Clonk_Prueba_selección.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Untitled4.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "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/JECaballeroR/EmployeeTurnover/blob/main/Clonk_Prueba_selecci%C3%B3n.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "07mlbrFDk_yN",
        "outputId": "094adbf7-14ee-4de1-b2bd-ee5267233aa3"
      },
      "source": [
        "pip install ppscore"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: ppscore in /usr/local/lib/python3.6/dist-packages (1.2.0)\n",
            "Requirement already satisfied: scikit-learn<1.0.0,>=0.20.2 in /usr/local/lib/python3.6/dist-packages (from ppscore) (0.22.2.post1)\n",
            "Requirement already satisfied: pandas<2.0.0,>=1.0.0 in /usr/local/lib/python3.6/dist-packages (from ppscore) (1.1.5)\n",
            "Requirement already satisfied: scipy>=0.17.0 in /usr/local/lib/python3.6/dist-packages (from scikit-learn<1.0.0,>=0.20.2->ppscore) (1.4.1)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.6/dist-packages (from scikit-learn<1.0.0,>=0.20.2->ppscore) (1.0.0)\n",
            "Requirement already satisfied: numpy>=1.11.0 in /usr/local/lib/python3.6/dist-packages (from scikit-learn<1.0.0,>=0.20.2->ppscore) (1.19.5)\n",
            "Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.6/dist-packages (from pandas<2.0.0,>=1.0.0->ppscore) (2.8.1)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas<2.0.0,>=1.0.0->ppscore) (2018.9)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.6/dist-packages (from python-dateutil>=2.7.3->pandas<2.0.0,>=1.0.0->ppscore) (1.15.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nzXQEXNfk8k8"
      },
      "source": [
        "import ppscore as pps\r\n",
        "import pandas as pd\r\n",
        "import numpy as np\r\n",
        "import seaborn as sns\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "\r\n",
        "\r\n",
        "from sklearn.pipeline import make_pipeline\r\n",
        "from sklearn.model_selection import  GridSearchCV, train_test_split\r\n",
        "from sklearn.preprocessing import StandardScaler\r\n",
        "from sklearn import metrics\r\n",
        "from sklearn.metrics import (f1_score, roc_auc_score, precision_recall_curve, \r\n",
        "                            roc_curve, confusion_matrix, classification_report, \r\n",
        "                            accuracy_score)\r\n",
        "from sklearn.linear_model import LogisticRegression\r\n",
        "from sklearn.ensemble import RandomForestClassifier\r\n",
        "from sklearn.svm import SVC\r\n",
        "from sklearn.utils import resample\r\n",
        "from sklearn.neighbors import KNeighborsClassifier\r\n",
        "\r\n",
        "from warnings import filterwarnings\r\n",
        "\r\n",
        "filterwarnings(\"ignore\")\r\n"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fLl1ddeDnRJc"
      },
      "source": [
        "# Preparación de datos\r\n",
        "##Carga y exploración básica\r\n",
        "\r\n",
        "\r\n",
        "Los features del dataset son:\r\n",
        "* **satisfaction_level**: Nivel de Satisfacción, de 0 a 1 (Flotante).\r\n",
        "* **last_evaluationTime**: Años desde la última evaluación de desempeño (Flotante)\r\n",
        "* **number_project**:Cantidad de proyectos terminados durante la vinculación laboral (Entera)\r\n",
        "* **average_montly_hours**: Horas promedio mensuales trabajadas (Entera)\r\n",
        "* **time_spend_company**: Años de vinculación en la compañía (Entera)\r\n",
        "* **Work_accident**: Si el empleado tuvo o no Accidente de trabajo (Binaria)\r\n",
        "* **left**: Si el empleado dejo o no el trabajo (Variable Respuesta) (Binaria)\r\n",
        "* **promotion_last_5years**: Si el empleado tuvo o no un ascenso en los últimos 5 años (Binaria).\r\n",
        "* **sales**: Departamento al que el empleado estuvo vinculado (Categórica).\r\n",
        "* **salary**: Nivel relativo del salario (low, medium, high) (Categórica)\r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        "Se cargan los datos y se validan ciertos elementos, a saber:\r\n",
        "\r\n",
        "*   Que no haya datos faltantes \r\n",
        "*   Que todos los datos sean coherentes con las variables\r\n",
        "*   Que la variable respuesta este balanceada \r\n",
        "\r\n",
        "Después de estas validaciones iniciales, se modifican los nombres de las variables para su interpretabilidad y se realiza un Análisis de Datos Exploratorio para conocer mejor el dataset.\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "FeTMfmxImpIb",
        "outputId": "20c503bb-15f3-419e-a2e0-ca96486e349d"
      },
      "source": [
        "#Se carga y presentan los datos\r\n",
        "df=pd.read_csv(\"HR_comma_sep.csv\")\r\n",
        "df"
      ],
      "execution_count": 3,
      "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>satisfaction_level</th>\n",
              "      <th>last_evaluation</th>\n",
              "      <th>number_project</th>\n",
              "      <th>average_montly_hours</th>\n",
              "      <th>time_spend_company</th>\n",
              "      <th>Work_accident</th>\n",
              "      <th>left</th>\n",
              "      <th>promotion_last_5years</th>\n",
              "      <th>sales</th>\n",
              "      <th>salary</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.38</td>\n",
              "      <td>0.53</td>\n",
              "      <td>2</td>\n",
              "      <td>157</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>sales</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.80</td>\n",
              "      <td>0.86</td>\n",
              "      <td>5</td>\n",
              "      <td>262</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>sales</td>\n",
              "      <td>medium</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.11</td>\n",
              "      <td>0.88</td>\n",
              "      <td>7</td>\n",
              "      <td>272</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>sales</td>\n",
              "      <td>medium</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.72</td>\n",
              "      <td>0.87</td>\n",
              "      <td>5</td>\n",
              "      <td>223</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>sales</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.37</td>\n",
              "      <td>0.52</td>\n",
              "      <td>2</td>\n",
              "      <td>159</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>sales</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14994</th>\n",
              "      <td>0.40</td>\n",
              "      <td>0.57</td>\n",
              "      <td>2</td>\n",
              "      <td>151</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>support</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14995</th>\n",
              "      <td>0.37</td>\n",
              "      <td>0.48</td>\n",
              "      <td>2</td>\n",
              "      <td>160</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>support</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14996</th>\n",
              "      <td>0.37</td>\n",
              "      <td>0.53</td>\n",
              "      <td>2</td>\n",
              "      <td>143</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>support</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14997</th>\n",
              "      <td>0.11</td>\n",
              "      <td>0.96</td>\n",
              "      <td>6</td>\n",
              "      <td>280</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>support</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14998</th>\n",
              "      <td>0.37</td>\n",
              "      <td>0.52</td>\n",
              "      <td>2</td>\n",
              "      <td>158</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>support</td>\n",
              "      <td>low</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>14999 rows × 10 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       satisfaction_level  last_evaluation  ...    sales  salary\n",
              "0                    0.38             0.53  ...    sales     low\n",
              "1                    0.80             0.86  ...    sales  medium\n",
              "2                    0.11             0.88  ...    sales  medium\n",
              "3                    0.72             0.87  ...    sales     low\n",
              "4                    0.37             0.52  ...    sales     low\n",
              "...                   ...              ...  ...      ...     ...\n",
              "14994                0.40             0.57  ...  support     low\n",
              "14995                0.37             0.48  ...  support     low\n",
              "14996                0.37             0.53  ...  support     low\n",
              "14997                0.11             0.96  ...  support     low\n",
              "14998                0.37             0.52  ...  support     low\n",
              "\n",
              "[14999 rows x 10 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NjNZpT7eo2I_",
        "outputId": "91532e1b-f400-4359-a6f0-9752926a0a35"
      },
      "source": [
        "#Verificación de datos nulos en el dataset\r\n",
        "df.isnull().sum()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "satisfaction_level       0\n",
              "last_evaluation          0\n",
              "number_project           0\n",
              "average_montly_hours     0\n",
              "time_spend_company       0\n",
              "Work_accident            0\n",
              "left                     0\n",
              "promotion_last_5years    0\n",
              "sales                    0\n",
              "salary                   0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EBTut0lspDgy",
        "outputId": "ac5e9064-4152-4e08-a8af-c7c1143888b5"
      },
      "source": [
        "#Para variables categoricas y binarias, verificación de que no hayan elementos mal escritos o similar.\r\n",
        "#Para esto, se imprimen los valores únicos de cada variable\r\n",
        "variables_bin_cat=[ 'Work_accident', \"left\", \"promotion_last_5years\", \"sales\", \"salary\"]\r\n",
        "[ (i, df[i].unique()) for i in variables_bin_cat]"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('Work_accident', array([0, 1])),\n",
              " ('left', array([1, 0])),\n",
              " ('promotion_last_5years', array([0, 1])),\n",
              " ('sales',\n",
              "  array(['sales', 'accounting', 'hr', 'technical', 'support', 'management',\n",
              "         'IT', 'product_mng', 'marketing', 'RandD'], dtype=object)),\n",
              " ('salary', array(['low', 'medium', 'high'], dtype=object))]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "JeGTpXF3neVz",
        "outputId": "af79d6b0-713e-4583-ce52-e1b561e47595"
      },
      "source": [
        "#Para la verificación de variables númericas continuas, se usa DataFrame.describe() para validar el rango de los datos\r\n",
        "df.describe()"
      ],
      "execution_count": 6,
      "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>satisfaction_level</th>\n",
              "      <th>last_evaluation</th>\n",
              "      <th>number_project</th>\n",
              "      <th>average_montly_hours</th>\n",
              "      <th>time_spend_company</th>\n",
              "      <th>Work_accident</th>\n",
              "      <th>left</th>\n",
              "      <th>promotion_last_5years</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>14999.000000</td>\n",
              "      <td>14999.000000</td>\n",
              "      <td>14999.000000</td>\n",
              "      <td>14999.000000</td>\n",
              "      <td>14999.000000</td>\n",
              "      <td>14999.000000</td>\n",
              "      <td>14999.000000</td>\n",
              "      <td>14999.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>0.612834</td>\n",
              "      <td>0.716102</td>\n",
              "      <td>3.803054</td>\n",
              "      <td>201.050337</td>\n",
              "      <td>3.498233</td>\n",
              "      <td>0.144610</td>\n",
              "      <td>0.238083</td>\n",
              "      <td>0.021268</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>0.248631</td>\n",
              "      <td>0.171169</td>\n",
              "      <td>1.232592</td>\n",
              "      <td>49.943099</td>\n",
              "      <td>1.460136</td>\n",
              "      <td>0.351719</td>\n",
              "      <td>0.425924</td>\n",
              "      <td>0.144281</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.090000</td>\n",
              "      <td>0.360000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>96.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.440000</td>\n",
              "      <td>0.560000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>156.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.640000</td>\n",
              "      <td>0.720000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>200.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>0.820000</td>\n",
              "      <td>0.870000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>245.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>7.000000</td>\n",
              "      <td>310.000000</td>\n",
              "      <td>10.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       satisfaction_level  last_evaluation  ...          left  promotion_last_5years\n",
              "count        14999.000000     14999.000000  ...  14999.000000           14999.000000\n",
              "mean             0.612834         0.716102  ...      0.238083               0.021268\n",
              "std              0.248631         0.171169  ...      0.425924               0.144281\n",
              "min              0.090000         0.360000  ...      0.000000               0.000000\n",
              "25%              0.440000         0.560000  ...      0.000000               0.000000\n",
              "50%              0.640000         0.720000  ...      0.000000               0.000000\n",
              "75%              0.820000         0.870000  ...      0.000000               0.000000\n",
              "max              1.000000         1.000000  ...      1.000000               1.000000\n",
              "\n",
              "[8 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uokCBPUHubQu"
      },
      "source": [
        "De este dataframe se puede observar que:\r\n",
        "1.   Las evaluaciones de desempeño se realizan juiciosamente de manera anual, dado que el máximo valor es 1 (eso, o quizá son más bien el nivel de desempeño en la última evaluación de desempeño)\r\n",
        "2.   Existen empleados que trabajan más de 10 horas al día en promedio (máximo de 310 horas al mes/30 días al mes= 10.33 horas al día). A priori, parece una posible causa de abandono.\r\n",
        "3. Se observa que la mayoría de empleados (al menos el 75%) no han abandonado a la compañía, recibido un ascenso en los últimos 5 años o tenido un accidente de trabajo. Esto sugiere que el dataset está no balanceado\r\n",
        "4. Igualmente, al menos el 75% de los empleados a trabajado con la compañía menos de 5 años, por lo que el punto 3 parece indicar que no hay muchos ascensos en la companía (otro posible factor de abandono)\r\n",
        "5. En general, parece no existir problemas con los datos.\r\n",
        "\r\n",
        "\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 293
        },
        "id": "p5lWOnRMn-XJ",
        "outputId": "557cad32-bacc-4f53-d772-a42f5922bb79"
      },
      "source": [
        "# Se verifica si la variable respuesta se encuentra balanceada en el dataset, mostrando el conteo de la misma\r\n",
        "df.groupby(\"left\")['sales'].count().plot(kind='bar')"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f0136d34e10>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        },
        {
          "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": "7smTWGtbzOx1"
      },
      "source": [
        "# Se renombra \"sales\" como \"department\", \"salary\" como \"salary_level\", \"left\" como \"employee_left\"\r\n",
        "# Esto es para dar nombres más interpretables de las variables\r\n",
        "df=df.rename(columns = {'sales':'department', 'salary': 'salary_level', 'left':'employee_left'})\r\n",
        "\r\n"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EwmjNZY91hFj"
      },
      "source": [
        "# Análisis Exploratorio de Datos (EDA)\r\n",
        "\r\n",
        "Se revisará a mayor profundidad los datos para detectar ciertas interacciones y guiar la futura interpretación de los modelos.\r\n",
        "\r\n",
        "Para ello, se revisarán interacciones, agrupaciones y visualizaciones de los datos, y se generarán ideas del algunas interacciones interesantes que se detecten. \r\n",
        "\r\n",
        "1. Observar si, al agrupar los datos por alguna variable, se observa que alguna otra tiene un nivel diferente (por ejemplo, si se ve un cambio en los niveles de motivación entre los diferentes niveles de salario).\r\n",
        "2. Generar algunos heatmaps para apoyar las posibles tendencias que se vean (por ejemplo, detectar los departamentos de mayor rotación de personal)\r\n",
        "3. Generar visualizaciones de apoyo para encaminar el primer modelo (elegir features)\r\n",
        "\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 142
        },
        "id": "-tUMfebk0o_s",
        "outputId": "f472e045-41d8-42d3-b544-cd1e7e595f65"
      },
      "source": [
        "df.groupby('employee_left').mean()"
      ],
      "execution_count": 9,
      "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>satisfaction_level</th>\n",
              "      <th>last_evaluation</th>\n",
              "      <th>number_project</th>\n",
              "      <th>average_montly_hours</th>\n",
              "      <th>time_spend_company</th>\n",
              "      <th>Work_accident</th>\n",
              "      <th>promotion_last_5years</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>employee_left</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.666810</td>\n",
              "      <td>0.715473</td>\n",
              "      <td>3.786664</td>\n",
              "      <td>199.060203</td>\n",
              "      <td>3.380032</td>\n",
              "      <td>0.175009</td>\n",
              "      <td>0.026251</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.440098</td>\n",
              "      <td>0.718113</td>\n",
              "      <td>3.855503</td>\n",
              "      <td>207.419210</td>\n",
              "      <td>3.876505</td>\n",
              "      <td>0.047326</td>\n",
              "      <td>0.005321</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "               satisfaction_level  ...  promotion_last_5years\n",
              "employee_left                      ...                       \n",
              "0                        0.666810  ...               0.026251\n",
              "1                        0.440098  ...               0.005321\n",
              "\n",
              "[2 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 266
        },
        "id": "I4rv4eqw1__t",
        "outputId": "6aa50a9a-1484-437a-f5f8-bcd56a8d6295"
      },
      "source": [
        "df.groupby(['employee_left','salary_level']).mean()"
      ],
      "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></th>\n",
              "      <th>satisfaction_level</th>\n",
              "      <th>last_evaluation</th>\n",
              "      <th>number_project</th>\n",
              "      <th>average_montly_hours</th>\n",
              "      <th>time_spend_company</th>\n",
              "      <th>Work_accident</th>\n",
              "      <th>promotion_last_5years</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>employee_left</th>\n",
              "      <th>salary_level</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th rowspan=\"3\" valign=\"top\">0</th>\n",
              "      <th>high</th>\n",
              "      <td>0.651879</td>\n",
              "      <td>0.708606</td>\n",
              "      <td>3.789610</td>\n",
              "      <td>200.348052</td>\n",
              "      <td>3.693506</td>\n",
              "      <td>0.166234</td>\n",
              "      <td>0.062338</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>low</th>\n",
              "      <td>0.668103</td>\n",
              "      <td>0.717028</td>\n",
              "      <td>3.784603</td>\n",
              "      <td>198.617418</td>\n",
              "      <td>3.253888</td>\n",
              "      <td>0.183709</td>\n",
              "      <td>0.010109</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>medium</th>\n",
              "      <td>0.668875</td>\n",
              "      <td>0.715461</td>\n",
              "      <td>3.788068</td>\n",
              "      <td>199.214272</td>\n",
              "      <td>3.435952</td>\n",
              "      <td>0.168259</td>\n",
              "      <td>0.034315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"3\" valign=\"top\">1</th>\n",
              "      <th>high</th>\n",
              "      <td>0.434512</td>\n",
              "      <td>0.644024</td>\n",
              "      <td>3.451220</td>\n",
              "      <td>193.097561</td>\n",
              "      <td>3.682927</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>low</th>\n",
              "      <td>0.441248</td>\n",
              "      <td>0.716994</td>\n",
              "      <td>3.836096</td>\n",
              "      <td>206.631215</td>\n",
              "      <td>3.874770</td>\n",
              "      <td>0.043738</td>\n",
              "      <td>0.006446</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>medium</th>\n",
              "      <td>0.438550</td>\n",
              "      <td>0.724571</td>\n",
              "      <td>3.912680</td>\n",
              "      <td>209.610478</td>\n",
              "      <td>3.891420</td>\n",
              "      <td>0.056188</td>\n",
              "      <td>0.003797</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                            satisfaction_level  ...  promotion_last_5years\n",
              "employee_left salary_level                      ...                       \n",
              "0             high                    0.651879  ...               0.062338\n",
              "              low                     0.668103  ...               0.010109\n",
              "              medium                  0.668875  ...               0.034315\n",
              "1             high                    0.434512  ...               0.000000\n",
              "              low                     0.441248  ...               0.006446\n",
              "              medium                  0.438550  ...               0.003797\n",
              "\n",
