzZpFSkoKFy9eZM2aNdSuXZuTJ0+SP39+Xn/9dV577TW2b9+e0VshHsK1hGt8+e+XPF3maeoVqXfHa8mpyfyx/w+WHVtmUnTC3ly2ZIPZYmJiePfdd4mMjMTHx4fSpUszadIk/Pz8mDx5Mu3atSM5OZlatWrx5ptv3nM/HTp0oHfv3hw/fhyALFmyMGfOHHr16kVUVBTJycm8//77VKpUie7du/Pmm2/i7+/Phg0b0l2vbNmydOnShaioKLTW9OrVi1y5cqU5bu3atXnhhRc4c+YMXbp0oWbNmpw4ceKOdZ544gn2799PvXpGYsiRIwfTp08nX758VKpUiejoaB599FEKFiyYZv+zZ89m2rRp+Pr6UqBAAQYMGECePHmoX78+lStX5qmnnuLLL79kw4YNVK1aFaUUX375JQUKFGDq1KmMHj0aX19fcuTIwS+//JKJd0qk55uN33Al7grDQoelea1nzZ6M2TiGfsv6sfn1zdL044aU0f7v3GrWrKnvnohl//79VKhQwaSIXNuUKVPYunUr48ePNzuUhybv/8OLToimyJgiNC7RmLkd5qa7ztSdU+n+Z3dmt51Nu0ppv9EJ16CU2qZvVU24ST7KhfAwAVkDmPfiPL5o+sU91+kS3IVKQZX4dOWnpKQ++JgT4dykqccDde/ene7du5sdhjBRaPHQDF/39vLmh5Y/4Ovli7eXt2OCEg4jiV8ID/Lpik+5nnSd/zzxn/sWu2tQtIGDohKOJk09QniI8Ohw/rPhP1yOu2x1hdPk1GR6zO/BV/9+ZefohCNJ4hfCQ3yx7guSUpIY1HCQ1dv4ePkQHhPOiLUjuBp31Y7RCUeSxC+EBzgVdYqJ2ybySrVXKJWn1ANt+3mTz4mKj2LkOpccZC/SIYk/E6Qs852mTJnCO++8A8APP/yQbv/72wu5CccZsWYEAJ82/PSBt62SvwpdgrswbvM4zlxLWz5EuB65ufuQbi/LnDVrVi5dukRiYqLV26ekpODt/WC9JaZMmULlypVvVgF1ZhkNWhOO99FjH/FYkccomrPoQ20/rPEwZu2dxfDVw5nYaqKNoxOOJlf8D8mVyzLXrVv3ZhkIgNDQULZu3UpsbCyvvPIKtWvXplq1avz555+A8YHTpk0bnnzyScqUKUOfPn1ubjt58mTKli1L7dq1Wb9+/c3lN0o13zifqlWrUrVq1TvKOcfHx/Pyyy9TpUoVqlWrxsqVK4H0SzqLzCmbtyzdQro99PbFcxVnSuspfNLwExtGJUyTXslOZ/uxpixzo8mN0vxM2GyUZY5NjE339ck7Jmuttb4YezHNa/fjymWZv/76az1o0CCttdbnzp3TZcuW1Vpr3b9/fz1t2rSb25UpU0bHxMToyZMn6xIlSujIyEgdFxenixYtqk+dOqXPnTunixQpoiMiInRCQoJ+7LHH9Ntvv6211nrw4MF69OjRWmutq1SpolevXq211vrjjz++WZr5q6++uhnz/v37dZEiRXRcXFy6JZ3vJmWZrXPg4gH9/Mzn9YmrJ8wORZgAE8oyuzVXLsvcvn37mzNlzZ49++Z0iUuXLmXkyJGEhIQQGhpKfHw8p06dAqBp06bkzJkTPz8/KlasyMmTJ9m0aROhoaEEBQWRJUuWdKt0RkZGEhkZScOGxuRrXbt2vfnaunXr6NKlCwDly5enWLFiHDp0yKqSzsI6Q1YPYenRpfj72uZ3eCrqFE9Me4LNZzfbZH/CHG7Txr+q+6p7vpbNN1uGrwdmC8zw9Xtx1bLMjz76KHnz5mX37t3MmjWLH3744eY+f//9d8qVK3fH/jZt2nSzSevGeScnJ2cYX2akV9K5SZMmdjueu9pzYQ8zw2bSv0F/8mXPZ5N95vbLzc7zO+m3rB/LX1pu9XgA4Vzkiv8huXpZ5g4dOvDll18SFRVFcHAwYJRh/vbbb2/MoMaOHTsy/B3UqVOH1atXc/nyZZKSkvjf//6XZp1cuXKRK1eumxO9zJgx4+Zrjz/++M3nhw4d4tSpU5QrVy7dks7iwQ1eNZhHsj7Cx499bLN9BmQN4NOGn7LyxEr+OfaPzfYrHEsS/0OKiYmhW7duN2+i7tu3jyFDhtxRlrlKlSp4eXndtyzz9OnTad++PXCrLHPfvn2pWrUqISEh/PvvvwA3yzKHhISQkpKS7nopKSl06dLl5g3Te5Vlbtu2LTNnzrx5XICBAweSlJREcHAwlSpVYuDAgRn+DgoWLMiQIUOoV68e9evXv2e1zMmTJ/P2228TEhJy80MF4K233iI1NZUqVarcbCrLmjUrs2fPpnLlyoSEhBAWFiaTsDyEbee2MffAXD6s+yF5/PPYdN9v1HiD4rmK029ZP1J1qk33LRxDyjILlyTvf8YuX7/M2E1j+ajeR+T0y2nz/U/bNY2X5r3ErLazaF+p/f03EKa4V1lmt2njF0LckjdbXoY1TjvJiq10qtKJyPhInir9lN2OIexHmnqEcDPvLX7P7tMment5826ddwnIGmDX4wj7cOnE7wrNVML25H2/t9UnVjNu8zh2X3DMDfHVJ1bTZGoTYhNjHXI8YRsum/j9/Py4fPmyJAEPo7Xm8uXL+Pn5mR2K09FaM3DlQArmKEjPmj0dckwfLx9WnljJuE3jHHI8YRsu28ZfuHBhzpw5w8WLF80ORTiYn58fhQsXNjsMp7Ps2DLWnlrL+KfG22zA1v3UL1qfVmVbMWr9KN6o+YbNexAJ+3DZXj1CiFu01tT9uS7h0eEcfvcwWX2y3n8jGwmLCCP4+2A+qvcRo58Y7bDjivuTydaFcGOpOpVuVbsxuvlohyZ9gMr5KtO1ale+3fytlG12ES7b1COEuMXby5u3ar1l2vGHhQ6jcfHGFMxR0LQYhPXkil8IF7f48GJ+2PoDyan2q590P8VyFaN7SHe8vR5sjglhDkn8QriwlNQUev/Tm7GbxqIwv2DaxK0T6TG/h9lhiPuQxC+EC5u1dxZ7L+5laOhQp7jajoiN4MftP7LpjH2nIRWZY/fEr5TyVkrtUEotsDyfopQ6rpTaafkJsXcMQrij5NRkhqwaQnD+YNpWbGt2OAB8UO8D8mXPR7/l/WSMjRNzxBX/e8D+u5b11lqHWH52OiAGIdzO9N3TOXzlMMNCh+GlnOPLe44sORjYcCCrTqzi76N/mx2OuAe7/rUopQoDzwA/2fM4QniiojmL0j2kO8+We9bsUO7Qo0YPSuQqQf/l/aVss5Oyd3fOb4A+wN2VnEYopQYBy4F+WuuEuzdUSvUAegAULVrUzmEK4XqalGhCkxLONzNZFu8sfP/M92T1yeo030TEnez2riilWgIRWuttd73UHygP1ALyAH3T215rPUlrXVNrXTMoKMheYQrhcuKT4/lszWdcibtidij31KJ0C0KLh5odhrgHe34c1weeVUqdAGYCTZRS07XW4ZYJ4BOAyUBtO8YghNuZtG0SA1cOZOd55749lpiSyHuL3+PHbT+aHYq4i90Sv9a6v9a6sNa6OPAisEJr3UUpVRBAGbM0PweE2SsGIdzN9aTrfL72c0KLhzplM8/tfL182XlhJwNXDiQmMe28z8I8ZjTAzVBK7QH2AIHAZybEIIRLmrB5AhdiLzC88XCzQ7kvpRQjm47kQuwFvtn4jdnhiNtIdU4hXER0QjQlxpagZqGaLOmyxOxwrPbczOdYeWIlR3sdJTBboNnheBSpzimEi7uWcI36RevbdS5dexjRZAQxiTF8sfYLs0MRFlKdUwgX8egjj/Lni3+aHcYDq5SvEhOenuD09yQ8iVzxC+ECft/3O4cvHzY7jIf2Zs03KZu3rNlhCAtJ/EI4uUvXL9H9z+58suITs0PJlNNRp2kzqw37Lu4zOxSPJ009Qji5L9d/SWxiLENCh5gdSqb4+/qz7NgyNJq5HeaaHY5Hkyt+IZzY+ZjzjN88nk5VOlExqKLZ4WRKYLZAej/Wm3kH5rHh9Aazw/FokviFcGIj140kMSWRwY0Gmx2KTXxQ7wPyZ88vZZtNJolfCCfm6+VLjxo9KJO3jNmh2MSNss1rTq5hyRHXGYvgbqSNXwgnNvqJ0W53Zfx6jdeJjI+kTuE6ZofiseSKXwgndCrqFGtOrgGM0gfuJIt3Fj5p+Al5/POYHYrHksQvhBMaumooT0x7gouxF80OxW7Wn1rP87OeJzEl0exQPI4kfiGczOHLh5m6ayo9a/YkKLv7zkURnRjNvAPzmLRtktmheBxJ/EI4maGrh5LVJyv9GvQzOxS7alHKmKxl+JrhUrbZwSTxC+FE9kbs5dc9v/JOrXfInyO/2eHYlVKKL5p+QURsBGM2jDE7HI8iiV8IJ3Ls6jGK5ypOn/p9zA7FIeoWrsvz5Z9n9L+j3fp+hrOR7pxCOJFW5VrxdJmn8fbyNjsUhxnRZARNSzQlp19Os0PxGJL4hXASK46voFGxRh6V9AEqBFWgQlAFs8PwKNLUI4QT2HRmE01/acr3W783OxTTTNk5hb7/9DU7DI8giV8IJzBo1SACswXSrWo3s0MxTVhEGKP/HU1YRJjZobg9SfxCmGzdqXUsPbqUvvX7EpA1wOxwTNO/QX8Csga4/LwDrkASvxAmG7hyIAVyFOCtWm+ZHYqp8mbLS5/H+vDXwb9Yf2q92eG4NUn8Qpjo8vXLhEeHM6DBALL5ZjM7HNO9X/d9CuQoIGWb7Ux69QhhorzZ8hL2VpgkOYvsWbIz/qnx+Pv6mx2KW5PEL4RJDl0+RMEcBT26XT89L1R8wewQ3J409QhhAq01HX/vSLNpzcwOxSklJCcwYPkAZoXNMjsUtySJXwgTzDswj+3h2+lZs6fZoTglX29flhxZQr/l/UhITjA7HLcjiV8IB0vVqQxaNYiyecvSJbiL2eE4JS/lxchmIzkReULKNtuBJH4hHGz23tmERYQxpNEQfLzkNtu9NC/ZnCYlmjB8zXCiE6LNDsetSOIXwsFWHl9JpaBKdKjcwexQnJpSipFNR3Lx+kXGbJSyzbZk98sNpZQ3sBU4q7VuqZQqAcwE8gLbgK5aa5l7TXiMia0mcjXuKl5Krrvup9ajtRjVbBRPln7S7FDciiP+8t4D9t/2fBQwRmtdGrgKvOqAGIQwXVJKEmeunQEgt39uk6NxHX3q9yE4f7DZYbgVuyZ+pVRh4BngJ8tzBTQB5lhWmQo8Z88YhHAWU3ZOodS4Uuy7uM/sUFzO2Wtn6Tq3KyciT5gdiluwd1PPN0Af4MYIlbxApNY62fL8DPBoehsqpXoAPQCKFi1q5zCFsK+E5ASGrxlOtQLVqBAotecflEYzZ98cFIpfnv/F7HBcnt2u+JVSLYEIrfW2h9leaz1Ja11Ta10zKCjIxtEJ4Vg/bf+J09dO81mTzzC++IoHUfiRwrxb+12m757Ongt7zA7H5dmzqac+8KxS6gTGzdwmwFggl1LqxjeNwsBZO8YghOnikuIYsXYEDYs1pGmJpmaH47L6NehHTr+cDFgxwOxQXJ7dEr/Wur/WurDWujjwIrBCa90ZWAm0tazWDfjTXjEI4QxWn1zNhdgLDG88XK72MyGPfx761u/LgkMLWHdqndnhuLQHTvxKKS+l1COZOGZf4EOl1BGMNv+fM7EvIZzek6Wf5MR7J2hYrKHZoaRv8mQ4dAhSUmDWLHDiSqG96vSiz2N9KJ2ntNmhuDRlTTlYpdSvwJtACrAFeAQYq7Uebd/wDDVr1tRbt251xKGEsKlL1y8RmC3Q7DDubcYM6NIFXnsNGjWCrl1hyBAYPNjsyIQNKKW2aa1r3r3c2iv+ilrraxhdLxcDJYCuNoxPCLcTFR9F2W/L8sXaL8wOJX1LlkD37hAaCt9+C507G8+HDIHvvjM3tvvYdGYTL//5MimpKWaH4pKsTfy+SilfjMT/l9Y6CXDe74NCOIExG8dwNf4qLUq3MDuUtDZuhBdegMqV4c8/wc8PlIIff4SWLeGdd2D2bLOjvKeTUSeZsnMKM/bMMDsUl2Rt4p8InACyA2uUUsWAa/YKSghXd/n6ZcZsHEObCm2oXrC62eGk9cUXULCgcdX/yG237Hx8jHb++vWhRw+IjDQvxgy0rdiW6gWrM2jlICnb/BCsSvxa63Fa60e11k9rw0mgsZ1jE8JlffXvV0QnRDM0dKjZoaTvt99gxQrInz/ta9mywfz5sHQp5Mrl+Nis4KW8GNl0JCejTvLD1h/MDsflWJX4lVL5lVI/K6UWW55XxOiKKYS4S2JKIlN3TeXFyi9SOV9ls8O55dIl4yZuVJSR3DMaEZ8rF9SubTyeMsXo9eNkmpdqTtMSTfls7WdcS5AGiAdhbcmGKcBk4BPL80PALKQrphBpZPHOwp6ee0hIcaImiJgYeOYZ2L3bSP5161q3XWQk9O0L/v7w779QqJB943xAo5qNYuWJlWTxzmJ2KC7F2jb+QK31bCAVwFJrR26nC3GX60nX0VqTN1teCgU4SZJMTIS2bWHbNqP93tqkD8aV/+LFcOUKtGgBV6/aL86HUKNQDT5+7GP8fPzMDsWlWJv4Y5VSebH05FFK1QWi7BaVEC6q99LePPbfx0hOTb7/yo6Qmmp00fz7b5g0CZ599sH3Ub06zJtnNPe0agXXr9s8zMz6bc9vjFw30uwwXIa1if9D4C+glFJqPfAL0MtuUQnhgk5GnuTH7T8SnC/YeaZUDA+HtWth5Eh45ZWH30+TJsZgrw0bYOFC28VnI6tOrGLQykEcv3rc7FBcgrUjd7NiNO2UAxRwEPDSWjukEVNG7gpX8Ppfr/PL7l848u4RiuQsYnY4t1y9ajTZ2KJO0IEDUL585vdjY+eiz1FqXCnaVmzLtOenmR2O08jsyN0NWutkrfVerXWYZQDXBtuGKITrOnLlCJN3TuaNGm84R9KfNAnefddo6smd2zZJH24l/XXrjLIOTlLXp1BAId6r8x4zds9g94XdZofj9DJM/EqpAkqpGoC/UqqaUqq65ScUyOaQCIVwAeM3j8fX25f+DfqbHQr88Qf07AlHjxqF1+xh7lwYNgxGO6Rcl1X61u9rlG1eLmWb7+d+DZEtgO4YdfO/vm35NUB+u0JYjGo2ivaV2lMwoKC5gaxcCR07Qp068L//ga+vfY4zejScO2d09QwKgpdfts9xHkBu/9x8/cTX5MiSA621lMDOgLVt/C9orX93QDzpkjZ+4cySU5Od42bujh1Ghc0iRYwbunny2Pd4iYlGL5/ly41vGQ/TY0jYVWbb+NffPXJXKfWqTSMUwgXtvrCbkmNLsvnsZrNDMa7ACxUyum7aO+kDZMkCv/8ONWrAnDn2P56V4pPj+WzNZyw6vMjsUJyWtYl/MvA3cGNEyiHgfbtEJIQLGbxqMNcSrlEmTxnzgki2jBl45hkIC4PChR137Bw5jJo+kyc77pj34ePlw7Td0+j9T28p23wPMnJXiIe09dxW5h2Yx0f1PiK3f25zgoiMNEbiTp1qPPcxockpZ07w9obTp6FZMzhubl96Hy8fRjQZwb6L+5i2W7p2pkdG7grxkAatHEQe/zy8V/c9cwKIi4PWrY36O85QQyc6GrZvhyeegIgIU0N5ocIL1CxUk0ErBxGfHG9qLM7I2sT/EWlH7r5rt6iEcHLbw7ez+Mhi+jzWh0eyZmYK6oeUnGz03lm7FqZNg+bNHR/D3SpWNEb1nj0LTz0F18yrmKmUYmTTkZy+dprvt3xvWhzOyqrvhVrrbUqpRtw2ctcyiEsIj1StQDUWdlpIo2KNHH9wreGNN4yZs779Fjp0cHwM91KvnnHD99ln4bnnYNEiY3YvEzQt2ZT+DfrToGgDU47vzKztzrkbmAnM0loftXtUd5HunMKZmN5HXGsYPty46h82zLw4MjJ9Onz1lXHjN18+s6PxWJntztkKSAZmK6W2KKU+VkplMIuDEO5Ja83Tvz7N1xu+vv/K9hAZaZRfGDTIeZM+QJcusGWLkfRTUkwt7RAeHU7PBT25EHPBtBicjbVTL57UWn+pta4BdAKCASmDJzzO0qNLWXJkCf4+/o4/+LRpULo07Nvn+GM/DF9fY5DXCy/AkCGmhRGdGM2P239k+JrhpsXgbKy94kcpVUwp1Qejyac80MduUQnhhLTWDFw5kGI5i/FqdQePX1y40CiLULUqlCrl2GNnhq8v5M1rfDsZP96UEMrmLctr1V9j4raJHLt6zJQYnI21c+5uAuZa1m+nta6ttf6PXSMTwsksOLSALee2MKjRIMdO9ffvv9CuHYSEGBOiZM3quGNnllIwcaLR7UVfjvYAACAASURBVLRXL2MGMBMMajQIXy9fBq4caMrxnc19E79Sygv4Q2tdXWs9UmstH5nC42itGbxqMKXzlOalqi857sCHD0PLlsZo3EWLICDAcce2FR8f+O03ePxx6NrVuOHrYDfKNv+651d2nt/p8OM7m/t259Rapyql2gGjHBCPEE5JKcV/W/+XyPhIxxZkK1IE2reHfv1cu3eMvz/89ZfR3p8rlykh9G3Ql6iEKAKzBZpyfGdibXfOkcAlYBYQe2O51vqK/UK7RbpzCo9z6ZJRBiG3SaUgHCEy0rQPAU+R2e6cHYC3gTXANsuPZGLhEX7d8ytd/ujCtQQHjUSNjjZGvrZoYcyg5Y6++gqCg+HMGYcfekf4Dj5Y8gHWXPS6K2u7c5ZI56dkRtsopfyUUpuVUruUUnuVUkMty6copY4rpXZafkJscSJC2ENSShKDVw1mT8QecmTJYf8DJiRAmzZGbf2BA8HL6o53rqVpU+OK/4kn4PJlhx56W/g2vtn0DX8e/NOhx3Um1vbq8VVK9VJKzbH8vKOUut/UPglAE611VSAEeNJS3A2gt9Y6xPIjd1qE0/pl1y8cuXKE4Y2H46XsnIRTUqBbN1i2DH7+2ZjkxF1Vq2a0+R87Zty8jo29/zY20j2kO+UDyzNg+QCPLdts7V/y90AN4DvLTw3LsnvShhjLU1/Lj+d+txIuJyE5gWFrhlGrUC1alXVAEh4+3OjuOHq08QHg7kJDjd4+mzcbN7Ad1PRyo2zz/kv7+WXXLw45prOxtntCLcuV+w0rlFK77reRUsob435AaWCC1nqTUqonMEIpNQhYDvTTWieks20PoAdA0aJSHUI43s87fuZU1CkmtZzkmNo8r79u3Ox834PmOHr+eZg0yRjo5cD6R8+Xf57aj9Zm8KrBdKzSET8fcwrJmcXaXj3bMQZuHbU8LwnM0VpXt+ogSuXCGAD2LnAZOA9kASYBR7XWGRYdkV49wgznos/x257f+LDeh/ZN/KtWGX3cvb3tdwxXceSIMTLZAR8Cq0+sZtWJVfSu35tsvtnsfjwzZLZXT29gpVJqlVJqFbACo0a/VbTWkcBK4EmtdbilGSgBY0rH2tbuRwhHKhRQiI8e+8i+Sf9//4MmTeA/MhCeXbugcmUY5ZghQ42KN2Jw6GC3TfoZyTDxK6VqKaUKaK2XA2WAPzCmX1wKZNjUo5QKslzpo5TyB5oDB5RSBS3LFPAcEJbpsxDChmITY2kzqw3bw7fb90DLlkHnzvDYY/DOO/Y9liuoUsUY4NW/P/z0k8MOO3f/XL7b8p3DjucM7nfFPxFItDyuA/QDJgAXMJppMlIQ41vCbmAL8I/WegEwQym1B9gDBAKfPWTsQtjF+M3jmXtgrn2n7Nu61WjfLlcO5s+HbJ531ZmGl5cxaXuLFsZEM/PmOeSws/fNpvc/vQmPDnfI8ZxBhm38SqldN27qKqUmABe11kMsz3dqrR3SB1/a+IWjXEu4RomxJajzaB0WdV5kn4MkJhoJX2ujAJszzJfrTGJjjX7+O3caH5CVK9v1cEeuHKHChAr0qN6DCc9MsOuxHO1h2/i9lVI3ev40xWjbv8GBBUuEcIyxG8dyJe4KwxrbcZKTLFmMboxLl0rST0/27EYZ6iFDjHl87ax0ntK8Xv11Jm2fxJErR+x+PGdwv8T/G7BaKfUnEAesBVBKlQai7BybEA51Ne4q/9nwH1qXa03NQmkukmxwgKtGwgeoWxfKlrX9MdxF3rxGYTovLzhxwhjoZUcDGw4ki3cWBq0cZNfjOIsMr9q11iOUUssx2uuX6lvtQl4YXTOFcBtZfbLyyeOf0KJ0C9vv/Pp1YyTuli3GzdxixWx/DHeUmmrU8o+NhXXroEABuxymYEBBPm/yObn93bgo3m2s6sdvNmnjFy4tKcmov7NwIcyeDW3bmh2Ra9m0yejyWrasMeYhZ06zI3IZme3HL4RbG7NhDL/u+dX2O9baGJG7YAFMmCBJ/2HUqQN//AFhYcbVf7z9elslJCfw9YavWXtyrd2O4Qwk8QuPFx4dzicrPmHJkSW23/nq1TB1KgwdCj172n7/nqJFC/jlF+P3OXSo3Q6TqlMZs3EMH//zsVuXbZaeOcLjfbHuCxJTEhnUyA439kJDYf16qFfP9vv2NB07GvMNN29ut0P4+/ozpNEQXpv/GnMPzKVNhTZ2O5aZ5IpfeLTTUaeZuG0iL4e8TOk8pW234xkzYK2lueCxxxxagMyttWljzDscGwvTptnlEN1CulE+sDyfrPiE5NRkuxzDbJL4hUcbsXYEWms+bfip7Xa6YIFRVnn0aNvtU9zpu+/gpZdg7Fib79rHy4fPm3zOgUsHmLpzqs337wykqUd4tGYlm1EiVwmK5bJR98p166BdO2OikRkzbLNPkdaHH8LGjUYJ68BAo+aRDT1X/jneqPEG5QPL23S/zkK6cwphK3v2QMOGkC+f8QEQFGR2RO4tPt6Ym3jdOqPe0ZNPmh2R05HunELc5uClg4xYM4KYxJj7r2ytSZOMYmtLl0rSdwQ/P/jzT6Oq51tvGTWQbOxi7EV6L+1NZHykzfdtJkn8wiMNXT2UL9Z9wfWk67bb6TffGM0PMirXcR55BBYvNj5ss2Sx+e7PRp/lqw1fMXq9e92vkcQvPE5YRBgzw2bybu13yZc9X+Z2du0adOoEp04ZM2gVKWKbIIX18ueH0qWNwXIjRxrvhY2EFAihU5VOjNk4xq3KNkviFx5n8KrBBGQNoHf93pnbUUKCUVN/9mzYv982wYmHd+oUfPGFMdjr0iWb7XZY6DCSUpMYttqOFVsdTBK/8Cg7wnfwx/4/+KDuB+Txz/PwO0pJgS5dYMWKW5OHCHMVK2bc5D1+HFq2hBjb3L8placUb9R4gx+3/8jhy4dtsk+zSeIXHsXHy4fnyj/HB3U/ePidaA1vvw1z5hhz5XbtarsAReY0bAizZhlVUNu2tdkN34ENB9KpSieyeNv+PoIZpDunEA8qOtpIMC1aGG3Kwvn8/LNRG2nFCmjQwOxoTCPdOYXH+2bjN5yOOp25nWhtlAxYt85oTxbO6dVX4dAhmyf9vRF73aKtXxK/8AhrTq7hg78/YPbe2Q+/k1mz4LnnjElVsmeX+jvOrnhx499Zs2z2zezvo38zeNVgVhxfcf+VnZgkfuH2tNZ8uuJTCuYoSM9aD1ka+Z9/jLb8q1cl4buaxYuhf39jgF0mvVXrLYo8UoR+y/q5dNlmSfzC7S07toy1p9Yy4PEBZPPN9uA72LLF6LZZoQL89Rf4+9s+SGE/P/4ITz9ttPn//numduXn48fQ0KFsObeFP/b/YaMAHU9u7gq3prWm3s/1OBd9jsPvHiarT9YH28GBA0Y78SOPGHX1Cxa0T6DCvq5fh2bNYNs2WLIEGjd+6F2lpKYQ/EMwKakphL0Vho+X89a6lJu7wiPFJcdRIagCQ0KHPHjSB6Pue4ECRkkASfquK1s2o1x26dLGLF6Z4O3lzahmo2hdrjWJKbavD+QIcsUvRHoSEozZngBSU8FLrpHcQkwM5MhhPNba7e/XyBW/8DjrTq1j67mHuGCIjTWaAm7M7SpJ333cSPo7d0L9+hCeufo7S44sYfru6TYIzLHkL1q4pZTUFN5c8CZd53YlVadav2FSkjGRyqZNRrlf4Z4SE2H3bqOGf+TDl1wet2kc7y5+l6txV20YnP1J4hduafbe2ey9uJchjYbgpaz8M09NhVdeMbr/ff+9Mb+rcE+1a8PcuUZxvWefhbi4h9rNF02/ICo+ii/Xf2njAO1LEr9wO8mpyQxZPYQq+arQrlI76zfs0wemT4fPPoMePewXoHAOzZsbE7avWwcdO0Lyg0+sXrVAVTpV6cTYTWM5F33ODkHah90Sv1LKTym1WSm1Sym1Vyk11LK8hFJqk1LqiFJqllLKPaoeCacxffd0Dl0+xNDQodZf7YMxT+6HH8KAAfYLTjiXDh1g3DhjGseHLOg2rPEwklOTGbpqqI2Dsx97XvEnAE201lWBEOBJpVRdYBQwRmtdGrgKvGrHGIQHup50ndDioTxX/jnrNoiIMP7t3NmotunmPT3EXd55BxYtMrp8JiQ88OYlc5fkk8c/oUahGnYIzj7slvi14UZBbF/LjwaaAHMsy6cCVv7vFMI6b9V6ixUvrUBZk8DnzYMSJWDVKrvHJZyYl5dxk7dBA+PD/wENDh1Mjxqu0zxo1zZ+pZS3UmonEAH8AxwFIrXWNxrTzgCP3mPbHkqprUqprRcvXrRnmMJNxCfH89fBv9BaW5f016yBF1+EypWhZpquzsLTBAQYhd0+/hh++eWBN09KSeKHrT+wPXy77WOzMbsmfq11itY6BCgM1AbKP8C2k7TWNbXWNYOCguwWo3Afk7ZNovXM1vx7+t/7r7xrF7RqZVztL1x4q3+38Fze3sbN/SZNjN5dCxc+0OZxyXEMXDmQPv/0sVOAtuOQXj1a60hgJVAPyKWUulHcojBw1hExCPd2Pek6n6/9nNDioTxW5LGMVz5/3ui//cgj8PffEBjomCCF88ua1Wj+CwkxxnP8a8VFhMUjWR/hk8c/Yfnx5Sw7tsyOQWaePXv1BCmlclke+wPNgf0YHwBtLat1A/60VwzCc3y35TsuxF5geOPh92/myZcPXn/dSPpFizomQOE6AgKMm71NmkChQg+0ac+aPSmasyj9lvV7sIGDDmbPK/6CwEql1G5gC/CP1noB0Bf4UCl1BMgL/GzHGIQHiE6IZtT6UbQo1YIGRTOYcenaNThxwriRN2wYVKzosBiFi8mXzyjqVry4MbDv0iWrNsvqk5VhocPYFr6N3/dlrgS0PdmtnqjWejdQLZ3lxzDa+4WwiaNXj5LdNzvDGmcwJV58vDF71tGjcPAg+Pk5LkDh2nr1MibiWbcOrLjf2CW4C4uPLCYou/Pem5TqnMItJKcm37suekoKtG8Pf/xh3Lzr3NmxwQnXtm6dMcq3cmVj8vaAALMjsppU5xRuaeOZjcQnx9876WsNb71lJP1vvpGkLx5cgwYwezbs2GHUb7JykNfVuKsMWz2M60nX7Rzgg5PEL1zW5euXeWLaE/Ra3OveK/3wgzHX6oAB8N57jgtOuJdWreDnn2HZMqNjgBXCIsIYvGow4zaNs3NwD8555wwT4j5G/zuamMQY3quTQULv0sUotfzuu44LTLinbt0gKsrqct2PF3ucZ8o8w6j1o3ijxhvk9s9t5wCtJ1f8wiVdiLnAt5u/pVOVTlTKVyntCsuXG7MtBQQYN+ek/o6whV69bs3Xu2fPfVf/vOnnRMVHMXLdSDsH9mAk8QuXNHLdSBKSExjcaHDaF5csMQZoffKJ4wMTnuHvvyE42GhKzEBw/mA6B3dm3OZxnLl2xkHB3Z809QiXk6pT2XF+By9VfYkyecvc+eKmTfDCC0YPjGEZdO8UIjOaNIGWLY2OA4GB0LbtPVcdFjqMmMQYklKSHBhgxqQ7p3BJWmvikuPI5pvt1sL9+40eGLlzw/r1kD+/eQEK93f9OjzxBGzZYoz0bdrU7IjS8NjunBGxEbjCh5uwzvmY80TERqCUujPpa23cfMuSBZYulaQv7C9bNpg/H8qWNQYH3mfi9iNXjjB241gHBZcxt0/8jac2pvjY4ry98G0WH15MfHK82SGJTPhk+SdUnFAxbd9opeC334y215IlzQlOeJ7cuY2/uW++gYIFM1x1xu4ZvP/3+2w5u8VBwd2bWyf+VJ3Kh3U/pFqBakzZNYWnf32awC8D+erfr8wOTTyEw5cPM3XXVLoEd7l1tR8bC+PHG1f8pUoZN9yEcKRCheBVy0SC27bB2fQLDn9Y70OCsgXRb3k/01sh3DrxeykvXq3+KvNenMflPpdZ1GmRcUMwj3FD8NjVY9T9qS6frfmMXed3mf5miIwNXT2ULN5Z6Negn7EgMdG4kfvee8Z/OCHMFB9vDPR68km4ejXNywFZA/i04aesOL7C9LLNbp34b+fn48dTZZ7iu2e+o3X51gBcun6JVJ3KwJUDCZkYcrNJ6Ow1mSLA2ey7uI9f9/zKO7XfoUCOAkbFxJdfNr5mT5woM2gJ8/n5wbRpcOiQ8QFwPW2phjdqvEHxXMXpt9zcss0ek/jTU/vR2mx+fTPhH4XzU6ufqFagGtN2TyOrT1YAFh5ayM/bf+Z8zHmTIxVLjy4lIGsAfer3MZp1PvgAfv0VPv8cXnvN7PCEMDRtahQC/PdfozBg0p1dOLP6ZOXzJp9Tr3A9U+83SnfOuySlJOHr7QtAx987MjNsJmB8SLQq24pWZVtRtUBVh8Qi7nTp+iUCswUa3TarVoW334avv5ZRucL5fP+90cd/3DhTy4XcqzunJP4MaK3ZfWE38w/NZ8GhBWw+u5n6Reuz9uW1AGw6s4mqBari5yO13e3pdNRpiuQscufC3buNQVpeHv2lVTizv/6Cp58Gn/THya49uZYrcVduNj3bg+cm/u+/NyZRLl8eKlQwRtk95BXihZgLRMRGUCV/FaLiowgcHUgW7yw0L9mcVmVb8XSZpykYkHGXLvFgtpzdQt2f6zKn3RyePwDExUGnTmaHJYT1zp837kV163ZzkdaahlMacuTKEY68e4TsWbLb5dAeO4CLMWPgjTegUSNjOrXAQHj//VuvL1tm3IxJTr7vrvLnyE+V/EZlPn9ff+Z3nE/3qt3ZHr6d1+a/RqGvCzF5x2QAUlJTpJeQDQxaNYjcfrlpdiYLvPgiTJhgTKwihKsYPRq6d4cpU24uUkoxsulIzsecZ+wmxw/qcv8r/tRUOH0aDhy49VOxotHulpRkjL5LTgZfXyhd2vhm0Lmz0U1Qa4iOhkceyfAQWmv2ROxh/sH5dKjcgdJ5SvO/vf/jw6Uf0rJMS1qVa0WTEk2kSegBrT+1ngaTGzC64vt8/OrPUKQIrF0LefKYHZoQ1ktMNOr6rFgBc+caPX4snv3tWVafXM2xXsfImy2vzQ/tuU09GUlJga1b7/xQOHDA6CXy0Udw5oyRbAoVMj4QypeHcuXgqaegTJkMd73m5Bq+2fgNS48uJTYplmy+2WhWshkz2swgR5Yctj8XN9RkahP2nw/j6DjI5u1n9JQoXNjssIR4cNHRRmG3sDBj/t4GDQBjspbg74P5qN5HjH5itM0Pe6/E79nVOb29oU4d4yc9WbPCyJG3PhBmzDAmYsib10j827ZBjx63PhRu+3BoWKwhDYs1JD45nlUnVrHg0AL2XtxLdl+jLW/46uFoNK3KtiKkQAhKeqbc4djVY2w4s4FR3k+SLWEtrFsqSV+4roAAo5BbgwbQp49RRFApKuerzHt13qNozqIODcezr/gflNZw4QJkz268kZs3w6BBxofCyZO31lu1yrinsH69Mdfr7R8KlpvLrX5rxcJDC9FoHg14lJZlW9KxckcaFW9k2uk5m/DocHL758bvcpQUXRPu4dQpI3/ktX2zTno89+auLSkFBQoYSR+gdm1j0o8TJ4yaMTt2GIXCQkKM18PC4LvvjG8FDRveurl89izzO84nvPkS/lusF3VyVWL67unMPTAXMG4MT94xmfDojKv9uaurV8+h27Wl4OFw476IJH3hLooWNZJ+QoLRnBwRARj/53/Z9QsHLh1wSBhyxW9vqanGp/yN5qKDB42iYt7e8OabRrkBIMHfl9hyJclTJpiNX71Pvcn1AaiZvxqtKjznMU1COimJWp8GUeFwFNM6zIQOHcwOSQjb27UL6tUzupivXMkln0RKji1J81LN+b397zY7jNzcdUaRkWlvLMfEoP/5h7CIMBZ82o75+iAbC4NWsCKyNY2rPseVDs/i7+OPv6+/2WdgW1ozr1dzng9czn+zd+Llj2eYHZEQ9rNoETz7rNEsvGgRwzaOYvCqwWx8dSN1Ct/jvuMDksTvipYtgy1biDi0g8VRW+m48hJZqoTQd2A9vt38Lc3C/Wl1LgfPBNSgUJnqxj2EkBCjW6oLSh3Qn5DIkcTny8O+QRfw8fLsvgfCA0ybBi+9BG3bEj31R0pNKEulfJVY8dIKm3y7l149rqhZM2jWjHxANzBuLsfE8OzV3VxPus78xF+YH3QKOEWTw3NZNhBUx45G8TKAjh2NnjA3uqHeuLnsjJKT+d/pJewpDTOeHydJX3iGrl3h0iUYOpSA0xcY2HAgvZb0YunRpbQo3cJuh5UrfhemtTaahA4tID7uGkOD2oOvL813fEDJHEVo9cNKmq4/h39s4q2NPv0Uhg83SsZ+992t3kbFi9+zpojdpaaClxehkxtxKe4yu97chbeXtzmxCGGG8+ehQAESUxJpPbM1H9X7iGYlm2V6t9LU4yHik+Pp8kcX/j76NzGJMfj7+NO0QD3ey96UZuH+Rk+k+vWNImdVb6symiWLMTZh1Ch45hljvMLhw8aHQg47DjhbtAiGDYP584nNmY0z185QLrCc/Y4nhLPS2vj/lyMHvPOOTXbp8KYepVQR4BcgP6CBSVrrsUqpIcDrwEXLqgO01ovsFYen8fPxY077OSQkJ7Dm5BrmH5rP/EPzCa/ZHV7sysnIk0xZNZSWZVtS/dIl1KFDd95czpXL2NGaNcaNJ7jVXFS+vFEHv2RJo9yFj0/mSiJv2ABt25JcoRzax4vsWbJL0heeKzXV+D8xfz4xeQP4qcRV3qr1Flm8s9j8UHa74ldKFQQKaq23K6UCgG3Ac0B7IEZrbfXEt3LFnzlaa1J1Kt5e3vy25zc6/9EZjaZQQKGbtYSal2x+cwIawOhfvG7drS6oBw4YdfA3bjRqHf3wA/TunXbU8lNPGfWP7mfvXnj8ccibl59+7MnI3d+x5uU1FAooZL9fhBDOLi4OWrRgScS/PNUxhQlPT+CtWm899O5Mb+pRSv0JjAfqI4nfVBdjL7Lo8CIWHF7A30f+JjYplgsfXyAwWyBhEWHk9svNo488mnbDG38rShkfCrNn3/qmcPq08VpkJOTMCWPHGr2Sbv9QqFDBKLB28qTR3JSSQsLalZRd0IICOQqw8dWNbj9OQYj7ioxEN2pIaK29HCyVi6MfnXross2m9upRShUHqgGbMBL/O0qpl4CtwEda6zQzEyulegA9AIoWdWwdC3cXlD2IbiHd6BbSjcSURHae32nMbAW8v+R9lh9fTvWC1WlVtpXRJFSwOl7K685mnQYNbhaaAiAmBo4cMZI+GBVPT540ClIlJBjLAgKMewdgNB9NmsRPV5ZzKuoUP7b6UZK+EAC5cqGW/M1/nqvN8WrPks3Xim/QD8juV/xKqRzAamCE1voPpVR+4BJGu/9wjOagVzLah1zxO86+i/v46+BfzD80nw2nN6DRtC7XmnkvzgMgMSXxwdocU1KMD4ADB+DKFejSxViuNXHJ8ZQaV4pSeUqxpvsaSfxC3C4+3pjAPRNMueJXSvkCvwMztNZ/AGitL9z2+o/AAnvGIB5MxaCKVAyqSL8G/bgYe5HFRxaTM6txFR8VH0XRb4ryeNHHaVm2JS3LtqTwI/epmOntbdwMLlnyzuVK8eueXwmPCee3F36TpC/E3TKZ9DNiz5u7CpgKXNFav3/b8oJa63DL4w+AOlrrFzPal1zxO4cLMRf4fO3nzD80n+ORxwGoVqAa454aR4OiDe6zdVqpOpUVx1fYpL+yECIth9/cVUo1ANYCe4BUy+IBQEcgBKOp5wTwxo0PgnuRxO9ctNbsv7Sf+QeNrqKTWk2iYlBFFh9ezNwDc2lZtiXNSjbLsG0yJTVFBmkJYWem9+rJDEn8rmH85vEMWD6A6MRo/Hz8aFqiKa3KtuLV6q/eUYIhKj6KGpNq8EXTL2hXqZ2JEQvh3qRWj7C7d2q/Q48aPYyBY5ZvA/sv7adHjR4A/LH/D4o8UoSFhxdy9OpRSuUpZXLEQngmueIXdqO15tL1SwRlDyJVp1LwPwWJiDUmnmhToY1N644LIdKSK37hcEopgrIHAeClvNj71l4WH17MulPr6Nugr8nRCeG55IpfCCHclMy5K4QQApDEL4QQHkcSvxBCeBhJ/EII4WEk8QshhIeRxC+EEB5GEr8QQngYSfxCCOFhXGIAl1LqInDyITcPxJj4xR3IuTgfdzkPkHNxVpk5l2Ja66C7F7pE4s8MpdTW9EauuSI5F+fjLucBci7Oyh7nIk09QgjhYSTxCyGEh/GExD/J7ABsSM7F+bjLeYCci7Oy+bm4fRu/EEKIO3nCFb8QQojbSOIXQggP4zaJXyn1pFLqoFLqiFKqXzqvZ1VKzbK8vkkpVdzxUVrHinPprpS6qJTaafl5zYw470cp9V+lVIRSKuweryul1DjLee5WSlV3dIzWsOI8QpVSUbe9H4McHaO1lFJFlFIrlVL7lFJ7lVLvpbOOq7wv1pyL0783Sik/pdRmpdQuy3kMTWcd2+YvrbXL/wDewFGgJJAF2AVUvGudt4AfLI9fBGaZHXcmzqU7MN7sWK04l4ZAdSDsHq8/DSwGFFAX2GR2zA95HqHAArPjtPJcCgLVLY8DgEPp/H25yvtizbk4/Xtj+T3nsDz2BTYBde9ax6b5y12u+GsDR7TWx7TWicBMoPVd67QGploezwGaKqWUA2O0ljXn4hK01muAKxms0hr4RRs2ArmUUgUdE531rDgPl6G1Dtdab7c8jgb2A4/etZqrvC/WnIvTs/yeYyxPfS0/d/e6sWn+cpfE/yhw+rbnZ0j7B3BzHa11MhAF5HVIdA/GmnMBeMHyNXyOUqqIY0KzOWvP1RXUs3xVX6yUqmR2MNawNBdUw7jCvJ3LvS8ZnAu4wHujlPJWSu0EIoB/tNb3fE9skb/cJfF7mvlAca11MPAPt64EhDm2Y9REqQp8C8wzOZ77UkrlAH4H3tdaXzM7nsy4z7m4xHujtU7RWocAhYHaSqnK9jyeuyT+s8DtV72FLcvSXUcp5QPkBC47JLoHc99z0Vpf1lonWJ7+BNRwftrlNAAAAwBJREFUUGy2Zs375vS01tdufFXXWi8CfJVSgSaHdU9KKV+MRDlDa/1HOqu4zPtyv3NxtfdGax0JrASevOslm+Yvd0n8W4AySqkSSqksGDc//rprnb+AbpbHbYEV2nKnxMnc91zuam99FqNt0xX9Bbxk6UVSF4jSWoebHdSDUkoVuNHeqpSqjfH/yhkvKrDE+TOwX2v99T1Wc4n3xZpzcYX3RikVpJTKZXnsDzQHDty1mk3zl8/DbuhMtNbJSql3gL8xesX8V2u9Vyk1DNiqtf4L4w9kmlLqCMaNuhfNi/jerDyXXkqpZ4FkjHPpblrAGVBK/YbRqyJQKXUGGIxx4wqt9Q/AIoweJEeA68DL5kSaMSvOoy3QUymVDMQBLzrpRQVAfaArsMfSpgwwACgKrvW+YN25uMJ7UxCYqpTyxvhgmq21XmDP/CUlG4QQwsO4S1OPEEIIK0niF0IIDyOJXwghPIwkfiGE8DCS+IWwUEr5KqXet/SuEMJtSeIX4pZhwHmtdcr9VrRUfVxgefysSqeKqhDOyi368QuRWZaBM2Fa65kPuq2ln/XdAwaFcFpyxS8EoLWOAyZaruRXWYrfHVBKzbht5OeTlmXbgTY3tlXG/AjjLY/zK6XmWoqC7VJKPWZZ3sVSc32nUmqiNCcJM0niFyKtasD7QEWMeRHqK6X8gB+BVhi1kQrcY9txwGpLUbDqwF6lVAWgA1DfUogrBehs31MQ4t6kqUeItDZrrc8AWEoBFAdigONa68OW5dOBHuls2wR4CYyKi0CUUqorxofFFsuXB3+M8rtCmEISvxBpJdz2OIXM/z9RwFStdf9M7kcIm5CmHiGscwAorpQqZXne8R7rLQd6ws3JNXJalrVVSuWzLM+jlCpm74CFuBdJ/EJYQWsdj9G0s9Byc/deTTXvAY2VUnuAbRhzwO4DPgWWKqV2Y0ye43RTGQrPIdU5hRDCw8gVvxBCeBhJ/EII4WEk8QshhIeRxC+EEB5GEr8QQngYSfxCCOFhJPELIYSH+T9mNq4qLG642gAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eFuM5PWDr0pV"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}