              "[6 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7ASgwbuN1vOR"
      },
      "source": [
        "Se observa que el nivel de satisfacción es mayor para los empleados que se mantienen en la empresa. Los que aun trabajan han tenido más ascensos y accidentes de trabajo.\r\n",
        "\r\n",
        "Los empleados de alto salario que abandonaron la empresa no tuvieron accidentes laborales o ascensos, siendo quizá este último una causa de su deserción."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        },
        "id": "Wd0Wn4vQ1LIQ",
        "outputId": "aa0970d5-d37b-488f-d247-565e5f1d30d5"
      },
      "source": [
        "df.groupby('salary_level').mean()"
      ],
      "execution_count": 11,
      "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>satisfaction_level</th>\n",
              "      <th>last_evaluation</th>\n",
              "      <th>number_project</th>\n",
              "      <th>average_montly_hours</th>\n",
              "      <th>time_spend_company</th>\n",
              "      <th>Work_accident</th>\n",
              "      <th>employee_left</th>\n",
              "      <th>promotion_last_5years</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>salary_level</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>high</th>\n",
              "      <td>0.637470</td>\n",
              "      <td>0.704325</td>\n",
              "      <td>3.767179</td>\n",
              "      <td>199.867421</td>\n",
              "      <td>3.692805</td>\n",
              "      <td>0.155214</td>\n",
              "      <td>0.066289</td>\n",
              "      <td>0.058205</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>low</th>\n",
              "      <td>0.600753</td>\n",
              "      <td>0.717017</td>\n",
              "      <td>3.799891</td>\n",
              "      <td>200.996583</td>\n",
              "      <td>3.438218</td>\n",
              "      <td>0.142154</td>\n",
              "      <td>0.296884</td>\n",
              "      <td>0.009021</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>medium</th>\n",
              "      <td>0.621817</td>\n",
              "      <td>0.717322</td>\n",
              "      <td>3.813528</td>\n",
              "      <td>201.338349</td>\n",
              "      <td>3.529010</td>\n",
              "      <td>0.145361</td>\n",
              "      <td>0.204313</td>\n",
              "      <td>0.028079</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "              satisfaction_level  ...  promotion_last_5years\n",
              "salary_level                      ...                       \n",
              "high                    0.637470  ...               0.058205\n",
              "low                     0.600753  ...               0.009021\n",
              "medium                  0.621817  ...               0.028079\n",
              "\n",
              "[3 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cApD4ybd2gqD"
      },
      "source": [
        "Se observa que el mayor salario parece inducir a menos probablidad de ascensos y a una menor tasa de abandono. Los demas factores lucen relativamente similares."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 390
        },
        "id": "CSE4apx81N7Q",
        "outputId": "91bdd68f-f937-4e52-d617-fd9de4542ffc"
      },
      "source": [
        "df.groupby('department').mean()"
      ],
      "execution_count": 12,
      "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>satisfaction_level</th>\n",
              "      <th>last_evaluation</th>\n",
              "      <th>number_project</th>\n",
              "      <th>average_montly_hours</th>\n",
              "      <th>time_spend_company</th>\n",
              "      <th>Work_accident</th>\n",
              "      <th>employee_left</th>\n",
              "      <th>promotion_last_5years</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>department</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>IT</th>\n",
              "      <td>0.618142</td>\n",
              "      <td>0.716830</td>\n",
              "      <td>3.816626</td>\n",
              "      <td>202.215974</td>\n",
              "      <td>3.468623</td>\n",
              "      <td>0.133659</td>\n",
              "      <td>0.222494</td>\n",
              "      <td>0.002445</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>RandD</th>\n",
              "      <td>0.619822</td>\n",
              "      <td>0.712122</td>\n",
              "      <td>3.853875</td>\n",
              "      <td>200.800508</td>\n",
              "      <td>3.367217</td>\n",
              "      <td>0.170267</td>\n",
              "      <td>0.153748</td>\n",
              "      <td>0.034307</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>accounting</th>\n",
              "      <td>0.582151</td>\n",
              "      <td>0.717718</td>\n",
              "      <td>3.825293</td>\n",
              "      <td>201.162973</td>\n",
              "      <td>3.522816</td>\n",
              "      <td>0.125163</td>\n",
              "      <td>0.265971</td>\n",
              "      <td>0.018253</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>hr</th>\n",
              "      <td>0.598809</td>\n",
              "      <td>0.708850</td>\n",
              "      <td>3.654939</td>\n",
              "      <td>198.684709</td>\n",
              "      <td>3.355886</td>\n",
              "      <td>0.120433</td>\n",
              "      <td>0.290934</td>\n",
              "      <td>0.020298</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>management</th>\n",
              "      <td>0.621349</td>\n",
              "      <td>0.724000</td>\n",
              "      <td>3.860317</td>\n",
              "      <td>201.249206</td>\n",
              "      <td>4.303175</td>\n",
              "      <td>0.163492</td>\n",
              "      <td>0.144444</td>\n",
              "      <td>0.109524</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>marketing</th>\n",
              "      <td>0.618601</td>\n",
              "      <td>0.715886</td>\n",
              "      <td>3.687646</td>\n",
              "      <td>199.385781</td>\n",
              "      <td>3.569930</td>\n",
              "      <td>0.160839</td>\n",
              "      <td>0.236597</td>\n",
              "      <td>0.050117</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>product_mng</th>\n",
              "      <td>0.619634</td>\n",
              "      <td>0.714756</td>\n",
              "      <td>3.807095</td>\n",
              "      <td>199.965632</td>\n",
              "      <td>3.475610</td>\n",
              "      <td>0.146341</td>\n",
              "      <td>0.219512</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sales</th>\n",
              "      <td>0.614447</td>\n",
              "      <td>0.709717</td>\n",
              "      <td>3.776329</td>\n",
              "      <td>200.911353</td>\n",
              "      <td>3.534058</td>\n",
              "      <td>0.141787</td>\n",
              "      <td>0.244928</td>\n",
              "      <td>0.024155</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>support</th>\n",
              "      <td>0.618300</td>\n",
              "      <td>0.723109</td>\n",
              "      <td>3.803948</td>\n",
              "      <td>200.758188</td>\n",
              "      <td>3.393001</td>\n",
              "      <td>0.154778</td>\n",
              "      <td>0.248991</td>\n",
              "      <td>0.008973</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>technical</th>\n",
              "      <td>0.607897</td>\n",
              "      <td>0.721099</td>\n",
              "      <td>3.877941</td>\n",
              "      <td>202.497426</td>\n",
              "      <td>3.411397</td>\n",
              "      <td>0.140074</td>\n",
              "      <td>0.256250</td>\n",
              "      <td>0.010294</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             satisfaction_level  ...  promotion_last_5years\n",
              "department                       ...                       \n",
              "IT                     0.618142  ...               0.002445\n",
              "RandD                  0.619822  ...               0.034307\n",
              "accounting             0.582151  ...               0.018253\n",
              "hr                     0.598809  ...               0.020298\n",
              "management             0.621349  ...               0.109524\n",
              "marketing              0.618601  ...               0.050117\n",
              "product_mng            0.619634  ...               0.000000\n",
              "sales                  0.614447  ...               0.024155\n",
              "support                0.618300  ...               0.008973\n",
              "technical              0.607897  ...               0.010294\n",
              "\n",
              "[10 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HrrRrmDT275J"
      },
      "source": [
        "Parece ser que el departamento en que un empleado trabaja afecta principalmente sus posibilidades de ascenso. La tasa de salida varía por deparamento también, pero en una menor medida comparado con el ascenso en los últimos 5 años.\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 266
        },
        "id": "cueMQjoO1b5h",
        "outputId": "04b1a07d-c4d7-4911-ad70-046adf686031"
      },
      "source": [
        "df.groupby('number_project').mean()"
      ],
      "execution_count": 13,
      "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>satisfaction_level</th>\n",
              "      <th>last_evaluation</th>\n",
              "      <th>average_montly_hours</th>\n",
              "      <th>time_spend_company</th>\n",
              "      <th>Work_accident</th>\n",
              "      <th>employee_left</th>\n",
              "      <th>promotion_last_5years</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>number_project</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.478769</td>\n",
              "      <td>0.568505</td>\n",
              "      <td>160.342546</td>\n",
              "      <td>3.180486</td>\n",
              "      <td>0.092965</td>\n",
              "      <td>0.656198</td>\n",
              "      <td>0.015494</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.687670</td>\n",
              "      <td>0.716572</td>\n",
              "      <td>197.507522</td>\n",
              "      <td>3.286560</td>\n",
              "      <td>0.175092</td>\n",
              "      <td>0.017756</td>\n",
              "      <td>0.023428</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.695132</td>\n",
              "      <td>0.740009</td>\n",
              "      <td>205.122108</td>\n",
              "      <td>3.454983</td>\n",
              "      <td>0.156472</td>\n",
              "      <td>0.093700</td>\n",
              "      <td>0.026804</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>0.678888</td>\n",
              "      <td>0.762010</td>\n",
              "      <td>212.061572</td>\n",
              "      <td>3.800072</td>\n",
              "      <td>0.150670</td>\n",
              "      <td>0.221659</td>\n",
              "      <td>0.018109</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>0.273458</td>\n",
              "      <td>0.786550</td>\n",
              "      <td>238.694208</td>\n",
              "      <td>4.192504</td>\n",
              "      <td>0.114140</td>\n",
              "      <td>0.557922</td>\n",
              "      <td>0.017036</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>0.118711</td>\n",
              "      <td>0.859609</td>\n",
              "      <td>276.078125</td>\n",
              "      <td>4.113281</td>\n",
              "      <td>0.015625</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                satisfaction_level  ...  promotion_last_5years\n",
              "number_project                      ...                       \n",
              "2                         0.478769  ...               0.015494\n",
              "3                         0.687670  ...               0.023428\n",
              "4                         0.695132  ...               0.026804\n",
              "5                         0.678888  ...               0.018109\n",
              "6                         0.273458  ...               0.017036\n",
              "7                         0.118711  ...               0.000000\n",
              "\n",
              "[6 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f99BAogo3geN"
      },
      "source": [
        "Podría decirse que los proyectos ejecutados podrían mostrar la responsabilidad asignada en la organización. Debe haber un equilibrio: Ni tan poca responsabilidad que parezca que la organización no confía en el empleado, ni tanta que parezca que el empleado es un solo departamento.\r\n",
        "\r\n",
        "Es interesante que aquellos empleados con el mínimo y máximo número de proyectos asignados son los empleados más propensos a abandonar la organización.\r\n",
        "\r\n",
        "Igualmente, los empleados con el mayor número de proyectos realizados no han tenido un ascenso en los últimos 5 años. \"Mucha carga, poca recomensa\" puede ser un motivo de abandono en esta organización."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "id": "7aJvmF0v63FD",
        "outputId": "87ce6f35-0d5a-48ab-88c9-faf5ee5824d6"
      },
      "source": [
        "\r\n",
        "def heatmap_count(df, var1, var2):\r\n",
        "  \"\"\"Generates a Count heatmap of any two columns of a data frame\r\n",
        "\r\n",
        "  Parameters\r\n",
        "  ----------\r\n",
        "  df : DataFrame\r\n",
        "      A DataFrame that has column features\r\n",
        "  var1 : str\r\n",
        "      The name of a column in the DataFrame\r\n",
        "  var2 : str\r\n",
        "      The name of a column in the DataFrame. Different from var1\r\n",
        "  \r\n",
        "  Returns\r\n",
        "  -------\r\n",
        "  plot : seaborn.heatmap\r\n",
        "      A heatmap of the 2 variables grouped by count, with preset configuration\r\n",
        "  \"\"\"\r\n",
        "\r\n",
        "  heat_df=[]\r\n",
        "  heat_df=pd.DataFrame(df.groupby([var1,var2]).count().iloc[:,1]).reset_index()\r\n",
        "  cols=heat_df.columns\r\n",
        "  value=[y for y in cols if y not in [var1, var2]]\r\n",
        "  heat_df=heat_df.pivot(columns=var1, index=var2, values=value[0])\r\n",
        "  plt.figure(figsize=(15,5))\r\n",
        "  return sns.heatmap(heat_df, cmap=\"Blues\",linewidths=0.5, annot=True, fmt='g')\r\n",
        "\r\n",
        "  \r\n",
        "heatmap_count(df, \"salary_level\", \"employee_left\")"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f01367c0da0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "L7SeLVPFC_mZ"
      },
      "source": [
        "Se observa que el alto salario parece \"prevenir\" el abandono en la empresa, dado que al incrementar el nivel de salario, existe un menor porcentaje de personas que abandonan la empresa."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "id": "bQAKIj9L_ZRn",
        "outputId": "6134cd10-7f4f-4a09-f5f1-effceb520887"
      },
      "source": [
        "heatmap_count(df, \"department\", \"employee_left\")"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f0136dbdc88>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAyMAAAE+CAYAAACeBM+pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeZxN9R/H8dfnzgzGOpaxk12h3Vb6lRKVdqVQoURFRagkSYv2UtqFaJGUFpUSCi0KIXso+xpjN5gZ398f9xhjn6u5c+Ze76fHebj3e7bPOffeM/dzv8sx5xwiIiIiIiLZLeB3ACIiIiIicmJSMiIiIiIiIr5QMiIiIiIiIr5QMiIiIiIiIr5QMiIiIiIiIr5QMiIiIiIiIr6I9TuAY9C4wyIiIiKSHczvADIj/sy7Q/5+nDzjtRx7bDk9GWH8gg1+h+C7RicXA2D2yu0+R+K/U8vmB2Dh2p0+R+K/aiXzArB4fbLPkfivSvF4ABat07moWiJ4LuboekEt73rx5/JtPkfiv9PLFwAg/sy7fY7Ef8kzXgOg/Yg5Pkfiv3duqEXDl3/1O4wcYUKXc/0O4YSV45MRERERERHxWHT1slAyIiIiIiISKSzHtrg6LkpGREREREQihWpGRERERETEF6oZERERERERX0RZzUh0HY2IiIiISDQzC3065iYtj5lNMbM/zWyumT3mlVc0s9/NbLGZfWxmubzy3N7zxd78Chm29ZBX/peZXXKsfSsZERERERGJFBYIfTq23cBFzrnTgTOAS82sPvAs0M85VwXYBLTzlm8HbPLK+3nLYWY1gBZATeBS4A0ziznajpWMiIiIiIhEijDUjLigfTeoivMmB1wEfOqVDwWu8R5f7T3Hm9/IzMwrH+6c2+2cWwIsBuoebd9KRkREREREIkV4akYwsxgzmwmsB8YCfwObnXOp3iIrgTLe4zLACgBv/hagaMbyw6xzWEpGREREREQixXHUjJhZBzOblmHqcPBmnXNpzrkzgLIEazNOzo7D0WhaIiIiIiKR4jhG03LODQAGZHLZzWb2I3AOkGBmsV7tR1lglbfYKqAcsNLMYoFCwMYM5ftkXOewVDMiIiIiIhIpwjOaVqKZJXiP44HGwHzgR+B6b7E2wJfe41Hec7z5PzjnnFfewhttqyJQFZhytH2rZkREREREJFKE5z4jpYCh3shXAWCEc+5rM5sHDDezJ4EZwCBv+UHA+2a2GEgiOIIWzrm5ZjYCmAekAp2cc2lH27GSERERERGRSBGGZMQ5Nws48zDl/3CY0bCcc7uA5kfYVl+gb2b3rWZaIiIiIiLiC9WMiIiIiIhEisCx+4BEEiUjIiIiIiKRIjx9RnyjZEREREREJFJkYnSsSKJkREREREQkUqhmREREREREfKGaERERERER8YVqRkRERERExBeqGREREREREV+oZkRERERERHyhmhEREREREfGFakZERERERMQXqhkRERERERFfqGZERERERER8oWRERERERER8oWZaIiIiIiLiC9WMiIiIiIiIL6KsZiS6UisREREREYkYqhkREREREYkUaqYlIiIiIiK+iLJmWkpGREREREQihCkZERERERERPygZERERERERf0RXLqJkREREREQkUqhmREREREREfKFkREREREREfKFkREREREREfKFkRERERERE/BFduYiSERERERGRSKGaERERERER8YWSERERERER8YWSERERERER8YWSERERERER8Ud05SJKRkREREREIoVqRkRERERExBfRlowE/A5AREREREROTEpGREREREQihJmFPGVim+XM7Eczm2dmc82ss1fex8xWmdlMb2qaYZ2HzGyxmf1lZpdkKL/UK1tsZj2OtW810xIRERERiRThaaWVCnRzzk03swLAH2Y21pvXzzn3wgEhmNUAWgA1gdLAODOr5s1+HWgMrASmmtko59y8I+1YyYiIiIiISIQIR58R59waYI33eJuZzQfKHGWVq4HhzrndwBIzWwzU9eYtds7948U63Fv2iMmImmmJiIiIiESIcDTTOmj7FYAzgd+9orvNbJaZDTazwl5ZGWBFhtVWemVHKj8iJSMiIiIiIhHieJIRM+tgZtMyTB2OsO38wEigi3NuK/AmUBk4g2DNyYtZfTxqpiUiIiIiEiGOp5mWc24AMOAY240jmIh86Jz7zFtvXYb57wBfe09XAeUyrF7WK+Mo5YelmhERERERkUhhxzEda5PBDGcQMN8591KG8lIZFrsWmOM9HgW0MLPcZlYRqApMAaYCVc2sopnlItjJfdTR9q2aERERERGRCBGmmx42AG4BZpvZTK+sJ9DSzM4AHLAUuAPAOTfXzEYQ7JieCnRyzqV58d0NjAFigMHOublH27GSERERERGRCBGm0bR+5vB1KKOPsk5foO9hykcfbb2DKRkREREREYkQYaoZ8Y2SERERERGRSBFduYiSEYAfvhrBL9+PAudo0OQqLrrqRr7+aBC/fD+KAoUSALjq5juoVfvc9HWS/l3LE3ffTNMWt9H42lZ+hZ7ldmzfxpsvPMHypYsxMzp2f5TqNU9j9OfD+e7LEQQCMZxd7zxuuaMzAEv/XsSAfn3ZuXMHgYDxzBvvkytXbp+PImu0u7Ep8fH5CMQEiImJod+AYQB8NfIjvvliBIFAgDr1/8etd3Vh4fw5vPbCEwA452jV9k7OOf8iP8PPUrc2v4z4vPkIBALExMTyysBhDHr9Jab8OonY2DhKlSlLl4ceI3+BgunrrF+3hrtuaUarW+/kupZtfIw+a912w2UZ3hexvPzOMN4f+Dq//zwBCxgJCUXo0vNxihYrzvZtW3n5mUdZu2olcbly0bnHY1SoVMXvQ8gyO7Zv440M14tO3R9l5rTJjPvmcwomBIeib9WuE2fXO4/1a1fT+dbrKV3uJACqnXIqd9zX08/ws8zqFUvp9+T+Y1m/dhU3tLmDCy6+nH59H+LftWtILFmK+3o9c8BnZPFfc+l17210ebgv9c+/2I/Q/7PcuWIZN6gLuXLFEhsTw+fjZvDkW6O588bzubvVhVQun0jZCx9k4+YdALS4rDZd2zbGzNi+cxf3PvUxsxcGB9oplD+eNx9tRY3KpXAO7nzsQ36ftcTPwwtZmzplOK1UAbbtTqXPmMXp5RdVKULDKkVxzjFrzTZGzlpHvfKFuKR6sfRlyiTk4cmxf7Ni8670sk4NypOYP9cB24oEiflz0fOSqhTOG4cDvp69jpEz11AgdyyPNq1GyYK5Wbt1N31G/8X23WlcXL0YLWuXwQx27kmj3w//8PeGneSKMV5pXou4mAAxAWPioo0M+W3FMfcf7VQzEmVWL/uHX74fxYMvDCQmNpbX+nSjVp0GAFx01Y1HTDRGDnqVGmfVz85Qs8Xg157njDrn0L3Pc6SkpLBn9y7mzJjK1F8n8uKA4cTlysWWTUkApKWl0v/pXtz70BNUqFyNbVs2ExMTXW+pvi8PoFBC4fTns6ZP5fdfJvDqoI+Jy5WLzd65KF+xMv3e/pCY2FiSNv7LvbfdSN1zzycmNnrOx9OvvHPAuTizTn3a3nEvMbGxDH7zZUZ8MJjb7uqSPn/gqy9ydr0GfoQadk8ddC6ua9mGW27vBMCoT4fx0ZAB3N29FyPeH0ilKtXp1bcfK5Yt4c1+T/PUy0cdWTGiDH7tec6scw73Z7hezJw2mSuub8XVN7Q+ZPkSpcvy4oCPfIg0vEqXq8Dzbwd/rNiblsYdLZtSt8GFfPHxEE49sy7XtGjLF8OH8MXwIdzc/t705T4c+Cqnn13Pz9D/s917Urm0Q392JO8hNjbAD4O78v0v85g88x9GT5rD9wM7H7D80tUbaXL7y2zelkyTBjV4vVdLzm/9AgAvPHA93/86j1b3DyIuNoa8eXL5cUj/ya9LNvHjoo3cVq9seln1xHycXqYgj3+/mNS9jgK5YwD4ffkWfl++BYAyhXLTscFJByQiZ5YpyO7Uvdl7AFkkba/jjUlLWfTvDuLjAgxodTrTlm/m0hrFmb5iC8OmraJV7TK0qlOWAT8vY83W3XT+dA7bd6dRt0IC3S6uTMfhs9mT5ug6ci7JKXuJCRiv3lCLKUs3MW/tdr8P0VfRloyEdWhfMzvZzB40s/7e9KCZnRLOfYZq7cqlVKhWk1y58xATE0vVWmcwc/LEo64z87dJFC1RilLlK2ZTlNljx/ZtzJ89g0ZNrwEgLi6OfPkLMOarT7m2RVvicgX/MBQqXASAP6f9xkmVqlKhcjUAChRKICYmxp/gs8noLz/h+la3pp+LBO9c5MkTn5547NmzJ+ouFIdzVt1z04/55JqnsfHf9KHImTzpB0qUKs1JFSv7FV62ypsvf/rjXbuS01//5Uv/4bSz6gJQ7qSKrF+7mk1JG32JMavt2L6NeYe5XpzoZs+YSslSZUgsUYqpv07kgsZXAHBB4yuY+uuE9OW+/fJj6p13EQUTivgTaBbakbwHgLjYGGJjY3DO8edfK1m+JumQZX/7cwmbtyUDMGXWEsqUCLY+KJg/D+edVZkhn08GICU1jS3bk7PpCLLOog072bEn7YCyhlWK8N38f0nd6wDYtjvtkPXqlk9g6vLN6c9zxwZoXK0o38z/N7wBh0nSzhQW/RusDUtO2cuypGSK5c9Fg0pF+G7eegC+m7ee8yoF3/9z12xju3de5q3ZRmL+/YlockowIYsNGLEBw2XngeRQ4b4De3YLWzJiZg8Cwwm2bJviTQZ8ZGY9wrXfUJUqX4m/5/3J9q1b2LN7F3P/mMymDcEvVRNHj+TJe1vzfv+n2Ll9KwC7kncy9rMPaNriNj/DDov1a1dTsFBhXn+uD93vaMWbLzzOruRk1qxczvzZM+jRqTW972vP4gXBEdpWr1yOmfHEg524/45WfDF8qM9HkNWM3t070qV9K74bNRKA1SuXMXfWDLrdeQs97m3Hwvn7R6v7a95sOra5jntubU7Hrg9HVa2ImfFI17u4t11Lvh316SHzx37zBWfXOw+A5J07+XTYEFrdemd2h5ktDKN3t7vofHtLvstwLt5751XaXncJE8aO5uZ2dwFQsUo1Jk8aDwTfH+vXrTkgaYtk+64Xr3nXize86wXAt1+M4L7bb+T15x9j+7atGdZZRfc7WvHIfe2ZN2uGX6GH1S8TxtDgwksA2LIpicJFg81wEooUTa9VTtqwnik/T6DJldf7FmdWCgSM34b3YPn4Z/jhtwVMnbMsU+u1veZcxvwyD4AKpYuyYdN2Bjx2M5M/epA3ereKyJqRwymRPxdVE/PxUKNKdG9YkQqF4w9Zpna5QkzxakkArq5VnLELN7AnQmtGMipZMDdVE/Mxf+12iuSLI2lnChBMWIrkiztk+ctrlmDK0v2JWcBg4E2n80WHOkxbvoX5J3itCCgZCUU7oI5z7hnn3Afe9AxQ15uXI5QqV4HGzW7i1T738VqfrpStWJVAIMD5l13L42+NoOfLQyhYuCgjB78GwDfDB3PRVTeSJz6vz5FnvbS0NP5ZtIAmV13PC28PI3eeeD4f/i5paWls37aVp18byi13dOalJ3rgnCMtLZUFc2bSueeTPPnKIKb8/COzpk/x+zCyzHOvvcsrAz+iz3Ov8c0XHzPnzz+C52LrFl548z1uu+s+nu3zAM4Ff6epXuNU3hg6kpfe+oBPPhzMnt27fT6CrPPc6+/Sf/BwHn/hdb75bARzZv6RPm/4e+8QExPDhU2aAvDhu29xzQ03EZ83+j4jAM++/i6vDBrOY8+/ztef7z8Xrdvfw5CRY2jYuClffzYcgOY33caO7du457Yb+Pqz4VSuWp1AIDruNbvvenHJQdeLS668ntff/5IXB3xEQpFiDH2rHwCFixTj7WHf8MLbw2h7V1defuphdu6Iri8VqSkp/DF5EvUvOLT/R8YvBEPeeJGbbr8nat4Le/c66rd4hiqX9KJ2rZOoUbnUMdc5v3ZV2lxzDr1e+RKA2NgYzji5HO988hPntHyWncm76X5b43CHni0CASNfrhieHv8Pn85ayx3nlDtgfsUi8exJ3cvqrcG/GeUS8pCYLxczVm3zI9wsFR8X4LHLq/PaxCXs3HNojZA7qJrjjLIFaVqrOG//vD+h3evg9g//pPmgaZxSIj8Vi0bn35aQhOGmh34K55VwL1D6MOWlvHmHZWYdzGyamU0bMCB72lY3aHwlD700mK5Pv0He/AUoXro8BROKEIiJIRAIcF6Tq1i6KPjrzdKFc/l86Bv0an8dP341gjGfvseEbw79pTgSFU0sTtHE4lQ75VQA6p9/MUsWLaBoYnHqnXchZkbVk2thZmzdspmixUpwyqlnUrBQYXLniefMeg1YsmiBz0eRdYomFgeCTbHO+d9FLJw/l2KJJTjn/EaYGdVOqUUgEGDrlk0HrFeuQiXi4/OybElkdTg8mmKJJQDvXJx/IX/ND96AdezoL5n660907/1U+hethfNmM/jNl7m1+WV8+cmHjHh/EF+NHO5b7FntgHPxvwtZOH/OAfMbNm7KLxODtSF58+Wny0OP8+rgEXR9+Em2bN5EydJlD9lmJDr4enHO+Rfzz6IFJBQpSox37Wx8+bUs8mpS43LlSh8QpHK1UyhZuiyrVy73Lf5wmDH1FypWOZmEwkWBYJPWTRs3ALBp44b0Tv1/L5rPK0/1pNPNV/LbT+MZ+OqzTPllgl9hZ5kt25OZOG0hTc6tcdTlalUtzZu9W9H8vgEkbQk25Vm1bhOr1m9Or1X5fNxMzji53NE2EzE27Uxh+spgDeHSpGT2Avlz72/SXKd8Iaau2F8rUqloXioUiefpy6vx4EWVKJE/F90bRl6z8JiA8dgV1Rm34F9++turFdyRQpG8wdqQInnj2OTVkgBUKpaX+y+uwsOjFrB1V+oh29u+O40ZK7dQ96SE7DkAyTbhbEfSBRhvZouAfUMflAeqAHcfaSXn3ABgXxbixi/YEMYQg7Zt3kSBhMIk/buWmZMncv9zA9iStIFCRYLV6zN/m0jp8pUA6Pb0m+nrff3RIHLniafh5dFR1V64SDGKJpZg1YqllClXgdkzplD2pEqUKF2WOTOnUevMOqxesYzU1FQKFkrgjDrn8OXHQ9m9K5nYuDjmzZrOFddFx8hiu5KT2ev2kjdvPnYlJzNj6mRatOlAfHw8s2ZM5bSz6rBqxTJSU1IoWKgwa9esIjGxBDGxsaxfu5qVy5dQvOThcvHIc/C5mD51Mi3b3sG0339h5LChPPvqQPLk2d/s4LnX301//OHgN8kTn5crr2vhR+hZ7nDvi5Zt72DVimWU8UaJ+v3nCZT1+pNt37aV3HniiYuLY8zXn1Hz9LMP6F8SyQoXKUaxw1wvNm38l8JFEwH4/ecfKV8h2G9oy+ZN5C9QkJiYGNauXsmalcspUaqMn4eQ5X75cX8TLYDa51zAxLFfc02Ltkwc+zV1zr0AgNffH5W+zOvP9eHs+udRt0HD7A43SxQrnJ+UlGD/jjy542hU72ReHDLuiMuXK1mY4S+0p90j77F4+fr08nUbt7Fy7SaqnlScRcvW07BudRb8szY7DiHsZq7eSvXi+fjr3x2UyJ+L2ICl948woHbZQjz34z/py0/8O4mJ3pf3onnjuOd/J/HChMgaVQzggYsrszwpmU9mrEkv+/WfJC6tUZxh01ZxaY3i/PJP8DiLF8jFE1dU56kxi1iZoRN/ofhY0vY6tu9OI1dMgNrlE/ho2qpsP5acJqc3uwpV2JIR59x3ZlaNYLOsfX9xVgFT990uPqcY8GxPdmzdSkxsLDfe0Y28+QswpF8/Vi5ZBBhFi5ekVccH/A4zW7S75wFeeaoXqSkplChVhk4P9CF3nnjeeP4x7mt3A7Gxsdz9YB/MjPwFCnLl9TfzYMfWmBln1W3A2fX/5/chZInNmzbSt1dXINgc5YKLL+Pseg1ISUmh/7N96NT2emJj4+jS83HMjHmzZvDpsHeJjY3FLMCd9/U8YLSlSLZp00b69tx3LlK5oPFl1K7XgNtbXElKyh4e7hrsG3JyzdO4u3svP0MNu82bNvLkw8FzsTctNf198VSvbqxcsZSABUgsWYpO3R4GYMWyJfR76hHMjPIVKtO5Rx8fo896+64XKd714u4H+jDotedZ+vdfgFG8ZGnu9IbvnTdrOsOHvOV9RowOXXpSoGAhfw8gC+1KTmbWH1Po0OXh9LJrWrSh3xMP8cO3X5JYohT39XraxwjDo2Sxgrzz+C3EBAIEAsbIsdP59qc5dGx5AV3bXEyJogWZOqIn3/08l46PD+OhDpdRJCEfLz90IwCpaXs576bnAOj67Ce8+1RbcsXGsHTVBjo8+oGfh3Zc2tcvS7XEfOTPHctzV1Rn1Nz1/LxkM23rlKHPJVVI3et4d8rK9OWrJuZjU3IKG3akHGWrkefU0gW4pEZx/v53BwNvOh2Ad35ZxrBpq3i0aTWa1izOum276fPNQgDa1CtHwTxx3HdR8IfftL2OOz6aRdF8uXioSRUCZgTM+HHRBiYv2XTE/Z4ooi0ZMXdwg72cJVtqRnK6RicHa2hmr4yu9tXH49SywV+VF67d6XMk/qtWMthudvH6yBtxJqtVKR6smVm0TueiaonguZij6wW1vOvFn8sjv+39f3V6+eBIZ/FnHrFhwgkjeUawD2j7EXOOsWT0e+eGWjR8+Ve/w8gRJnQ5F3J874qgKt2/DfnL++IXLsuxxxY9w/2IiIiIiES5aKsZUTIiIiIiIhIhoiwXUTIiIiIiIhIpVDMiIiIiIiK+iLJcRMmIiIiIiEikCASiKxtRMiIiIiIiEiFUMyIiIiIiIr5QnxEREREREfFFlOUiSkZERERERCKFakZERERERMQXSkZERERERMQXUZaLKBkREREREYkUqhkRERERERFfRFkuQsDvAERERERE5MSkmhERERERkQihZloiIiIiIuKLKMtFlIyIiIiIiEQK1YyIiIiIiIgvoiwXUTIiIiIiIhIpVDMiIiIiIiK+iLJcRMmIiIiIiEikUM2IiIiIiIj4IspyESUjIiIiIiKRQjUjIiIiIiLiiyjLRZSMiIiIiIhEimirGQn4HYCIiIiIiGSOmYU8ZWKb5czsRzObZ2ZzzayzV17EzMaa2SLv/8JeuZlZfzNbbGazzOysDNtq4y2/yMzaHGvfSkZERERERCKEWehTJqQC3ZxzNYD6QCczqwH0AMY756oC473nAJcBVb2pA/BmMDYrAjwK1APqAo/uS2CORMmIiIiIiEiECEfNiHNujXNuuvd4GzAfKANcDQz1FhsKXOM9vhp4zwX9BiSYWSngEmCscy7JObcJGAtcerR9q8+IiIiIiEiECHeXETOrAJwJ/A6UcM6t8WatBUp4j8sAKzKsttIrO1L5EalmREREREQkQhxPzYiZdTCzaRmmDkfYdn5gJNDFObc14zznnANcVh9PppMRM6uYmTIREREREck5nHMDnHO1M0wDDl7GzOIIJiIfOuc+84rXec2v8P5f75WvAsplWL2sV3ak8iMKpWZk5GHKPg1hfRERERER+Q/C0YHdgh1LBgHznXMvZZg1Ctg3IlYb4MsM5a29UbXqA1u85lxjgCZmVtjruN7EKzuiY/YZMbOTgZpAITNrlmFWQSDPMY9ORERERESyRCA8nUYaALcAs81splfWE3gGGGFm7YBlwA3evNFAU2AxsBO4FcA5l2RmTwBTveUed84lHW3HmenAXh24AkgArsxQvg1on4n1RUREREQkC4QjF3HO/QwcacuNDrO8AzodYVuDgcGZ3XdmkpHrnXO3mFlP59xTmd2wiIiIiIhkrWi7A3tmkpGzzaw0cKOZvclBWdOxql5ERERERCRrBKIrF8lUMvIWwTsuVgKmHzTPeeUiIiIiIhJmJ1zNiHOuP9DfzN50zt2VDTGJiIiIiMhhRFkukvmhfZ1zd5nZeWZ2K4CZFdN9RkREREREso8dx7+cLDPNtAAws0eB2gRH13oXyAV8QHAoMBERERERCbMTsc/IPtcCZ+L1G3HOrTazAmGJSkREREREDnHC9RnJYI9zzpmZAzCzfGGKSUREREREDiPKcpGQkpERZvY2kGBm7YHbgHfCE5aIiIiIiBwsTHdg902mkxHn3Atm1hjYSrDfSG/n3NiwRSYiIiIiIgeIslwkpJoRvORDCYiIiIiIiA9OuD4jZraN4M0ND5kFOOdcwSyPSkREREREDhFluUimbnqYqRGzzKywc27Tfw9JREREREQOJ9r6jGT6poeZMD4LtyUiIiIiIlEupD4jxxBdaZqIiIiISA4TbV+4szIZOVy/EhERERERySInXAd2ERERERHJGQLRlYuomZaIiIiISKSItpqRkDqwm9l5Znar9zjRzCpmmN0oSyMTEREREZEDmIU+5WSZrhkxs0eB2gTvvv4uEAd8ADQAcM4lhSNAEREREREJiraakVCaaV0LnAlMB3DOrTazTN2DRERERERE/rsTuc/IHuecMzMHYGb5whSTiIiIiIgcRrTVjITSZ2SEmb0NJJhZe2Ac8E54whIRERERkYPZcUw5WaZrRpxzL5hZY2ArwX4jvZ1zY8MWmYiIiIiIHCAQZTUjoQ7tuxBwzrlxZpbXzAo457aFIzARERERETlQlOUimW+m5TXN+hR42ysqA3wRjqBERERERORQZhbylJOF0mekE8FhfLcCOOcWAcXDEZSIiIiIiBzqhL3PCLDbObdnX3ZlZrGAC0tUIiIiIiJyiBO5z8hEM+sJxHsd2TsCX4UnLBEREREROViU5SKYc5mr3DCzANAOaEJwlLAxwECX2Q0cH9W8iIiIiEh2iIiv+Z0+nx/y9+PXrz0lxx5bKDUjlwODnHO6t4iIiIiIiA9C6fAdCUJJRm4EXjazkcBg59yCMMV0gA3bU7NjNzlasfzBl2lniiqK8sYFE/styXt9jsR/heKDl6PNyWk+R+K/hPgYAHbpckEe76quz8j+z4jOxf5zMXruep8j8V/TmsGxdxau3elzJP6rVjIvfy7XHRoATi9fwO8QTliZTq6cczcDZwJ/A0PMbLKZdTAzvXoiIiIiItngRB7aF+fcVoL3GhkOlAKuBaab2T1hiE1ERERERC6zvYEAACAASURBVDIIWOhTTpbpZlpmdhVwK1AFeA+o65xbb2Z5gXnAq+EJUUREREREIOcnF6EKpc/IdUA/59ykjIXOuZ1m1i5rwxIRERERkYPl9GZXoQqlz0gb4C8zu8KbimeYNz4s0YmIiIiISLpwNNMys8Fmtt7M5mQo62Nmq8xspjc1zTDvITNbbGZ/mdklGcov9coWm1mPTB1PZg/czJoDU4DmwA3A72Z2fWbXFxERERGR/8Ys9CkThgCXHqa8n3PuDG8aHdy/1QBaADW9dd4wsxgziwFeBy4DagAtvWWPKpRmWr2AOs659V4gicA4gh3aRUREREQkzAJhaKblnJtkZhUyufjVwHDn3G5giZktBup68xY75/4BMLPh3rLzjraxUEbTCuxLRDwbQ1xfRERERET+g8BxTP/B3WY2y2vGVdgrKwOsyLDMSq/sSOVHFUp835nZGDNra2ZtgW+A0SGsLyIiIiIi/8HxNNPy7g04LcPUIRO7ehOoDJwBrAFeDMfxZLqZlnPufjO7DmjgFQ1wzn0ejqBERERERORQx9NMyzk3ABgQ4jrr9j02s3eAr72nq4ByGRYt65VxlPIjCqXPCM65kcDIUNYREREREZGskV0j+5pZKefcGu/ptcC+kbZGAcPM7CWgNFCV4CBXBlQ1s4oEk5AWQKtj7eeYyYiZbQPc4WYBzjlX8FjbEBERERGR/y4cNz00s4+AhkAxM1sJPAo0NLMzCOYBS4E7AJxzc81sBMGO6alAJ+dcmredu4ExQAww2Dk391j7PmYy4pwrcBzHJCIiIiIiWSxMo2m1PEzxoKMs3xfoe5jy0YTYpzykZlpmdhZwHsEM6Wfn3IxQ1hcRERERkeMXZTdgD+mmh72BoUBRoBgwxMx6hSswERERERE5UDjuwO6nUGpGbgJOd87tAjCzZ4CZwJPhCExERERERA5k5PDsIkSh3GdkNZAnw/PcZGK4LhERERERkcMJpWZkCzDXzMYS7DPSGJhiZv0BnHP3hiE+ERERERHx5PRmV6EKJRn53Jv2mZC1oYiIiIiIyNGcsMmIc25oOAMREREREZGjsygbTiuU0bSuMLMZZpZkZlvNbJuZbQ1ncCIiIiIist+JPJrWy0AzYLZz7nB3ZBcRERERkTCKsoqRkJKRFcAcJSIiIiIiIv4Ixx3Y/RRKMvIAMNrMJgK79xU6517K8qhEREREROQQOb3ZVahCSUb6AtsJ3mskV3jCERERERGRI4myipGQkpHSzrlaYYtERERERESOKnAC34F9tJk1CVskIiIiIiJyVGahTzlZKDUjdwHdzGwPkAIY4JxzBcMSmYiIiIiIHOBE7jNSCLgJqOice9zMygOlwhOWiIiIiIgcLNpG0wqlmdbrQH2gpfd8G/BalkckIiIiIiKHdSI306rnnDvLzGYAOOc2mZlG1RIRERERySbRVjMSSjKSYmYxgAMws0Rgb1iiEhERERGRQ0RZLhJSM63+wOdAcTPrC/wMPBWWqERERERE5BCB45hyskzXjDjnPjSzP4BGBEfSusY5Nz9skYmIiIiISFQLpZkWzrkFwIIwxSIiIiIiIkdhUdZOK6RkRERERERE/BNdqYiSERERERGRiHEij6YlIiIiIiI+iq5URMmIiIiIiEjEiLKKESUjIiIiIiKRQh3YRURERETEFzn9viGhUjIiIiIiIhIhVDMiIiIiIiK+iK5URMmIiIiIiEjEUM2IiIiIiIj4Qn1GRERERETEF6oZERERERERX0RXKqJkREREREQkYkRZxYiSERERERGRSBGIsrqRaOsDIyIiIiIStcxCn469TRtsZuvNbE6GsiJmNtbMFnn/F/bKzcz6m9liM5tlZmdlWKeNt/wiM2uTmeNRMiIiIiIicmIbAlx6UFkPYLxzriow3nsOcBlQ1Zs6AG9CMHkBHgXqAXWBR/clMEejZEREREREJELYcfw7FufcJCDpoOKrgaHe46HANRnK33NBvwEJZlYKuAQY65xLcs5tAsZyaIJzCCUjIiIiIiIR4niaaZlZBzOblmHqkIldlXDOrfEerwVKeI/LACsyLLfSKztS+VGpA7uIiIiISIQ4ng7szrkBwIDj3adzzpmZO971j0Y1IyIiIiIiESIcHdiPYJ3X/Arv//Ve+SqgXIblynplRyo/KiUjIiIiIiIRIhuTkVHAvhGx2gBfZihv7Y2qVR/Y4jXnGgM0MbPCXsf1Jl7ZUamZloiIiIhIhMhMh/SQt2n2EdAQKGZmKwmOivUMMMLM2gHLgBu8xUcDTYHFwE7gVgDnXJKZPQFM9ZZ73Dl3cKf4QygZERERERGJEIEw3PPQOdfyCLMaHWZZB3Q6wnYGA4ND2beSERERERGRCBGOmhE/KRkREREREYkQ/6EPSI6kZEREREREJEKoZiTKrFu7hid6P8SmpI1gxtXXNueGVrfwSI9uLF+2BIDt27aRv0ABhn70GfPmzOLZvn2CKzvHbR06ccFFF/t3AFmoT6+eTJo0gSJFivLpF18B0O+F55g08UfiYuMoW648jz35FAUKFmTz5k3cf19n5s6Zw1XXXEOPh3v7HH3WWrd2DX169SApaSMA1153Ay1uas2WLZt5+IGurFm9ilKly/DU8/0oWLBQ+nrz5symXZuWPPnMizRqfIlf4Wep4Ll4iKSkDRjGNdfdQIubbmH899/xzluvs3TJP7z7wcecUrMWAKtXraJFsysof1IFAGqddjo9evXx7wDC6MP3hzLy009wznHd9c25uXVbvh/zLW++/hpL/vmbD4d/Qs1ap/odZliE+hmZ+ON43n6jP2YBYmJj6Hr/Q5xx5tk+H0XWeOLRh/l50gQKFynC8JHBa+fCvxbwTN8+JO/cSanSZXj8qefJnz8/qSkpPPnYI/y1YB5paWk0veJq2rbLzL3Hcq6PXnuaedN+JX+hwjz4ynsA7Ni2lfdefJSkf9dSJLEkbbo/Tt78BVi3chkfvfY0K/9ZyOWt2nPhNQc2U9+blsZLD7SnUJFitH/4OT8OJ0u1u7Ep8fH5CMQEiImJod+AYQx79y3GfP0ZhRIKA9C6/d3Urv8/1q1ZTcfWzShT/iQAqtc4lU7devkZfpbZsX0bb730BCuW/o1h3NW9N7ly5+GdV55mV/JOEkuW5t4eT5A3X35+Gv8to0a8n77u8iWLePaND6hQpbqPR5CzhKPPiJ9O+GQkJiaWe+57gOqn1GDHjh20u7k5deqfwxPPvJi+zKsvPUe+/PkBqFS5KoPeH0FsbCwb/v2XNi2b0eD8hsTGRv6pvPKaa7mx1U080rNHeln9c87lni5diY2N5ZWXXmDwwAF07tqd3Lly0/GezixetIi/Fy/0MerwiImJoXO3Bzj5lJrs2LGD1i2vo279c/l61OfUqXcObW5rz9DB7zB08Dvc06U7AGlpabz6yovUq3+uz9FnrZiYWO9cBD8jbVpeT93651CpSlWefak/zzzR55B1ypQtxwcjPs/+YLPRokULGfnpJ3w4/BPi4uLoeMftnH/BhVSpUo1+r7zKE4896neIYRXqZ6ROvfqc3/AizIxFC/+i5wP38ckXo/0+jCxx+VXX0LxFK/r02n/t7PvYI3Tuej9n1a7LqC9G8sHQQdzZqTPjxo4hJWUPH306il3JydzY7AqaXHo5pcsc8ybFOVbdCy/jvMuaMax/3/Sy8Z9/QNXTzubiZjcz7rMPGP/ZB1zZ+i7y5i9Is3admT3lp8Nua9I3n1Ci7Ens2rkju8IPu74vD0hPPPa5uvnNNGvR+pBlS5YpS/9BH2dXaNnm3Tde4Iza59Kt93OkpqSwe/cunnywE7d06EyN08/mh+++ZNQn79Oi7V38r9Fl/K/RZQAsX7KY5x/tpkTkINFWM3LC32ekWGIi1U+pAUC+fPk4qWIl/l2/Pn2+c44fxo2h8aWXA5AnPj498dizZzcWRQ33zq5dh0KFCh1Qdk6D89KP99TTTmfdurUAxOfNy5lnnU3u3LmyPc7sUCyxOCefUhMIvi8qVqrMv+vXMWnCD1x+5dUAXH7l1Uz8cXz6OiM++oCLGjWmcJGivsQcLsUSEzk5w2ekQqXgZ6RipcqcVKGiz9H5Z8k/f3PqaacR710Tzq5dh/HjvqdS5cpUqFjJ7/DCLtTPSN68+dKvl8nJO6Pq2nnW2XUoWDDhgLLly5dy5tl1AKhX/1x+HD8WADMjOTmZ1NRUdu3eRWxcHPny58v2mLNS5ZpnkK9AwQPK5kz5mToNLwWgTsNL05OPAgmFKV/1FGJiDv0Bb/OG9cz7YzL1L74i/EFLttm5YzvzZ8/gosuC14Xge74Aq1cu45TTzgLgtLPq8ftPPxyy7s8/jOHchk2yNd5IkI33GckWviQjZnarH/s9ljWrV7FowXxq1jotvezPGX9QuEhRynnVpgBzZ8/ipuZX0frGa7j/od5RUSuSGV9+PpIG553vdxjZbvWqVfy1YD41Tz2dpI0bKZZYHICixRJJ2hhsorJ+3Tom/DiO62440sh40WH1qlUsXDCfmqeedszlbrmxGXe2a82M6dOyKbrsVaVKNab/8QebN28iOTmZn3+axNq1a/0OyxeZ+YwA/PjDWJpf05Su99xFrz5P+hVutqhUqUp6IjZu7BjWrV0DQKOLmxAfH0/Txudz1aWNuLn1bRQqlHC0TUWkbZs3UahIMQAKFi7Kts2bjrnO54P7c2XrjphF0++kRu/uHenSvhXfjRqZXvrN58O559YbeOWZPmzftjW9fN2aVXRu14Ie97Zj7p/T/Qg4y61fs4qChRJ44/nHeODOVrz14hPsSk6mXIXKTP11IgC/TRrHxn/XHbLu5Inf0+DC6GjynJXsOKaczK9P/GM+7feIdu7cwcP3d+He7j3Sm2QBjP1uNI0vaXrAsjVPPY0PPxnFwPc/5v0h77B79+7sDjfbDXz7LWJiYml6xZV+h5Ktdu7cQY/u99L1/h7kz/C+gOAvnPt+3X3p+ae5u3M3AoFo+iN6oOC56Mx99z90yLnIqFhiIqO+G8/7H39G524P0vuhB9i+fXs2Rpo9KlWuzK3tbufO9u3oeMftVD/5ZGKi+PU/ksx+RgAuvKgxn3wxmuf6vcrbb/TP7lCz1SOP9WXkiI9o3fI6du7YQWxcHABz58wmEIhh9PcT+WL0WD58/11WrVzhc7ThFXwfHH2ZudN+oUChwpSrHF3NcZ577V1eGfgRfZ57jW+++Jg5f/7BZVc3Z8Cwr3hl0HAKFy3GoNdfAqBI0WIMHvEtrwwazu2duvHCEz3ZuSPyr51paWksWfQXTa68nufeGkbuPPF88fEQ7urWm+9HfcKDHW8mOXknsbFxB6y3aP4ccuXOQ/mKVXyKPOcKmIU85WRh+0nfzGYdaRZQ4ijrdQA6ALz99ts0a3VbGKI7UGpKCg/f34Uml11Ow4sa7y9PTWXij+MY/MGIw65XoWJl4uPz8s/fizilRq2wx+mXUV98xqRJP/L2wCFR1bTiWFJTUniwW2cuaXolFzYKVhMXKVqUDf+up1hicTb8u57CRYoAMH/eHHo92A2AzZs38+vPk4iJiaFhlAxukJqSQo9uXbi06RVc2KjxUZfNlSsXuXIFm++dUqMmZcuWY8Wypekd3KNJs+ua0+y65gD0f/klSpQ44qUtKoXyGcnorLPr8PjKlWzetImEwoUPmR8NKlSsxKtvDQJg2bIl/PJT8BfgMd9+HWz+GhdHkSJFOf2Ms5g3dw5lypbzM9wsVyChMFuSNlCoSDG2JG0gf6Gjv85LFsxmztRfmDf9N1JT9rBr5w4+ePlxbu4S2YOjFPVqCRMKF+Gc/13EwvlzqXX6/oEbLrmiGY8/dC8AcblyEeddO6tUr0HJMmVZtWIZVU+umf2BZ6GiicUpmlicqqcE/wbUP78RXwwfQou2d9Hr2dcBWL1yGdN///mA9X6ZMEa1IkcQbd/EwvkzXgmgNXDlYaaNR1rJOTfAOVfbOVe7Q4fwjzDinOPpJ3pzUsVKtLi57QHzpk2ZzEkVKlK8RMn0stWrVpKamgrA2jWrWbZ0CaVKRW7Hw2P55eefGDJ4EC+/+ibx8fF+h5NtnHM88VgvKlasxE23tE0vP/+Ci/jmqy8B+OarLzm/4UUAfDl6HF9+O54vvx3PRRc34YGevaMmEXHO8eRjj1ChYiVaZTgXR7IpKYm0tDQAVq1cwYrlyyhdtmyYo/THRq8J0prVqxk/7nsuu/zEqTkM9TOyYvkygjfthQXz55KyZw+FEqKvedI++0YZ27t3L4PfeYtmzW8EoESpUkyb8jsQ7DszZ/afUdnHqFadBkyd8B0AUyd8R6265x11+StuvpM+Az+j99uf0LprH6qeelbEJyK7kpPZ6XXE35WczIypkzmpYmWSNv6bvszkn37gpIqVAdiyef+1c+3qlaxeuZySpSP/2plQpBhFE0uwesVSAGbPmELZkyqxZVMSEPyMfPbhIBpfcV36Onv37mXyxHE0uFD9RU4E4ezs8DWQ3zk38+AZZjYhjPsNyayZ0/num1FUrlKNNi2bAXBHpy6ce975jBvzLRcf1ERr1szpvD9kILGxsQQsQPcej0TNL3s97u/KH1OnsnnzJi5pdAF3dryHdwcOYM+ePdzVPlhDdeppp9Pr0WAru6ZNLmLH9h2kpKTw4w/jeWPAICpXjo7q1D9nTufbr0dRpWo1brrhWgA63tOF1rfdTs8HujLq808pWbo0Tz3Xz+dIwy/jubjZOxd33dOFlJQUXnimL5s3JXHfPXdRrfrJ9H/zHWZMn8aAN14NfkYCAR7s9WhUtokH6NblHrZs3kxsbCw9ez1KwYIFGT9uLM889QSbkpK4u+MdVK9+Cm+9M8jvULNcqJ+RH8Z/z+ivviQ2No7ceXLT97mXoqamtVePbvwxbQqbN2/miiYNaX/X3STv3MknHw8D4MJGjbny6uDfl+Y3tuLx3g9zY7NgJ+0rrrqWqtUiu2nSey/1YfGcGezYtoU+tzfj0ha30ajZzQx9oTe/j/+GwoklaNPtcQC2btrIS/e3Z1fyDswCTPz6E3r0f588eSO7E//hbN60kb69ugLBpkoXXHwZZ9drwItP9mLJ4r8wM4qXLEWn7sHhe+f8OZ0PB79JbGwsZgE6dX2YAgULHW0XEeO2TvfT/+lHSE1NoXipMnTs/iiTxn7DmFGfAFD3vAu58JKr0pefP3s6xRJLUKJU5CdjYREdl850tu+XqhzKbdie6ncMviuWP5gz7kzJ0a9VtsgbF/wEbkne63Mk/isUH6zY3Jyc5nMk/kuIjwFgly4X5PF+YtJnZP9nROdi/7kYPXf9MZaMfk1rBptOLVy70+dI/FetZF7+XL7N7zByhNPLF4AI+Zr/+99bQv5CWK9yoRx7bCfGMFAiIiIiIlEgSiqV0ykZERERERGJEFGWiygZERERERGJGFGWjSgZERERERGJEBZl2YiSERERERGRCKE+IyIiIiIi4osoy0WUjIiIiIiIRIwoy0aUjIiIiIiIRAj1GREREREREV+oz4iIiIiIiPgiynIRJSMiIiIiIhEjyrIRJSMiIiIiIhFCfUZERERERMQX6jMiIiIiIiK+iLJcRMmIiIiIiEjEiLJsJOB3ACIiIiIicmJSzYiIiIiISIRQB3YREREREfGFOrCLiIiIiIgvoiwXUTIiIiIiIhIxoiwbUTIiIiIiIhIh1GdERERERER8oT4jIiIiIiLiiyjLRXSfERERERGRiGHHMWVms2ZLzWy2mc00s2leWREzG2tmi7z/C3vlZmb9zWyxmc0ys7OO93CUjIiIiIiIRAg7jn8huNA5d4Zzrrb3vAcw3jlXFRjvPQe4DKjqTR2AN4/3eJSMiIiIiIhECLPQp//gamCo93gocE2G8vdc0G9AgpmVOp4dKBkREREREYkQYWqlBeCA783sDzPr4JWVcM6t8R6vBUp4j8sAKzKsu9IrC5k6sIuIiIiIRIrjqOnwkosOGYoGOOcGHLTYec65VWZWHBhrZgsyznTOOTNzoe/96JSMiIiIiIhEiOO5z4iXeBycfBy8zCrv//Vm9jlQF1hnZqWcc2u8ZljrvcVXAeUyrF7WKwuZmmmJiIiIiESIcPQZMbN8ZlZg32OgCTAHGAW08RZrA3zpPR4FtPZG1aoPbMnQnCskqhkREREREYkQYbrPSAngcwtmLrHAMOfcd2Y2FRhhZu2AZcAN3vKjgabAYmAncOvx7ljJiIiIiIhIpAhDNuKc+wc4/TDlG4FGhyl3QKes2LeaaYmIiIiIiC9UMyIiIiIiEiGOpwN7TqZkREREREQkQvzHmxjmOEpGREREREQiRJTlIkpGREREREQihWpGRERERETEJ9GVjSgZERERERGJEKoZERERERERX0RZLqJkREREREQkUqhmREREREREfKH7jIiIiIiIiD+iKxdRMiIiIiIiEimiLBdRMiIiIiIiEinUZ0RERERERHyhPiMiIiIiIuKP6MpFlIyIiIiIiESKKMtFlIyIiIiIiESKaOszEvA7ABEREREROTGpZkREREREJEKoA7uIiIiIiPhCzbRERERERESygGpGREREREQiRLTVjCgZERERERGJEOozIiIiIiIivlDNiIiIiIiI+CLKchElIyIiIiIiESPKshElIyIiIiIiEUJ9RkRERERExBfqMyIiIiIiIr6IslxEyYiIiIiISMSIsmxEyYiIiIiISISItj4j5pzzO4ajydHBiYiIiEjUiIhv+btSQ/9+nCc25x5bwO8AjsFywmRmd/gdQ06ZdC50LnQedC50LnQudC50LqL0XESEPLFYqJPfMR9NTk9GcooOfgeQg+hc7KdzEaTzsJ/OxX46F/vpXOync7GfzsV+OhcnMCUjIiIiIiLiCyUjIiIiIiLiCyUjmTPA7wByEJ2L/XQugnQe9tO52E/nYj+di/10LvbTudhP5+IEltNH0xIRERERkSilmhEREREREfGFkpHDMLPtZnaqmc30piQzW+I9Hud3fP+FmaV5xzHHzL4ys4Qs2m5bM3vNe9zHzFZ5+1lkZp+ZWY2s2E+kMLMKZtYqw/PaZtbfz5iyindsc/yOQ/Y7+P3mw/4bmtnXISzf1sxKZ3g+8ES7RpyIzGyImV3vdxzRwsy6mFnebN5ngpl1PM51Q379j/faEOo1SfylZOQInHOznXNnOOfOAEYB93vPL/Y7tv8o2TuOWkAS0ClM++nn7acq8DHwg5klhmlfOVEFIP3LoXNumnPuXv/CyV5mFut3DCeYCmR4v2Wn43yt2wLpyYhz7nbn3LwsCyqMMv7wcpzrlj72kiJHZ2YxQBcgW5MRIAE4rmTkeETStUGOn5KRE9tkoAyAmdU1s8lmNsPMfjWz6l55W69m4zuvluO5fSub2a1mttDMpgANjrQT59zHwPdkw5clM/vCzP4ws7lm1sEru9TMppvZn2Y23ivLb2bvmtlsM5tlZtd55S29sjlm9myG7W7P8Ph6MxviPR5iZv29c/ZPhl99ngH+59UO3ZfxVxqv5miwmU3w1rk3w7YfMbO/zOxnM/vIzLqH+ZQdrxgze8c7z9+bWbx3PC+b2TSgsx9BeTUEC7zXZaGZfWhmF5vZL977t+5xvtffNLNp3vE+lqG8qbe/P7z3wb7XOJ/3Gk/x9nN1hn18YWZjzWypmd1tZl29ZX4zsyLecpW9OP4ws5/M7GSvPFPvt2w6V6PM7Adg/EHbreMtX9nMzjazid5xjDGzUl7MtYEPvXj3vXdqe+tvN7O+3uf1NzMrkeGc/OZ9Pp/M+JnMChb8chdubcmQhEUD773+jfd6zTGzG82st5lN9Z4PMLNDbrh2uPeGV36vmc2z4HV5ePYf0bEd4ZiXmlkxb35tM5vgPe5jZu97n6NFZtbeK29oZpO87fxlZm+ZWcCbd8S/Q2b2opn9CTxM8L30o5n9mI2H/wxQ2fvsPm9m93uv9ayDro2tvbI/zez9DOuff/D1yzsXE8zsU++a9OG+98xB14bD/S0/7DVKIoxzTtNBE7D9oOdDgOv9jisrjw2I+X97dx8sV13fcfz9CUGSNiGApE6nhSYgGjBIQB4EFK8OZKo1ggNRMmgFmXYy5aHYMoVqZxoHa6RqqQ5FFIfGKcpooBAbKiGjBCQQ8/xAwADmpsSSAQWS8pAESb794/vb3JPl3r27N9m7l+Tzmrlzz549Z8/vd85vf89nDzAb+NPy+mBgeFk+G7izLF8MrAfGACOA/wGOAP4QeBoYC7wFWAjcWPaZAVxdd9yrgG8NQvwOK/9HAo8CbwM2AuPr3r8e+NfKfoeSGXstTsOBnwHn1acJ4AJgViVtzCYb9scBT5X1XcDcyj67Xpfz8zBwEHA48DxwIHAKsLKc59HAk/XncSj8kb3wrwOTyusfAZ8CFgA3DZGwHV+uyTLgVvLJuucCd7ea1uvSzQElnu8u21TT1u2Va/xl4FNl+RDgCeD3yzGeKtd3LLAFmF62uwG4qiz/FDimLJ8G/KyV9DZI5+rXlfPSBcwFziifc2RJ0w8DY8s2nwRuLcsLgJMrYdn1GghgSln+Z+AfyvJcYFpZnk5dPt1EXH8JfB94HLiD7FHeQOYFy4ELgWnAGjLvuL6y/yXlGi4GbqEnr5tFpWxg93zimvJZq8gK3AXAy8A68ns+so+wbgBmlm2WAicB84BfVdJKVzlnd1TiVftBmo+UdcuAb7aaLgbwnTsfuKXyekwtXZTX/1G5nrPKeWiUNp4BDqp9dzqZn7QY5w3A4eX1ycCCsjyjpIGRZH6/kSxruoBtwFFkvjK/nJtG5VAAn6hLK4cPctzHAY+W5cnkr2CJzEPmAmcB7yrfl9r5qOUTs+g7/9oC/HF57xHgfeW9BeV8jqX3sryvPKqr3Wnff3vvz1Mp9j8jJa0kR0QeJzNAyMz0e5KOITO8Ayv7/DQitgBIegz4EzJTXRARvynrfwi8o8Fxqi6hIgAACelJREFU39Az1iZXSvp4WT6CfKrrgxHRDRARL5T3ziYrH5T1L0o6i93j9H0yY727n2PeHRE7gcdqvbhNuCcitgPbJT1HNprOBOZExDZgm6T/avKzOqE7IlaW5WVkAQU5Ja/TuiNiDYCktWT6DUlryHC2mtY3Ap9QjrQNJxvix5GF5vpa2iIbI7WnCE8GPqaeka0RZAUd4P6IeAl4SdIWoHad1wDvljSKrNTPrnQoH1QJ40DSW1/25FzNr3yfAI4lKyaTI+IZSROBicD8Eo8DgE1NhOk1slIDmbbOKcunA+eV5R8AX2slosA7gUsjYqGkW+mZavJ8RJyknD61CHgP8CJwn6TzgF8AXyzrtwD3AysaHUjSh8kG3WkR8aqkwyLiBUmXkx0MS/sJ69MRMUnSDWQF7kwyDT0K3Fy2OZGs9D1DdgadqRyV/DZwVkR0S7q9uVOzR9YAXy89+HMj4ueSzpf0d2SD7zBgLT3pHPJa9JU2VpOjZnfTf97bKb3FudH2cyJiK7C1jGKcCmwGFkfEeoByrd4H/I6+y6EdwJ1titNATC5/te/DKOAY4ARgdkT8FnYrd6Hv/GtxRPwaoNRRxgEPVd5/L72X5Y3yKHuTcGNk/7O1FHK/R/a2XUb2nl1HVpI+Lmkc2RtRs72yvIOBpZsTyV6+tpHURTYyTi8VgAVk7+KEvfDx1d/AHlH3XvX8NNvo2hvntJPqwz+yLL/SgbDUq4ZtZ+X1TvI8t5TWJY0HrgZOKY3WWbwxDdQTcH5ErNttpXRaE+EbBmyOvF+tv/jtaSN/T85V/bXeRJ6XE8kKsoC1EXF6i2H6XUTUvm9787uxMSIWluXbgNr0yFoD+hR6rwRCax0vkPnQv0fEq/CGylgzflz+rwFGVRqv29XzoyO9Vd5epu8GcltExBOSTiJHZL5Ups9cRo50bZQ0gzd+XxqljT8jz/sU4AuSjo+I19sXg9b1EefX6Zn6Xh/f+mcoRD/r+7ItIna0Gt42EjAzIr6920rpigb79JV/DbRMbJRH2ZuE7xnZT5VC8krgb5U3oI4B/re8fXETH/EL4AOS3irpQGBqXxsq78eYTBaM7TQGeLE0RCaQPSkjyDmq40tYDivbzqdy876kQ8kpGB+QdLhy/vg04IGyybOSji1zemsjL428RE7FacVCYIqkEaV3/KMt7m/NaTWtH0xWvLeUnrwPl/XrgKNKAQg51aRmHnBFZd7zic0GLiL+D+iWNLXsK0kn9LPbQNJbM1o5V5vJiuTM0jGwDhgr6XQASQdKelfZdiDhXUROj4HKqGYL+qr47UkDelcFtOQNb9mDz6qqNgrrG4zD67aBDnZolBGlVyPiNuCr5LQygN+WfKy3X0/qNW2Uc3hERNxPTnMbQ/a2Dyl9xHkDOXoGPem05tySr7+VnD60pKw/VdL4Eu9PkiMBjcqheu363jdSPeY84LPlOiPpjyT9ATm1bGqJb7Xc3ROL6L0sbzU/tyHIjZH9WESsIIfEp5Fzs2dKWkEThVpEbCLnwj5CVqIfr9vkcyo/7UveT/ChWs9iG91L9mQ/Ts7RXgT8huwZ/E/lTX+1XtAvAYcqbxBcBXywxOlachrGKmBZRMwp219LTh15mOammqwGdihvtGvqhuKIWEL2iK4GfkL2im5pZl9rSatpfRU5DeGX5PSghWX9VnKqz72SlpGFdO16XUdOF1hdpj9d12IYLwIuLWlzLTnlp5GW01uTWj1Xz5KN6H8jR0guAK4v8VhJTj+DnHp0c8kjRvb2Wb24CvgbSauBt9P6d+PIWuWX/DGNh+re76sS2KjjZQM9FdCP0TNFZD5wSRmBrlac2l15bNRAbpfjgcVldOYfybz1FnJK2Tx6Kt67RMRr9J42DgBuU04TXAF8MyI2D0IcWtVbnL8IfKNMlasfvVhNliuLgOsi4pmyfglwI1l+dgN39VMO1fsOmf8M2g3sEfE8sFD58+7nkHniI+Wa3QGMjoi1wD8BD5Tr+y974bh9leUt5VE2NPkJ7GZDiKRREfFyqcQ8CPxlRCzvdLisd5XrJbIC/mRE3NDpcO2Lyndia7mn5ULyZvb+Gmm1fceRnRVLycbDY8Cny/+Ta3PbJU0DPk9OH7knIq4p6y8B/p4c/VkJvBYRl5eRsjnkNMV7gcsiotZLfC3w5+Q9MP8dEZ8vo8RfBraS00m39hLWDbUwSbq4LF9efY+83+LqiPhoWX8jsDQiZkmaQvbWv0JWdkdHxEXNnCfb+8o0tZcj4mt167uoXEOz/ZkbI2ZDiKQfkDdHjwC+FxEzOxwka6CMQnyGnJ6zAviL2n0CtndJej/ZiyyyUfDZiHiqyX3HkTcaT2xbAIcIN5CHFjdGzPrnxoiZme3T9rPGiBvIZvam4saImZnZIJN0FzC+bvU1ETGvE+ExM+sUN0bMzMzMzKwj/GtaZmZmZmbWEW6MmJmZmZlZR7gxYmY2iCTNkHR1Gz//EEl/1a7PH+zjmJnZvs2NETOzfYSk4cAh5MMY222wjmNmZvswN0bMzNpM0hckPSHpIeCdZd3Rku6VtEzSzyVNKOtnSbpZ0tKyT+3BduPKdsvL3xllfVdZ/2PyIX5fAY4uTzf/ann/AUlzJK2X9BVJF0laLGmNpKPL54yVdKekJeXvzLJ+hqRbJS0o+19ZorXbcQbzfJqZ2b5jeKcDYGa2L5P0HuBCYBKZ5y4HlgHfAaZHxJOSTgNuAj5UdhsHnAocDdwv6e3Ac8A5EbFN0jHA7eTTuAFOAiZGRHd5psbEiJhUjt8FnAAcC7wArAe+GxGnSvpr4ArgKuAbwA0R8ZCkI4F5ZR+ACcAHgdHAOknfAq6tHsfMzGwg3BgxM2uv9wN31R48V0YwRgBnALPzQdkAHFTZ50cRsRN4UtJ6sjHQDdwoaRKwA3hHZfvFEdHdIAxLImJTOf6vgPvK+jVkIwPgbOC4SngOljSqLN8TEduB7ZKeA97WdOzNzMwacGPEzGzwDQM2NxhVqH8AVACfA54lRzmGAdsq77/Sz/G2V5Z3Vl7vpKccGAa8NyKqn0tpnFT334HLDjMz20t8z4iZWXs9CJwnaaSk0cAU4FWgW9JUAKUTKvtMlTSs3M9xFLAOGANsKiMmnwYO6ON4L5HTqVp1HzllixKm/qZfDfQ4ZmZmu7gxYmbWRhGxHPghsAr4CbCkvHURcKmkVcBa4NzKbk8Di8v208toxU3AZ8r2E+hjNCQingcWSnq0xRvLrwROlrRa0mPA9H7iNdDjmJmZ7aKI+tkAZmbWKZJmAXMj4o5Oh8XMzKzdPDJiZmZmZmYd4ZERMzMzMzPrCI+MmJmZmZlZR7gxYmZmZmZmHeHGiJmZmZmZdYQbI2ZmZmZm1hFujJiZmZmZWUe4MWJmZmZmZh3x/14WICM5R+SWAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LhdDrCvRDhp3"
      },
      "source": [
        "Se logra ver que los departamentos de Sales, support y technical poseen mayor munero de empleados. Además, se ve que existen departamentos como management que tienen una menor tasa de salida (salen 91 de 539+91, vs digamos, en sales que salen 1014 de 1014+3126- que es casi un tercio-)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "id": "g7qCxD5BCBPD",
        "outputId": "b265ff18-cfac-4010-bf72-9406cb87d24b"
      },
      "source": [
        "heatmap_count(df, \"department\", \"salary_level\")"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f012de4abe0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QY05V-0zEBmS"
      },
      "source": [
        "Los niveles de salario Low y Medium tienen un comportamiento similar entre los departamentos.  Los departamentos Sales, technical y Management son departamentos que presentan un número relativamente mayor de personas de mayor salario."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 458
        },
        "id": "JU_7jozyvwC8",
        "outputId": "8ba0bd0c-e975-4316-a809-112d1c1e88de"
      },
      "source": [
        "#Se realiza un mapa de calor de Predictive Power Score (PPS).\r\n",
        "#El PPS es una métrica que mide el poder predictivo de una variable sobre otra.\r\n",
        "#Esta se utiliza para observar correlaciones y posibles variables de interés para los modelos\r\n",
        "matrix_df = pps.matrix(df)[['x', 'y', 'ppscore']].pivot(columns='x', index='y', values='ppscore')\r\n",
        "plt.figure(figsize=(15,5))\r\n",
        "sns.heatmap(matrix_df, vmin=0, vmax=1, cmap=\"Blues\", linewidths=0.5, annot=True)"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f012ddb4080>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "W0cMQNbb4ba5"
      },
      "source": [
        "El Predictive Power Score es una medida del poder predictivo de una variable. \r\n",
        "\r\n",
        "Esta gráfica nos muestra que la variable \"employee_left\" puede ser predecia por las variables \"number_projects\", y \"satisfaction_level\", pues la fila de \"employee_left\" posee valores en estas columnas.\r\n",
        "\r\n",
        "Esta gráfica parece apoyar algunas de las hipótesis que se han ido formando:\r\n",
        "1. El número de proyectos si tiene poder predictivo sobre la satisfacción del empleado.\r\n",
        "2. La variable que mayor poder predictivo tiene es la satisfacción del empleado.\r\n",
        "3. Es posible que los Features más relevantes sean satisfaction_level,number_project, average_monthy_hours, time_spend_company y last_evaluation. Esto son, qeullos features que directamente tienen poder predictivo sobre la variables respuesta; y los features que tienen poder predictivo sobre aquellos que predicen la variable respuesta."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "v66FEZnLEhst"
      },
      "source": [
        "# Modelado preliminar\r\n",
        "## Preparación de los datos\r\n",
        "Se aplica cambios a los datos originales, haciendo las variables departmetn y salary_level categóricas.\r\n",
        "\r\n",
        "Se probarán inicialmente los modelos:\r\n",
        "\r\n",
        "* Regresión Logística\r\n",
        "* Random Forest (Clasificador)\r\n",
        "* KNN (Clasificador)\r\n",
        "* Red Neuronal para clasificación binaría\r\n",
        "\r\n",
        "Para las pruebas iniciales, se hará un split 80-20 para Entrenamiento y Prueba, se tomarán más muestras de la clase minoritaria (upsampling) para manejar la clase no balanceada y se tomará la versión \"por defecto\" de los modelos en sklearn con todos los features del set de datos. \r\n",
        "\r\n",
        "Para comparar los modelos, se usarán la curva ROC-AUC y los datos de la matriz de confusión. Esto es puesto que, al no estar balanceados los datos de las variables respuesta, el acuraccy no es una medida idónea para comparación de modelos.\r\n",
        "\r\n",
        "El objetivo de este paso es la generación de un insight inicial de que tan buenos predictores para el set de datos son cada uno de los modelos.\r\n",
        "\r\n",
        "\r\n",
        "Posteriormente, se hará un tuneo de los hiper parámetros de cada modelo. En este paso, se compará la mejor versión encontrada de cada modelo entre si, y se elegirá aquel que posea una mejor ROC-AUC y Matriz de confusión.\r\n",
        "\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KfkjA_-SE7L-",
        "outputId": "19c8a47b-221a-4d94-84b4-c3095199cd0b"
      },
      "source": [
        "df=df.astype({\"department\":\"category\", \"salary_level\":\"category\"})\r\n",
        "df.dtypes"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "satisfaction_level        float64\n",
              "last_evaluation           float64\n",
              "number_project              int64\n",
              "average_montly_hours        int64\n",
              "time_spend_company          int64\n",
              "Work_accident               int64\n",
              "employee_left               int64\n",
              "promotion_last_5years       int64\n",
              "department               category\n",
              "salary_level             category\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "veSJMRamHWFM"
      },
      "source": [
        "def get_X_y(df, y_name):\r\n",
        "  \"\"\"\r\n",
        "  Splits a DataFrame in X (Features) and y (response variable)\r\n",
        "  Parameters\r\n",
        "  ----------\r\n",
        "  df : DataFrame\r\n",
        "      A DataFrame that has column features.\r\n",
        "  y_name : str\r\n",
        "      The name of the target varriable in the DataFrame.\r\n",
        "\r\n",
        "  \r\n",
        "  Returns\r\n",
        "  -------\r\n",
        "  X : DataFrame\r\n",
        "      DataFrame with the features used to predict y.\r\n",
        "  y : Array(int)\r\n",
        "      Array with the response variable's values.\r\n",
        "  \"\"\"\r\n",
        "  y=[y_name]\r\n",
        "  X=[col for col in df.columns if col not in y]\r\n",
        "  y=df[y].copy().values.flatten()\r\n",
        "  X=pd.get_dummies(df[X].copy())\r\n",
        "  return X, y\r\n",
        "\r\n",
        "\r\n",
        "def data_preprocessing_up_or_down_sample(X, y, sample=\"up\", test_size=0.2):\r\n",
        "  \"\"\"\r\n",
        "  Applies downsampling or upsampling, and returns the Train-Test split \r\n",
        "  of data.\r\n",
        "  Parameters\r\n",
        "  ----------\r\n",
        "  X : DataFrame\r\n",
        "      DataFrame with the features used to predict y.\r\n",
        "  y : Array(int)\r\n",
        "      Array with the response variable's values.\r\n",
        "  sample : str(optional)\r\n",
        "      Chooses the method to apply. Downsampling (\"down\"), \r\n",
        "      upsampling (\"up\") or no method (anything else). Default is \"up\". \r\n",
        "  test_size : float(optional)\r\n",
        "      Sets the test_size parameter of sklearn.model_selection.train_test_split.\r\n",
        "      Default is 0.2\r\n",
        "  \r\n",
        "  Returns\r\n",
        "  -------\r\n",
        "  splitting : list, length=2 * len(arrays)\r\n",
        "      List containing train-test split of inputs, with the method defined by\r\n",
        "      the parameter sample applied.\r\n",
        "  \"\"\"\r\n",
        "\r\n",
        "  # Use the sample parameter to define local variables to select the correct \r\n",
        "  # method\r\n",
        "  a,b=0,0\r\n",
        "  if sample==\"up\": \r\n",
        "    a,b=1,0\r\n",
        "  if sample==\"down\":\r\n",
        "    a,b=0,1 \r\n",
        "\r\n",
        "  \r\n",
        "  # Apply the normal train_test_split to the data\r\n",
        "  X_train, X_test, y_train, y_test = train_test_split( X, y, \\\r\n",
        "                                        test_size=test_size)\r\n",
        "  # Using the a and b local variables, apply downsampling or upsampling only\r\n",
        "  # if the sample parameter is \"up\" or \"down\".\r\n",
        "\r\n",
        "  if a+b>=1:\r\n",
        "    X_train_temp, y_train_temp = resample(X_train[y_train == a],\r\n",
        "                                    y_train[y_train == a],\r\n",
        "                                    n_samples=X_train[y_train == b].shape[0])\r\n",
        "    X_train = np.concatenate((X_train[y_train == b], X_train_temp))\r\n",
        "    y_train = np.concatenate((y_train[y_train == b], y_train_temp))\r\n",
        "  return (X_train, X_test, y_train, y_test)\r\n",
        "\r\n",
        "\r\n",
        "def rocauc_plot(model, model_name, y_test, X_test):\r\n",
        "  \"\"\"\r\n",
        "  Plots the ROC-AUC curve for a model. \r\n",
        "  Multiple consecutive calls will allow to display multiple curves on the\r\n",
        "  same plot\r\n",
        "\r\n",
        "  Parameters\r\n",
        "  ----------\r\n",
        "  model : estimator object.\r\n",
        "      Either from sklearn or keras interfaces\r\n",
        "  model_name : str\r\n",
        "      Name of the plot\r\n",
        "  X_test : {array-like, sparse matrix} of shape (n_samples, n_features)\r\n",
        "      Input values\r\n",
        "  y_test : {array-like, sparse matrix} of shape (n_samples, n_features)\r\n",
        "      Target values\r\n",
        "  test_size : float(optional)\r\n",
        "      Sets the test_size parameter of sklearn.model_selection.train_test_split.\r\n",
        "      Default is 0.2\r\n",
        "  \r\n",
        "  Returns\r\n",
        "  -------\r\n",
        "  plot : matplotlib.pyplot\r\n",
        "      Plot of the ROC-AUC score\r\n",
        "  \"\"\"\r\n",
        "  try:\r\n",
        "    auc = roc_auc_score(y_test, model.predict_proba(X_test)[:, 1])\r\n",
        "    fpr, tpr, thresholds = roc_curve(y_test, model.predict_proba(X_test)[:, 1])\r\n",
        "  except:\r\n",
        "    auc = roc_auc_score(y_test, model.predict(X_test))\r\n",
        "    fpr, tpr, thresholds = roc_curve(y_test, model.predict(X_test))\r\n",
        "  plt.plot(fpr, tpr, label=model_name+\" AUC = {:.5f}\".format(auc))\r\n",
        "  plt.title(\"Curva(s) ROC\", fontdict={\"fontsize\": 21})\r\n",
        "  plt.xlabel(\"False positive rate\", fontdict={\"fontsize\": 13})\r\n",
        "  plt.ylabel(\"True positive rate\", fontdict={\"fontsize\": 13})\r\n",
        "  plt.legend(loc=\"lower right\")\r\n",
        "  plt.plot([0, 1], [0, 1], \"r--\")\r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        "def plot_roc_conf_matrix(y_test,X_test, model, model_name):\r\n",
        "  \"\"\"\r\n",
        "  Print the classification report and plots the ROC-AUC curve and confusion \r\n",
        "  matrix for a given model.\r\n",
        "\r\n",
        "  Arguments\r\n",
        "  ---------\r\n",
        "  y_test : {array-like, sparse matrix} of shape (n_samples, n_features)\r\n",
        "      Target values\r\n",
        "  X_test : {array-like, sparse matrix} of shape (n_samples, n_features)\r\n",
        "      Input values\r\n",
        "  model : estimator object.\r\n",
        "      Either from sklearn or keras interfaces\r\n",
        "  model_name : str\r\n",
        "      Name of the model. Used as part of the plot's titles.\r\n",
        "  \r\n",
        "  \"\"\"\r\n",
        "  try:\r\n",
        "    y_pred=model.predict_classes(X_test)\r\n",
        "  except:\r\n",
        "    y_pred=model.predict(X_test)\r\n",
        "  cm = metrics.confusion_matrix(y_test, y_pred)\r\n",
        "  plt.figure(figsize=(15,5))\r\n",
        "  plt.subplot(1,2,1)\r\n",
        "  sns.heatmap(cm, annot=True, fmt='g', cmap='Blues')\r\n",
        "  plt.title(model_name+ \" - Matriz de confusión\", y=1.1, \\\r\n",
        "            fontdict={\"fontsize\": 21})\r\n",
        "  plt.xlabel(\"Predicted\", fontdict={\"fontsize\": 14})\r\n",
        "  plt.ylabel(\"Actual\", fontdict={\"fontsize\": 14})\r\n",
        " \r\n",
        "  print(classification_report(y_test, y_pred))\r\n",
        "  plt.subplot(1,2,2)\r\n",
        "\r\n",
        "  rocauc_plot(model, model_name, y_test, X_test)\r\n",
        "\r\n",
        "\r\n",
        "def apply_model_to_df(data, model, model_name):\r\n",
        "  \"\"\"\r\n",
        "  Applies a sklearn estimator to a DataFrame. \r\n",
        "  It returns a fitted model and shows relevant information of the model's\r\n",
        "  performance (applies the plot_roc_conf_matrix function to it)\r\n",
        "\r\n",
        "  Parameters\r\n",
        "  ----------\r\n",
        "  data : List of Arrays\r\n",
        "      List of arrays, equivalent to the output of the function\r\n",
        "      sklearn.model_selection.train_test_split  \r\n",
        "  model : estimator object.\r\n",
        "      Either from sklearn or keras interfaces\r\n",
        "  model_name : str\r\n",
        "      Name of the model. Used as part of the plot's titles.\r\n",
        "  \r\n",
        "  Returns\r\n",
        "  -------\r\n",
        "  model : estimator instance\r\n",
        "      Fitted classifier or a fitted Pipeline in which the last estimator \r\n",
        "      is a classifier.\r\n",
        "  \"\"\"\r\n",
        "  X_train, X_test, y_train, y_test=data\r\n",
        "  model.fit(X_train, y_train)\r\n",
        "  y_pred = model.predict(X_test)\r\n",
        "  plot_roc_conf_matrix(y_test,X_test, model, model_name)\r\n",
        "  return model"
      ],
      "execution_count": 314,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TDrBYzm5GNqc",
        "outputId": "414f238f-8f8b-4735-d95d-354a996d2c28"
      },
      "source": [
        "nn.predict_proba(X_test)"
      ],
      "execution_count": 320,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.8689878 ],\n",
              "       [0.1676667 ],\n",
              "       [0.90926147],\n",
              "       ...,\n",
              "       [0.00184131],\n",
              "       [0.9460629 ],\n",
              "       [0.58358693]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 320
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4FlxsVdOTioG"
      },
      "source": [
        "X,y=get_X_y(df, \"employee_left\")\r\n",
        "data_up = data_preprocessing_up_or_down_sample(X, y ,\"up\", test_size=0.2)\r\n",
        "data_down = data_preprocessing_up_or_down_sample(X, y ,\"down\", test_size=0.2)\r\n",
        "data_as_is= data_preprocessing_up_or_down_sample(X, y ,\"None\", test_size=0.2)"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 514
        },
        "id": "KnTi0CXiKSgJ",
        "outputId": "3dc64ffd-81d7-48b4-f4ed-57cd6679cbaf"
      },
      "source": [
        "rf_us=apply_model_to_df(data_up,RandomForestClassifier(), \"Random Forest - Upsampling\")\r\n"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      1.00      0.99      2266\n",
            "           1       0.99      0.97      0.98       734\n",
            "\n",
            "    accuracy                           0.99      3000\n",
            "   macro avg       0.99      0.99      0.99      3000\n",
            "weighted avg       0.99      0.99      0.99      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 514
        },
        "id": "3b5ID6VihVNW",
        "outputId": "9d13c51d-e51f-4584-e1b9-79b470ec9794"
      },
      "source": [
        "rf_ds=apply_model_to_df(data_down,RandomForestClassifier(),\"Random Forest - Downsampling\")\r\n"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      0.99      0.99      2308\n",
            "           1       0.97      0.97      0.97       692\n",
            "\n",
            "    accuracy                           0.99      3000\n",
            "   macro avg       0.98      0.98      0.98      3000\n",
            "weighted avg       0.99      0.99      0.99      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 509
        },
        "id": "DRqkohtKhrqh",
        "outputId": "d555e88b-ce34-4e86-b566-5dd86c0ed289"
      },
      "source": [
        "rf_asis=apply_model_to_df(data_as_is,RandomForestClassifier(),\"Random Forest - Data As Is\")\r\n"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      1.00      1.00      2283\n",
            "           1       1.00      0.98      0.99       717\n",
            "\n",
            "    accuracy                           0.99      3000\n",
            "   macro avg       0.99      0.99      0.99      3000\n",
            "weighted avg       0.99      0.99      0.99      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OHbRY5GZTsol",
        "outputId": "19a499e9-0034-4143-f10c-b14859ddeb94"
      },
      "source": [
        "rf_us.get_params(deep=False)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'bootstrap': True,\n",
              " 'ccp_alpha': 0.0,\n",
              " 'class_weight': None,\n",
              " 'criterion': 'gini',\n",
              " 'max_depth': None,\n",
              " 'max_features': 'auto',\n",
              " 'max_leaf_nodes': None,\n",
              " 'max_samples': None,\n",
              " 'min_impurity_decrease': 0.0,\n",
              " 'min_impurity_split': None,\n",
              " 'min_samples_leaf': 1,\n",
              " 'min_samples_split': 2,\n",
              " 'min_weight_fraction_leaf': 0.0,\n",
              " 'n_estimators': 100,\n",
              " 'n_jobs': None,\n",
              " 'oob_score': False,\n",
              " 'random_state': None,\n",
              " 'verbose': 0,\n",
              " 'warm_start': False}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 514
        },
        "id": "BKbXzFstY6H0",
        "outputId": "d338c814-5424-4e4b-a0bc-0cb86a633a78"
      },
      "source": [
        "lr_no=apply_model_to_df(data_up,LogisticRegression(class_weight=\"balanced\"), \"Regresión Logística\")\r\n"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.92      0.75      0.83      2266\n",
            "           1       0.51      0.81      0.63       734\n",
            "\n",
            "    accuracy                           0.76      3000\n",
            "   macro avg       0.72      0.78      0.73      3000\n",
            "weighted avg       0.82      0.76      0.78      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 513
        },
        "id": "_lx0DFJPZCwx",
        "outputId": "b2df3b82-d057-415d-f936-cb85ddacf14c"
      },
      "source": [
        "knn_no=apply_model_to_df(data_up,KNeighborsClassifier(), \"KNN Classifier\")\r\n"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      0.89      0.94      2266\n",
            "           1       0.75      0.97      0.84       734\n",
            "\n",
            "    accuracy                           0.91      3000\n",
            "   macro avg       0.87      0.93      0.89      3000\n",
            "weighted avg       0.93      0.91      0.92      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 853
        },
        "id": "8a-zVBtyMnef",
        "outputId": "47660e6d-d551-45eb-f419-e27189426f3f"
      },
      "source": [
        "import keras\r\n",
        "from keras.models import Sequential\r\n",
        "from keras.layers import Dense, Dropout\r\n",
        "X_train, X_test, y_train, y_test=data_up\r\n",
        "nn=Sequential()\r\n",
        "nn.add(Dense(20, input_dim=20, activation='relu'))\r\n",
        "nn.add(Dropout(0.15))\r\n",
        "nn.add(Dense(1, activation='sigmoid'))\r\n",
        "nn.compile(optimizer='adam',loss='binary_crossentropy',metrics=['binary_accuracy'])\r\n",
        "nn.fit(X_train,y_train,epochs=10,batch_size=5)\r\n",
        "plot_roc_conf_matrix(y_test,X_test, nn, \"Red Neuronal Sencilla\")\r\n"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "3665/3665 [==============================] - 5s 1ms/step - loss: 3.3594 - binary_accuracy: 0.5481\n",
            "Epoch 2/10\n",
            "3665/3665 [==============================] - 5s 1ms/step - loss: 0.5714 - binary_accuracy: 0.7399\n",
            "Epoch 3/10\n",
            "3665/3665 [==============================] - 5s 1ms/step - loss: 0.4603 - binary_accuracy: 0.8114\n",
            "Epoch 4/10\n",
            "3665/3665 [==============================] - 5s 1ms/step - loss: 0.4320 - binary_accuracy: 0.8266\n",
            "Epoch 5/10\n",
            "3665/3665 [==============================] - 4s 1ms/step - loss: 0.4145 - binary_accuracy: 0.8333\n",
            "Epoch 6/10\n",
            "3665/3665 [==============================] - 4s 1ms/step - loss: 0.4166 - binary_accuracy: 0.8322\n",
            "Epoch 7/10\n",
            "3665/3665 [==============================] - 4s 1ms/step - loss: 0.4115 - binary_accuracy: 0.8354\n",
            "Epoch 8/10\n",
            "3665/3665 [==============================] - 4s 1ms/step - loss: 0.4036 - binary_accuracy: 0.8384\n",
            "Epoch 9/10\n",
            "3665/3665 [==============================] - 4s 1ms/step - loss: 0.4045 - binary_accuracy: 0.8398\n",
            "Epoch 10/10\n",
            "3665/3665 [==============================] - 4s 1ms/step - loss: 0.3969 - binary_accuracy: 0.8402\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      0.82      0.90      2266\n",
            "           1       0.64      0.96      0.77       734\n",
            "\n",
            "    accuracy                           0.86      3000\n",
            "   macro avg       0.81      0.89      0.83      3000\n",
            "weighted avg       0.90      0.86      0.87      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA54AAAFXCAYAAADK95GyAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdedyVc/7H8denzVYh1SBSiLFOiBiDLEOWkczY+cm+jrGvM8RYYjLGYIyEGBNmhJiRXcaSVMq+RZYWipBKpfr8/vh+j67Ofc69dd3nus+538/7cT3Oua/lnM/Zruv6XN/N3B0RERERERGRhtIs6wBERERERESksinxFBERERERkQalxFNEREREREQalBJPERERERERaVAtsg5ARESkoZnZUcA6wB3u/knW8YiIiDQ1KvEUEZGKZmZ7A7cDnZV0ioiIZKNWiaeZeYFplpmNM7Pfm9mKDR2omXWJzzuyDtuMTMS7Zw2POzG1YJuQ+N59XMdt2prZJWb2qpl9Z2bzzWyymY0ys4FmtmMDhZsqMxsSX3+vvPm5712XvPkfm1mjHL/IzPolfisfm5lVs+7tiXUHlzLOYop9FpX+3I2Bme1hZi+Z2ez4PnxTouftFZ9vSA3rrQ7cAbwAnFiK2ESaOjNbwcxOM7MnzWyamS0ws2/jeeNAM9so6xgbgpmNMLPP63tebGZbx/3aufXYNnkcz00LzewLM/uvme1Ti8dY0czOMrMXzezLxPnZv4udRxd4jK3M7BYzezfmCvPN7DMzG25mR5hZq7q+Nqkcda1qOwyYDRiwNrAdsCXwGzP7hbvPTjm+NF0KjMg6iKbOzDoDzwFdgDnAaOALoB3QA9gW2BT4X0YhSqiO+Avg+fwFZrYC8Os0n8zM+gOXAEe5+5A0H1salpmtAzwAtASeBGYAczMNKiFeQBlC2Nfs7+4Lso1IpPKZ2TaE/UInYB7hOD8VaANsAZwFnGFmR7v7nZkFmjIz+yXQGzjT3eu1H3T3MWb2KHCBmQ1295n1eJgvgMfi/eWBTYC9gL3M7GJ3/2OhjczsZ8AjhPP774AXgW+AdQnH/d+Y2b+BI9x9foHtWwA3sOQC34fAM4TvwNoxhn0Jx/v16/G6pALUNfE8290/zv1jZt0IV5F/BvwOuCK90FL1PbC1me3r7g9nHUwTdyMh6XwcODS5UzWzZkAvYPNMIqu7C4ABwKdZB5Ki8YQTg8MpkHgCfYC2wKuEi06NRSV+FuVgN2BF4I/ufnGJn/sVYCPg22rW6QK8TDgRnFGKoESaMjPbDHiWsF/4M9Df3b9LLDdgV2Ag0DWTIBvOAOBr4O/L+DhXEs6tz4tTXb3r7v2SM8zsd8BfgIvN7K78JgfxIuKzwKrAzcA57j4nsXxT4B7gAKA5hS9A30E4d/gQONrdlypAMLN2wDmECw/SRC1TG093/4CwY4FwlaexuineXlZdFUJpWLG0LFdV49T8K3nuvtjdn3H3v5Q+urpz92nu/m59r2w2Uq8CbwMHFKkOcwSwELi3pFHVoEI/i3LQKd5+VOondve58TOfVs06k9y9v7u/XcrYRJqieH51NyHpvNzdz0omnQAePAX0BP6bQZgNwsxyNQD/7e7fL8tjufuLhH3q0WlVS3X364F3CQVOuxVY5RZC0jnE3U9OJp1x+zcJFwymA/ub2UHJ5Wa2PyHpnAHskJ90xseY6e4XALuk8JKkTKXRudBb8bZj/gILDjGzZ8zsazObZ2bvmFn/YvXfzWxtM/uHmc0ws7mxPcDhyxjjcGAsoWS2TtUEzWyj2Ibrs1hP/Qszu9fMNimwbv9Yp75fkceq0vbPEm1XLbR9/LOZTTKzH8zsL4n1NjKzu81samwrMTX+/9MCz/Nj2yczW9XMbjKzKTH+92L9/SoJuJntaGY3mNlrZjYzfl4TzeyvFtpJLatVWVLKPr2uG5tZz9jOINdeZLKZDY7Vd/PX/fGzMLNNzeyh+JrmmtnL1bVViO/1bRbaOs43s+kW2jucHauS5NZLpW2fme0TX8fbFtrAzI33rzCztsvy2PX0D8JntXdenB2B3Qml1QVLj+L37VQzeyy+f/Pib/8ZM6vy27PQPviS+O8dtnTblF5xnVy7lf5mtkH8/X1hZovNbL+4TpXPIjGvuqlfbd8UMzvazCaY2fcW2vAMqel3YWbtzOyq+Hl+Hz/fZ6wWbW2KPF5vM3s4vv5cu5n/FHlvG2yfkftMCE0YYOnPrl9cp2Bb58RjFGwfbqHN6GPx9z0//t5fNrMrLXHcsGraeJpZi/g9HGuh7elsMxtjZqckf8OJ9X+M1cz6WGizOsfMvrGw79iw0GsQkaXsQaixNAW4rLoV3X2+u4/J/V/T8dQK9JGQt99aPR5HJ1to13h63Fe6mf2qyGNafNzFFkr8MLNmFs5bh8Z9X27/8aqZnWvFE8Hj4+0/izzXtmY2zJYcF6fHx7zOzH5SYJOhQHugb5Hnq4+C5+sWzmf3IFSJPbvYxu4+HchV080vtcy1Sb2kuouB8XFeqG3AUnnSSDzbxNulEgkL1Sb/SfjxbA1MAB4FViKcaD5roQQsuU1XQvWpw4FZhIRxDnAXy140n6sG1j/GViMLJ7XjgSOBr4CHgUnAgcArlm4nOCsQ2j72I7xXDxOqbGBmuxIS58OAzwltbb+I/48ttqMGVgFGEZLtlwjtJtchVHG5pMD6fwaOA+YTqlw8Tmi79VtgTE0n2bXwJWHHBnXs5MPMTia8hv0J1SkfInwmxxDeg2IdFWxFaF+yAfAE8AbhSut/4vua/zwHED7zowlVtB8ExhHaJ/wJaF2XuGtpCHAQoS3F44T3vh1wIfC8ma3UAM9ZnX8CTvh+JR1CuHBwdzXbbk9o4/FTYCLh/Xsd2AG438x+n7f+/cBr8f6LwJ2J6fO8dTcExgDbEN6jJ4EfqonlhbzHS065K9KLqtn+R2Y2ALgN2JjwO/ofoZbHaMJnVWibDQi/5fMJJQCPE37HPYFHzKzoAb7I411LaKe+N+G9fYBQpWl74A956zb0PmMi4X0s9NnVu6M2MzuJ0DZpN8IV/wfic3QkVKcu+F7nPUZzwrHjBsLv/kngKcL350bggWqOASfF51xEKI35klC9/HkLF15EpLjcxcp/u3t1++a0dSAcG/Ym7L9GENqa545V+ceynO0J+7cXElVPVySct+5BOK/9D2H/1hW4mnDuUGj/sTfh/GZU/gILFxpfJJy/TCecv4whnPedDnQr8HjPJh43LQXP1wltLwEed/evaniMoYTzgx5m1h4g3m4T59+TUqxSqdy9xonwZXKgS4Fld8ZlF+XNPyfOfw5YIzG/FTA4LhuQt82IOP82oEVi/j6E6n0OjKxNzHG7kXGbX8T/X4r/H5pYp0ucNzFv2y6EjpRmA7vnLesNLCAkQK0S8/vHx+pXQzxdCjy/x/hWydtmJcKJowOn5C07I86fCqyYmN8r8ZgPAMsnlvWM7+VsoHXe4+1d4PmbJ17X4CLfjY/r8Jn8PRHbK4ST2b2ADtVss22MeSqwdd6yY+JjvZw3v3/iec7NW3Z2oe8SYef/PSGZOSJvmRFK+5ZLzBsSH6dXTZ9znP8xoaZR/uvrm/z84rzlWfI7+X1t39/6ToQLHj9+xoSD3jxg5cQ6YwkXhFbIXz+xznrAtgUefz3gk/g5rlPksyr2u8k9lxOSieYF1in4WRR5vIsS378VarH+tsBiwoWBLRLzWwNPJ2LrlVjWnJBwOyHxbJ5Ytj4hqVoIbFrLz+dwlvzWt8xbtgLwy8T/pdxnFP3siv0OEsur7Dvib2QxsE2B9XsCKxWIeUjeemfF+W8CP0nMX4NQ1cyB04vE+j2wU2J+S8LFEye0VWvQ36EmTeU8EfoFcODwemxb7T6cAsfP6vZbcfnyhDbgc4E2BR7z5rjt8Yl5rYD9gJZ567YlXIyq8voIF1odGFUk9tz+5YACyzZJ7qfynm8R8Ekd3sN+FDlPJly0+5rCx+C7qcO5BuH45cCu8f9dKXAerUlToal2K+UlnoST8M6Ek47FhISpdWL9FoSqeHNIJJ2J5SsA04CZQLM4b934HN+SONlNbHNvsR9UNXHnfuy5xDP343iXeCJI8cTzL3H+aUUe+/q4vG9iXn+WLfHsUWCbo+Kyl4o85pi4/MjEvF5x3iwKJHSEK3hO4gSrFu/lZODLIt+Nj+vwOCsQxtNbnHjdHv8fDRxUYJuH4jr7FnnM4XF5MinIfRZVDgSEk8mZhIsHLRPz/xa3uaaWr2UIKSSeNbxXPwDjartNfSeqJp5Hx/+Pjf9vROIkP3/9Wj7HcXGb3+bNr+l3k3uu6eQl6DV9FgXW2z9+1z6jwL6pyDa5i2uXFli2ceK73Csxf78474Eij9k3Lr++ljG8HdffqxbrlmyfUd1nV+x3kFheZd9BOGZ8Xcv3JBfzkLz5n8T5uxfYZu+4bFKRWAcU2KZHXPZsbb/rmjQ1xQl4J/5W9qjHttXuw6k+8ZwHdCqy3e35+7s4vyWhRsN8YNVaxtgtPtawvPkHxvl3FNnurbi8yrltDc83OW5X2/j6kXeeTEi+exAKgRw4q8B2uUKfE2r5PC/H9Q+K/x9ENYm3Jk3Jqa5VbSfFOvaLCQf3SwjVZ3fypYdS2ZJQN/0lL1DX20PD63GEdmS5Kga/iLePuXuhXgqXufje3Z8m/Pg2JJQgVGf3eDusyPJcj5/bLGtc0TR3H1tg/g7xtmC7AUI15OR6SeO8cE+O78XbNfMXmNkaZnZ8bHdwW2w7MYSwk17NQq9k9ebu37v70YQqcOcTTmi/IFzM2Aa418yuT8TTjHDBYC7Fh8Op7rN4LH+GhypAkwivqX1iUa7B/ZBavpzUmNl6ZvZbC+1pb4/v+c2E5LhQNZyGdj/hYJ77nRwRb/9R04axjcyuZnaxmd1sZnfE1/ObuEp9X89TvgydB5nZFoT4vwf6FNo3FZH7bVXpUMlDpzWv5c8nxf2Hma1JSPynu/ujNa1PifcZKRsHrBLbam1c140ttPfuDMxw9yfyl7v7fwlV9LuY2VoFHqLQPqZUr11E6udVd59SZFmx6ra9gdWAR9396/yNzGwzC/063Jg4huWaiuQfw3LV8IsNfTIu3t5lZj3Mat3JZe7xOtRy/ZydYttWJxzvxhDOsQ9392vr+FgiqarvOJ6tCFULtiBcQb6QJZ1MQCjFA9gtvzF4Ae0JB/bcQf2TIut9XMdYi/k94aTvYjMrdmIGS17D5Br2Ee2rW1gHxYaByL0vHxdZPinediqw7LMi2+R6mVsuOdNCd9vXED7fYtpQfOdaa+4+kdBe4ur43FsSSk9+BZxmZv/y0LNbe5a0q1xQj8+iLu/B2vH2/ZriT5OZXUOoHrhMba5jm+T98ud7XrfqteHus8zsYULvtusQDtpTWNLupFgMnQjtk6sbaqVNNcuqU++hUsxsDUJcKxCqO71ah81rs2/qnjevS7y928yqaxNbm/1H7nv5QS3WhRLuMxrAyYQaDscAx5jZdELbqIeAe73mcThr+qwgvC+rEV7/5LxlVV6/u38X9zsN/dpFyt2X8bauidKyqu7YMJLwO9/FzFZ391zfAblEdKn9s5m1JJSSVlc4kX8MWznefpe/YnQBoUrtvnH6xsxGEaru3plXcJM0K96uUk0shSTH8VwZ+DkhOf6bmb3u7m/krZ9r11nbzy2XaH+Zd1vqz13K0LKO43kI4ar6xWb2qC/poSx3Aj2RcNJQnZoaMqfK3V8wsycIJRJHETqeKCT3Gmoa3Hh0HZ6+usRiXjXL6mtxbVc0s20J1YtnETr+eRb43N3nxeUvAdsRSiZT5+6vxsRpNKFayN6E707uPZtN8dKjnLcKzKv1e5AFC12Sn0NI6s4gdEwwPXeCbWZTCW3TaqM7oSOsfP3qGd7dhCpENxNKkQa6e03v52BC0vkQ4aLCe8Asd19kZrkecev7HarXb8TMlo/xrAX8wd1r+h6lIfe9fYxwElDMl9Usy0JJfi/FOvdx9zdjD4u7E4Ze2olQLbkvcJ6Zbe/u3zRgaI16fyHSyE0glKxtRfWd0NVHvc6f3H2xmd1DOM4eAlxnZm0ICeC3hFpXSWcSks63CGNojgVmuvsPsUfb+VQ9huX2SQV7oXf3KWa2DbAzoU+LnQidF+0JXGBmv0ieWyfkEtq67vOWGsczHgPvIVyY/oeZbZl3LH+NkIj3qOmBY623LontcrcOrGtmqzTwPlrKXF0Tz6W4+z2xd8TjCSeZubF5cleRqwxiW41ctbd1iiwvNr8+Liac2PyeUPW2kMmEDlHO8pp7+crJXY0v1vPp2kXmV2dqvO1SZHlufrFqJrWV67L7Ine/o8Dy9Zfx8WsUDxDPEXZ+uZKgXE+4i4Gj3L2mEvRl8RmhCk03QluVUtg/3p7o7ksdAGNvtrXuSdjd+xNKjdPyGOH9zw09U+2JRIx3d0Ki9Rt3z+8xtsG/Q0XcQajSOtTdL6/H9tMIv7N1KPy9KLRvyu0DB6eQ6OZK4Wr7/pVqn1GT6vaHRfeF7j4feCROmNm6hOrvOxBOHi+q5jlzr72640WXeNvQr1+kqfkvcCqhpsw57r6wDtsW3V9YGAKpthdgC7mbsO84DLiOcL6zAuGYMD9v3dwx+WAPY1cmFdsH53qJLdoUKR4Pn4oTcZSAGwk9iF9B4Z53V423BYcvqy13n2dhmKv3CcMKHs6SJhcQmsxdA+xhZu08b4z1PAcTEu8x7v5lfPwvzewVQgdwhxAuVosUlMZwKv0JicHOZrZ9nDeGcCVppzq0CcyN69PbCo9dePAyRZng7qMJV7k6Ezo8KSRXElqXMZRyyXOVMd8sjJ1XZbzJWsi1BSvWHfgReevVV24HV6WqWRyaoVRVKHI79ikA8cA1knAlscrwJyl7Kt4WKjVsKEXfd5bs4DMR28IOIdRKeMndC7VlTFqZsE+ZViDphOK/4dwJxzJdCCvEzC6Oz/syofpmfeR+WwcWePyfUrWaLdRv/1GQu08lJLw/MbPetdikVPuMmhTdH7KkDWyN3P0jwnAuAJvWsO6nhGp3HWIJ+1IsjN+7GqFTo/xqtiKybB4n9ObdiSVD2BVkZq3MLFnCVt3+YmeW4fjg7q8ThlLbysKYvAWr2UbVHZMPKfIUuWNjlTGSq4npc5aMdVplvxbPg9cEPi3UBrWuYt8pV8V/L0zWOnH3twjDzS1PGDauIDPrwJLPNb+t6DXxtn9NQ+8lcgVpgpY58YwddPw9/vv7OG8+4UvYBnjQzNbL387MOpnZEYnH+ZDwxW8LXBvHYsutuxdwwLLGmudiQtWAk4ssv5bQKHugFR6cfTkz+01eBxW5XsOOSL5mC4MD30b93u9/EUqQtrMwlmUyhtMIJTnTgH/X47GTch1oHGeJAZItDP6eytUrM1vFzF6J71urvGXNzOxYQvWXxYQhDHKuiPPuMLOdCzxuazM72vLGha2HvxAuopxpZofmPYeZ2S/NLO12Xrn3/aRkhwNm1p0lB4nMuPs57t7e3WtzoPiCUCVoUzP7seOa+N5dQOHObGBJKVWhE456szAma3/CCUTfXLXxesjt3043s58lHn8lwvAuhS4ODCP0RHuYmf0h/3sT35Pt63AAHhBvB8fvRvKxljezXyZmlWqfUZOR8fZsS4xFa2ZbsWQQ8mRsK5rZaWa2av4ylowzV6wNatIN8fZaS4y9GU+GcidL11fZSkSWSayRdAShM8A/mNnAWK11KbGm3CjCUHk5I+PtyXm/2/VZ8pteFrkk80zCRezJFK7xljsm5+87d6PIePLu/j5hCKvu+ec2cdszLfQzkK+6/drWhGNLsVp59XELYd+/IVXPqU8gHL+PNrMbzGzF5EILnb09BfwEeNDd70sud/cHCO9xR8K4x1WO9/Ec8I/U0FeEVLjadH1L3nAqBZavTtjROHGMOUKSdVect4DQdu9ewgnZm4REYkLe46zLkvHnJhLqpD8X172RZRxOpcDyYYnXVmX8IcLA4XPi8g8InZMMJQyqPjvO7563Ta7r7lmEUtXHCT/m51kyjmiXxPpdanpdLOnV1Qm9ow0FXo3/z6HqcB69KDDMQGJ5f/KGQCBUEZkS539COHkdQUi+nye0t6zyHaAOw6kQGsjn3u/v4uczlFClbhJLhlU5v8C2J7JkLNc3CWN23UsoyZoX569S3Wss8t3Ifz0Hx++rExKHewjVUD4t8BxDWMbhVAjVemfF9d+Nr+mZ+FrvLbRNQ0zUcXiUYusTeir2GP9T8f17L/4/sND3knBV9/u4zgjCRZrBwIZ5z9W/mniqfBbAh3HeqLi80FRw31Dg8f/Ekn3ZY8B9hAP4J4T9QqHvQTeWjHf2BaEU9G6WtPt08saTrCGGv8ZtFhFqiAwlHMC/puq+tMH3GTX9zghXz3PDwEwm7G9fJAwRdDV5+w6W7B+Sx4t/E44FTqjO1rWmmAljqObG2/uWcBHrQZb8zh4mbyxY6jH0iyZNmgpPhCqXuaFA5hKOaf+Mv73csXQhifGyCUlWblzkGYR2+c/G7QseC2vab+Wtu1bcd+bOQa4ust4v4j7K4z7zHpacuw0oti8Abo3Ldiqw7Jv43BMI51b3smSc5zkUHrf4chJDltTyfe9HzeeTp8V1XgMsb1l3QhKcO4d9lHDseJklw4b9m7zxUhPbt2Dpsdo/IJyr3UM4ZuXOrd7L+juqKbupdivVkHjGda6N6zyQN39flgyZsSDejiWceGxZ4HHWiTuoLwkno+MJVR+71PSDKvBYI6k+8dw0sSMqOPAtoZ3nTYS68d/HH+O78Yd0ANAqb/2WhOoTk+Lr/SS+1hWofhzPal8XYbzAfxJOdn+It/8ENiqwbi/qdxK5JqEzpU8Jydz7cee3fKHYE9+Nj2v5eRiwLWEYnmfje/R9nCYSLlT8vJrtuxOShY8JDfy/JiShtxE6I7KaXmOB70aV7zSwOWHYjcmJ7+wLhCulLRLrDWEZE884fwPCznka4SD7BuHKavNi26Q9kVLiGZcdRviNzyb0gPwY4WBe9HtJqHr5AuGCRG5/0yvvufpXE0+VzyL33tUwFfx+FHmOYwkH63nxO/EPwm+m4PcgbrMyoU3iuPjavid87x8jXFFvX8fPad+47VfxN/AZ4cJN3wLrlmKfUXB+YvnqhH3KjPi+vQ4cF5flJ54t4ntyP+GEZTYhcXyDcMK3Rm1jjo/12/i+z4nTOEL7sxYF1h+JEk9NmlKbgBUJSc5TcX/5Q/w9587/uhXYpg2hdHNq3L+9T7iYWfBYWNN+q8DjP5PY929WzXpbE2rgfRn3Q2MIfUwU3RfEbRwYVGDZEYSLju/E92B2vP83YL0iMXwY95utavPa4jb9qDnxXJ4lFwWqjI0eP7ezCcn2zPg5TIn75T1rGUcPQunqe4TjXu5Y9RChunKVfbCmpjNZ/JKIiIiIiEg9xA521gc6eRivvr6P8wtCTbNr3P28tOITaQzS6FxIRERERKQpu5DQOdFJy/g4FxCq5169zBGJNDIq8RQRERERWUZmNgLYAljX3efWY/sehKq957n7NTWtL1JulHiKiIiIiIhIg1JVWxEREREREWlQSjxFRERERESkQbXIOoBKscIWp6rOsjSYUcOvyjoEqXDdO7extB6rPvvD78ffmNrzS+PTvn1779KlS9ZhiIhIAxs3btyX7t6h0DIlniIiItKgunTpwtixY7MOQ0REGpiZfVJsmRJPERFJl6kVh4iIiCxNiaeIiKTLVGtWRERElqbEU0RE0qUSTxEREcmjxFNERNKlEk8RERHJo8RTRETSpRJPERERyaPEU0RE0qUSTxEREcmjxFNERNKlEk8RERHJo7MDERFJl1ndJ2kUzOx2M5tuZm8WWW5m9lczm2hmr5vZlqWOUUREypMSTxERSZc1q/skjcUQoHc1y/cEusXpeODmEsQkIiIVQFVtRUQkXSrBLFvu/j8z61LNKn2Au9zdgZfNbBUzW8Pdp5UkQBGRCjB09KcMnzAl6zCWstKcWazTbS0u+dUmDfYcSjxFRCRdKsGsZJ2AzxL/T47zqiSeZnY8oVSUzp07lyQ4EambxpgANQWjJ80EoGfXdhlHArZ4EX2e+Cf7jbiLOwb+E1DiKSIi5aIBSjzN7HZgH2C6u28a590HbBhXWQX4xt27xxK7d4D34rKX3f3EuM1WhOqkKwCPAr+LpXeSMncfBAwC6NGjh95jkWpklQA2pgSoKenZtR19unfi0J4ZX5SbOhUOPRSeew4OPJCTD9upQZ9OiaeIiKSrYUo8hwA3AnflZrj7QT8+pdm1wLeJ9T909+4FHudm4DhgNCHx7A2MaIB4K9UUYO3E/2vFeSKSpy7JZFYJYKNJgCQbLVrA5Mlwxx1w5JEN3lRGiaeIiKSrARLP6toempkBBwK7VBuW2RpAW3d/Of5/F7AfSjzr4mHgVDO7F+gJfKv2nZK2Sqn+WZdkUgmglMzs2XDjjXD22dCxI7zzDrRsWZKnVuIpIiLpalbyzoV2AL5w9w8S87qa2XhgFvB7d3+e0BZxcmKdXPtEiczsHqAX0N7MJgOXAC0B3P3vhFLivYCJwFzgqGwilXJQ3wSyUqp/KpmURmfMmFC19sMPoWdP2HnnkiWdoMRTRETSVo8Sz2RHNNGg2EawNg4B7kn8Pw3o7O5fxTadD5lZw/WWUEHc/ZAaljtwSonCkUakPklkfRNIJWwiKVu0CK65Bi6+GNZYA0aOhB13LHkYSjxFRCRzyY5o6sLMWgD7A1slHms+MD/eH2dmHwIbENoirpXYXO0TRfIUSzDrk0QqgRRpJI47LrTjPPBA+PvfYdVVMwlDiaeIiKSrtON47ga86+4/VqE1sw7ATHdfZGbrAt2Aj9x9ppnNMrNtCZ0L/R9wQymDFUlLQ7WDLJZgKokUKUOLF0OzZnDCCbDDDtCvX6ZjbSvxFBGRdDVA50KF2h66+23AwSxdzRZgR+AyM/sBWAyc6O4z47KTWTKcygjUsZA0MrVNKBuqHaQSTJEKMHs2nHYarLQS3HBDaM/Zs2fWUSnxFBGRlDXA1dRibQ/dvV+BecOAYUXWHwtsmmpwIikZOvpTLnzwDaDmhFIJoogUlOxA6MILwT3TUs4kJYeUk5EAACAASURBVJ4iIpKuhhnHU6SiJZPOK/tupoRSROomvwOhZ5+FnXbKOqqlKPEUEZF0NZIrqyLlIFe1Nld1VkmniNTLp5/C5ZfD/vtn2oFQdZR4iohIulTiKVJrwydM4e1ps1R1VkTq56WXYLvtoGtXmDAB1l+/0V4A1tmBiIiky6zuk0gTNHT0p4yeNJON12jLfSdsp6RTRGpv9mw4+mjYfnt46KEwr1u3Rn1MVYmniIikSyWeItXKr17bp3unjCMSkbKS34HQPvtkHVGtKPEUEZF0NeKrrSJZy++5VtVrRaRObrwRzjgDVl+9UXYgVB0lniIiki6VeIosJTk2pzoREpFlsvba0Lcv3HJLo+xAqDpKPEVEJF1KPEV+lF/CqVJOEamz+++Hzz+HU0+FPn3CVIaUeIqISLpU1VaaOJVwikgqZs+G006DO+4InQiddBI0b551VPWmxFNERNKlEk9pwlTCKSKpyO9AqH//sk46QYmniIikTSWe0gTl91SrEk4RqbfPP4cddoAOHcquA6HqKPEUEZF0qcRTmhj1VCsiqfjuO2jTJvRYe/fdsOuuZdeBUHV0diAiIukyq/skUqaSSeeVfTfjvhO2U9IpInV3//3QpQs88UT4/ze/qaikE5R4ioiIiNRbrhMhVa0VkXqZPRuOPhoOOADWWw+6ds06ogajqrYiIpIqUwmmNBFDR3/K6Ekz6dm1nZJOEam7Qh0ItWyZdVQNRomniIikSomnNBW50s4+3TtlHImIlKVXXoF58yqqA6HqqKqtiIiky+oxiZQplXaKSJ1MngxPPhnun3wyvPVWk0g6QYmniIikzMzqPImUm1w1WxGRWhs2DDbfHPr1g/nzQ+d6bdtmHVXJKPEUEZFUKfGUpkDVbEWk1mbPhmOOCT3Vrr8+jBwJyy2XdVQlpzaeIiKSKiWSUsmGjv6U4ROm8Pa0WapmKyI1+/Zb2HprmDgRLrgALr20ojsQqo4STxERSZUST6lkuaRz4zXaqrRTRGq28sqhpHP33aFXr6yjyZQSTxERSZfyTqlwG6/RlvtO2C7rMESksZo8GY47DgYMgJ/9DK68MuuIGgW18RQRkVSpjaeIiDRZuQ6Enn8ePvoo62gaFSWeIiKSKiWeUomGjv6Ug24ZxdvTZmUdiog0RrNnw7HHLulAaPx46Ns366gaFSWeIiKSKiWeUonUtlNEqvW3v8Htt4cOhF58Ebp1yzqiRkdtPEVEJFVKJKXS5Mbs7Nm1ndp2isgSixbBZ59Bly5wxhmw446w7bZZR9VoqcRTRETSZfWYRBqpoaM/5cIH3wA0ZqeIJEyeDLvtBr/4BXz3XRgiRUlntVTiKSIiqVKJp1SS4ROmAHBl3800ZqeIBMOGhV5rFyyAG26A1q2zjqgsqMRTRERS1RBtPM3sdjObbmZvJub1N7MpZjYhTnslll1gZhPN7D0z2yMxv3ecN9HMzk/9xUtFSVaxVdIpIsyfX7UDoaOOAl1wrRUlniIikqoG6lxoCNC7wPzr3L17nB6Nz78xcDCwSdzmb2bW3MyaAzcBewIbA4fEdUUKypV2qoqtiAChOu3UqepAqJ5U1VZERBo9d/+fmXWp5ep9gHvdfT4wycwmAtvEZRPd/SMAM7s3rvt2yuFKBVFpp0gTt2gR/OUvcOCBsPba8Mgj0Lx51lGVJZV4iohIukrbudCpZvZ6rIq7apzXCfgssc7kOK/YfJEqctVsRaQJy3UgdPbZcNddYZ6SznpT4ikiIqmqT1VbMzvezMYmpuNr8VQ3A+sB3YFpwLUN+sKkSVE1W5Embtgw2HxzGDMmjM954YVZR1T2VNVWRERSVZ9ebd19EDCojtt8kXjOW4H/xH+nAGsnVl0rzqOa+SI/UqdCIk3c7bfDMcdAjx4wdKjacqZEiaeIiKSqVMOpmNka7j4t/tsXyPV4+zAw1Mz+DKwJdANeIVTq7WZmXQkJ58HAoSUJVsqGxu0UacIWLoQWLUKvtdOnw1lnhQ6FJBVKPEVEJFUNkXia2T1AL6C9mU0GLgF6mVl3wIGPgRMA3P0tM/sXodOghcAp7r4oPs6pwONAc+B2d38r9WClrGncTpEmaNEi+NOfQvXa55+Htm3hfI24lTYlniIikq4GKPB090MKzL6tmvWvAK4oMP9R4NEUQ6s4ZtYbuJ6QnA929wF5yzsDdwKrxHXOzw1lUylUxVakCZk8GY44AkaODCWdCxbA8stnHVVFUudCIiKSqgYax1NKoJZjnf4e+Je7b0Gorvy30kYpIpKS/A6E/vWvUNopDUIlniIikiolkmVtG2oe69SB3JnZysDUkkYoIpKGhQvh8sthvfXUgVCJKPEUEZFUKfEsa4XGOu2Zt05/4Akz+y2wErBbaUITEUnBq6+GZHPlleG//4X27aFVq6yjahJU1VZERNJl9ZiknBwCDHH3tYC9gH+YWZXzieTYrDNmzCh5kPWRG0ZFRCrQokVw9dXQsyf84Q9h3pprKuksIZV4Smr+fslh7LnjpsyY+R09DrgSgM036MQNFx3Mcsu1ZOGixZx+5X2MfesTAHbYqht/OufXtGzRnK++mc3ux14PwMqtV+DmSw5l4/XWwB1OvPSfjH59UmavSxqvxYsWccEpR9CufUfOu/wvTJ82heuvvJDvZn3Lut024tTzLqNFy5b8sGABN11zCR998A5t2q7M7y66io6rr5l1+BVLJZ5lrboxUHOOAXoDuPsoM1seaA9MT66UHJu1R48e3lABpynXo62GURGpMPkdCPXvn3VETZJKPCU1/3jkZfqcctNS8644fT+uGDSCbQ8ewB9v/g9XnL4fEJLL6y88kANOv4WtfnMFh52zpHPKgef+hideepvu+1/ONgddxbsffV7S1yHl49EH76FT564//v/PwTew1/6H8tc7H2Kl1m145rHhADzz2HBWat2Gv975EHvtfyhDB9+QVchNgjoXKmtjiGOdmlkrQudBD+et8ymwK4CZbQQsD5RHkWYtqEdbkQozcmTVDoTatcs6qiZJiWdkZj81s/PM7K9xOi8eUKWWXnz1Q2Z+O3epee7QdqXQJfXKrVdg2oxvAThozx4Mf/o1Pvv8awBmfD0bgLatl+cXW67HkAdHAfDDwkV8O/v7Ur0EKSNfzfiC8aNfZJc9w8UMd+etCWPYdsddAdhp930Y8+JIAMa+9Bw77b4PANvuuCtvjn8F97IogClLSjzLl7svBHJjnb5D6L32LTO7zMz2jaudBRxnZq8B9wD9XD8oEWms1l8fttkGxo+Ho44CHXMyo6q2gJmdR2izci/wSpy9FnCPmd2bP4aZ1N45A+/nkZtO4aoz+tKsmbFzv2sB6LZOR1q0aM7jt/6O1isux033jGTof16hy5qr8eXXsxl06eFstkEnxr/zGWdfcz9z5y3I+JVIY3Pnzddy2HGn8f33cwD4bta3rNi6Dc2bh91au/YdmflVqPk386vprNbhJwA0b96CFVdqzXezvqXtyqtkE3yFUyJZ3gqNderuFyfuvw1sX+q4RERqbexYGDwY/vY3WGsteOyxrCMSVOKZcwywtbsPcPe74zSA0K38MRnHVtaOP2AHzr32Abrt+QfOHTiMmy85DIAWzZux5UZr0/e3N7PvKTdxwXG9Wb9zSEa7/3Rtbv3382x3yNXM/X4+Zx/9y4xfhTQ2415+nrartGPdDVQpoVFS50IiIpKFxYtDB0LbbRd6rP3ss5q3kZJR4hksBgr1NLJGXFZQsse+hV++1WDBlbPD9unJQ09PAGDYk+Ppsck6AEyZ/g1PjnqHufMW8NU3c3jh1YlsvkEnpnzxNVOmf8OYN0MHRA8+NYHuP1276ONL0/TeW68xbtT/OPXwX3H9FRfx5oQxDPnbQObO/o5FixYCMPPL6bRbrSMA7VbryFczvgBg0aKFzJ0zmzZtV84s/kqnqrZSjtSjrUiZmzwZdtsNzj8f9tsPXnsN1lkn66gkQYlncDrwtJmNMLNBcXoMeBr4XbGN3H2Qu/dw9x4t2m9SsmDLybQZ37LDVmFA3l7bbMDET0P/E4+MfJ2fd1+P5s2bscLyLdl60y68O+lzvvjqOyZ//jXd1ukYt9lQnQtJFYcecyo33/MoN979CL+76Ao27b41p11wORv/rAcv/+9pAJ574j/0+PlOAPTYbkeee+I/ALz8v6fZpPvWSnZEZCnq0VakjLnDr34Fr7wCt92mDoQaKbXxBNz9MTPbgFC1NnfEmQKMcfdF2UVWXu68qh87bNWN9qu0ZuJjf+SPf3+UU/44lD+d8xtatGjG/PkLOfXyewB4b9IXPPnS24z51wUsXuwMefAl3v5wGgBnXv1v7riyH61aNOfjKV9y/CV3Z/mypIwcdtxvuf6KC7lvyM10WW9DdundB4Cd9+zDjQMu5rQj96N1m7b87qIrM460simpl3KlHm1FysycOdCyZRiL85ZbYNVVoVu3rKOSIkwd0aVjhS1O1RspDWbU8KuyDkEqXPfObVLLFtc/e0Sd94cTB+6pbLWC9ejRw8eOHZt1GNU66JbQm/p9J2yXcSQiUitjx8Khh0LfvqFdpzQKZjbO3XsUWqaqtiIikiq18RQRkQaT7EBo7lzo3TvriKSWlHiKiEiqzOo+iWRJHQuJlIlkB0J9+sDrr8POO2cdldSS2niKiEiqVIIp5UYdC4mUiZkzQ7J5221w1FG6cllmlHiKiEiqdB4g5UgdC4k0UnPmwP33w5FHwuabwyefwEorZR2V1IOq2oqISKqaNbM6TyIiIlWMHQtbbhlKN994I8xT0lm2lHiKiEiq1MZTRESWSbIDoTlz4OmnYbPNso5KlpESTxERSZV6tZVyoo6FRBqhAw5QB0IVSG08RUQkVcojpZyoYyGRRsQ9HEQOPhj22guOPloHlQqixFNERFKlEsxsmdm6wMFAJ3c/xcw2BFq4+1sZh9ZoqWMhkYzNmQOnnw4/+xmcemoo8ZSKo6q2IiKSKlW1zY6Z/RJ4DdgWOCLObg8MzCwoEZHqjBsXOhC67TaYMSPraKQBKfEUEZFUqXOhTA0ADnD3fYFFcd6rwJbZhSQiUsDixXDNNUt3IHTppVlHJQ1IVW1FRCRVKsHM1Hru/li87wDu/r2ZtcwwJhGRqsaOhfPOg1//GgYNgnbtso5IGphKPEVEJFUq8czUZ2a2aXKGmf0M+DibcERE8rz7brjdZhsYPRr+/W8lnU2EEk8REUmV2nhm6q/AA2Z2ONDczH4N3A1cl21YItLkzZkDxx0Hm2wCr7wS5m2zja4+NiGqaisiIqnSOUR23P3WmMifBzQHLgX+4u7/yDQwEWnaxo2DQw+FDz6Ac8+F7t2zjkgyoBJPERFp9MzsdjObbmZvJub9yczeNbPXzexBM1slzu9iZt+b2YQ4/T2xzVZm9oaZTTSzv1oFFre6+63uvpm7t3b3Td19cNYxNVZDR3/K6Ekzsw5DpLJdd93SHQgNGACtWmUdlWRAiaeIiKSqgaraDgF65817EtjU3TcH3gcuSCz70N27x+nExPybgeOAbnHKf8yyZmbvFJn/RqljKQfDJ0wBoE/3ThlHIlLBFi6EX/0KXnsNdt4562gkQ6pqKyIiqWqIMkR3/5+Zdcmb90Ti35eB31Qfl60BtHX3l+P/dwH7ASNSDTZba9VxfpPXs2s7Du3ZOeswRCrLAw+EUs199oGzzlJPcgIo8RQRkZRlVHv1aOC+xP9dzWw8MAv4vbs/D3QCJifWmRznlT0zuzDebZG4n7M+8FmJQxKRpmjOHDj9dBg8GHbfPSSezVTBUgIlniIikqr65J1mdjxwfGLWIHcfVMttLwIWAv+Ms6YBnd39KzPbCnjIzDape1Rl5ZfxtmXiPsBi4HNCYi4i0nCSHQiddx5cdlnWEUkjo8RTRERSVZ8Sz5hk1irRzHuufsA+wK7u7vGx5gPz4/1xZvYhsAEwhaWrnK4V55U9d98ZwMxucPffZh2PiDQxb78dOhDq2DF0IKS2nFKAyr5FRCRVuaY8dZnq9zzWGzgX2Nfd5ybmdzCz5vH+uoROhD5y92nALDPbNvZm+3/A8GV8uY2Kkk4RKakFC8LtRhvBwIHqQEiqpcRTRERS1RC92prZPcAoYEMzm2xmxwA3Am2AJ/OGTdkReN3MJgD3Aye6e27MjJOBwcBE4EMqq2MhzGwFM7vCzF42sw/N7KPclHVsjY2GUhFZRg88AOutB+++G64gnnYarLZa1lFJI6aqtiIikqoG6tX2kAKzbyuy7jBgWJFlY4FNUwytsbkO+AVh2JirgfOAU1nS/lUiDaUiUk/JDoS22gqaN886IikTKvEUEZFUNdA4nlI7vyJUPb4JWBhvfw2o7lsBGkpFpI7GjYMtt4TbbgsdCL30EnTrlnVUUiZU4ikiIqlSIpmp1u6eq1a7wMxaufvbZrZ1plGJSGW4++5Q4qkOhKQeVOIpIiKpKlXnQlLQJDPbKN5/FzjazA4Gvs0wJhEpZ1OmhE6DAK66Sh0ISb2pxFNERFKlEs9MXQV0Bt4B/gg8CLQCTsoyKBEpUw88AMceC2utBRMmwPLLh0mkHpR4iohIqpR3ZiMOEfMs8CWAuz9pZqsCrdx9TqbBiUh5ye9AaOhQaKaKkrJslHiKiEiqVOKZGQM+IQwxsxjA3X8AfsgyKBEpM1OmwC67wAcfhA6ELrsMWrXKOiqpALp0ISIiqVIbz2y4+2LgI2DVrGNp7DSGp0g1Vl89lHI+/TQMGKCkU1KjxFNERFLVzKzOk6TmWmComW1vZmuZ2Zq5KevAGhON4SmSZ8oUOPhg+PzzMC7n0KHqQEhSp8RTRESkcgwGdgWeJ1S7/QyYHG8lQWN4ikQPPgibbw6PPBI6EBJpIGrjKSIiqVIBZqa6Zh2AiJSJOXPgjDPg1luXdCC0wQZZRyUVTImniIikSp0LZcfdP8k6BhEpExddFHqtVQdCUiKqaisiIqlqZnWfpPEws95m9p6ZTTSz84usc6CZvW1mb5nZ0FLHuCzUsZA0aYsXw1dfhft/+AM884w6EJKSUYmniIikSiWe5cvMmgM3Ab8ktA0dY2YPu/vbiXW6ARcA27v712bWMZto627o6E+58ME3AHUsJE3QlClw5JEwaxa8+CKsthr06pV1VNKEqMRTRERSpeFUyto2wER3/8jdFwD3An3y1jkOuMndvwZw9+kljrHecr3ZXtl3M3UsJE3LQw+FDoRGjYLjj4cWKnuS0lPiKSIiqbJ6/Emj0Ymle8CdHOclbQBsYGYvmtnLZta7ZNGlQL3ZSpMydy6ccAL07Qtdu8L48XDssbriJ5lQ4ikiIqlSG89smdnhZvakmb0e/9/RzPZP8SlaAN2AXsAhwK1mtkqBOI43s7FmNnbGjBkpPr2I1MkLL4QOhF56Sb3WSqaUeIqISKrMrM6TpMPMzgQuBUYAuWK9GcC5tXyIKcDaif/XivOSJgMPu/sP7j4JeJ+QiC7F3Qe5ew9379GhQ4c6vIqGoU6FpMlYvBhuuSUMl7LiijBunDoQkkZBiaeIiKRKbTwzdRKwp7v/GfA4731g/VpuPwboZmZdzawVcDDwcN46DxFKOzGz9oSqtx8tY9wNLte+U50KSUWbMgV23x1OPBHuuivMW375bGMSidSyWEREUtVMmWSW2rn7+/F+LvG0xP1quftCMzsVeBxoDtzu7m+Z2WXAWHd/OC7b3czeBhYB57j7V6m+igai9p1S0R56CI45BubNg1tvDfdFGhElniIikirlnZl628z2cff/JOb1Bl6r7QO4+6PAo3nzLk7cd+DMOIlIY/DnP8NZZ8FWW8HQoWrLKY2SEk8REUmV2mxm6kLgv2b2L2A5M7uBUF12n2zDEpEG4R6u9u23H8ycCRdfrLac0miVbeJpZm9Q+6pDmzdwOCIiEinvzI67P29m2xLaej5L6Muhl7u/lW1kIpKqxYvh2mvhlVfgX/+CddeFyy/POiqRapVt4gncn3UAIiJSldp4Zsvd3wZ+m3UcItJApkyBI4+Ep58O43POmwcrrJB1VCI1KtvE090vzToGERGpSmlndsxsInA7MMTdp2Ydj4ikrFAHQrrYJ2VCw6mIiEiqNI5npi4Hdgc+NrMRZnaAmbXMOigRScHs2XDSSdC1K7z6Khx7rJJOKSsVk3ia2VFm9oSZvWtmHyWnrGMTEREpBXcf4u69gI2AscCfgGlmdn2mgYlI/b31FixcCK1bwzPPwEsvwYYbZh2VSJ1VROJpZucA1wLjgC6Ewa3fBNoRqhyJiEiJNLO6T5Iud//Q3f8AbAuMBk7NOCQRqavFi2HgQNhiizBcCsBGG6nXWilbFZF4AscBx7v7BcAPwI3uvi8hGV0n08hERJoYVbXNlpk1N7M+ZvYQ8DGwCnBCtlGJSJ1MmQK77w7nnAP77BPacoqUuUpJPNcCXon3vwfaxvv3AL/OJCIRkSbKrO5TzY9pt5vZdDN7MzGvnZk9aWYfxNtV43wzs7+a2UQze93Mtkxsc2Rc/wMzO7IhXn+WzOw6YApwE/AO8DN3397dB2cbmYjU2uOPw+abw6hRoQOhYcNgtdWyjkpkmVVK4vk50D7e/wTYLt5fn1qO9SkiIulooBLPIUDvvHnnA0+7ezfg6fg/wJ5AtzgdD9wc42oHXAL0BLYBLsklqxWkE3AU0NndL3D397IOSETqqF072GADdSAkFadSEs9ngH3j/duAP5vZs8B9wAOZRSUi0gQ1RBtPd/8fMDNvdh/gznj/TmC/xPy7PHgZWMXM1gD2AJ5095nu/jXwJFWT2bLm7ge6+wh3X5x1LI3J0NGfMnpS/tdHpBEZPx4GDAj3t95aHQhJRSrbcTzzHE9Mot3972b2NbA9MAy4JcvARESamhK22fyJu0+L9z8HfhLvdwI+S6w3Oc4rNr+smdnZ7j4w3r+w2HrufmXpompchk+YAkCf7mX/cUulWbw4dBx04YXQsSOccAKsuqpKOaUiVUTiGa/sLk78fx+htFNEREqsPqdLZnY84SJiziB3H1Tb7d3dzaypNq3YBRgY7/+yyDoONNnEE6Bn13Yc2rNz1mGILDF1Khx5JDz1FOy/PwwaFJJOkQpVEYlnsuOIQtz91VLFIiLS1DWrx5X6mGTWOtGMvjCzNdx9WqxKOz3OnwKsnVhvrThvCtArb/7IOgfbyLj7Xon7O2cZi4jU0oIF8POfw4wZoQOhY45RKadUvIpIPAmDZDtLX2hPXvluXtpwRESarhKeOz0MHAkMiLfDE/NPNbN7CR0JfRuT08eBKxMdCu0OXFCyaEvAzAa5+/EF5t/s7idlEZOIJMybB8stF8bivP56+OlP1ZZTmoxK6VyoK7BuvO0KbAAcDLwB7JNhXCIiTU5D9GprZvcAo4ANzWyymR1DSDh/aWYfALvF/wEeBT4CJgK3AicDuPtM4I/AmDhdFudVkoOLzD+wpFGISFXjx0P37nDHHeH/Pn2UdEqTUhElnu7+SYHZE83sW0LX+SNKHJKISJPVECWe7n5IkUW7FljXgVOKPM7twO0phtYomNnP491mZrYdS9cA6gbMKX1UIgJU7UCoa9esIxLJREUkntWYBHTPOggRkaakPm08ZZm9EG8deDEx34FpwEUlj0hECncgtNpqWUclkomKSDzjoOBLzQLWAPoDGjxbRKSElHeWnrs3AzCzCe6uC64JuTE8e3bNP1UQKYHx42HUKHUgJEKFJJ7AlyzdmRCE5PMz4KBSBPD1mBtL8TTSRN034bOaVxJZBt07t0ntsUo4jqfkUdJZlcbwlJKbOxeefx722AP23hsmTYIOHbKOSiRzlZJ45ncfvxiYAUx094UZxCMi0mRVSq915cLMznb3gfH+hcXWc/cmO46nxvCUkhk/Hg45JCSbkybBmmsq6RSJKiXxnAR8FjuUWIqZdXb3TzOISUSkSVKJZ8ntAgyM939ZZB0HmlziqWq2UjLJDoQ6dIARI0LSKSI/qqTEMzl4OABmtlpcpnE8RUSkIrn7Xon7+TWAmjRVs5WSWLw4VKl97DHo2ze051QHQiJVVEriaVRt4wnQGphX4lhERJq0ZirwbDTMrBew0N1fqGndSqVqttLgmjWDnXcOvdYee6w6EBIpoqwTTzP7a7zrwFVmNjexuDmwDTCh5IGJiDRhSjyzY2ZPAJe7+//M7HfAVcAiM7vY3a/LODyRyjF3Lpx5Zijh3GMPOPfcrCMSafTKOvEENou3BmwELEgsWwC8ypJ2LyIiUgJq45mp7sCoeP84YHfgG+AhQImnSBpyHQi9/z506RISTxGpUVknnrm2LGZ2B/A7d5+VcUgiIk2eSjwz1crdfzCznwAdc1VszaxjxnGJlL/8DoSeegp22SXrqETKRlknngkXAG2BpRJPM1sL+MHdv8gkKhGRJkgFnpn6yMyOBNYDnoEfO9pTfwciy+rhh+Gcc9SBkEg9Vcpwa3cDexaYvwfwjxLHIiLSpDUzq/MkqTmXMGzK4cAVcd7ewJjMIspIbigVkWU2dWq47dMHHn0Uhg1T0ilSD5WSePYA/ldg/vNxmYiIlEizekySDnd/yt07ufu67v5GnH0PsF+WcWVBQ6nIMpszB044AX76U/j441CdY889Va1DpJ4qpaptC2C5AvOXLzJfREQaiM7JsmVmrQmlnGsDnwH/dffZ2UaVDQ2lIvWW60DovfdC9do118w6IpGyVykXmkcDJxWYfwpNsHqRiEiWVNU2O2a2CfA+oUf3PvH2fTPbNNPARMqFO1x7LfTsCd99FzoQuuYaaNUq68hEyl6llHheBDxjZpsTO1MAdgG2BHbNLCoRkSZIeWSm/gLcAlzm7m5hbJvfA9ej46FIzczg3Xdh771h8GC15RRJUUUknu7+spltR+hUYf84+1XgZKBDZoGJiDRBGk4lU1sCe7m7A8TkcwBwerZhiTRyw4fDOutA9+5w003QsqWuoomkrFKq2uLu3BBedgAAIABJREFUr7n7Ye6+CaE32/eBB4HHs41MRKRpUVXbTH0DdMmb14W84cZEJJo7F048EfbbD66+Osxr1UpJp0gDqJjE08yam9n+ZvZfYBKhB7+/A+tnG5mISNNiVvdJUnMn8F8zO9rMdjGzo4FHgCHZhiXSCI0fD1tuCbfcEjoQuvPOrCMSqWhlX9XWzDYEjgX+D5gDDCWUeB7h7m9nGZuISFOkqraZugL4ATiPJb3aDgH+lGFMIo3PCy/ALrtAhw6hA6Fd1QRapKGVdeJpZs8DmwLDgAPd/bk4/7xMAxMRacIMZZ5ZcfdFwFVxEpF8ixdDs2ah19qzzoKzz1YHQiIlUu5VbbcD7gKuyyWdIiKSrWZW90nSY2atzexgMzvbzA4yszZZxyTSKAwfHqrWfvVV6DzoqquUdIqUULknnlsTSm1fMLPxZnaGma2edVAiIiJZMLMewEfA1YRxPK8BPozzRZqmZAdCzZrBLPW1JZKFsk483X28u58CrAH8GdiX0J6lGbC3ma2aZXwiIk2RSjwz9TfgWndfx913cPd1gIHAzRnHJZKN8eNhq62WdCA0ahR07Zp1VCJNUlknnjnuPs/d/+HuOwMbETpROAP43MxGZBudiEjTYmZ1niQ1GwHX5s37M/DT2j6AmfU2s/fMbKKZnV/Ner82M1dpqjRql14aSjifegquuQaWWy7riESarIpIPJPcfaK7n0/oze9AYEHGIYmINCkq8czUBEKne0mbxfk1MrPmwE3AnsDGwCFmtnGB9doAvwNGL1O0Ig1h6lSYPDncHzQIXn9dvdaKNAJl3attdWLPfsPjJCIiJaICzEw9AfzHzAYDnwBdgKOBQWZ2aG4ldx9aZPttgInu/hGAmd1LaCuaPzzZHwntSM9JNXqRZTV8OBxzDGy9NYwYAR07Zh2RiEQVm3iKiEg2minzzNLRhHE8j0zMWxjn5zhhzOtCOhH6SsiZDPRMrmBmWwJru/t/zUyJpzQOc+fCmWeGtpxbbAHXXZd1RCKSR4mniIikqiGqzprZhsB9iVnrAhcDqwDHATPi/Avd/dG4zQXAMcAi4DR3fzz9yBoXd2/QXlPMrBmhzWi/Wqx7PHA8QOfOnRsyLGnqJk6EX/0K3n03jMt5+eVqyynSCCnxFBGRVDVEgae7vwd0D49vzYEpwIPAUYSxnAcuHYNtDBwMbAKsCTxlZhvEZhhS3BRCHwk5a8V5OW0IbUhHxk6hVgceNrN93X1s8oHcfRAwCKBHjx7ekEFLE9exI3ToADfcALvtlnU0IlJExXUuJCIi2WqG1Xmqo12BD939k2rW6QPc6+7z3X0SMJHQflGqNwboZmZdzawVIXl/OLfQ3b919/bu3sXduwAvA1WSTpEGN3UqnHwyzJsHbdvCc88p6RRp5JR4iohIqszqPtXRwcA9if9PNbPXzez2xPjNhdoqdqr3i2oi3H0hcCrwOPAO8P/t3XeYFGXW9/HvYUZAAUWCKEFBBSTOoAQxIEGCPgKiiGACUTDhiqK7uq6Krq646vqI4iqurqwrwcy86CKsyGMEJUcDShYByQhIOu8fVTM2wzABuruGmd/Hq67prqquOl3OUH36vu9zv+7u883sITPrEm10+Tdy6jKmLl4fdRiSKGPHQuPG8MorMH16sE5jy0UKPSWeIiISVwcznYqZ9TezaTFL/5yOHbbCdQHeCFf9HTiFoBvuKvafw1IKyN3fd/c67n6Kuz8Srrvf3TNy2Ld1YWztHDsr6B3cNV3fNRQp27bBjTfCxRfDiSfCjBlw9tlRRyUi+aQxniIiElcHU9U2djxgHi4AZrj76vB1qzM3mNmLwLjwaV5jFYs0CwZgHu/uq6KOJSotalXgihYqalSkXHcdjB6tAkIihym1eIqISFwluKttL2K62ZrZCTHbugHzwscZQE8zK2VmtYDawJeH9s4KPzMra2YvAdsJxrViZheb2QPRRiZykPbuDVo6AR54ACZOhMcfV9IpchhS4ikiInFVwqzAS36YWRmgPfB2zOq/mtlcM5sDtAFuB3D3+cDrwAJgPHBLMalo+yRQBTgb2Bmu+wq4PLKIRA7WqlXQqVPQ0ukOp52mAkIihzF1tRURkbhKVI0Pd/8FqJht3dW57P8I8Ehioim0LgLqu/smM3MAd19pZlUjjkukYDIyoG/foLXzf/836mhEJA7U4ikiInFV4iAWiZsSBN1ss5hZWWBrNOGIFNC2bXDTTdC1628FhPr3V9VakSJA93sREYkrMyvwInHzKXBPtnW3Ah9FEItIwW3cCG++GRQQ+uKLoHutiBQJ6morIiJxpTQyUncAk8zsKqCsmc0FSgJtow1LJBd798Ibb8Bll0HVqvDtt3DssXm/TkQOK0o8RUREigh3X25mDYHOQE1gKTDO3bfn+kKRqKxaBX36wIQJkJoKl16qpFOkiFLiKSIicXUw83hK/Lj7r8CbUccRlZFTlzF18Xpa1KoQdSiSl9gCQi+8AJdcEnVEIpJASjxFRCSulHZGx8yGH2ibu/dPZixRGTtrJQBd06tFHInk6qGHgnk5mzSBkSM1llOkGFDiKSIicaUGz0gdke15VaAVxawFtEWtClzR4sSow5DctGkDW7bAww9DqVJRRyMiSaDEU0RE4kpVaqPj7tdmX2dm3YAOEYQj8pu9e4P5ONetg0cegXPPDRYRKTY0nYqIiMSV5vEsdN4FLo86CCnGVq2CCy6AQYNg4ULYsyfqiEQkArrfi4hIXGkez0LnAkBVbSUaGRnQuDF88klQQOittyAlJeqoRCQC6morIiJxpTQyOmb2HeAxq8oAxwG3RRORFGurVsHll0O9eiogJCJKPEVEJL7Ughmph7M93wLMcvcfoghGiqlly+DEE+GEE2DiRGjWTAWERERdbUVEJL40xjMaZpYKVAHGuPuIcHlbSackzd698NRTULs2vPFGsO6cc5R0igig+72IiMSZxnhGw913A3909x1RxyLFUGYBoTvuCH62aRN1RCJSyCjxFBGRuLKDWCRuPjKz86IOQoqZ99/ft4DQO+9ApUpRRyUihYzGeIqISFypATNSS4CxZvZm+Hhv5gZ3/0tEMUlRt3Ur1KihAkIikislniIiElcl1IaZdGa22d2PBtKBmcAp4ZLJASWeEj+zZ8O8eXDlldCjB1xyCaTqY6WIHJj+hRARkbhSi2ckDMDdi/XAupFTlzF18Xpa1KoQdShF19698PTTcPfdUK0adO8eFA9S0ikiedAYTxERiSs7iP/kkHneuxRtI6cu44/vzAWga3q1iKMporIXEPryS1WsFZF809dTIiISV2rxjERpM3s5tx3cvW+ygonC2FkrAfhLt0Zc0eLEiKMpgjZtgvR02LIFnn8e+vfXH7uIFIgSTxERkaJhT9QBRK1FrQpKOuNtzx5ISYFjjoEHHgimSalXL+qoROQwpMRTRETiSsWFIrHD3ftFHYQUMbNnw1VXwbPPwnnnwc03Rx2RiBzGNMZTRETiyqzgi4gUInv3wlNPQfPmsG4deLEfQiwicaDEU0RE4kqJZySK9VXMrGYrcZC9gNCcOdC6ddRRiUgRoK62IiISV6pSm3zuXi7qGKKUWVhI1Wzj4M034ZNPVEBIROJOiaeIiMRVCX1OlSSKnbtThYUO0rZtsHAhnHEG3HILXHQR1KoVdVQiUsSoq62IiMSV5vGUZFJr5yGaPRuaNoWOHYOpUkqUUNIpIgmhxFNEROJKYzwl2dTaeRBiCwht3AijRkG5Yt1jW0QSTF1tRUQkrhLVgmlmS4AtBPNV7nb3pmZWARgD1ASWAD3cfYOZGfA0cCGwDejj7jMSEpjI4Wb7drj4YpgwAbp2hX/8AypVijoqESnilHhKwv20ahX33vN71q9bB2Z0v6wHV17dG4CRr73KmFGvUaJECq1ancftd/4+4mjlcLDux+W888zDWc83rllFq+69aXRue9555mE2rV3NMZWr0O1393FkmXLs2LaVjOeGsHndGvbu2UOL/7mMtPM6RfgOirYEj/Fs4+4/xzy/G/jQ3YeY2d3h8z8AFwC1w6UF8PfwpxQhseM7pQCOPBJOPFEFhEQkqZR4SsKlpKZw5+/vpl79Bvzyy1Z6XnYpZ7Y8m3XrfmbypA954+0MSpYsybp166IOVQ4TFavW4PpHXwBg7949PDOgJ3WbnsMXGaOp2aAJZ3XpxecZo/giYzRte/Vj+sQMKlU7iR53Pswvmzfywp3X0vDsdqSkHhHxOymakjxmsyvQOnw8AphMkHh2Bf7l7g5MMbPyZnaCu69KZnCSWBrfWQDbtsE998ANN0D9+vDii1FHJCLFjMZ4SsJVrnwc9eo3AKBMmbKcfPLJrFmzmjfGjKLv9f0pWbIkABUrVowyTDlMLZk3k2OPq8oxlavw7YzPaXxuBwAan9uBb6d/BgQTHO7csQ13Z9eO7RxZthwlSqREGHXRlsAxng5MMLPpZtY/XFclJpn8CagSPq4GLI957YpwnRQRqmZbAJkFhIYOhQ8/jDoaESmmlHjmwcyujTqGomTlyhV8vXAhjRqnsXTJEmZMn8aVPS+jb++rmDd3TtThyWFowZSPqH9WGwB+2bSBsscGX2CUKV+BXzZtAOCMDhfz88plDB1wOS/e3Y/2V9+MldA/f4liB7OY9TezaTFL/xwOfY67n07QjfYWM2sVuzFs3fQEvS0pZNTamQ/ZCwhNmAC33hp1VCJSTKmrbd4eBP4ZdRBFwbZffmHQwN9x191/pGzZsuzes4dNmzbx71GvM2/uXO4aNJD3P/gQ01gTyac9u3fx3fQvaH359fttM/ttmo4f5kyjykmncOW9T7Bh9Y+MGvIHatRtRKmjyiQ75GKhxEH8Dbv7cGB4HvusDH+uMbN3gObA6swutGZ2ArAm3H0lUCPm5dXDdVKEqLUzD8OHwx13QJcu8NJLKiAkIpHSV/6Amc05wDKX37pt5fS6rG/oX3ox189Lxd6uXbu4Y+DvuPB/OnN++6ArZJUqVWh3fnvMjEaNG1OiRAk2bNgQcaRyOPl+1pccX7M2ZY85FoAyxxzL1g3BWOGtG9Zx1DHlAZjz8XjqNjsXM6PC8dUoX/l41q1afsDjyqE5mBbPPI9pVsbMymU+BjoA84AMoHe4W29gbPg4A7jGAmcCmzS+U4qNzZuDn9deC6+9Bu++q6RTRCKnFs9AFaAjkD3rMeDzA70o9hv6HbvVvetA3J3B99/LySefzDV9fuu53Kbd+Xz15VSatziTJUsWs2vXLo499tgII5XDzfwvfutmC1D79JbM+WQCZ3XpxZxPJlDn9LMAOLricSyZP4MTT2vE1k0bWLdqOeWPOyGqsIu+xHRaqAK8E/aISAVGuvt4M/sKeN3MrgOWAj3C/d8nmEplEcF0Kho2UYSomu0BbNsGd94J48fDzJlwzDFwxRVRRyUiAijxzDQOKOvus7JvMLPJyQ+naJk5YzrjMsZSu04delzSFYBbB95Bt26Xcv99f+SSrhdxxBFH8OdHhqibreTbzh3bWTJvOhdcNzBrXcvOPXnnmYeZPXk8x1Q6jm6/uw+Ac7pdxbjnH+fFP1yPA2179uOocsdEFHnRl4iqtu7+A5CWw/p1QLsc1jtwS9wDKQbMrBPBHKgpwD/cfUi27XcA1wO7gbVAX3dfmswYNb4zB7NnQ69esHAhDBoEpUtHHZGIyD4suDfLoVKLpyTSmFnqFiqJ1btpjbhli1O/31Tgfw9bnHKMvnUqBMwsBfgWaE9QCfgroJe7L4jZpw0w1d23mdlNQGt3vzy34zZt2tSnTZsWtzgvf+ELAMbc0DJuxzxs7d0LTz8Nd98NFSvCiBHQvn3UUYlIMWVm0929aU7bNMZTRETiKoHTqUjiNQcWufsP7r4TGE0wJ2oWd//I3beFT6cQFG6SqLhDRgZ06gRz5ijpFJFCS4mniIjEVSKKC0nSFHT+0+uA/yQ0omwyx3cWe++9B6tWQUpKkHiqgJCIFHJKPEVEJL6UeRYLZnYV0BR4/ADbsyq/r127Nm7nLfbjO7dtg1tugYsugkcfDdaVK6euAyJS6CnxFBGRuLKD+E8KjXzNf2pm5wP3Al3c/decDuTuw929qbs3rVy5clyDLLbzd86eDU2bwnPPBfNzPp5jzi8iUiipqq2IiMSVGl4Oa18Btc2sFkHC2RPYZz4OM2sCvAB0cvc1yQ+xmHr/fejWDSpUgA8+gA4doo5IRKRA1OIpIiJxpZ62hy933w0MAD4AFgKvu/t8M3vIzLqEuz0OlAXeMLNZZpYRUbjFS8uWcM01QQEhJZ0ichhSi6eIiMSXMsnDmru/D7yfbd39MY/PT3pQoczCQi1qVYgqhOQaNy7oVvvuu3DssfDii1FHJCJy0NTiKSIicaUxnpIoxaaw0PbtQQGhzp3hxx8hjsWZRESiosRTRETiSvN4SiIV+cJC2QsITZ0K1Yp4oi0ixYK62oqISFwpjxQ5SO7Qty+sX68CQiJS5CjxFBGR+FLmKVIwP/0EZcsGy6hRwXjOOE9BIyISNXW1FRGRuNIYT5ECGDcOGjWCO+8Mntepo6RTRIokJZ4iIhJXGuMpkg+xBYSqVYPbbos6IhGRhFJXWxERiSvlkSJ5WLgQuneHBQvg9tvh0UehVKmooxIRSSglniIiEl/KPEVyd+SRsGePCgiJSLGirrYiIhJXGuMpkoOffoI//zmoXFuzZtDaqaRTRIoRJZ4iIiJS6I2cuoypi9dHHcbBySwg9Je/wPz5wboS+ggmIsWL/tUTEZG4UnEhSYSxs1YC0DW9WsSRFED2AkIzZkDDhlFHJSISCY3xFBGRuFIeKYnSolYFrmhxYtRh5F/XrjBxogoIiYigxFNEROJNmacUZ3v3BuM4U1LgnnuC+Tk1llNERImniIjEl4oFSbH100/Qpw+0aAEPPght2kQdkYhIoaExniIiElca4ynF0rhx0Lgx/N//QdWqUUcjIlLoKPEUEZG4soNYRA5b27fDgAFBAaGqVWH6dLjhhqijEhEpdJR4iohIfCnzlDgr1FOpfPstvPhiUEBo6lSoXz/qiERECiWN8RQRkbjSGE+Jt0I3lYo7fPQRtG0LaWnw3Xdw4mFUbVdEJAJq8RQRkbjSGE9JhEIzlcpPP8GFF0K7dvD558E6JZ0iInlSi6eIiMSV8kgpssaNg759YcsWeO45aNky6ohERA4bavEUEZH4SsAYTzOrYWYfmdkCM5tvZreF6web2UozmxUuF8a85h4zW2Rm35hZx/i+SSl27r573wJCN92k5noRkQJQi6eIiMRVgsZ47gYGufsMMysHTDezieG2p9z9iX1iMKsP9AQaAFWB/5pZHXffk4jgpBioVy8oIPToo1CqVNTRiIgcdpR4iohIXCWiEcjdVwGrwsdbzGwhkFulma7AaHf/FVhsZouA5sAX8Y9OiiR3GDoUjj4arr0WeveOOiIRkcOautqKiEhcJXo2FTOrCTQBpoarBpjZHDN72cyODddVA5bHvGwFuSeqIr9ZvTooIDRwIEyYEHU0IiJFghJPERGJr4PIPM2sv5lNi1n653hos7LAW8BAd98M/B04BUgnaBF9MqHvTYq+996DRo1g8uSggNDIkVFHJCJSJKirrYiIxNXBjPF09+HA8FyPa3YEQdL5mru/Hb5udcz2F4Fx4dOVQI2Yl1cP14kc2IIFcNFFwdycI0dC/fpRRyQiUmSoxVNERAo9MzPgJWChu/8tZv0JMbt1A+aFjzOAnmZWysxqAbWBL5MVrxxm1q8PftavD2+/DVOnKukUEYkzJZ4iIhJXZgVf8uFs4GqgbbapU/5qZnPNbA7QBrgdwN3nA68DC4DxwC2qaCv7cYenn4YTTwySTYBu3VS1VkQkAdTVVkRE4ioRk6m4+6cHOPT7ubzmEeCRBIQjRcHq1dCnD4wfH3SvPfnkqCMSESnSlHiKiEhcJWI6FZG4ev/9IOncsiUoIHTjjfrFFRFJMCWeIiISZ/oAL4XcjBlQtaoKCImIJJHGeIqISFwlaIynyKGZOzeYIgXgnntgyhQlnSIiSaTEU0RE4uogpvEUSRx3GDoUmjWDgQOD5ykpULp01JGJiBQrSjxFRCSu1OIphcbq1XDhhXDbbdC+PUycqF84EZGIaIyniIjElakNUwqDZcugadOggNCwYXDTTUo6i5hdu3axYsUKduzYEXUoIsVO6dKlqV69OkcccUS+X6PEU0RE4kuf7SVK7kGCWaMG9O0LV18NDRpEHZUkwIoVKyhXrhw1a9bE9KWCSNK4O+vWrWPFihXUqlUr369TV1sREYkrjfGUyMydC2efDYsWBcnnkCFKOouwHTt2ULFiRSWdIklmZlSsWLHAvQ2UeIqISFxpjKckXWwBocWLg7GdUiwo6RSJxsH87SnxFBGRuLKD+E/koGUvIDRnTtDqKZIEKSkppKen07BhQzp37szGjRsL9PqyZcvmuN7MGDRoUNbzJ554gsGDBx9KqAnRp08f3nzzzf3WT5kyhRYtWpCenk69evXiHntGRgZDhgwBYPDgwTzxxBO5xpOX9PR0evbsuc+61q1bM23atKznS5YsoWHDhlnPv/zyS1q1akXdunVp0qQJ119/Pdu2bTuYt5Nl8eLFtGjRglNPPZXLL7+cnTt37rfPzp07ufbaa2nUqBFpaWlMzpwmCujUqRNpaWk0aNCAG2+8kT179gAwe/ZsWrZsSaNGjejcuTObN2/Oeg/p6emkp6eTlpbGO++8k3Ws8ePHU7duXU499dSsa32olHiKiEh8qa+tJNPjjwfzcw4bBhkZULly1BFJMXLkkUcya9Ys5s2bR4UKFRg2bFhcjluqVCnefvttfv7557gcL5O7s3fv3rgeMye9e/dm+PDhWdemR48ecT1+ly5duPvuu+NyrIULF7Jnzx4++eQTfvnll3y9ZvXq1Vx22WU89thjfPPNN8ycOZNOnTqxZcuWQ4rlD3/4A7fffjuLFi3i2GOP5aWXXtpvnxdffBGAuXPnMnHiRAYNGpT1//T1119n9uzZzJs3j7Vr1/LGG28AcP311zNkyBDmzp1Lt27dePzxxwFo2LAh06ZNY9asWYwfP54bbriB3bt3s2fPHm655Rb+85//sGDBAkaNGsWCBQsO6b2BEk8REYkz5Z2ScNu3ww8/BI8ffBBmzICbb1a/bYlUy5YtWblyJQDff/89nTp14owzzuDcc8/l66+/BoIWrcyWpz/96U8HPFZqair9+/fnqaee2m/b2rVrufTSS2nWrBnNmjXjs88+A/Zt+YMgqViyZAlLliyhbt26XHPNNTRs2JDly5dz11130bBhQxo1asSYMWMAmDx5Mq1bt6Z79+6cdtppXHnllbg7AA899BDNmjWjYcOG9O/fP2v9gaxZs4YTTjgBCFqF69evD8Avv/xC3759ad68OU2aNGHs2LEAvPLKK1xyySV06tSJ2rVr8/vf/z7rWOPHj+f0008nLS2Ndu3aZe0/YMCAXGPIb8yjRo3i6quvpkOHDlnx5GXYsGH07t2bli1bZq3r3r07VapUydfrc+LuTJo0ie7duwNB8v7uu+/ut9+CBQto27YtAMcddxzly5fPapk9+uijAdi9ezc7d+7M6g777bff0qpVKwDat2/PW2+9BcBRRx1FampQa3bHjh1Z+3/55ZeceuqpnHzyyZQsWZKePXvm+9rkRlVtRUQkrvTZ//BmZp2Ap4EU4B/uPiTb9lLAv4AzgHXA5e6+JGkBzp0LV1wBu3cHj8uUgXr1knZ6KZwe/H/zWfDj5rges37Vo3mgc/6KU+3Zs4cPP/yQ6667DoD+/fvz/PPPU7t2baZOncrNN9/MpEmTuO2227jpppu45ppr8mwdveWWW2jcuPE+SRjAbbfdxu23384555zDsmXL6NixIwsXLsz1WN999x0jRozgzDPP5K233mLWrFnMnj2bn3/+mWbNmmUlJTNnzmT+/PlUrVqVs88+m88++4xzzjmHAQMGcP/99wNw9dVXM27cODp37nzA891+++3UrVuX1q1b06lTJ3r37k3p0qV55JFHaNu2LS+//DIbN26kefPmnH/++QDMmjWLmTNnUqpUKerWrcutt95K6dKl6devHx9//DG1atVi/fr1uf+PiJHfmMeMGcPEiRP5+uuveeaZZ7jiiivyPPa8efPo3bt3nvt98803XH755Tlumzx5MuXLl896vm7dOsqXL5+VCFavXj3ri4xYaWlpZGRk0KtXL5YvX8706dNZvnw5zZs3B6Bjx458+eWXXHDBBVlJbIMGDRg7diwXX3wxb7zxBsuXL8863tSpU+nbty9Lly7l1VdfJTU1lZUrV1KjRo2sfapXr87UqVPzfL95UYuniIjElcZ4Hr7MLAUYBlwA1Ad6mVn9bLtdB2xw91OBp4DHkhJcbAGhtWvhf/8XUvX9uURr+/btpKenc/zxx7N69Wrat2/P1q1b+fzzz7nssstIT0/nhhtuYNWqVQB89tln9OrVCwiSodwcffTRXHPNNQwdOnSf9f/9738ZMGAA6enpdOnShc2bN7N169Zcj3XSSSdx5plnAvDpp5/Sq1cvUlJSqFKlCueddx5fffUVAM2bN6d69eqUKFGC9PR0lixZAsBHH31EixYtaNSoEZMmTWL+/Pm5nu/+++9n2rRpdOjQgZEjR9KpUycAJkyYwJAhQ0hPT6d169bs2LGDZcuWAdCuXTuOOeYYSpcuTf369Vm6dClTpkyhVatWWVN2VKhQIdfzxspPzNOmTaNSpUqceOKJtGvXjpkzZ2YltzkVzyloQZ26desya9asHJfYpLMg+vbtS/Xq1WnatCkDBw7krLPOIiUlJWv7Bx98wKpVq/j111+ZNGkSAC+//DLPPfccZ5xxBlu2bKFkyZJZ+7do0YL58+fz1Vdf8eijjyZ0Xlz9iy0iInGlFs/DWnNgkbv/AGBmo4GuQOzgnq4gPtdEAAATRElEQVTA4PDxm8CzZmaeV9+7Q3DUti387qXBMP8LuOgiePlljeWUfeS3ZTLeMsd4btu2jY4dOzJs2DD69OlD+fLlmTVrVo6vKUjyMnDgQE4//XSuvfbarHV79+5lypQplC5dep99U1NT9xm/GZtAlClTJl/nK1WqVNbjlJQUdu/ezY4dO7j55puZNm0aNWrUYPDgwflKTk455RRuuukm+vXrR+XKlVm3bh3uzltvvUXdunX32Xfq1Kk5nvtg5TfmUaNG8fXXX1OzZk0ANm/ezFtvvUW/fv2oWLEiGzZsyNp3/fr1VKpUCQhaEKdPn07Xrl1zjaMgLZ4VK1Zk48aN7N69m9TUVFasWEG1atX2e11qauo+XbDPOuss6tSps88+pUuXpmvXrowdO5b27dtz2mmnMWHCBCDodvvee+/td9x69epRtmxZ5s2bR7Vq1fZpFT1QLAWlFk8RERHJVA1YHvN8Rbgux33cfTewCaiY/UBm1t/MppnZtLVr1x5SULVqVaF8iT0qICSF1lFHHcXQoUN58sknOeqoo6hVq1ZWYRd3Z/bs2QCcffbZjB49GoDXXnstz+NWqFCBHj167FNkpkOHDjzzzDNZzzMT3Jo1azJjxgwAZsyYweLFi3M85rnnnsuYMWPYs2cPa9eu5eOPP87qppmTzIStUqVKbN26NV9VY997772sMZXfffcdKSkplC9fno4dO/LMM89kbZs5c2auxznzzDP5+OOPs95Lfrva5ifmvXv38vrrrzN37tyssbBjx45l1KhRQFDV9t///ndWrCNGjKBNmzZA0I13xIgR+3Q/ffvtt1mdbSqngrR4mhlt2rTJinXEiBE5Jrbbtm3LKoI0ceJEUlNTqV+/Plu3bs1qWd+9ezfvvfcep512GhCMuc18zw8//DA33ngjEIw5zkzwly5dmpWEN2vWjO+++47Fixezc+dORo8eTZcuXfJ17XOjxFNERETizt2Hu3tTd29a+RATxfsvTqPW7CkqICSFWpMmTWjcuDGjRo3itdde46WXXsqa2iKzMMvTTz/NsGHDaNSoUY7j93IyaNCgfarbDh06lGnTptG4cWPq16/P888/D8Cll17K+vXradCgAc8+++x+rWCZunXrRuPGjUlLS6Nt27b89a9/5fjjjz/g+cuXL0+/fv1o2LAhHTt2pFmzZnnG/Oqrr1K3bl3S09O5+uqree2110hJSeG+++5j165dNG7cmAYNGnDfffflepzKlSszfPhwLrnkEtLS0g7YengwMX/yySdUq1aNqlWrZq1r1aoVCxYsYNWqVfTv359y5cqRlpZGWloaW7du5c477wSgSpUqjB49mjvvvJO6detSr149PvjgA8qVK5ev+A7kscce429/+xunnnoq69atyxoznJGRkTVedc2aNZx++unUq1ePxx57jFdffRUICjd16dKFxo0bk56eznHHHZeVYI4aNYo6depw2mmnUbVq1awW9E8//ZS0tDTS09Pp1q0bzz33HJUqVSI1NZVnn32Wjh07Uq9ePXr06EGDBofeq8AS2DOmWNmxG11ISZgxs5bnvZPIIejdtEbcPs1v3L6nwP8elj8yRdlEIWBmLYHB7t4xfH4PgLs/GrPPB+E+X5hZKvATUDm3rrZNmzb12PnwROJh4cKF1FNhKZHI5PQ3aGbT3b1pTvurxVNEROJKxYUOa18Btc2slpmVBHoCGdn2yQAyyzl2ByYlcnyniIgUDSouJCIicaWekIcvd99tZgOADwimU3nZ3eeb2UPANHfPAF4CXjWzRcB6guRUREQkV0o8RUQkrpR3Ht7c/X3g/Wzr7o95vAO4LNlxiYjI4U2Jp4iIxJcyTxFJEncv8NyKInLoDmaEhcZ4iohIXGmMp4gkQ+nSpbPmhhSR5HF31q1bt99csnlRi6eIiMSVGh9EJBmqV6/OihUrONR5YkWk4EqXLk316tUL9BolniIiElfKO0UkGY444ghq1aoVdRgikk9KPEVEJL6UeYqIiEg2SjxFRCSuNGZTREREslPiKSIicaUxniIiIpKdqRKYRMHM+rv78KjjkKJLv2MihYeZrQWWHuJhKgE/xyGcokTXJGe6LvvTNdmfrsn+4nFNTnL3yjltUOIpkTCzae7eNOo4pOjS75hI0aK/6f3pmuRM12V/uib70zXZX6KviebxFBERERERkYRS4ikiIiIiIiIJpcRToqKxd5Jo+h0TKVr0N70/XZOc6brsT9dkf7om+0voNdEYTxEREREREUkotXiKiIiIiIhIQinxlKQzs05m9o2ZLTKzu6OOR4oWM3vZzNaY2byoYxGRgsvrHmFmpcxsTLh9qpnVTH6UyZWPa3KHmS0wszlm9qGZnRRFnMmU388SZnapmbmZFfnqpfm5JmbWI/xdmW9mI5MdYxTy8fdzopl9ZGYzw7+hC6OIM1ny+pxkgaHh9ZpjZqfH69xKPCWpzCwFGAZcANQHeplZ/WijkiLmFaBT1EGISMHl8x5xHbDB3U8FngIeS26UyZXPazITaOrujYE3gb8mN8rkyu9nCTMrB9wGTE1uhMmXn2tiZrWBe4Cz3b0BMDDpgSZZPn9X/gS87u5NgJ7Ac8mNMuleIffPSRcAtcOlP/D3eJ1YiackW3Ngkbv/4O47gdFA14hjkiLE3T8G1kcdh4gclPzcI7oCI8LHbwLtzMySGGOy5XlN3P0jd98WPp0CVE9yjMmW388Sfyb4YmJHMoOLSH6uST9gmLtvAHD3NUmOMQr5uS4OHB0+Pgb4MYnxJV0+Pid1Bf7lgSlAeTM7IR7nVuIpyVYNWB7zfEW4TkREJD/3iKx93H03sAmomJToolHQ++Z1wH8SGlH08rwmYffAGu7+XjIDi1B+fk/qAHXM7DMzm2JmxaF3UH6uy2DgKjNbAbwP3Jqc0AqthH1WT43HQUREREQkWmZ2FdAUOC/qWKJkZiWAvwF9Ig6lsEkl6D7ZmqBV/GMza+TuGyONKnq9gFfc/Ukzawm8amYN3X1v1IEVNWrxlGRbCdSIeV49XCciIpKfe0TWPmaWStA1bl1SootGvu6bZnY+cC/Qxd1/TVJsUcnrmpQDGgKTzWwJcCaQUcQLDOXn92QFkOHuu9x9MfAtQSJalOXnulwHvA7g7l8ApYFKSYmucErYZ3UlnpJsXwG1zayWmZUkGMSdEXFMIiJSOOTnHpEB9A4fdwcmedGelDzPa2JmTYAXCJLO4jBuL9dr4u6b3L2Su9d095oE4167uPu0aMJNivz87bxL0NqJmVUi6Hr7QzKDjEB+rssyoB2AmdUjSDzXJjXKwiUDuCasbnsmsMndV8XjwOpqK0nl7rvNbADwAZACvOzu8yMOS4oQMxtFcGOtFI7XeMDdX4o2KhHJjwPdI8zsIWCau2cALxF0hVtEUCCjZ3QRJ14+r8njQFngjbDO0jJ37xJZ0AmWz2tSrOTzmnwAdDCzBcAe4C53L8q9BfJ7XQYBL5rZ7QSFhvoU5S+zcvqcBBwB4O7PE4xzvRBYBGwDro3buYvwdRUREREREZFCQF1tRUREREREJKGUeIqIiIiIiEhCKfEUERERERGRhFLiKSIiIiIiIgmlxFNEREREREQSSomnSCFjZt3NzGOe9zGzrRHFMs7MXoni3CIiIrHMbLKZ/SnqOArKzLaaWctD3UfkcKfEUySfzOwVM/Nw2WVmP5jZE2ZWJsGnHgOcnN+dzWyJmd2ZwHhEREQOWphA/homW5nLP6KOK1Hcvay7fwFgZq3NbHdu+xQW4eedc6KOQ4qO1KgDEDnM/Be4mmCi3XOBfwBlgJtidzKzVGBPPCYgdvftwPZDPY6IiEgh8md3fzjqIIojMzvC3XdFHYcUP2rxFCmYX939J3df7u4jgdeAi81ssJnNC7vFfg/8CpQxs2PMbLiZrTGzLWb2f2bWNPaAZnaNmS01s21mNg6okm37fl1tzexCM5tqZtvNbJ2Z/T8zK21mk4GTgMczW2djXnNWeP5tZrbSzP5uZkfHbD8qbNXdamarzeyP8b54IiIiB2JmPc1stpltNrNVZvbCgXoVmVmpmPvrZjP7zswui9l+rpl9ambrzex7MxtkZnaAY9UM75nXm9m3ZrbJzMaa2XEx+1Q0s3+Z2U/hMsLMKsRs/52ZLQ7v9SvN7C8x29zMzjGzqsB/gJSYlt7e2fZJDd/7xdlifMXM/hnzvF/4uWOTmc00sw65XNc+ZrbIzO4ysxXArHD9X8LeW1vDazQw5jWzw4cTYlukw88KT4Tvdb2ZjTezUw90bpFYSjxFDs12gtZPgFrAFcBlQBpB8vkeUA24CGgCfAxMMrMTAMysBfAKMBxIB/4f8FBuJzSzTkAGMBE4A2gD/B/B3/MlwIrwGCeEC2bWCJgQvi4t3C8deDnm0E8A7YFLgXZhvK0KeD1EREQO1iaC+2h5gl5F5wIHGtPZG2gG1HP3o4G2wHwAM6sPvA88DlQG/gcYQNBjKTfXENz3agB7gX/HbHsNOBaoFy6VgFfD89UBhgAXuXs5oAHB/XYf7v4jcAFBj6iy4TIi2z67w+P2yVxnZmWB7oT3bDPrB/wBuDKM6V7g7TwSwJpAVaA2wXUDWACcA5QD+gGPmlnHMI60cJ8OYZzXh89fBE4DzgSOB6YC48ws87OQyAGpq63IQTKz5gQ3yA/DVSWBq919dbi9LUFyVznsLgtwn5l1Jrj5/RW4DfjQ3R8Jt39rZs2A63I59X3Am+4eezOeE/7cZmZ7gC3u/lPM9ruAMe7+ZEz8NwEzw290t4Xn7OvuH4TbryVIYkVEROLtXtu3HkEnd/9PzPNFZvYcQTKYk51AWaC+mX3h7stjtt0MvOHuY8PnX5vZs+Gx/pVLTA9m3jvN7C7gu7CVEqAjUMfdN4Tb7wiPewKwGzCggZktdfeNwJTc336u/gnMNrPj3H0N0AP40d0/CbffBjzk7pmtku+b2UdAT+BA3Zd3AXe7+6+ZK9w9NrGeZGbvEXzx/EFOBzCzSgSfe06K+azzIDAQaAF8WvC3KsWJWjxFCqZT2OVkB/AFQQvmreG2FZn/EIfOAI4C1sZ0qdkKNAROCfepFx4nVl7FBZrwW7KbX2cAV2WL47Nw2ynhUjL23O6+FZhbwPOIiIjkxyPuXj5mmWJm7c3sEzNba2abgccIWixz8m+COgtPAevMLLbFrxbQy8w2Zi7AA4S9gHKxJIfH1QlaQAEWx2z/PvxZw91/IGh97Af8GHbxPWDX17y4+0JgBnBVuOpagmQ0Uy1gWLb314agh9WBrIpNOiGre/BcM9sQHqMzB77emecFmBNz3vUEPb9qHPhlIgG1eIoUzMdAf4JvDn/MHJwfDhv5Jdu+JYDVBF2FstucwBhzUoLfbtDZrQTqJDccERGR35hZSeBd4PfAy+6+3cwGADlWaQ+7pD4GPGZm5YFnCbqitgKWhse4pYBh1OS3hLJm+HNFtu2LwseZ1eaXh/G8TdDdtSRwIzDWzCq6+7Zs59ibz1j+CdxiZhkE3Vp7xmxbCjzg7m/k81j7ndfMzia4fu2Aqe6+x8zeJGi5zZS9QOLS8Gdtd19bgHOLAGrxFCmobe6+yN2X5qMi3AyCQkF7w9fELmvCfRYS3FBiZX+e3UyCG8WB7ARScoilQQ5xLAq7AX9PkExnnduCgg4N84hFREQkHkoCpYANYdJZn2BcZo7MrK2ZnRGOLdxO8OXvnnDzc0BPM+tsZkeEBXvqm9l5ecRwn5lVsaDw3mPAf939x3Bs5gTgSTMrb2bHAk8C/3H3VWZW18w6mdlRBPfSTQRJW05J5k8ExYVq5bAt1mjgVGAoMNHdV8ZsewoYbGbpFjgyLEx0Wh7HjHU0wfVaC7iZ/Q/B+NPssdbOfBJ+dhkJPGdm1QDC69EtHIcqkislniKJ81+C7qxjzewCM6tlZi3N7EEzy2wFHQqcb2b3mFntsGBAtzyO+whwmZk9HN5IG5jZ7eEND4LuQeeaWbVwPAYEN9DmZva8mTUxs1PN7CIzewGyutW+RPDNcXsza0DwzXH2BFZERCTuwvvQTcBfw+EgwwiSnAOpQlCEZwOwiqCie//wWPMIivoNDLetISjkl1s3Ugi6735C0IpZkn2LEV0FbAG+Ab4GNvLb+NOSwP3huTYCvwMudfcdObzPb4G/A1+G3VVzLHjk7puAdwiSwZezbXuRoE7EP8P3v4yg/kNBCvx8QDDe9UvgZ4LiRe9k2+de4KGwK+4L4bp+BNdgspltIRiScxn7t46K7MfiMM2gSLFgZq8Aldz9ohy2DQa6u3vDbOvLEQz0vxQ4jqDr7WfAve7+fbjPtcCDBDfEyQSVcJ9xdwu39wGedfeyMcftQjBepSHBjfBzoIe77zCzM4EXgLpAqZjjNA1jOYsgofwBeMfd7w+3lyG4GV5CUGzoGYJiAT+7e5+Du2oiIiKFm5nVJBi/WcPdVVRPJEGUeIqIiIhIsaXEUyQ51NVWREREREREEkotniIiIiIiIpJQavEUERERERGRhFLiKSIiIiIiIgmlxFNEREREREQSSomniIiIiIiIJJQSTxEREREREUkoJZ4iIiIiIiKSUEo8RUREREREJKH+P6G9i04bSBL1AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 964
        },
        "id": "mR5v5bIvZBI4",
        "outputId": "6a1f7e87-2581-4332-b8e7-0a450178d2e1"
      },
      "source": [
        "plt.figure(figsize=(10,10))\r\n",
        "\r\n",
        "rocauc_plot(rf_us, \"Random Forest\", y_test, X_test)\r\n",
        "rocauc_plot(knn_no, \"KNN\", y_test, X_test)\r\n",
        "rocauc_plot(nn, \"Red Neuronal sencilla\", y_test, X_test)\r\n",
        "\r\n",
        "rocauc_plot(lr_no, \"Regresión logística\", y_test, X_test)"
      ],
      "execution_count": 313,
      "outputs": [
        {
          "output_type": "error",
          "ename": "IndexError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-313-8240bf82282c>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mrocauc_plot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrf_us\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"Random Forest\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mrocauc_plot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mknn_no\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"KNN\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0mrocauc_plot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"Red Neuronal sencilla\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0mrocauc_plot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlr_no\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"Regresión logística\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<ipython-input-312-938c714ecb78>\u001b[0m in \u001b[0;36mrocauc_plot\u001b[0;34m(model, model_name, y_test, X_test)\u001b[0m\n\u001b[1;32m     99\u001b[0m   \u001b[0;31m# para que funcione con estimadores de sklearn y keras.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    100\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 101\u001b[0;31m   \u001b[0mauc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mroc_auc_score\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict_proba\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    102\u001b[0m   \u001b[0mfpr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtpr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mthresholds\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mroc_curve\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict_proba\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    103\u001b[0m   \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfpr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtpr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmodel_name\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;34m\" AUC = {:.5f}\"\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mauc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mIndexError\u001b[0m: index 1 is out of bounds for axis 1 with size 1"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VXnbRl9BDUyV"
      },
      "source": [
        "# Tuneo de Hiper Parámetros y selección de mejor modelo\r\n",
        "\r\n",
        "Para esto, se define un pipeline general para todos los modelos: \r\n",
        "\r\n",
        "1. Se aplica un StandardScaler, esto es, se estandarizan los datos. \r\n",
        "2. Se realiza un Grid Search con Cross Validation, usando:\r\n",
        "  1. Una lista de parámetros, con sus posibles valores\r\n",
        "  2. 10-fold Cross Validation\r\n",
        "  3. Selección del mejor modelo por el f1 score.\r\n",
        "\r\n",
        "La estandarización de datos se realiza para mejorar el desempeño de los modelo. Dada la naturaleza original no balanceada de los datos, se usa el f1 score y la curva ROC-AUC en lugar del Acurracy cómo medida de mejor desempeño.\r\n",
        "\r\n",
        "\r\n",
        "El mejor modelo entonces se guarda, y se simula su carga y evaluación sobre datos nunca vistos (con los datos de prueba). Adicionalmente, se presenta el gráfico de Feature Importance del modelo seleccionado."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SN6ZCehkI1RM"
      },
      "source": [
        "X,y=get_X_y(df, \"employee_left\")\r\n",
        "X_train, X_test, y_train, y_test =data_preprocessing_up_or_down_sample(X, y , \"up\",test_size=0.2)"
      ],
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JhflliYY1pM0"
      },
      "source": [
        "def pipeline_classifier(X,y,model, param_grid):\r\n",
        "  \"\"\"\r\n",
        "  Creates a general Pipeline for sklearn classifiers.\r\n",
        "  Applies GridSearchCV to optimize hyper parameters of the model.\r\n",
        "\r\n",
        "  Parameters\r\n",
        "  ----------\r\n",
        "  X : {array-like, sparse matrix} of shape (n_samples, n_features)\r\n",
        "      Input values\r\n",
        "  y : {array-like, sparse matrix} of shape (n_samples)\r\n",
        "      Target values\r\n",
        "  model : sklearn.estimator object\r\n",
        "      Instance of an estimator form sklearn.\r\n",
        "  param_grid: dict or list of dictionaries\r\n",
        "      Dictionary with parameters names (str) as keys and lists of parameter \r\n",
        "      settings to try as values, or a list of such dictionaries, in which case \r\n",
        "      the grids spanned by each dictionary in the list are explored. \r\n",
        "      This enables searching over any sequence of parameter settings.\r\n",
        "    \r\n",
        "  Returns\r\n",
        "  -------\r\n",
        "  model : sklearn.estimator instance\r\n",
        "      Fitted classifier or a fitted Pipeline in which the last estimator \r\n",
        "      is a classifier.\r\n",
        "  \"\"\"\r\n",
        "  pipe = make_pipeline(StandardScaler(), model)\r\n",
        "  clf= GridSearchCV(pipe,\r\n",
        "                    param_grid=param_grid,\r\n",
        "                    cv=10,\r\n",
        "                    refit=True,\r\n",
        "                    scoring=\"f1\",\r\n",
        "                    n_jobs=-1)\r\n",
        "  clf.fit(X,y)\r\n",
        "  return clf\r\n"
      ],
      "execution_count": 223,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hi0sanNKKQ1p"
      },
      "source": [
        "rf_param_grid = {\r\n",
        "        \"randomforestclassifier__min_samples_leaf\": [1, 3, 5],\r\n",
        "        \"randomforestclassifier__n_estimators\": [50,100, 250],\r\n",
        "        \"randomforestclassifier__criterion\": [\"gini\", \"entropy\"]}\r\n",
        "rf_gs=pipeline_classifier(X_train,y_train,\r\n",
        "                          RandomForestClassifier(class_weight=\"balanced\")\\\r\n",
        "                          ,rf_param_grid )"
      ],
      "execution_count": 224,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 513
        },
        "id": "4d7PaBws-qmB",
        "outputId": "6292cde2-c8e7-444d-8744-320e6fb182d9"
      },
      "source": [
        "plot_roc_conf_matrix(y_test,X_test, rf_gs, \"Random Forest (GS)\")"
      ],
      "execution_count": 225,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      1.00      0.99      2275\n",
            "           1       0.99      0.96      0.98       725\n",
            "\n",
            "    accuracy                           0.99      3000\n",
            "   macro avg       0.99      0.98      0.99      3000\n",
            "weighted avg       0.99      0.99      0.99      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TP7F0L8fJlrP"
      },
      "source": [
        "lr_param_grid = {\"logisticregression__penalty\": [\"l1\", \"l2\"],\r\n",
        "                   \"logisticregression__C\":  np.arange(0.5, 20, 0.5),\r\n",
        "                   \"logisticregression__fit_intercept\": [True, False]\r\n",
        "                  }\r\n",
        "lr_gs=pipeline_classifier(X_train,y_train,\r\n",
        "                         LogisticRegression(class_weight=\"balanced\"), lr_param_grid )"
      ],
      "execution_count": 33,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ykfeWO0FF-k2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 514
        },
        "outputId": "d3d7ae73-6f34-4b35-f921-31095bfeafae"
      },
      "source": [
        "plot_roc_conf_matrix(y_test,X_test, lr_gs, \"Regresión Logística (GS)\")\r\n"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.92      0.74      0.82      2275\n",
            "           1       0.50      0.80      0.61       725\n",
            "\n",
            "    accuracy                           0.76      3000\n",
            "   macro avg       0.71      0.77      0.72      3000\n",
            "weighted avg       0.82      0.76      0.77      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k_7PgGeqOjDM"
      },
      "source": [
        "knn_param_grid = {\"kneighborsclassifier__n_neighbors\": range(1, 20),\r\n",
        "                  \"kneighborsclassifier__weights\": [\"uniform\", \"distance\"]}\r\n",
        "knn_gs=pipeline_classifier(X_train,y_train,\r\n",
        "                         KNeighborsClassifier(), knn_param_grid )"
      ],
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sHYFXgkKTDIT",
        "outputId": "141e67c1-188f-40bf-df73-b3269fd4d14f"
      },
      "source": [
        "knn_gs.best_params_"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'kneighborsclassifier__n_neighbors': 1,\n",
              " 'kneighborsclassifier__weights': 'uniform'}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9AzUn3hiF690",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 513
        },
        "outputId": "047a54c4-ccfe-402c-ccad-49990daa0e91"
      },
      "source": [
        "plot_roc_conf_matrix(y_test,X_test, knn_gs, \"KNN (GS)\")\r\n"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      0.97      0.98      2275\n",
            "           1       0.91      0.96      0.93       725\n",
            "\n",
            "    accuracy                           0.97      3000\n",
            "   macro avg       0.95      0.96      0.95      3000\n",
            "weighted avg       0.97      0.97      0.97      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3YAAAFXCAYAAADuwtw3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd5xU1fnH8c+zdBBbREUR0Whs2IklaqKxYW+JERMVS4iFWGOsiS0ao6LR2JMoUSPGxESJgigafxpjwy4IiogCioAFlYWtz++Pc4Ydhpnd2TJz9+5+37zua2bOLXNml713nnvOeY65OyIiIiIiIpJeFUlXQERERERERFpHgZ2IiIiIiEjKKbATERERERFJOQV2IiIiIiIiKafATkREREREJOW6Jl0BEZE0M7P1gGOA9939L0nXR0RERDonBXYiIi1kZl2Be4HNgO8kXB0RERHpxNpVV0wz87gMKrD+0rj+czPbLpbtmrXf840ce3Tc5icFyt3MTmqibrUt+ExdzOwtM3vdzKzANv3jZ3vOzOabWY2ZfWFmr5jZTWa2ayPH383MHjCzOWZWHX8208zs72Y20sxWytrWzOzNuJTtd29mM7N+xsc2st26Zlafte2ActWxkToNinV5qjO9d3tgZivE//+zzaw2/ixOL9N7P9XYuSjLr4HtgCPd/a2SV0xEOj0z62Vmp5rZ42b2cbz2LzSzl83sGjPbJOk6loKZjTezuWbWu4X7fzue13/Zgn2HZ303ySy1ZvaJmT1iZvsXcYzeZnaWmT1rZgvMrCpe3/5uZvsUWY9tzew2M5tqZl/GY8wys4fM7Cgz697czyYdS7sK7BpjZlcAvwI+A3Z39xfzbLa9me3Xirc538x6tGL/fI4n3M2/2PPMBm9mPwXeI3y2jYFXgPuBZ4CVgJOB/5jZI3n2/TXwJHAosBB4GHgMWBzL/gAsPcHH978UGAwMb6sP2Ew/aWJd3uC3peLJd2ZbHlPK5krC///FwN+AvwBTEq1RFjPbGTgfOMfdH066PiLS8cWb2u8C1wM7A9OAfwBPA2sAZwFvmdkxiVWyBMxsT2Ao8Dt3r2zJMdz9JWAccJ6ZrdrCqnxCuBb9hfBznwfsC/zbzH5VaCcz2xKYClwDbA68BPwTmAMcBowzs/sLfQc1s65mdgswCRhB6HH3JPAv4MNYh7toR9dISYi7t5sF8LgMyim/KpbPB7bMWbdrXFcZH18ucOzRcf1PCpRn9j+1kbrVNvPzdAc+JvyhWZ71P4vHXQT8FOiaZ5vBwN3Auznl2wL1QDVwcJ791gR+AWycU27AO8DsfO9Xot/rzPg5XwHqgLULbDeVcNKcHbcf0Eb/p2a2Yv9uhIB7YDl+VjnvPSjW/6lyv3d7WIBZ8e9yhQTee2D8vXdrZJtjgZOT/jlp0aKlcyyEgGBRvC6MAvrmrDdgD+A1ws3kxOvchp/9ZcKN/V6tPM5O8ef3u2buN7zQ9Rg4La6rAdbNs37dWHcHbgb65KwfDLwZ1z9Q4P3vjuunA9/Ns35V4LdAddK/Ky3JLu2+xc7MrgXOJtwV+b67v15g0+eAV4FtzOyQFrzVzYQ/mvPMrFeLKru8wwgB1t3uvkxrnZkNJNxxqwcOcPc/uvtyXT3d/S13PwoYlrPqUMJJ/H53fzDPfnPd/Rp3n5pT7sA9wNrAgS3/aC1yN6GV+MjcFWb2bWAj4D6g2V1eS8Xda9x9qrt/mHRdOqG1gXnu/nW539jdP4y/95pGtrnT3W8uZ71EpHOKQznuAXoDv3H3s9z9q+xtPJgIbA8s18snrcxsR2Ab4O/uvrg1x3L3Z4EZwHFt1W3R3a8n3JjuSgisc90GrAKMdveT3X1Rzv5vAbsTvuceamY/yl5vZocSejTNB3Zx96fz1OEzdz8P+H4bfCRJsXYd2JnZ9cAZwFxgN3d/s5HNnTDeBeCSQuPZGvEaoVl8TeCU5ta1gBHx8d48634O9AD+5u5PNnUgd5+UU9QvPs5rQb3+Gh9/2oJ9W+PfhC6j+bpjHhUf7y60s5l918z+YGG84mdmtsTMppvZDWa2Zs62w80sE0yvm9Mv/qms7TLj/8zMfh6PXWlmr8X1y41zyyprbJlZ7A/FzNYxs7stjK+sjOMkGuuymtlvaOzbPz/2s59hZtea2TeKfe+sY33DzC63MP5yUey7/6aZXWVm/XO27Wph/OYkM/s6Li+Z2SkWkonkHnvpeDUzO8jM/hff4wsze9DMNsrZfmb83RnL/u5mxvWZcbWjC3yWi+P64Tnlq8XPONnMvoqfcbqZ/c3M9sjZtuAYOzPbxMzuMbOPLIxt+Si+3jjPtkvramarWBgzOCf+vqZZGG/Rpt2PRaRD2hvYgtB179LGNnT3Kg/dDoFlcgnsmm/7rHNudln2uWtNM/uTNYx3Pt3Mxsb1BxQ4psXj1pvZurGswsyGmdm98fyXuX68Yma/tMKBVua71F/zrTSzHSzkGphp4XvBvHjM68xsjTy73AusBrSkEaCQyfFx9Zy6bUb43S0h9KLKy93nAZfFl2flrM6MCbzI3T9urBLu/t9iKywdU3sN7MzMbgJOBT4CdnX3JvsNexjn8iKhu8LhLXjfiwgtaOeY2Qot2H8pM1sR2AWY4e4f5NkkM1B2TAvfYlZ8PMzM+jW6ZQ53fy/u/31ru9bJYiwh9EnfwswGZwpjMHAEMDVPAJvtWkIwWgX8B5hA6Cr5c+AlWza4m07oAw+h68pfspZH8xz7VkLXlnnAWMIdvUK+zjle9pJJ4FPXyP5LWUiV/yIh2P0SeCjW9y6WP7ln73clMJ5wd3BarHMt4UbICwUuZoWOtQnhxsb5hBsGE4CJhMDqbMLd38y2XWId/wB8C3g8brsRcCPwTyucmOckws2TOsLd5AXAQcAzZpZ9MfwH+X93/yj2M+X5jCsQfjfnE8auTiT8P5hPaLk+osjj7E4Y4/Bjwg2nBwjdh38MTCr0xQlYmdCr4DDgf4TxMOsSxltc1JLPJCKdSiZ/wN8b60lQAv0I48H2I5zDxhO6yN8T1/+4wH47Ec5x/836DtSbEFTtTbjWPgw8C6wH/A54uMD1Yz/C94fncldYSFryLKEX0zzgwVjfXsDpwIZ5jvefrOO2lb7xMfdm+77xcYK7f9rEMe4lNFIMMbPVINyQJCToclr+fVE6k6T7gmYvNIyx+3d8nAVs2MQ+u8ZtJ8bXe8XXbwMVWduNpvExdj+JrzN/WOfnqVvRY+wIg3wdGJNnXTdCAOkUGG9WxPHXp2Fc4ML4OU4Atga6FLH/P+K+u5Xh9zozvtcA4Hvx+ZVZ6/ePZRfmbp9znP2AlXPKugAXx+3/VOD/1Mwi6jYf2CzP+kEUOc6N0Nr7Yfzd/qDIn834ePw/kzXmMf5MavO9N/DDWD45+++DEIhdEtfdV+T7dyV0IXHgBqBHzvrNgG9mvT4rbvsWsEZWef+s45yec4ynYvli4Hs5fwf/iusuLvZ3R8Pf/OgCnynz/2F4VtnwWDY29++DEHRtU6DOg7LK+hCCOQdOydn+jFj+EdA7T12dENT2zFq3ffwdf00C4wi1aNGSnoWQUG257zBF7js67rtrgfUziSM1ssoKnrvi+p6E7x6V5Iz1i+tvifuOyCrrDhxMzthlYEXCzb5839E2juXPFah75lz9wzzrNsu+TuW8Xx3wQTN+hplryFN51q0KfB7P5+vmrLuHrO83RbzPjLj97vH17vH19KT/D2pJx9JeW+wyaWNHuvu7zdnR3R8D/ks4GSw3lqsIlxD+4M+KrW4ttUV8nJpn3ao0ZH9ckLvSzNaO3R9yl6V3ntx9BnAAIfhdkTBB8h8JCUoWmNnNltOFLkemXls161O13tPAB8CRWV3QjiKcuPJ2s8hw90fc/Yucsjp3v5jQPeXgVtTrd+4+uenN8jOznoQ7hesQuks02bpkZusTbgB8CZzpWWMsPbQ+FzrGBfHxyOy/D3d3QlDzGvCDzB2/JhxKaG17ATjN3auyV7r7ZA8tvBmnxscz3f2TrO0+pqGF8bQC73W9u/9f1j41wOXx5feKqGtrZFq1n3D3ZVpT3f0Ld3+liGMcTsg695y735RzjOsILXn9CYF3rq+An7n7kqx9XiC0GvYhJEMSESkkcz6fX+b3rQJ+nn3uAoivHyC0jB2avc7MuhHOg9XA37P2qXb3Bz2nxdHdvyS0rsHy3SMb+y4FDef2x3JXxOvXJ3nKvyQkthtoZqsUOG6TzKynmQ0h3KBcmZAhObeHVmZoRLG/t0yL32o5j+X+vUtKtdfALtNH+DYz+1YL9s+knL3I8oz5aYy7TyPcYVmVcBe+pTJdyz5rwb6rEAK13GWZ7nXu/gSwAeGkeishqKslnGBOAl7LHb+UJVOvZnXjbK0YfNxLCIC+Z2GevQOBZ939/ab2tzDn34jYd/7PmaCX0PrzDWt5CuOxLdwv4w5CC8wYd7+sqY2jnePjo+6+MM/65bpdxC6LWxLu3i2XSCj+fJ8ltGQWEyxkxpbdFfctyELCn4HA/HgDJfe9HwE+BQZZ/jkIx+cpmxYf1yqirq3xcnz8pZkd3sKu1rvEx0I3IO7K2W6Z93f3fBfmcn1+EZGWeMXd5xRYV6g75lBCQDPO3T/P3cnMNjezX5jZjWZ2Z7yGXxhX53adbOq7VObcfpeZDWnGmOWWfgf6XhxbmOmF8hLhWv4Tdx/VzGOJtLlmBT1ldBzwe0Lf5IlmtkueuyAFuftTZvYEoQn7aMKX7ua4hNDad4aZ3ZDvxFSEzMTgX+VZl0l7a4S7McucND1kSFp6crKQuCNvi4a7VxPuFv0rbrsyYbzQFYQT4o3Annl2/TI+rlzMh7H8iSoe9DwZOYtwN3AeYVzZNwldOu5pdI9Qh9MIU180lsmqLy0Lpluc9dLC3DXDCGPljmvGrpkv84X+b8/MUzYoPm6QO9g9j2Ja7NaJj+8UsW1T9YVQ528QMlrOzlk3K3djd/8qXofbev7I3Pd50syuJrQq/g2oNbM3gCcIXTqLmfsn8/lnFlifuTGxdp51y332KHN+KOnnF5HUy/TuKevNWBq/Nj5FOM9/38zWdPe5sTwT6C1zXY8teXfQ+Hy2fXNeN/ZdCsJ3ic0IN4gPBL4ws+cIXTv/4oWzKjfrO1CWT2gYp78S8B3Cd62bzewNXz7JX2ZcXbG/t0wguyDnsdy/d0mp9tpiV0NIMvAk4YvnE010K8wn02r3q3gyKVpsObqT8EdbMItREzJdBpfrzhm7IbwdX27TwuPnFbuV3UrDBOS7mVnvPJtmTpZf5FmXT74WxBZ143T3twmtiz8gTOBeTZiUvSAz24EQ7C8hBE/rEeazMXc3GgZVtyjDYG43k2KZ2Q8INwJmAwe19DjNkPmbnUvhBC6ZpeibIWVSX6b3yXtec/dfErqd/oKQ9GUDQnKYN83sxBLXqVyfXUQ6ptfiYym6bTf2XbDgNc3d6wk9S7oQp2Qys76EAGshITlKtjMJQd1kwpCbNYHu8RqeubmVew0v+F0q1mEOIbnInsB1hDFqexNuak+1PJmNo+Z+B8qY6u7D43IIIUHMg7F+d+dJ/pLpWTOkqQPHHkeZ+r6e9ejA+vHGvUij2mtgl/mifSAhg9w3CS13xbRAZPbPZG8aRAgemus3hL7lpzbnfbNk+kkX6ho4Lj7mzk/XVjJTKHQh/x2pTL/yovptZwKonOXiVtTvbsKJdUcKdNfIkel3f4GH+cNm5gRRG7SiLi1iZtsQAqhK4MCsu5XFyqQtXrfA+nzlmVawBVkXl0JLMWmPMy1J+TKH5fqoifpCw0WpUNedtlAdHwt1p1ynQDnuPt3dR7n7voQWzcy54ffxC0ljMp9/UIH1mfJSfnYR6Zwy89L9sLlDTGjknBmP1dwb59lyu2MeQhh394/cMds0jMU7Io6Z/yRrvF2ha3hT36UyY+0nuvuZ7r4todfEA/Hx8gK7Nes7UCPvvYRwI30eYZhEbmtk5rve3kUMFTmCENi+5O4L4vEXEHoDGaX7vigdSLsN7AA8TOK4L6F1Z1PgsTgmq1iZee0uoJldndx9FiEZyQrAOc3ZN8rcbVlubqvoD4TA8UdmtltzD15EP/LMSbKaPAlagE3i42t51pXDGMKJ8FNCxq6mZE7Cy3VpiynoC3VTqKEEXY5jC/JYwgXsKHd/tQWHyQReQwsk6lkuBb+7zyYMIt+0heNPc02Mj0c39X/KwyTtHwL9zGyv3PVmtg+hG+bMWM9SyQTEy40fjfMg7VrMQTxMPn8H4W+1B03fHHgmPhZK731UznYiIm1lAvAGIVj5dWMbmln3mNQjo+A5E9iNVlwj3f0N4E1g2zimP283zKjgdZzCQUtT36Xy1WkuDXP9Dc5dH6+3awEftnCoTe77LQR+G1+en91qF5OyPUYYcnJ1oWPEaasyv9fcsXpXxceLc6Z1ynecnZpRdemA2nVgB0v/YPYiNN1vDYwzsz5F7juJMOfWAELrX3NdQRgce3IL9v0fIZHJdgXq9iEhC1QF8G8z+2m+u3AxYUW+xAqXmdnVZvbNPPusDdwWX46N4/BybUcI+p7Ps67k4p26Ndx9NXd/qIhdMkkmfmpZk5jGbha3NLLfR8AabdmFwcLcfw8RLrAXuvu/WnKcmG3yMUIXjlFxjrjMe+xL/uyKECYxrQAeMLOt89TvG2ZW7OTz/ySMr9sBuNZyJog1s81i9s6MP8THUdlzz8WLTeZidH2R790isav0B8BgMzssqw7dCd11B+XuY2YHm9mOeco3JLRW1tPQIlfI/YTxFTua2TLnBDM7lfA39TFZWeBERNpCTG51FKGHyK/M7Jp8vQwszKX5HA3ZxSGMhQM4Oee8vQEN5/TWyARxZxJyG8wG/i/PdpnreO75cw8KzNvq7u8Qhh5slXt9ivueWWCoTmb+uHxB5LcJLWD56thStxHO/xux/LX7Z4Qun8eZ2R9yh8eY2aaEm6xrAP9y979lr3f3fxJ+xqsT5n1dLkGXma1sZpfRMEefdFbeDuZcyCw0zJkyKM+6NQlfQJ2Q8KBnLN+VrHns8uy3JQ1zxuWbI2V0vvKs9aOy9i16Hru47+OFPk/WNiMIwaMT5kGZQMi69zDhTlVdXPc/oH/Wfr/Pqtc0QvKUMYTWgupY/i555skjtEw4ML5Mv9eZ5JmXrjnbE7phzInlHxC+ZI+PP7tnCJkgl/tZE+Zmc0K/+3uAPwFn575XI3UZRM7cNTRMz1AZ///kW64p8rOuT8PcaNPj7/D/4v/ZG3PfO2u/y+O6OkJWsPsJAUUmM+oXzfj9DCZckDw+PkAI+N6MZQdnbduFhvmGFtKQuOdLCs8T91S+303O3/3MYsvjuqOzPv9TsQ6zCIHXnSw/j13m7+Ujwt/WPYS/z6pYfk0xdSZ8acnMH/kyIcPrK/H1InLmiaIFc+5p0aJFS6GFkH15dtY16EnCd4axhB4VHq8BR2XtY4TvTU7oevggIQCoBO6j8XnsRhdRpwE0fFdxwvRB+bbbmdCLxuN5cwzhu40DVzZyLfhjXPe9POu+iO/9GuE6eB+hZTNzTt4uzz6/iet/1Iyf+3AKXI+ztjk1bvM6YDnrtorXKCdcL8fF68fzNHxH/Ts58wVm7d+VkP088zN+l3CdHkPo/ZP53jct6f+jWpJdEq/AMpVpJLCL6wfS8KX/YUKK+8zJJ29gF/e7P+vYzQ3s+hEmEG5JYJeZSPr8JrbrT+g28Dyh22RNPFm9TrgLtNwk4oTxQT8hjFV7I2u/T+Mf+dlAnwLv9+tYr0PL9HvN/M5aHNjF8rUIY9o+JAzofieeoHtS+It4H8IdyQ9puKA8lftejdRlUJ59hmf9fyq0zGzGz2ddwoV5ASFQfZWQnGa5987Z77vx//YcGrrcvh4/73eb+TtandBNZFqswxfx/9WVwJo523YFfk4IbBbF5WVgJFmTrGdtn/d3k7W+2YFdXH8k4WK+JH72MYRzxMUsH9htFT/f84RAuir+n3iUMD6y6DoTuoX/lRAE18THvwKb5Nl2VxTYadGipQ0XoDchiJhIuJlVQ7jRNgn4HbBhnn36xmvDR/H89w5wLuFm3UxaEdjF7Z/Muv5t3sh23yb0VFlA+G71EnBsXFfoWvDtuO72POuOItyoezv+DL6Oz28GvlmgDu8RAtzuzfiZD6fpwK4nDUF3vutKb0Lyrv8RsndXEa7f/wD2KbIeQwjfC6cRMoVWEQLGBwndWZe7BmvpXIu5O1IaMRvnTMLdmU29Hfyw4ziqaYQTzCDPmhRbREREpL0xsxcJvY3WdvfFrTjOzoRePle5e0vyJ4i0a+1+jF2aecj2dBFh0O8hTWxeLj8gjCn6tYI6ERERSYHzCclXTmrlcc4j9Ej5XatrJNIOqcWuxGJCjMxYua2SbLWLrXWvE/rbb+lhDhoRERGRds3MxhOS6K3v7pUt2H8IoevnOe5+VVPbi6SRAjsREREREZGUU1dMERERERGRlGvziZtLpdfWI9W0KCWz4IW2mMpHpLA+3RufAL45WnI+XPzqjW32/tL+rLbaaj5o0KCkqyEiIiX28ssvL3D3fvnWpSawExERkfwGDRrEpEmTkq6GiIiUmJl9UGidAjsRkbQx9aIXERGRZSmwExFJm7br1SkiIiIdhAI7EZG0UYudiIiI5FBgJyKSNmqxExERkRwK7ERE0kYtdiIiIpJDgZ2ISNqoxU5ERERyKLATEUkbtdiJiIhIDn07EBFJG7PmL9IumNkdZjbPzN4qsN7M7AYzm25mb5jZNuWuo4iIpJMCOxGRtLGK5i/SXowGhjayfh9gw7iMAG4pQ51ERKQDUFdMEZG0UQtcarn702Y2qJFNDgLucncHnjezlc2sv7t/XJYKiohIUdyd6rp6qmrrqaqpZ0lNXXheW8eSmvAY1oVHX/AZ/Qb1Z6cNVitZnRTYiYikjVrgOrK1gVlZr2fHsuUCOzMbQWjVY+DAgWWpnIhIe1NbV8+SrACqIcDKel5TV3CbqnwBWU09S+LjstsvG7S5N12/HrXVnPvUnRzw9tOMuupv7LTB3iX7WSiwExFJG7XYCeDutwO3AwwZMqSIrxciIqVRX++NBkBLapoOkhqCsOygKuuxQNBWV9+601/PbhX06Npl6WOPrhX07BYee3SrYKVe3eiRZ5semW2W2b4LPbPWrTRzOuuNPJ6eb0/m65+dzLlHfqeNfuL5KbATEUkbtdh1ZHOAdbJeD4hlIiKNcvfWt0I1uU1OsBW3r6lrXXDVvUvF0uApBErxebcKenbtwip9ui8bQOULxLpVLLdNjzzBWva23btUYKW8WbrkE1i8CMaNY4V99ind+0QK7ERE0kaBXUc2FhhpZvcB2wMLNb5OJD3cnZo6b7IVKt8YrOW6DjbRUrX0OFnrWqNrhS3X+pT9uGKvbsu3UGUHWN0qlmmtKrzNskFY9y4VVFR0oJ4o8+bBvffC6afDppvCO+9At25leWsFdiIiadORLoCdjJmNAXYFVjOz2cBFQDcAd78VGAfsC0wHKoFjk6mpSLrV1XvhFqnGgqQ8rVZ5W78aCdpa0zPQjKWBT24A1KNrBb26dWGV3t2W7w64zPYFuhUu01qV062wawVdu+imYauNHw/Dh8PChbDffrDhhmUL6kCBnYhI+qjFLrXcfVgT6x04pUzVESmpzLir5rReNWub7KArJzlGbSvHXWWCnR7d8gdJK/TounyL1HItVdnjsBpvtcps062LlbZroJTGkiVwzjlwww0weDA88UQI6spMgZ2IiIhIB5VJyb4kJxHFkmJao1qU7KLhmNV1resa2K2LLdMKlTsGa+Ve3ejRt8cyiS6abLVqZJvsZBgKrqRo7jB0KPzf/8Gpp8Lvfgc9eyZSFQV2IiJpoy8cIqni7tTmZg0sZgxVzmNzuxBmtmmNCoOe3bos350vtjCt0KMr3+jTeCtUwQCrkW26d62gi7qdS3vmHpaKCjjzzNBiV4YEKY1RYCcikjbqiinSInX13obZAXODr8a7DrayZ2CjKdmXtl61MCV73qAtbqNxVyJ5zJsHxx4Le+wBZ5wBBx6YdI0ABXYiIumjFjtJsaUp2RsdO9WcZBf5Jx+uzrN9q1OyZ8ZNLQ2elg2kVu3TvelWq5yWrbxjsMqdkl1EipedIGX//ZOuzTIU2ImIpE0JWuzMbB3gLmANwIHb3f16M1sV+BswCJgJHO7un1v4lnk9IYNjJTDc3V+JxzoGuDAe+jfu/pc2r7Ak5vEpn/DG7C+aneQi87y6DVKy57Ywdc9qYVqxVzf6ZY+7yhtgFU6KUSgI63Ap2UWkefIlSBk8OOlaLUOBnYhI2pTmzn0tcJa7v2JmfYGXzexxYDjwhLtfaWbnAucC5wD7ABvGZXvgFmD7GAheBAwhBIgvm9lYd/+8FJWW8jv7H6+zcHENPRtpherToyur9ik8QXBLk1x076KugSKSkFdegRtvTDxBSmMU2ImIpE0JWuziJNgfx+dfmdnbwNrAQYR51wD+AjxFCOwOAu6K6fmfN7OVzax/3PZxd/8MIAaHQ4ExbV5pScTXS2o58Xvf5JyhGyddFRGR0nKH556D73wnLFOnJjKNQbF020tEJG3Mmr806/A2CNgaeAFYIwZ9AHMJXTUhBH2zsnabHcsKlUsHUB3nB+vTvUvSVRERKa1588IYup13Dq110K6DOlBgJyKSPlbR7MXMRpjZpKxlRN5Dm60APACc7u5fZq+LrXOtzO0naVZZXQtA7+7q8CMiHdj48bD55mEc3fXXw9ZbJ12joujMLCKSNi0YY+futwO3N35Y60YI6v7q7v+MxZ+YWX93/zh2tZwXy+cA62TtPiCWzaGh62am/KlmV1japUXVdQD0VoudiHRU55wDV10VEqNMnBgCvJRQi52ISNq0oMWuyUOGLJd/Bt5292uzVo0FjonPjwEeyio/2oIdgIWxy+YEYC8zW8XMVgH2imXSASzOtNj10H1hEemg1loLfv5zePHFVAV1oBY7EZH0Kc0E5TsBRwFvmtlrsex84ErgfjM7HvgAODyuG0eY6mA6YbqDYwHc/TMzuwx4KW53aSaRiqTfoqrQYqcxdiLSYbjDTTdB//5w2GFw2mlJ16jFFNiJiKRNCT43DsgAACAASURBVKY7cPf/AoUOvHue7R04pcCx7gDuaLvaSXuxSGPsRKQjmTcPjj0Wxo2DI48MgV2K6cwsIpI2pWmxE2nSYo2xE5GOYvx4GD4cFi4Mk46PHJl0jVpNgZ2ISNqUZoJykSZlkqf06aHATkRS7JVXYN99U5kgpTEK7ERE0kYtdpKQyip1xRSRFPvqK+jbF7bZBu66C37wA+jVK+latRl9OxARSZsST1AuUoimOxCRVHKHG2+EddeFyZND2VFHdaigDhTYiYiISJEWK3mKiKTNvHlwwAFhCoMddoDVVku6RiWjM7OISMqYWuAkIYuq6+jWxejeVfeFRSQFHn00JEj54ouGBCkd+BqqwE5EJGUU2ElSKqtq6dVN3TBFJCUmToR+/eDxxztMgpTG6JabiEjaWAsWkTZQWV1Hnx66Jywi7djkyfDii+H55ZeH550gqAMFdiIiqWNmzV5E2kJldZ0Sp4hI++QON90EQ4aE8XTu0KNHh0uQ0hjddhMRSRkFapKURdW1arETkfZn3jw47jh45BHYZx+4884OPZauEJ2dRURSRoGdJKWyuk5j7ESkfXnvPdhpp06TIKUxCuxERFJGgZ0kpbK6ltX79ky6GiIiDdZbDw49FE46qdOMpStEY+xERNJGyVMkIZVVGmMnIu3A5Mmw++7w0UdQUQE339zpgzpQYCcikjpKniJJWVRdq8BORJKTnSDlrbdg5syka9SuqCumiEjKKFCTpISsmPrqICIJmDcPjj8eHn64IUHKGmskXat2RWdnEZGUUWAnSXD3OI+dWuxEJAEXXRQmGu/kCVIao8BORCRlFNhJEqpq66mrd7XYiUj5LFkCn34Ka68NV1wBJ5+ssXSN0NlZRCRtFNdJAhZX1wFojJ2IlMfkyXDkkdCtG7z4IqyySlikICVPERFJGSVPkSQsqq4FoI9a7ESklLITpMydC5deGjJfSpN0dhYRSRkFapKEykyLncbYiUipfP45HH20EqS0kAI7EZGUUWAnSahUV0wRKbWePUMrnRKktIgCOxEREWlSZVXoiqnkKSLSppYsgauvhtNPh7594fnnoYtuILWEzs4iImmjG5iSgEWxxU5j7ESkzUyZAsOGwRtvwDe/GZKlKKhrMY1EFBFJmVIkTzGzO8xsnpm9lVX2NzN7LS4zzey1WD7IzBZnrbs1a59tzexNM5tuZjeY+o12GJUxeUovdcUUkdZyh5tvhm23DV0vH3kkBHXSKrrtJiKSMiWKlUYDNwJ3ZQrc/UdZ7zkKWJi1/XvuvlWe49wC/BR4ARgHDAXGl6C+UmaZMXaaoFxEWu2ii+Cyy5QgpY0psBMRSZlSBHbu/rSZDSrwfgYcDny/iXr1B1Z09+fj67uAg1Fg1yEs0hg7EWmt2lro2hVOOCEEcyefrAQpbUhdMUVEUqYlXTHNbISZTcpaRjTjLXcBPnH3d7PK1jOzV83s/8xsl1i2NjA7a5vZsUw6AGXFFJEWW7IEzjgDDj44dMMcOBBOOUVBXRtTYCcikjbW/MXdb3f3IVnL7c14x2HAmKzXHwMD3X1r4EzgXjNbsbUfq7Mws6FmNi2OQzw3z/qBZvafGDi/YWb7JlHPXJXVdXTvUkG3LvrqICLNMGUKbL89/P73sP76UFOTdI06LPWnEBFJmXLmIzGzrsChwLaZMnevAqri85fN7D3gW8AcYEDW7gNimURm1gW4CdiT0KL5kpmNdfcpWZtdCNzv7reY2aaEsYqDyl7ZHJXVtZqcXESK5w633AJnnQUrrhgSpOzbLu5TdVi67SYikjKlyIrZiD2Aqe6+tIulmfWLAQpmtj6wITDD3T8GvjSzHeK4vKOBh1rz5h3QdsB0d5/h7tXAfcBBOds4kGkBXQn4qIz1K2hRVZ2mOhCR4n35JVx+Oey2W5jOQEFdyekMLSKSMqVosTOzMcCuwGpmNhu4yN3/DBzBst0wAb4LXGpmNUA9cKK7fxbXnUzIsNmLkDRFiVOWtTYwK+v1bGD7nG0uBh4zs58DfQjBdeIW19RqqgMRadozz8AOO8BKK8Fzz8E662gsXZkosBMRSZsSXB/dfViB8uF5yh4AHiiw/SRgcJtWrvMZBox291FmtiNwt5kNdvf67I1iApwRAAMHDix5pUKLnQI7ESlgyRI477wwlu666+D000OSFCkbBXYiIimjOb9TbQ6wTtbrfOMQjyfM/4e7P2dmPYHVgHnZG8UEOLcDDBkyxEtV4YzK6lpNdSAi+U2ZAsOGhS6XI0fCz36WdI06JZ2hU2LAGivzp8uOZvVv9MUd7njgWW4a8xRXnH4w+353MNU1dbw/ewEjLrqHhV8vBuAXx+3F8IN2pK6+nrOu+gcTn3t76fEqKoxn//pLPpq3kMNOuzWpjyUpMPP9GZx79plLX8+ZPYsTTzmV1Vdfg9tuuZH3Z7zH3WPuZ9PNNk+wlp2LArtUewnY0MzWIwR0RwBH5mzzIbA7MNrMNgF6AvPLWss8FlXV0X+lbklXQ0Tam/vug2OPVYKUdkCBXUrU1tVz7rX/5LWps1mhdw/+d+85PPHCVJ54fiq/+sNY6urq+c2pB3H2cXtx4Q0PsfH6a/LDvbdhmx9cTv9+KzHu1pFsfvCl1NeHm7ojj9yNae9/Qt8+PRP+ZNLeDVpvfe77x4MA1NXVMXT377Hb7nuwZMkSrrnuBi6/9KKEa9j5KLBLL3evNbORwASgC3CHu082s0uBSe4+FjgL+KOZnUFIpDLc3UveIteUxTV19O6hrw0ikmOjjWDvveG228Kk45KYsp2hzWxjQuavzGS1c4Cx7v524b0kY+6CL5m74EsAvq6sYur7c1mr38o88fzUpdu8+Ob7HLLH1gDsv+sW/H3CK1TX1PLBR5/y3qwFfHvwIF54433WXn1lhu68Gb/78wRO/cn3E/k8kk4vvvAcA9ZZh7XW0pzTSVJgl27uPo4whUF22a+znk8Bdip3vZqyqKpWY+xEJHjsMXjqKbjiCth6a3jwwaRrJJRpugMzO4eQ0tmAF+NiwJh8k7NK4wb2X5WtNhrAS2/NXKb86IN2ZMKzYSqktfutxOy5ny9dN2fe56y1+koAXH32YVxw/YNLW+9EijVh/Dj23me/pKshLZigXKS1KqvrlBVTpLOrqoIzzggtdP/+N3z9ddI1kizlmsfueODb7n6lu98TlysJ8/kcX2gnMxthZpPMbFLtgsllqmr71qdXd8ZccwJnX/MAXy1asrT8l8fvTV1dPfeNe6nR/ffZZTDzPvuKV9+e1eh2Irlqaqp5+qkn2XOvoUlXpdMr8zx2Irg7ldW1msdOpDObMgW22y5kvRw5El58EVZYIelaSZZynaHrgbWAD3LK+8d1eWVn/Oq19chO37zUtWsFY675KX8bP4mHnnx9aflPDtiefb87mH1+dsPSsjnzFzJgzVWWvl579VX4aN5C9vve5uz/vc0ZuvNm9OjejRX79OSO3xzNcRfeVdbPIunz7DPPsPEmm/KN1VZLuioiUmZVtfXUO/TuoRY7kU5p8eIw0bg7PPww7KfeO+1RuQK704EnzOxdGiZmHQhsAIwsUx1S79aLfsy09+dywz1PLi3b8zubcObwPdjrhOtZvKRmafkjT73B6N8O54a7n6R/v5XYYGA/XnprJi+88T6//sNYAHbZdkNOP3p3BXVSlEfHP6JumO2EWuCk3BZV1QKoxU6ks/n8c1h5ZejVC/76Vxg8GNZcM+laSQFlOUO7+6Nm9i1C18vs5CkvuXtdOeqQdt/Zan1+vP/2vPnOHJ6/LwxLvOjGsYw6+4f06N6Vh28J8fGLb87k1Mvv4+0Zc3ngsVd59YELqK2r5/Qr79eYOmmxxZWVvPDcs1zw60uWlj35xONcdcVv+Pzzzzj15BP51sYbc/Ntf06wlp2H4jopt8rqcKnWGDuRTmTCBDjmGLjkkjAv3R57JF0jaYK1gwzKRVFXTCmlBS/8IekqSAfXp3vbhWMbnv1os8+H7149VOFgBzZkyBCfNGlSyY4/be5X7P37p7npyG3Yb4v+JXsfEWkHqqrg3HPDWLrNNoN774Uttki6VhKZ2cvuPiTfOvWpEBFJGbXYSbktqg5dMTXGTqSDmzIFjjwSXn89JEi56qrQDVNSQYGdiEjKaIydlFtlVeiK2bubAjuRDu3DD+Hjj5UgJaUU2ImIpIziOim3ythi16eHvjaIdDjz58N//gOHHw5Dh8KMGdCnT9K1khbQGVpEJGUqKhTZSXllkqf0VvIUkY7lscdCgpSvvgrTGfTrp6Auxco1QbmIiLQRs+YvIq2xdIydpjsQ6RiqquCMM2DvvWHVVeF//wtBnaSaztAiIimjMXZSboszLXZKniKSfrW1sPPOMGkSnHIKXH21EqR0EArsRERSRnGdlNsiJU8RST/3cAHp2jV0v7zoIth//6RrJW1IXTFFRFLGzJq9iLRGZXUtPbpW0LWLvjaIpNL8+XDQQSHbJYSpDBTUdTg6Q4uIpIwCu2SZ2fpmdr6Z3RRfb2RmmyVdr1KqrK5T4hSRtHrssTDB+IQJMHdu0rWRElJgJyKSMkqekhwz2xN4HdgBOCoWrwZck1ilymBRda0Sp4ikTVUVnHlmQ4KUl16CE05IulZSQgrsRERSRi12iboS+KG7HwjUxbJXgG2Sq1LpVVbV0UeJU0TS5aGH4LrrQoKUSZNCq510aLr9JiKSMorTEvVNd380PncAd19sZt0SrFPJLaqupZda7ETaP3eYNg023hh++EMYNAi22y7pWkmZqMVORCRlStFiZ2Z3mNk8M3srq+xiM5tjZq/FZd+sdeeZ2XQzm2Zme2eVD41l083s3Db/8MmbZWaDswvMbEtgZjLVKY/F1XX00Rg7kfYtkyBlyBD48MNwF1BBXaeiwE5EJGVKNMZuNDA0T/l17r5VXMaF97dNgSOAzeI+N5tZFzPrAtwE7ANsCgyL23YkNwD/NLOfAF3M7DDgHuC6ZKtVWouq6zTGTqQ9y06Q8pvfwIABSddIEqCztIiI4O5Pm9mgIjc/CLjP3auA981sOpC5LTzd3WcAmNl9cdspbVzdxLj7H2ML6DlAF+AS4PfufneiFSuxyupaZcUUaY/c4Re/gGuvhU03hUcfhS23TLpWkhC12ImIpExLumKa2Qgzm5S1jCjy7Uaa2Ruxq+YqsWxtYFbWNrNjWaHyDsXd/+jum7v7Cu4+2N3/lHSdSq2yWslTRNolM1i8uCFBioK6Tk2BnYhIyrSkK6a73+7uQ7KW24t4q1uAbwJbAR8Do0r6wVLAzN4uUP5muetSTpVVmu5ApN1wh1tuCYEcwI03hqVXr2TrJYlTYCcikjLlmu7A3T9x9zp3rwf+SEN3yznAOlmbDohlhco7kkIDVzrsgBZ3p7JGyVNE2oVMgpSTT4Y//zmUVejrvAS6/SYikjLlmu7AzPq7+8fx5SFAJmPmWOBeM7sWWAvYEHgRMGBDM1uPENAdARxZntqWlpmdH592zXqesQHLdkHtUJbU1OOOpjsQSdpjj8Exx8Bnn4X56U49NekaSTujs7SISMqUYsJxMxsD7AqsZmazgYuAXc1sK8J8bTOBnwG4+2Qzu5+QFKUWOMXd6+JxRgITCIlF7nD3yW1e2WTsGR+7ZT0HqAfmAseVvUZlsqi6FkBj7ESSNGECDB2qBCnSKAV2IiIpU4oWO3cflqf4z41sfzlweZ7yccC4Nqxau+DuuwGY2R/c/edJ16ecKqvqADTGTiQJ1dXQvTvsvjtcfXVIkqKxdFKAOuWKiKRMucbYyfI6W1AHDS12mu5ApIwyCVI23jiMq+vaNUxroKBOGqHbbyIiKaM4LTlm1gu4ENgd6EcYVwiAu6+fVL1KqbI602KnwE6kLObPhxNOgLFjYe+9ob4+6RpJSqjFTkQkZdRil6jrCJOu3w2sQZgCogq4I8lKlVLl0jF2uhcsUnKPPQZbbBHG0V13HYwbB2uskXStJCUU2ImIpIwCu0QdABzo7jcBtfHxMGC3ZKtVOoviGLte3dRiJ1JyN98Mq64KL74Ip5+uqQykWXT7TUQkZRSnJWoFd58Rn1ebWXd3n2Jm3060ViW0uEYtdiIlNWUK9O4NgwbBnXdCz54aSyctotsAIiIpoxa7RL1vZpvE51OB48zsCGBhgnUqqUyLnSYoF2ljmQQp224Lp50WylZZRUGdtJhuv4mIpIzitET9FhgIvA1cBvwL6A6clGSlSikzxq63WuxE2k5ugpTbbku6RtIB6CwtIpIyaoFLhoUf/H+ABQDu/riZrQJ0d/dFiVauhDJZMTXGTqSNvPFGCOY++ywkSDn1VI2lkzahwE5EJGUU1yXGgA+AvkA9gLvXADVJVqrUKqvr6Nmtgi4V+o8n0ibWXx+23x4uuQS23DLp2kgHotsDIiIpU2HW7EVaz93rgRnAKknXpZwWVdXSp7vuA4u0yttvwxFHQGUlrLACPPiggjppcwrsREREijcKuNfMdjKzAWa2VmZJumKlUlldRy8lThFpGXe49daQIOWJJ+Cdd5KukXRgugUnIpIyaoBL1J/i4+6Ax+cWn3fI6KeyWi12Ii2SmyBl9GhYc82kayUdmM7UIiIpo+QpiVov6QqUW2V1Hb17dMiYVaS0TjgBHn1UCVKkbBTYiYikjHJYJMfdP0i6DuW2qKqW3uqKKVKcqqqwrLgijBoFl16qsXRSNrp1ICKSMpqgPL3MbKiZTTOz6WZ2boFtDjezKWY22czuLXcdc1VW19FbXTFFmvb227DDDnDcceH1BhsoqJOyUmAnIpIyZs1fJHlm1gW4CdgH2BQYZmab5myzIXAesJO7bwacXvaK5qisrqOPWuxECstOkDJ7NhxzTNI1kk5KgZ2ISMpYC/5Ju7AdMN3dZ7h7NXAfcFDONj8FbnL3zwHcfV6Z67icyupaevdQi51IXp9+CoccAiedBLvsEiYfP+CApGslnZQCOxGRlKmw5i/SLqwNzMp6PTuWZfsW8C0ze9bMnjezoYUOZmYjzGySmU2aP39+CaobVFbX0bubWuxE8qqpgZdfDglSxo+H/v2TrpF0YgrsRERSRmPskmVmPzGzx83sjfj6u2Z2aBsdviuwIbArMAz4o5mtnG9Dd7/d3Ye4+5B+/fq10dsvq77eY1ZMtdiJLFVVBTfdBHV1YfqCd9+F009X1ktJnP4HioikjMbYJcfMzgQuAcYDA2PxfOCXRew+B1gn6/WAWJZtNjDW3Wvc/X3gHUKgl4jFNXUAGmMnkjF1akiQMnIkPP54KOvZM9k6iUQK7EREUqbCrNmLtJmTgH3c/VoaJih/B9igiH1fAjY0s/XMrDtwBDA2Z5sHCa11mNlqhK6ZM9qg3i2yqLoWQNMdiLjDbbfBNtuEBCljx8LQgj2lRRKhwE5EJGXUYpeoVd39nfg8E9hZ1vOC3L0WGAlMAN4G7nf3yWZ2qZkdGDebAHxqZlOA/wBnu/unbfoJmmFxdWix03QH0umddhqceKISpEi7pjO1iEjKlGLMnJndAewPzHP3wbHsauAAoBp4DzjW3b8ws0GEwGRa3P15dz8x7rMtMBroBYwDTnP3JoOeFJliZvu7+8NZZUOB14vZ2d3HEX4u2WW/znruwJlxSdyiqtgVs4da7KSTcg93x4YNg/XWCwGextJJO6X/mSIiKVOiFrvRhAAl2+PAYHffgtDd8Lysde+5+1ZxOTGr/BZCyv4N49LR+iqdD9xrZn8CepjZH4A7gQuSrVZpVMaumL3UYiedTVUV/OIXcNZZ4fWOO8IZZyiok3at4JnazN6kiK4lAPGiLyIiZVCKMXPu/nRsicsueyzr5fPADxo7hpn1B1Z09+fj67uAgwmJRjoEd3/GzHYgjLX7D+EG6a7uPjnZmpVGZbWSp0gnNHVqaKF77TU45ZSGVjuRdq6xW3D/KFstRESkaC35emFmI4ARWUW3u/vtzTjEccDfsl6vZ2avAl8CF7r7M4Q52WZnbZNvnrbUc/cpwM+Trkc5VC5NnqIWO+kE3OH220PLXJ8+IUGKxtJJihQ8U7v7JeWsiIiIFKclY+xiENecQC77/S4AaoG/xqKPgYHu/mkcU/egmW3WkmOnjZlNB+4ARrv7R0nXp9Q0xk46lVmzQlC3yy4werQmG5fUUUdhEREpyMyGE5Kq/DiTBMXdqzKZGt39ZUJilW8R5mQbkLV7vnna0u43wF7ATDMbb2Y/NLNuSVeqVCrjPHa91BVTOrI33gitdQMHwvPPw/jxCuoklYoO7MzsWDN7zMymmtmM7KWUFRQRkWVVWPOXljCzoYSJtw9098qs8n5m1iU+X5+QJGWGu38MfGlmO1hoVjwaeKiVH7ddcffR7r4rsAkwCbga+NjMrk+0YiVSWRW6YvZRV0zpiDIJUrbcEh54IJRtsYUSpEhqFfU/18zOBkYBLwODCBOovgWsSuiSIiIiZWJmzV6KOOYY4DlgIzObbWbHAzcCfYHHzew1M7s1bv5d4A0ze40wHvtEd/8srjsZ+BMwndCS12ESp2Rz9/fc/VfADsALhPnpOpxFMXlKr25qsZMOZupU2GEHGDUKTj4Z9t036RqJtFqxt+B+Coxw93+Y2UjgRnefYWa/AtYtXfVERCRXKZKzufuwPMV/LrDtA8ADBdZNAga3YdXandhauT9wLGE6h5eBnyVaqRJZXF1Lr25dqGhps69Ie3TXXWGycSVIkQ6m2MBuAPBifL4YWDE+HxPLf9rG9RIRkQJKMUG5FMfMrgOGEZLJ3A2c4+7TGt8rvRZV1ylxinQ8ffsqQYp0SMUGdnOB1YAPgQ+AHYHXgA0ocq47ERFpG2o8SdTahJa6Ce5en3RlSq2yqlZTHUjHMHEizJgBI0bAIYfAwQdrbjrpcIodHfokcGB8/mfgWjP7D2FOo3+WomIiIpJfKcbYSXHc/XB3H98ZgjoILXa9lRFT0iyTIGXPPeGWW6A2JARSUCcdUbG34UYQg0B3v9XMPgd2IoyxuK1EdRMRkTz0daS8zOwX7n5NfH5+oe3c/Yry1ao8FiuwkzSbOhWOPBJefTUkSLn6auiqFmjpuIr63x3vTNZnvf4bobVORETKrEJ3msvt+8A18fmeBbZxoMMFdouqa1mhh74ISwotWADf/jb07AkPPQQHHtj0PiIpV9TZ2sy2aWy9u7/SNtUREZGmKK4rL3ffN+v5bknWpdwqq+pYvW+PpKshUrwlS0Iwt9pqcPPNsMceSpAinUaxY+wmAS/Fx0lZrzOLiIiUicbYJcfMbi9Qfku561IOlTVKniIpMnEibLABPPFEeH3UUQrqpFMpNrBbD1g/Pq4HfAs4AniTMJePiIiUiVnzF2kzRxQoP7ystSiTyiqNsZMUyE6QsuKKobVOpBMqdozdB3mKp5vZQuAiYHyb1kpERArSGLvyM7PvxKcVZrYjy+aw2RBYVP5ald6i6lr6aIydtGf5EqT07p10rUQS0dqz9fvAVm1RERERKY7iukT8Nz468GxWuQMfAxeUvUYlVlfvLKmpp1c3tdhJOzZxIsyapQQpIhSfPGXV3CKgP3AxMK2N65TX5y/dWI63kU7qyWnzkq6CdHD7brZ6mx1LY+bKz90rAMzsNXfvFDc0F9fUAdCnhwI7aWcWLIApU+C734VTToEjjlD3SxGKb7FbQLgrmc2AWcCP2rRGIiLSqGIHR0vb6yxBHUBlVZjIWclTpF2ZOBGOPhrq6mDmTOjVS0GdSFTs2To3vXM9MB+Y7u61bVslERFpjFrsyquzTlC+qDq02Cl5irQLVVVwwQUwahRssgnce28I6kRkqWIDu/eBWe6e22qHmQ109w/btloiIiLtRqecoLyyWi120k58/XXodvnqq3DSSXDNNUqQIpJHcwK7/sAyA5HM7BtxnW7niYiUSYUa7Mqqs05QXlmtMXbSTqywAuy2G1x8sRKkiDSi2KEaxvJj7ABWAJa0XXVERKQpFdb8RUrDzHY1s52TrkcpLNIYO0nSggXwox/Bm2+G16NGKagTaUKjZ2szuyE+deC3ZlaZtboLsB3wWonqJiIieWiMXXLM7DHgN+7+tJmdBvwWqDOzX7v7dQlXr00t1hg7SUomQcqnn8IBB8DmmyddI5FUaOo2XOYvyYBNgOqsddXAKzSMOxARkTJQC1yitgKei89/CuwFfAE8CHSowC6TPKWPWuykXKqq4MILwxi6jTeGceNgq06TiFak1Ro9W2fGEpjZncBp7v5lWWolIiIFqcEuUd3dvcbM1gBWd/f/AphZ201U2E4sTZ6iMXZSLjffHII6JUgRaZFib8OdB6wILBPYmdkAoMbdP2nriomISH4ViuySNMPMjgG+CTwJSxOJdbjx5pXqiinl4A5z50L//mGy8c03hz32SLpWIqlUbPKUe4B98pTvDdzddtUREZGmVLRgkTbzS8K0Bj8BLo9l+wEvJVajEqmsqsUMenZVYCclsmABHHIIbL89LFwI3bsrqBNphWKv90OAp/OUPxPXiYhImZg1f5G24e4T3X1td1/f3WO6PsYABydZr1JYVF1H725dqNCgTimFiRNhiy3COLrTT4e+fZOukUjqFRvYdQV65CnvWaBcRERKpMKs2UtTzOwOM5tnZm9lla1qZo+b2bvxcZVYbmZ2g5lNN7M3zGybrH2Oidu/G7ssdjhmtoKZ/cjMfmFmPwJ6uHtN0vVqa5XVtfRS4hRpazU1cPbZsOeesNJK8MILcOaZUKG+BSKtVexf0QvASXnKT6EDdj8REWnPStRiNxoYmlN2LvCEu28IPBFfQ+iav2FcRgC3hHrZqsBFwPaE6XAuygSDHYWZbQa8Q8gIfVB8fMfMBidasRKorK7T5OTS9rp0gddfhxNPhJdfhq23TrpGIh1GsbfiLgCeNLMtiIPFge8D2wC7l6JiIiKSXyl6xsV52QblFB8E7Bqf/wV4Cjgnlt/l7g48b2Yrm1n/uO3j7v4ZgJk9ts94iwAAIABJREFUTggWx7R9jRPze+A24FJ3dwuTCl4IXE8Hux4uqqrT5OTSNtzhjjtg771hwAB4+OEwnk5E2lRRLXbu/jywIzATODQuM4AdAOWiFREpo1J0xSxgDXf/OD6fC6wRn68NzMrabnYsK1TekWwDXBGDWuLjlYT57TqUyupa+igjprTWggVw6KFwwglhOgNQUCdSIkXfinP314Efw9JpDo4F/gWsC+jMLyJSJi2J08xsBKHbZMbt7n57sfvH1ilv/jt3OF8Ag4B3s8oGkTMdUEdQWV1H355qsZNWmDgRjj46BHejRoUkKSJSMkWfsc2sC6H7zfHAXsAbwK3A30tTNRERyaclXTFjEFd0IBd9Ymb93f3j2NVyXiyfA6yTtd2AWDaHhq6bmfKnml/bdu0vwCNmdiWhF8sgwhQIo5OrUmlUVtey5oo9k66GpNV998GwYbDxxvDIIxpLJ1IGTXbFNLONzOxq4CPCIPFXAQOOcver3P39EtdRRESyWAv+tdBYIJPZ8hjgoazyo2N2zB2AhbHL5gRgLzNbJSZN2SuWdSSXA3cSxho+HB//QsOcdh3Goqo6eit5ijRXfX143GcfOO88JUgRKaNGAzszewZ4HlgFODzO23MhoO44IiIJqbDmL00xszHAc8BGZjbbzI4njB3b08zeBfaIrwHGEcZZTwf+CJwMEJOmXEbIlvwSIcHIZ2364RPm7nXu/lt338jde8fH37p7bdJ1a2uLa+rorTF2Uix3uO022GUXqKoKUxlccQX0VioGkXJpqivmjsBNhLEYk8tQHxERSYC7DyuwarlMjzFhyCkFjnMHcEcbVq3dMbMVgP0JXU1nAePc/atka9X2FlXV0kdZMaUYn34akqM8+CDssQd8/TX00DTHIuXWVFfMbxOCv/+a2atmdoaZrVmGeomISAGlaLGT4pjZEEJr5e8I486vAt6L5R1GbV09VbX1mu5AmjZxImyxRRhHd801MGECfOMbSddKpFNqNLBz91fd/RSgP3AtcCDh7mQFsF9Hm3hWRCQNzKzZi7SZm4FR7r6uu+/i7usSxp/fUuwBzGyomU0zs+lmdm4j2x1mZp5E0FhZUwegrpjSuPp6OPtsWHFFeOEFOOssqChqJi0RKYFi57Fb4u53u/tuwCbA1cAZwFwzG1/KCoqIyLLUYpeoTYBROWXXAhsXs3PMMH0TsA+wKTDMzDbNs11f4DTghVbVtoUWV8fATslTJJ9p02DhwhDEPfigEqSItBPNvq3i7tPd/VxCquvDgeo2r5WIiBRk1vxF2sxrwOCcss1jeTG2A6a7+wx3rwbuI3TpzHUZobvnkpZWtDUWVYVcMBpjJ8vIJEjZeuuQ8RJg3XWVIEWknWjxGdvd6wiprx9qalsREWk7FYrUkvQY8LCZ/Qn4gDCP3XHA7WZ2ZGYjd7+3wP5rE4Y0ZMwGts/ewMy2AdZx90fM7Ow2rHvRKqvVFVNy5CZIufDCpGskIjl0K05EJGXUtTJRxwE1NMzvB1AbyzMcKBTYNcrMKghdO4cXse0IYATAwIEDW/J2BTUEdvqaIMCLL8Ihh8D8+SFByhlnaCydSDukM7aISMqowS457r5eKw8xhzCUIWNALMvoS+jq+VRMerMmMNbMDnT3STl1uR24HWDIkCFtOr/sourQFVNj7ASA/v1Dl8uHH9ZYOpF2TLdbRERSpgJr9iLtxkvAhmb2/+3dfZzVc/7/8cermaaa0qZLuqCWQtelb8hu5KryJVdFIhchLD9aZbHLau3ylYvatditlKssclkIG7Ebi0QplZIUk7bSJDQ1pzO9fn98PjNO09TM1Jw58znzvO/tc+ucz+XrfHacz+d13u/369PGzLKAwcD0woXuvsndG7t7a3dvDbwP7JTUJVteftBipzF21dhnnwUtc9u3Q6tW8O67SupEqjgldiIiEaPiKdHl7nHgauB1YAkw1d0XmdltZjYgtdH9JK+wxU5j7Kofd5gwAbp3h8cfhxUrgvn6IhGp8vRTnIhIxGiMXbS5+wxgRrF5v9/FusdWRkzFqXhKNVW8QMqjj0Lz5qmOSkTKSImdiEjEqCqmJFvhGLu6tXSbUG24w//+L3z8sQqkiESUvrFFRCJGeV3qWVDZZD93X5PqWJIhL78AM6iVqRv7tBeLBV8qNWvC2LFQp47G0olElL6xRUQipoZZuSepGGZWz8wmAVuA5eG8083s1tRGVrHyYgXUzcrE9LeT3pYuhaOOglvDP99evZTUiUSYEjsRkYhR8ZSUuhdoBhwNxMJ5HwLnpCyiJMiLxTW+Lp25w8SJQYGUVaugZ89URyQiFUBdMUVEIka/yKXUKUB7d99kZg7g7qvNLK0qTGyOFWh8XbrasAEuuwxeeEEFUkTSjO4PREQixszKPUmFqUHQDbOImdUDfkxNOMmxJRanTk212KWlnBx4442gQMrrryupE0kjSuxERCLG9mCSCvMOcFOxef8PeCsFsSTN5vwC6tZSYpc2YjF49tngdZcuQffLkSNV9VIkzaifhYiISNldB8wys/OBema2EMgCjkttWBUrLxanQXZWqsOQirB0KQwZEjzG4KOPgnF1++6b6qhEJAmU2ImIRIyqXKaOu39tZh2BU4HWwCrgZXffstsNIyYvVkDzBmqxizR3eOghGDEieITBCy8ESZ2IpC0ldiIiEaO0LrXcPR94NtVxJFNerIDsLN0iRNqwYfDII3D88fDYYxpLJ1IN6FtbRCRi1GCXOmY2YVfL3H14ZcaSTJtjcY2xi7qTToIOHeC66zSWTqSaUGInIhIxqnKZUjWLvW8O9CbNWvDy8guoo+fYRUssBrfcAq1bw5VXwrnnpjoiEalkSuxERCJGv72njrtfXHyemZ0BnJSCcJJiW8F2YgXbqauumNGxdCmcd15QHGXEiFRHIyIpovsDEZGIScZz7MzsEDObnzB9b2YjzGy0ma1OmH9ywjY3mdlyM1tqZn2T+qGrtheBc1IdREXJixUAkK0Wu6rPHSZODIqirFwZFEgZNy7VUYlIiujnOBGRiElGR0x3Xwp0BTCzDGA18AJwMTDO3e/ZIQaz9sBgoANBd8Q3zKyduxckIbyqrj/FHloeZXmxOAB1a+kWocqbNw+GD1eBFBEBlNiJiEROJYyxOx74wt1X7eZYpwFPhRUivzSz5UBP4L1kB5dKZvY54Amz6gJNgWtTE1HFU4tdBHz1FRxwQNBS99Zb0Lu3CqSIiBI7EZGo2ZPbNzMbDiRWbZzg7ruq8DgYeDLh/dVmdgEwFxjp7huBFsD7CevkhPPS3Z+Kvf8BmO/uK1IRTDLk5RcmdrpFqHIKC6SMHQuzZ8ORR8Kxx6Y6KhGpIvStLSISMXvSYhcmcbss1Z+w7yxgAHBTOOtvwB8JWqn+CNwLDCt3AGnAzDKBZsB97r411fEky+bCrphqsataEgukXH45dO6c6ohEpIpRu72ISMTYHkzl0B/42N3XArj7WncvcPftwESC7pYQjMFrlbBdy3Be2nL3OPDbdE7qALaEXTH1uIMq5OGHdyyQ8ve/Q3Z2qqMSkSpGiZ2ISMSYlX8qh3NJ6IZpZvsnLDsD+DR8PR0YbGa1zKwN0BaYs3efLBLeMrNjUh1EMm1W8ZSqZ906OOooWLAATj891dGISBWlb20RkYipkZS6mGBmdYETgcsTZt9lZl0JumKuLFzm7ovMbCqwGIgDV1WTipgrgWlm9mz4envhAne/I0UxVaifxtipxS6lZs0KxtT16wfXXx9MKpAiIruhxE5EJGKSVRTT3TcDjYrNG7qb9W8Hbk9ONFWLmX3v7vUJHgkxDzgonAo5kBaJXWGLnYqnpEhhgZS774ajj4a+fZXQiUiZ6FtbRCRiLEktdrJbBuDufVIdSLLpcQcpVLxAytixyfslR0TSjhI7EZGI0X1eSnjpq6SHvFicjBpGrUy1ElWq5cuDAim1awcFUjSWTkTKSYmdiIhI6Wqb2eTdreDuafEYiM35BWRnZezRYzVkDxQUQEYGHHRQ0AXzggugefNURyUiEaSf40REIqYGVu5JKkRBKVNa2BIrUDfMyjJrFrRvH3TBNIMbb1RSJyJ7TC12IiIRo4aUlNjq7pelOojKsDkWp64KpyRXYoGUdu2C9yIie0nf3CIiEaPETpIpL1ZAdi212CVNSQVS9LBxEakASuxERCJGVTFTotqc9LxYnOyauj1ImvHj4csvVSBFRCqcxtiJiERMDSv/JHvH3fdJdQyVRS12SZCbC4sXB69vvx0WLlRSJyIVTomdiEjE2B78T6SsNudrjF2FmjULOneGgQNh+3aoU0cFUkQkKZTYiYhEjFn5J5GyyosVUEdVMfdeLAY33AAnnAD16sGUKVBDt10ikjz6SU5EJGLUAifJlBcroK4Su72zbh2cfPJPBVLuvRfq1k11VCKS5pTYiYhEjMbMSTLlxeJk19LtwV5p1AhatYKbb9ZYOhGpNPrmTgP5+flcfMF5bIvFiBcUcOJJffnV1ddw029GsmjRp2Rm1qRjp07ccutt1KxZM9XhSkRs2fwDTz0whv9+/SVgnHv1jdTMqs0z4+8htnUL+zbdj6Ejfk/t7Lqs+nwxU/92d7ChO33PGUbnI3unNP50phY7SZZYfDvbClwtdnsiNzfoevmnP0GzZkHVSxGRSqTELg1kZWXx0ORHya5bl23btnHR0CH84pe9OfmUAdwx5h4Abrx+JC889wxnDx6S4mglKp6fdB+HdTuCi3/zJ+LbtrEttpW/jb6OARf9ioM7dOODN19h1otPcvKQS9n/gJ9z3d0TycjIZFPut9xz3cV0+J9eZGToKyYZNGZOkmVLrACAOiqeUj5vvQVDh/7UBfOMM1IdkYhUQxrFmwbMjOyw7348Hicej4MZv+x9DGaGmdGxU2fWrl2b4kglKrZs/pEViz/hiBNOASCzZk3q1N2H9Wu+5qD2XQFo16UHC95/G4CsWrWLkrj4tpgyjySzPZhEymJzLA6gFruyisXgxhvh+OODAinvv6+kTkRSJuWJnZldnOoY0kFBQQFnn3kafX7ZiyOP6kXnzl2Klm3bto2XX5rG0b/4ZQojlCjJXbeGevUb8OT9d3DPyGE89cCd5G/dwn6t2vDpnNkAfPKft/ju23VF26xatog7rx3KXb++iEGXj1JrXRLVMCv3JFIWeWFipzF2ZXTLLTBmDFx2WVAopXv3VEckItVYyhM74A+7WmBmw81srpnNnTRxQmXGFDkZGRlMfX4a/5z1Lz5duIDPP19WtOyOP/6Bww/vQffDe6QwQomSgoICclYs4+i+pzPq3slk1a7Dm88/weCrbuSd117k3lGXsHXLFjIyfxqzeWC7Dtz4l8e57q4JvPn8FLbF8lP4CdKbWuwkWfLCrpjZNdVit0vu8P33wevrr4dp02D8eFW9FJGUq5Sf5Mxswa4WAc12tZ27TwAmAGyN40kILe3Ur1+f/+l5BP95ZzZt27bj7w/ez8aNudwy+v5UhyYR0qBRE37WqAkHtusAQJejjuXN56dw8pBLufLWsQCs++Yrlnz03k7bNmvZmqzadVjz1ZcccPChlRp3taFMTZJkc36Y2NVSYlei3NygdW71apg9Gxo3hgEDUh2ViAhQeS12zYALgFNLmDZUUgxpKzc3l+/DXw+3bt3K++/9h9Ztfs7zzz7Df959hzvvHksNPRRVyqH+vo1o0Lgp61Z/BcDnCz5iv1at+eG7jQBs376dmc88Rq++pwGwYe03FBQEXbhy1/2XdatX0bDpfqkJvhqwPfifSFnkFY2xU1fMncyaBZ07w0svwVlnQYaSXxGpWirrm/tloJ67zy++wMzerqQY0ta369dx829vZPv2ArZvd07q249jju1D987t2b95cy4Ycg4Ax51wIlf86uoURytRcdalI3j8z7dREN9Go2bNOffq3/Lh26/x7qvPA9DpyGPoedzJAKxYsoA3X3iCjIxMzIyBw6+jXv0GqQxfpEoys37AX4AM4CF3v7PY8uuAS4E4sB4Y5u6rKiu+zYVdMVU85SexWDCW7u67oW3boECKxtKJSBVk7tHo4aiumJJMs5auK30lkb1wcoemFdZsNmfFpnJ/H/b8+c/UbJdiZpYBLANOBHKAD4Fz3X1xwjp9gA/cPc/MrgSOdfdzStt3jx49fO7cuXsd49MffsUNzy3k3RuPo0WDOnu9v7SweTMcfjgccwyMHauxdCKSUmb2kbuXWDhD/fNERCJGxVMiqyew3N1XuHsMeAo4LXEFd3/L3fPCt+8DLSszwMIxdtX+cQfu8OSTQVJXty7MmaMCKSJS5SmxExGJmiRldma20swWmtl8M5sbzmtoZjPN7PPw333D+WZm95nZcjNbYGbqm1a6FsDXCe9zwnm7cgnw6q4WJlaOXr9+fYUEWPS4g+o8xi43FwYOhCFDYEJYkbt+/dTGJCJSBkrsREQiJsnFU/q4e9eEbh43Am+6e1vgzfA9QH+gbTgNB/5WQR9PADM7H+gB3L2rddx9grv3cPceTZo0qZDj5sUKyKxhZGVW09uDxAIpd90F116b6ohERMqsmn5zi4hEl1n5p71wGvBo+PpR4PSE+Y954H2ggZntv1dHSn+rgVYJ71uG83ZgZicAvwMGuHulPhAyL1ZQfQunjB8PJ5wQdLd8773gGXWqKC0iEaJvLBGRiNmTnpiJ3fbCaXgJu3bgn2b2UcLyZu6+Jnz9X3569mh5uxVKUCylrZm1MbMsYDAwPXEFM+sGjCdI6iq9qtPm/Dh1a1XTbpgnnABXXQUffxwUSxERiZhq+u0tIhJhe9AC5+4TgAmlrPYLd19tZk2BmWb2WbF9uJmpQvEecve4mV0NvE7wuIPJ7r7IzG4D5rr7dIKul/WAZyxoav3K3SvtCdh52wqoU11a7Nxh8mR46y14/HE46CD4619THZWIyB5TYiciEjHJeuC4u68O/11nZi8QVHFca2b7u/uasKtlYStSmboVyo7cfQYwo9i83ye8PqHSg0qQlx+vHg8nz82F4cPhuefguOOC6pf16qU6KhGRvaKumCIiEZOMMXZmVtfM9il8DZwEfErQVfDCcLULgWnh6+nABWF1zCOBTQldNiWiNleHMXaFBVKmTw8KpMycqaRORNJCNfhZTkQkvSTpuXTNgBfC7n+ZwD/c/TUz+xCYamaXAKuAs8P1ZwAnA8uBPODi5IQllSkvFqdJvVqpDiN5tmyB884LHl/w3nsaSyciaUWJnYhI1CQhs3P3FUCXEuZvAI4vYb4DV1V8JJJKebECstOxeMqXX0KrVlCnDrz6KrRtq4eNi0jaUVdMEZGISfJz7KQay8svoG46dcV0h4cego4d4d57g3lduyqpE5G0lIY/y4mIpLe9fC6dyC5tjsXJTpfiKcULpJx/fqojEhFJKrXYiYhEzJ48x06kNO7OlnQpnvLOO0GBlGnTYMyYoEBKCz1mUUTSW5r8LCciUo0oU5MkiBVsJ77d0+MB5TVrQoMGQWKnAikiUk2oxU5EJGI0xk6SIS+/ACC6LXbLlsGf/xy8PuIIWLBASZ2IVCtK7ERERIS8bRFN7AoLpHTrBn/6E6xfH8yvoVscEale9K0nIhIxyXhAuUhefhwgWsVTcnNh0CC47DI48kj45BNo0iTVUYmIpESEvr1FRAQ0xE6SY3MsaLGrWysiLXbxOBx1FKxYERRIGTVKrXQiUq0psRMRiRpldpIEhS12dWpW8VuDeBwyMiAzM+h6+fOfayydiAjqiikiEjkqniLJkBeFFrtly4Iul48+GrwfNEhJnYhISImdiEjEaIydJMPmWBUeY5dYIGXFiuBRBiIisgMldiIiEaMHlEsyVNkWu+IFUhYsgNNPT3VUIiJVjhI7EZGoUWYnSVCY2GVXtTF2//kPTJ8eFEiZORNatkx1RCIiVVIV+/YWEZHSaMycJENR8ZSq8By7WAzefx9694ZTToHly+GAA1IdlYhIlaYWOxGRiNEYO0mGzbECsjJqkJWZ4luDZcvg6KPhhBNg1apgnpI6EZFSKbETEYkY9cSUZNgSi6e2tc4dJk0KCqR88QU89RQceGDq4hERiRh1xRQRiRplapIEm2MF1E1VYucOgwfD1KnQpw889pjG0omIlJMSOxGRiNEYO0mGvFic7Fopui0wg06doHt3GDUqeAC5iIiUixI7EZGI0Zg5SYbN+QVkV2aLXSwGt94Kxx0HJ54IN99ceccWEUlDSuxERCJGeZ0kw5ZYJSZ2n38OQ4bA3LnBLxUnnlg5xxURSWNK7EREokaZnSTB5lic/erXTu5B3OHhh+GaayArC557Ds48M7nHFBGpJpTYiYhEjMbYSTLkxQqSP8bu1VfhkktUIEVEJAn0uAMREcHMWpnZW2a22MwWmdm14fzRZrbazOaH08kJ29xkZsvNbKmZ9U1d9FIR8mJxsmsmqStmbm7wb//+8PTTMHOmkjoRkQqmxE5EJGKS9IDyODDS3dsDRwJXmVn7cNk4d+8aTjOCGKw9MBjoAPQDHjQzlTKMsLz8ArJrVfD/hbEY3HQTHHwwrFwZ/DGefbaqXoqIJIG6YoqIREwyOmK6+xpgTfj6BzNbArTYzSanAU+5ez7wpZktB3oC7yUhPEkyd2dzLE7drAq8LUgskHLppdCkScXtW0REdqIWOxGRiElSi13C/q010A34IJx1tZktMLPJZrZvOK8F8HXCZjnsPhGUKiw/vp3tDnUqqirm5MnQrRt88QU8+yxMnAh161bMvkVEpERK7EREIsfKPZnZcDObmzANL3HPZvWA54AR7v498DfgIKArQYvevUn9aJISebECAOpWVGL3zjvQsycsWABnnVUx+xQRkd1SV0wRkYjZkweUu/sEYMLu92s1CZK6J9z9+XC7tQnLJwIvh29XA60SNm8ZzpMI2pwfB9i7qphvvw0NG0LnzvDgg1CzpsbSiYhUIrXYiYhETPnb68qwTzMDJgFL3H1swvz9E1Y7A/g0fD0dGGxmtcysDdAWmLMXH0tSqLDFbo8eUL5tW1Ag5bjj4Pe/D+bVrq2kTkSkkqnFTkQkYvakxa4MjgaGAgvNbH4477fAuWbWFXBgJXA5gLsvMrOpwGKCippXuXtBUiKTpMuLBS125S6eUrxAyp//nIToRESkLJTYiYhETDIeUO7u71By496M3WxzO3B7hQcjlW6PWuw+/DB40HhWVlAgRWPpJI1s27aNnJwctm7dmupQpJqqXbs2LVu2pGbNmmXeRomdiEjUJKfFTqqxwjF2dcsyxs49aDbu0gWGDYPf/EYPG5e0k5OTwz777EPr1q2xJHWTENkVd2fDhg3k5OTQpk2bMm+nMXYiIhGTjDF2Ur1t2Ra02JX6uIO334ajj4bc3KCl7r77lNRJWtq6dSuNGjVSUicpYWY0atSo3C3GSuxERCIm2c+xk+pnc37h4w520WKXWCDl229h3bpKjE4kNZTUSSrtyd+fumKKiERMMsbYSfVWWDwlu1YJLXYlFUjRw8ZFRKoctdiJiESN+mJKBSsqnlKzhMTuhhvgiy+CAikTJyqpE6kk9erVK3o9Y8YM2rVrx6pVqxg9ejTZ2dmsS2g5T1zXzBg5cmTR+3vuuYfRo0eXeIwXX3yR2267rej9lClT6Ny5Mx06dKBLly5ceumlfPfddwC8/PLLdOvWjS5dutC+fXvGjx8PwP3338/kyZN3+1m6du3K4MGDd5h37LHHMnfu3KL3K1eupGPHjkXv58yZQ+/evTnkkEPo1q0bl156KXl5ebs9Tmm+/PJLjjjiCA4++GDOOeccYrHYTuvEYjEuvvhiOnXqRJcuXXj77bd3WDZ8+HDatWvHoYceynPPPQfA2LFjad++PZ07d+b4449n1apVAKxatYru3bvTtWtXOnTowN///veiffXr148uXbrQoUMHrrjiCgoK9r6wtBI7EZGIUV4nFW1zLE5WZg0yM8Lbgo0bYc2a4PWDD8KCBap6KZIib775Jtdccw2vvvoqBx54IACNGzfm3nvvLXH9WrVq8fzzz/Ptt9+Wuu+77rqLX/3qVwC89tprjBs3jldffZVFixbx8ccf06tXL9auXcu2bdsYPnw4L730Ep988gnz5s3j2GOPBWDYsGH89a9/3eUxlixZQkFBAbNnz2bz5s1l+sxr165l0KBBjBkzhqVLlzJv3jz69evHDz/8UKbtd+WGG27g17/+NcuXL2ffffdl0qRJO60zceJEABYuXMjMmTMZOXIk27dvB+D222+nadOmLFu2jMWLF3PMMccA0K1bN+bOncuCBQsYOHAgv/nNbwDYf//9ee+995g/fz4ffPABd955J9988w0AU6dO5ZNPPuHTTz9l/fr1PPPMM3v12UBdMUVEIkfDPqLNzPoBfwEygIfc/c5iy2sBjwGHAxuAc9x9ZTJjyssvoG5h4ZS334ahQ+Gww+Cf/4T99kvmoUWqvD+8tIjF33xfofts37w+t57aodT1/v3vf3PZZZcxY8YMDjrooKL5w4YN45FHHuGGG26gYcOGO2yTmZnJ8OHDGTduHLffvusn0ixbtoxatWrRuHFjIEha7rnnHlq0aAFARkYGw4YNAyA3N5d4PE6jRo2AIHk85JBDAMjOzqZ169bMmTOHnj177nScJ598kqFDh7JkyRKmTZvGkCFDSv3cDzzwABdeeCFHHXVU0byBAweWut3uuDuzZs3iH//4BwAXXngho0eP5sorr9xhvcWLF3PccccB0LRpUxo0aMDcuXPp2bMnkydP5rPPPgOgRo0aReeuT58+RdsfeeSRTJkyBYCsrKyi+fn5+UUJIkD9+vUBiMfjxGKxChnTqRY7EZGIsT34n1QNZpYBPAD0B9oTPAC+fbHVLgE2uvvBwDhgTLLj2hyLs08GPxVIqVMH7rgj2YcVkd3Iz8/n9NNP58UXX+TQQw/dYVm9evUYNmwYf/nLX0rc9qqrruIghrF3AAAUlUlEQVSJJ55g06ZNu9z/u+++S/fu3YveL1q0aIf3iRo2bMiAAQM48MADOffcc3niiSd2SFJ69OjB7NmzS9z26aefZvDgwZx77rk8+eSTu4wn0aeffsrhhx9e6npLly6la9euJU6FXUgLbdiwgQYNGpCZGbRrtWzZktWrV++0zy5dujB9+nTi8ThffvklH330EV9//XXR/m655Ra6d+/OoEGDWLt27U7bT5o0if79+xe9//rrr+ncuTOtWrXihhtuoHnz5kXL+vbtS9OmTdlnn332OnEFtdiJiESOWuwirSew3N1XAJjZU8BpwOKEdU4DRoevnwXuNzNzd09WULW/Wc3E8SPh66VBgZRx4yBhzI5IdVaWlrVkqFmzJr169WLSpEklJnDXXHMNXbt2ZdSoUTstq1+/PhdccAH33XcfderUKXH/a9asoUmTJiUuW7hwIUOHDuWHH37gjjvu4JxzzuGhhx5i4cKFvPHGG9xzzz3MnDmTRx55BAhatgpbshLNnTuXxo0bc8ABB9CiRQuGDRtGbm4uDRs2LLGFqrytVocccgjz588v1zalGTZsGEuWLKFHjx4ceOCB9OrVi4yMDOLxODk5OfTq1YuxY8cyduxYRo0axeOPP1607ZQpU5g7dy7/+te/iua1atWKBQsW8M0333D66aczcOBAmjVrBsDrr7/O1q1bOe+885g1axYnnnjiXsWuFjsREZHK0wL4OuF9TjivxHXcPQ5sAhoV35GZDTezuWY2d/369XsV1IbMOmTATwVSlNSJpFyNGjWYOnUqc+bM4Y4SWtAbNGjAkCFDeOCBB0rcfsSIEUyaNGmX49rq1Kmzw3PSOnTowMcffwxAp06dmD9/Pv3792fLli1F63Tq1Ilf//rXzJw5s6hwCATP/SspgXzyySf57LPPaN26NQcddBDff/990XaNGjVi48aNRevm5uYWdW3s0KEDH3300S7PTaHytNg1atSI7777jng8qAKck5NT1O00UWZmJuPGjWP+/PlMmzaN7777jnbt2tGoUSOys7M588wzARg0aFDR+QJ44403uP3225k+fTq1atXaab/NmzenY8eOO7Vs1q5dm9NOO41p06aV+nlLo8ROREQkgtx9grv3cPceu/rVvax+d+6R8OEcFUgRqWKys7N55ZVXeOKJJ0os9HHdddcxfvz4omQlUcOGDTn77LNL3A7gsMMOY/ny5UXvb7rpJkaNGkVOTk7RvMKk7scff9yhOuT8+fOLCrlAMF4vsaIlwPbt25k6dSoLFy5k5cqVrFy5kmnTphV1xzz22GOZMmUKhZ0RHn300aKxaldffTWPPvooH3zwQdH+nn/++Z26Pha22JU0NWjQYId1zYw+ffrw7LPPFh3vtNNO2+m85OXlFSXDM2fOJDMzk/bt22NmnHrqqUXn4c0336R9+6An/bx587j88suZPn06TZs2LdpXTk5O0TncuHEj77zzDocccgg//vgja8ICVfF4nFdeeWWn7rZ7Ql0xRUQiRl0xI2010CrhfctwXknr5JhZJvAzgiIqSXNAo+xk7l5E9kLDhg157bXX6N27905dJxs3bswZZ5zBuHHjStx25MiR3H///SUu6927NyNHjsTdMTNOPvlk1q9fT//+/SkoKKBBgwZ07NiRvn374u7cddddXH755dSpU4e6desWdcOEYLxe8UcqzJ49mxYtWuwwpqx3794sXryYNWvWMHz4cD777DO6dOmCmdGjRw/+7//+D4BmzZrx1FNPMWrUKNatW0eNGjXo3bs3/fr124Mz+JMxY8YwePBgbr75Zrp168Yll1wCwPTp05k7dy633XYb69ato2/fvtSoUYMWLVrs0NVyzJgxDB06lBEjRtCkSRMefvhhAK6//np+/PFHBg0aBMABBxzA9OnTWbJkCSNHjsTMcHdGjRpFp06dWLt2LQMGDCgqqNKnTx+uuOKKvfpsAJbELvsVamucaAQqkTRr6brSVxLZCyd3aFph6dimLdvL/X34szo1lA5WAWGitgw4niCB+xAY4u6LEta5Cujk7leY2WDgTHc/e3f77dGjhyc+D0pE9s6SJUs47LDDUh1G0l177bWceuqpnHDCCXu8j3nz5jF27NgdEiCpGCX9HZrZR+7eo6T11RVTRCRizMo/SdUQjpm7GngdWAJMdfdFZnabmQ0IV5sENDKz5cB1wI2piVZE0t1vf/vbvX7o97fffssf//jHCopI9oa6YoqIRIzytGhz9xnAjGLzfp/weiswqLLjEpHqp1mzZgwYMKD0FXdjbys5SsVRYiciEjXK7EREkq5w7JlIKuzJcDl1xRQRiRg9oFxEJLlq167Nhg0b9ujmWmRvuTsbNmygdu3a5dpOLXYiIhGjH5BFRJKrZcuW5OTksLfPiBTZU7Vr16Zly5bl2kaJnYhIxCivExFJrpo1a9KmTZtUhyFSLkrsRESiRpmdiIiIFKPETkQkYjRmTkRERIpTYiciEjEaYyciIiLFmar9pCczG+7uE1Idh6Qn/X2JVC1mth5YtZe7aQx8WwHhpBOdk53pnJRM52VnOic7q4hzcqC7NylpgRK7NGVmc929R6rjkPSkvy+R9KP/rnemc7IznZOS6bzsTOdkZ8k+J3qOnYiIiIiISMQpsRMREREREYk4JXbpS+OfJJn09yWSfvTf9c50Tnamc1IynZed6ZzsLKnnRGPsREREREREIk4tdiIiIiIiIhGnxC4NmVk/M1tqZsvN7MZUxyPpw8wmm9k6M/s01bGIyJ4p7RphZrXM7Olw+Qdm1rryo6xcZTgn15nZYjNbYGZvmtmBqYizMpX1XsLMzjIzN7O0r35YlnNiZmeHfyuLzOwflR1jZSvDfzsHmNlbZjYv/O/n5FTEWZlKu1eywH3hOVtgZt0r6thK7NKMmWUADwD9gfbAuWbWPrVRSRp5BOiX6iBEZM+U8RpxCbDR3Q8GxgFjKjfKylXGczIP6OHunYFngbsqN8rKVdZ7CTPbB7gW+KByI6x8ZTknZtYWuAk42t07ACMqPdBKVMa/k5uBqe7eDRgMPFi5UabEI+z+Xqk/0DachgN/q6gDK7FLPz2B5e6+wt1jwFPAaSmOSdKEu/8byE11HCKyx8pyjTgNeDR8/SxwvJlZJcZY2Uo9J+7+lrvnhW/fB1pWcoyVraz3En8kSPy3VmZwKVKWc3IZ8IC7bwRw93WVHGNlK8s5caB++PpnwDeVGF9KlOFe6TTgMQ+8DzQws/0r4thK7NJPC+DrhPc54TwREZGyXCOK1nH3OLAJaFQp0aVGea+blwCvJjWi1Cv1nITdx1q5+yuVGVgKleXvpB3QzszeNbP3zSzde7iU5ZyMBs43sxxgBvD/Kie0Ki1p9+qZFbETERERkXRnZucDPYBjUh1LKplZDWAscFGKQ6lqMgm61x1L0Kr7bzPr5O7fpTSq1DoXeMTd7zWzo4DHzayju29PdWDpSC126Wc10CrhfctwnoiISFmuEUXrmFkmQfepDZUSXWqU6bppZicAvwMGuHt+JcWWKqWdk32AjsDbZrYSOBKYnuYFVMryd5IDTHf3be7+JbCMINFLV2U5J5cAUwHc/T2gNtC4UqKrupJ2r67ELv18CLQ1szZmlkUwUHV6imMSEZGqoSzXiOnAheHrgcAsT++H3pZ6TsysGzCeIKlL93FTUMo5cfdN7t7Y3Vu7e2uCcYcD3H1uasKtFGX5b+dFgtY6zKwxQdfMFZUZZCUryzn5CjgewMwOI0js1ldqlFXPdOCCsDrmkcAmd19TETtWV8w04+5xM7saeB3IACa7+6IUhyVpwsyeJLhoNQ77y9/q7pNSG5WIlNWurhFmdhsw192nA5MIukstJygAMDh1ESdfGc/J3UA94JmwjsxX7j4gZUEnWRnPSbVSxnPyOnCSmS0GCoDr3T1tW7vLeE5GAhPN7NcEhVQuSvMfikq8VwJqArj73wnGGp4MLAfygIsr7Nhpfm5FRERERETSnrpiioiIiIiIRJwSOxERERERkYhTYiciIiIiIhJxSuxEREREREQiTomdiIiIiIhIxCmxk2rHzAaamSe8v8jMfkxRLC+b2SOpOLaIiEghM3vbzG5OdRzlZWY/mtlRe7uOSDpQYidVhpk9YmYeTtvMbIWZ3WNmdZN86KeBn5d1ZTNbaWajkhiPiIjIHgkTtPwwmSmcHkp1XMni7vXc/T0AMzvWzOK7W6eqCO91fpHqOCS96AHlUtW8AQwleJDjL4GHgLrAlYkrmVkmUFARD7l09y3Alr3dj4iISBXxR3f/U6qDqI7MrKa7b0t1HFI9qcVOqpp8d/+vu3/t7v8AngBON7PRZvZp2G3yCyAfqGtmPzOzCWa2zsx+MLN/mVmPxB2a2QVmtsrM8szsZaBZseU7dcU0s5PN7AMz22JmG8zsJTOrbWZvAwcCdxe2LiZs0ys8fp6ZrTazv5lZ/YTl2WGr5I9mttbMflvRJ09ERKQkZjbYzD4xs+/NbI2Zjd9Vjxgzq5Vwbf3ezD43s0EJy39pZu+YWa6ZfWFmI83MdrGv1uH18lIzW2Zmm8xsmpk1TVinkZk9Zmb/DadHzaxhwvJrzOzL8Dq/2szuSFjmZvYLM2sOvApkJLRUXlhsnczws59eLMZHzOzhhPeXhfccm8xsnpmdtJvzepGZLTez680sB5gfzr8j7Hn0Y3iORiRs80n48p+JLarhfcI94WfNNbPXzOzgXR1bpDgldlLVbSFovQNoAwwBBgFdCJK7V4AWwClAN+DfwCwz2x/AzI4AHgEmAF2Bl4DbdndAM+sHTAdmAocDfYB/Efz3ciaQE+5j/3DCzDoB/wy36xKu1xWYnLDre4ATgbOA48N4e5fzfIiIiOyJTQTX0AYEPWJ+CexqTN2FwP8Ah7l7feA4YBGAmbUHZgB3A02A/wWuJuhtszsXEFzzWgHbgSkJy54A9gUOC6fGwOPh8doBdwKnuPs+QAeCa+0O3P0boD9Bb5564fRosXXi4X4vKpxnZvWAgYTXazO7DLgBOC+M6XfA86UkWK2B5kBbgvMGsBj4BbAPcBnwf2bWN4yjS7jOSWGcl4bvJwKHAkcC+wEfAC+bWeF9kMhuqSumVFlm1pPgIvRmOCsLGOrua8PlxxEkT03C7pQAt5jZqQQXmLuAa4E33f32cPkyM/sf4JLdHPoW4Fl3T7zgLQj/zTOzAuAHd/9vwvLrgafd/d6E+K8E5oW/SuaFxxzm7q+Hyy8mSBJFREQq0u9sx7Hg/dz91YT3y83sQYJkqyQxoB7Q3szec/evE5b9CnjG3aeF7z8zs/vDfT22m5j+UHjdNLPrgc/DVjaAvkA7d98YLr8u3O/+QBwwoIOZrXL374D3d//xd+th4BMza+ru64CzgW/cfXa4/FrgNncvbFWbYWZvAYOBXXVv3Qbc6O75hTPcPTFxnWVmrxD8qPt6STsws8YE9zwHJtzn/AEYARwBvFP+jyrVjVrspKrpF3ZL2Aq8R9AC9//CZTmFX3ahw4FsYH1Ct4sfgY7AQeE6h4X7SVTaAOpu/JRMltXhwPnF4ng3XHZQOGUlHtvdfwQWlvM4IiIipbnd3RskTO+b2YlmNtvM1pvZ98AYgha3kkwhGOM+DthgZoktVm2Ac83su8IJuJWwB8turCzhdUuCFjyALxOWfxH+28rdVxC0nl0GfBN2Ad1l18jSuPsS4GPg/HDWxQTJXqE2wAPFPl8fgt5Bu7ImMamDou6jC81sY7iPU9n1+S48LsCChOPmEvRaarXrzUR+ohY7qWr+DQwn+PXrm8IByGHX/c3F1q0BrCXoTlLc90mMsSQ1+OkiWNxqoF3lhiMiIhIwsyzgReA3wGR332JmVwMlVngOuyyOAcaYWQPgfoKuir2BVeE+ripnGK35KWFrHf6bU2z58vB1YaXqr8N4nifoDpkFXAFMM7NG7p5X7BjbyxjLw8BVZjadoNvj4IRlq4Bb3f2ZMu5rp+Oa2dEE5+944AN3LzCzZwlaHgsVL/62Kvy3rbuvL8exRYqoxU6qmjx3X+7uq8pQVepjgkIo28NtEqd14TpLCL60ExV/X9w8gi/jXYkBGSXE0qGEOJaH3US/IEhWi45twaD1jqXEIiIisreygFrAxjCpa08wLq5EZnacmR0eju3aQvDDakG4+EFgsJmdamY1w4Ik7c3smFJiuMXMmllQVGwM8Ia7fxOOjfsncK+ZNTCzfYF7gVfdfY2ZHWJm/cwsm+A6uokgKSopifsvQfGUNiUsS/QUcDBwHzDT3VcnLBsHjDazrhaoExZeObSUfSaqT3C+1gNuZv9LMP6veKxtC9+E9y3/AB40sxYA4fk4IxwHKFIqJXYSZW8QdHecZmb9zayNmR1lZn8ws8JWvPuAE8zsJjNrGw6KPqOU/d4ODDKzP4UXqw5m9uvwogJBF5JfmlmLsE88BBepnmb2dzPrZmYHm9kpZjYeirpdTiL49fNEM+tA8Otn8QRRRESkQoXXoCuBu8KhAg8QJBG70oygyMhGYA1BNejh4b4+JShYNiJcto6gSNnuuhlC0L1zNkErXBY7Fls5H/gBWAp8BnzHT+P/soDfh8f6DrgGOMvdt5bwOZcBfwPmhN0ZSyzo4u6bgBcIkq3JxZZNJBij/3D4+b8iGHtfngImrxOMN5wDfEtQnOWFYuv8Drgt7Ko5Ppx3GcE5eNvMfiAYrjGInVv3REpkFfAYMJEKYWaPAI3d/ZQSlo0GBrp7x2Lz9yEYzHwW0JSga+a7wO/c/YtwnYuBPxBcdN4mqKT5V3e3cPlFwP3uXi9hvwMIxgx0JLjY/Ac42923mtmRwHjgEKBWwn56hLH0IkjYVgAvuPvvw+V1CS44ZxIUU/krwYDob939oj07ayIiIlWXmbUmGD/Xyt1VMEwkiZTYiYiIiEhSKLETqTzqiikiIiIiIhJxarETERERERGJOLXYiYiIiIiIRJwSOxERERERkYhTYiciIiIiIhJxSuxEREREREQiTomdiIiIiIhIxCmxExERERERiTgldiIiIiIiIhH3/wHD6K7eH0TGawAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HqexVS7C_OZY"
      },
      "source": [
        "from keras.wrappers.scikit_learn import KerasClassifier\r\n",
        "def nn_base_model():\r\n",
        "  nn=Sequential()\r\n",
        "  nn.add(Dense(20, input_dim=20, activation='relu'))\r\n",
        "  nn.add(Dropout(0.15))\r\n",
        "  nn.add(Dense(1, activation='sigmoid'))\r\n",
        "  nn.compile(optimizer='adam',loss='binary_crossentropy',metrics=['binary_accuracy'])\r\n",
        "  return nn\r\n",
        "nn_base=KerasClassifier(build_fn=nn_base_model)"
      ],
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "y7K8tL3ZDzlS",
        "outputId": "ed662002-b365-4eb5-b0fb-7bca3f9ab67a"
      },
      "source": [
        "nn_param_grid = {\"kerasclassifier__epochs\": [5,10,15],\r\n",
        "                 \"kerasclassifier__batch_size\": [5, 10,20]}\r\n",
        "nn_gs=pipeline_classifier(X_train,y_train,\r\n",
        "                         nn_base, nn_param_grid )"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "3662/3662 [==============================] - 4s 1ms/step - loss: 0.4651 - binary_accuracy: 0.7740\n",
            "Epoch 2/10\n",
            "3662/3662 [==============================] - 4s 1ms/step - loss: 0.2296 - binary_accuracy: 0.9193\n",
            "Epoch 3/10\n",
            "3662/3662 [==============================] - 4s 1ms/step - loss: 0.2082 - binary_accuracy: 0.9295\n",
            "Epoch 4/10\n",
            "3662/3662 [==============================] - 4s 1ms/step - loss: 0.2121 - binary_accuracy: 0.9228\n",
            "Epoch 5/10\n",
            "3662/3662 [==============================] - 4s 1ms/step - loss: 0.2059 - binary_accuracy: 0.9294\n",
            "Epoch 6/10\n",
            "3662/3662 [==============================] - 4s 1ms/step - loss: 0.2027 - binary_accuracy: 0.9314\n",
            "Epoch 7/10\n",
            "3662/3662 [==============================] - 4s 1ms/step - loss: 0.1962 - binary_accuracy: 0.9311\n",
            "Epoch 8/10\n",
            "3662/3662 [==============================] - 4s 1ms/step - loss: 0.1975 - binary_accuracy: 0.9303\n",
            "Epoch 9/10\n",
            "3662/3662 [==============================] - 4s 1ms/step - loss: 0.1915 - binary_accuracy: 0.9352\n",
            "Epoch 10/10\n",
            "3662/3662 [==============================] - 4s 1ms/step - loss: 0.1980 - binary_accuracy: 0.9335\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rwjvNwvJJTT8",
        "outputId": "e1180b70-bb4d-46aa-b7d7-4746910f4ecf"
      },
      "source": [
        "nn_gs.best_params_"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'kerasclassifier__batch_size': 5, 'kerasclassifier__epochs': 10}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 513
        },
        "id": "a0c3AiCqIAYJ",
        "outputId": "56cb6343-a276-401c-e14e-dd6e70a9125f"
      },
      "source": [
        "plot_roc_conf_matrix(y_test,X_test, nn_gs, \"NN (GS)\")\r\n"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.97      0.94      0.96      2275\n",
            "           1       0.84      0.92      0.88       725\n",
            "\n",
            "    accuracy                           0.94      3000\n",
            "   macro avg       0.90      0.93      0.92      3000\n",
            "weighted avg       0.94      0.94      0.94      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3YAAAFXCAYAAADuwtw3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdebxc8/3H8dc71thKal8iUdpaawmhSqO2WIPWEq1SNNYSVK0/VEsXQu2kqorGUmopIfZaSkisRWgQxJYQ+xJLPr8/vmdkMpm5d+7N3Dn33Pt+5nEeM/csM5+ZezNzPuf7/X6+igjMzMzMzMysuHrkHYCZmZmZmZnNGid2ZmZmZmZmBefEzszMzMzMrOCc2JmZmZmZmRWcEzszMzMzM7OCmz3vAMzMOjtJcwK/BAIYFhGf5RySmZmZ2QzcYmdm1rqTsuU1J3VmZmbWGRUmsZMU2fKBpIVr7LNHts+FNdaHpCtbeI67s32+1474zpP0iaSlamzvKWl/STdLek3SVEkfSnpO0ghJu2StAtWOXUHS+ZL+J+nT7LgXJN0q6VhJy1fsf3YWS++2vo72knRx2Xv8t1b2fb5s3580K8aWSJogKZdJHfN87s5A0iGSnsn+T4Sk65r0vCdkz7dHK/ttAhwG/DEiWvzbNjPraNn5xEGSbpP0uqTPJL0naaykUyWtmHeMHSE7f3pD0jztPH7t7DP/V+04tvw8srR8IelNSTdJ2rqOx5hH0mGS7pf0VvadN1HSPyRtUWcca0m6QNI4Se9nj/GKpOsl7VbrPNK6j8IkdmXmA46YheN3lLRqo4IByD5Efw4Mj4hXq2xfF/gfcA7wA+AF4FrgVuAjYBfgcuB5SQtVHLsZ8DiwDzAvcCdwAzAR+B7wG+BHFU/5e0DAbxvzCttse0k9q22QtD6wXCOfrCwh79PIx7WOJ+lHwGnAosD1wN9If+OdgqSvk2K6ETgq53DMrJuTtA7pfOIM0jnAs8DVwD3AYqSLUP+VtHtuQXYASZsCA4E/RMTH7XmMiHgYGAkcJalXO0N5k/Sd8DfS+z4J2BL4l6T/q3WQpO8A44BTgVWBh4F/Aq8CPwRGSrpK0lw1jp9d0nnAGGAIaSjVnaRzyZezGC4Bnm7n67KuIiIKsZDGtkwDPgE+Bhavss8e2X4X1lj/cXb7zxrPcXe2/XttjO0a4AtgqSrbVi973tOABavssxhwIinJW7psfU/Sh0iQuoHNUXHcvMCuwKAqj/nn7P1aqUm/n4uzOB/JbgfX2O/8bPvY7PYnDXju0u+tzyw8xjeAbzf77zp77gnpv2LznzvvBbg0+939IIfnXhj4NvC1FvbZCPg/YP683ysvXrx074WUEHyUfWYOq/xcIl3Q3QR4DDgh73gb/NrHAlOAnrP4OOtn798f2nhc6Tzy7irbDs62fQ4sW2X7slnsAZwLzFuxfRXgyWz7NTWev/RdOR7YsMr2XsDvgM/y/l15yXcpWovdNFJi0JP2XT2/Fnid1KK0ZiMCkrQ0MAi4Iypa6yQJuIwU7wkRcWhEvFv5GBHxZkQcB6wBvFe26XukloxXI+KYiPi84riPImJERFxfJbRLSR/y+87Cy2uPEcCXwExdLLMuAjsBT5A+xDqNiHg+IsblHUc3VOq6/EKznzgi3oqIcRHxXgv73BURv4mID5oZm5lZubLziXmA30bEYZWfS5HcDvQHbsohzA4haT1gTeAfEfHJrDxWRNxP+r7Zs1HdFiPiDFJr3OykxLrSBcBCwMURsX9EfFRx/H+BjUmtfztI2rl8u6QdSOdUk4ENIuKeKjFMiYijSL3CrBsrWmIHqZvhx8A+WVLVFp8AJ2f3T2xQPHsBswF/r7JtS2Bl4BXq6BYZEc9VfFAvkt1Oakdc95Ka538iae52HN9ebwC3A5tJWqRi29akD7dLax2cjSc8QdJ/ysYOvJZ1UVizYt8+2di072erXizv/162X2n83wBJm0u6S9K72boFs31mGudWWtfK0qeeNyUbE3GSpBeVxkk+L+nXrX2xSFoxi/+VrC/9m5KukLRyPc9b8VhzSNpX0n3Z6/9E0nhJf5W0VpX9t1Aaxzkli/k5SX+s7C6c7fvVeDVJq0i6LjvuY0kPVo4fKP1OSC1iMOPvrk+2T0iaUOO1DMi2X1zlNe4jabSkydlrfDl7HftW7FtzjJ2kXtlrfS577VOyxxhYI57I/l5mk/RLpfEPn2Z/u+dIWqDacWZmddocWI3Uda/F85eImBqp2yEw43dgtf1rfP999RkraXFJFyqNB/tC0lBJN2Tbt6nxmMoed5qkZbN1PSQNVqor8KxSvYAPJT0i6VctfB8OyW6rnWchaV1J12TP96mkSdljni5psSqHjCD12Ni+xvO1x1PZ7aIVsa1M+t19SqqsXFVETCINrYHUnbZcaUzg8RHxektBRMR99QZsXVPhEruIeBM4G5gLOKYdD/FnUqK1laT+DQipNGD2rirbSiez/4iIL9vx2K9kt6tK+m5bDoyIAP5NSqTadGwDXEa6crVzxfrdSK2uI1o4dh/gOGB+UteL60iJ7Y7AfyRtXLbvh6R+7m9mP1/D9L7v1Ypc7ArcTOrCejOpj3tLRUuurni80lJe3KPV32v2ZTUKOBr4GulK6jPA4dlzqMZx2wGPArsDb5PGVr5IavV8SNKGrT132WPNS0q4zyO1DD9IGtP2FvBj0u+mfP+jSGMRfpDFcB2p5fnw7LmXqPFUawGjgW+SxpA+Sbp6fGPF7+4+av/uPqz3dVVxGalVfyXS7/c60tXZNWnhS7WcUgGkh0mvde7sMR4lvRc3Szq8hcP/DvwaeB64BZgD2B+4XlLV37OZWR22ym7/Udl7p4MtQvo83Ap4gPTd+THpsxbS90c165O6IN4XES9l6+Yhff9vTvpevxG4H+gL/IH0PVHtvHQrUmL0QOUGpaIl9wM7ZI95XRZvT2AosEKVxyudr21VZVt7zZ/dVl6I3zK7HRURb7fyGCNI5yT9lBUJzG7XydZf3qBYrSvLuy9ovQvpj/qL7P7XgfeBzyjrz0zrY+wuzH4ekv18a8V+d9OGMXakQi5fAK/X2H5f9ng/budrno10FSiy5/kXcAiwITBPHccfmB376yb8fi7OnusnpMTpQ+DBsu29gKnA7ZX7VzzOd4Hlqjz+ltnv+zlANX5vfVqJLYCda+wzgTrGuWW/k5uzxzqrzvfmCKaPP/x62fpvkK6+RuVzA32y9/BDYLOKbQOz9+JlYM46Y7gwe557gEUrti0G9C/7eW1SwvoBsG7Z+rmAq7LHua7iMU4oe49/VbGtNP/b3VXiqvm7y9ZPqPF6BmTbL654zwJ4CVi4Yv85SF1YqsW8R8X6f2Xr/17+/pK6Rn+UvTdrVYk1sr/PZcrWL0pKLAMY0NH/D7148dI1F1IvnHaNSy/7DhxQY/tM339ln7FBKvIxd8X2uUlDRz6myhhk0kXEAIaUrZsT2I6Z6wUsQLrgWe2c4NvZ+gdqxF76DtmxyraVgcWqrF8g+xx/qQ3v4R4tfI/1At4hnactW7Htsuy4Y+t8ntL3xcbZzxtnP4/P+2/QSzGWwrXYAUS66nEG6WStZhWiFvyV1PKxqdoxtUGZlUgn+rXGZn09u32r2kZJZ2XdHMqXr7o1RGrl2wr4T/Y8W5MKsPwbeDfrCtGvhfhKca1e9ytqgEj9x/8J9Nf0qRh2Jn2oX1bzwHTsfyJipvFWETES+Afp6tsq7QztpoioOd1FnU4jJVa3kq4G1mP/7PawKLtiFxHPM73rRaWhpAT56Ii4tXxDRNxC+tJchjquOEpakvSl9BHww0hdPsof782IGF226kBSa/5ZEfFg2X5TgQNIX+Tbqno31Acj4o8V684gfel9V9IcrcU7C0pdfx+JiBn+z0XE5xFxb2sPIGk50v+zD4FfRNmcdZG6uJxLem8OrPEQB0XEK2XHTMqOgeldhs3M2qo0zdPkJj/vVNJn4aflK7OfryG1jO1Qvi37nN+RdAHyH2XHfBYR18XM9QLeZ/r3aWX3yNWy21rnWaXP/VsrN0TEU5F6eVWuf59Ub6F3taEF9ZI0d3YOdi2wIHBETG+dLCmdB9b7eyt9Py9ccdvs37sVVCETu8ww4F1gd0nfaMuB2YdKqY96rRPrepT6Uk9p5/GDSd3sypfvlO8QERMiYn1St4Y/kJK6D0lJ7TbAA5J2qvH4pbgqx7o1QymBKxVR2Y00xvGa1g6UNK+knST9XtKfS0kv0xO6al0r6nFDO48rxTUEOIj0BbNT1NG9Vmkuwd7ApIio1l23VteKzbLbWu9XKUlZp7UYSFdeZwNGRkQ9Xw4bZLczjWfIjr+F1H202kWRW6oc8znpQsocTP+S6gjPkpLXrZTmClq8HY9Rek23RES1/9eXZLcbVNn2Oam7a7W4AJZsRzxmZnl6JKpM45Sp1R1zICmhGRkR71QeJGnVbCzy2UpjvC8Gjs02V36/t3aeNTa7vURSvzZ0eW/v+dH3NX0c/yekbp/fI7U0DmvjY5k13Ox5B9BeEfGupGGkxOx44KdtfIhLSZU1B0j6QUS0Z+6sr2W3tSrmlVpnqp7MRsRX6yWdQHodVUXEf0gtd6UxW5sCfyS1Gl4gaWREVI5Nej+7XbD2S5hO0pGkbg/lxkXE7+s5vsIdpCtiP5Z0GbAecEW0Ul0wG9x9Bal7YC3zt7CtJS+38zgkbUQa2zkF2CZaqKRYoXQyX3kVD4CIeE/Su8z8O+qT3U5s5XuqnkRpmez2uTr2hekxT6ix/cXsdqkq216psg6m/x+pOkdPI0TE+5L2JHU7PRU4VdLzpIshl0eqFteaWXntb0TEF1XWd/hrN7Mur9QLodkXalv63rybNKfuDyQtHhFvZOtLid4MPXSylryLqFI1u0zl93tr51lHkbpcbpst70p6gNS1829VzotK2nR+VOZNpl/A/Bpp+MiiwLmSnoiIyorfpfPAen9vpUT2rYrbPC7QWwEVucUOUhevt4FdJVUmJC3KWltOyH5sb6tdaeqCWhXvHs9uGzK1QknWneEmUt/rj0kfTNUKpJQ+EGeaYqGGgczcgli1CmAdMX5Jao1aHjgzW91iN8yswMc/SEndSaQWuvmBHhEh0hwtUKPYSB0+bX2XqnEtTypyAvCjiBjfzudvi9L/zWrFW8qX0VWPzs+0Jj1P1c+uiLiKNBB/T9LfX8/s/m2SqlZUa6BmvXYz634ey25nqmDcAC2dC9b83oyIaaTP2dlIPZCQND8pwXqPVByl3KGkpO4pUpf3xUnjmMX0C1+V3+8tnmdlrYnrkC52n04ao7Y56ULsuBpDBqDt50cl4yJij2zZnlQg5rosvkurFH8pnQe2NGwGSNWYmX5R9/Gy2wCWU1bF26wlhU7sstafP5I+VE5ox0NcSfqA+a5qlDFvRakvdK8a20dmtztKmq0dj9+i7OrYM9mP1VpuSn3H6+qbHREDIkIVy4BZCLE0rcEWWQyjWtl/Q9LruCYijs36x38YEaXKlcu3cGyHyD5IbyT9jg+s0Z2yJaXSxMvWePwFqH7FcGJ2e1jZl0i15cI6Yii1otXbhfW17LZPje2l9bW65zTK56QCRdUsU2M9EfF2RPw1InaNiKVI3SbfIF0A2rSV5+wsr93MrFxpXrodJbW1t1VprPBMn6fZY9WqclyPyu6Y25MuqF2djcsuVxqLt0tE3JSN7y6Nt6v1/d7aeRYR8WVE3B5pruC1SD0qrsluT6pxWJvOj1p47k9JY9gnkYbSVLZGls4DN88St5bsQkpsHy6NE89uH8rWD56VWK17KHRilzmb1DS+E20sqpFdbSp1f2zPvHZPk04+v1Vj+8hsn2WY3n+8bq31Fc+Sxb7Zj9VONFfMbh+rsq3DRcRjpPLEb5OqF1brplau9EE7U3e+rORvrZPy0pdWQ7sWZ+/vVaTf7xkRMbytj5ENpH4FWFRSteIZu9Q49LbsthHz7NxNqgC2ZamEcitK4/dmKmOdHT+QdAWxo+fLeR34eo2YN6uyrqqs6ElpLGNrnxGl1zSwxpdwaVqIVguxmJk10CjgCVKyclxLO0qas6KwWukCY7VzlY2Yhe/OiHiCNK3NWpK+RY1umJma3/HUTlpKLVd198rKLnqXzulm+szPLqguCbxcbQxgW2VDM0o9io4ub7WLiKdIhV3mBk6p9RhK8/6Wfq+VY/VKBclOaG3suKT12xC6dUGFT+wi4mPSpOUC9mvHQ/yTlPisnS1tee6PSAN3F8+KZFRuL5X//5T0H/K0ahWYJM1H9RPObZQm5p6pm2XWbfE80lWs16gyvwvTC2v8u86X1HAR8d2IWDgiftX63l8Vmfhh+aSi2Wu9kNp94UutLLUS7PY6g5RM3sLME4a2xXnZ7bDyZEFSX2p/QQ8jDcw+VdIPKzdKmkvSjyQt3dqTR8RrpKIf8wFXq2LieEmLVszpeA6pW+EvytdnYzvPJlXrvCEiJrT23LPo7ux2hvdI0k+pchIgaQ1JP1TFJLeS5mF6RcpaYwAByCqy3kR6r84sr+IpaT1SVdBppPfBzKwpsvOJ3UjDL/5P0qlZt8cZZOPUH2D6HLsw/bN0f0mLlu27PHBWA8IrJXGHkoaITKT6eUfpO37/8pWSNqHGd2xEPEfqcbF65Wd7duyhqj6vamn+uGqf+WuTzhkbeW50ASmB/hapImi5fUhdPvfMqqHPU75R0kqkwluLAddGRfXuiPgn6T1eFLhX0kzFuyQtKOk3VJ9T2bqRwhZPqXA+aTLhNledi4iQdBypYuI8re1fxY3AuqSrXjNNih0Rj2aFN64mzUF3gKSHSR98c2Yxr0bquvAKM37Q9CB9QOwo6Q3SJMnvkAbR9iNd/foI2K28LDt81dr3fdKHyX/a8bqaLiLGShpF6h//nKS7SfPCbEg6mb6Y1OWh0g2k8YAjJN1K6ttPROzd3lgkLUM6iYeUmP+lRgPqLytL61cxjDQtwfrAeEl3kn73GwN3klrTZrgwEBHjJQ0mTVh6taTxpG63HwJLk8ZtzkuabHwirTuY9IXzfeBFSfeS3qdls8c6j2y8XkQ8JOlY4GTgvuz38FYW/zLAeNp3EaWtfg/8iJRgDiCdFHyLVDDodNJJRLllSf/PPpA0hvQlOx+pYlkv0snO9XU87z6kFrkfAxtI+g/p/9wAUrfvIyJibO3DzcwaLyKekPQDUjfDw0iJ2oOkz7r5SVMbLUP6TvlT2aF3k75rfgA8Jel+0hiz/qTvz7mpMVygTiNILVZDSj+XDaEo9wdSj4+TJe1I+kxfllRc7Q+kOV+ruRHYO9uvMhk7DjhF0pOkAmHTSN8Rq5KS4Gq9sTbKbm+qsq1dIuITSb8nXRA+WtJVpfcgIiZk54H/Ik2Vs7uk+0jnZ8uRLsKL9P21W9UngJ+Rzvf2Ae7JzgmeJE1HsUz2GHNQf5E066qiE0ymV89C2QTlNbbvz/TJNFucoLzG8aPLjq9rgvLsuCVI3TFvbWW/nqREYRTpQ/gz0kn6eNJYv12pPgHolqQPiodILVOfk6o5PU5KGJat8XwbZq/lzCb9fi6mDZOn1to/e82/Jk0p8Gn2mi8mfXCdQJUJpbPjhpLGS35a+j1Wea4BLcQzoeKYPmV/Dy0tfep8vfOQEqWXSB/EL5L6/s9Z+dwVx32D1IL2HKkF7/3svbmclPTXNUF59lhzkqZrGE2qMPZx9vd3EbBmlf23JHUJfSeL+X+kriS9quxb83eTbb+72vtVa33Z9rWzGD7IlrtIyekAZp6gfHHg6Gz/l7O/hUnAg6Qv0571xkwq1X1K9v5Mzd6D24AtasQZtGEydS9evHhp75J9nxxEauV5k3Re8B4whpQgrVDlmPlJrXOvZZ9pzwFHki5WzfQd1NbPLVLiWPpeXLWF/dYmdU18i3QO9DDws2xb1c/R7JgAhlfZthupNeuZ7D34MLt/LvCNGjE8Txpb15bvzz2oMUF52T5zky60BrBtjd/bL0kX26dkv4dXSQld1e+WKo/Rj9Q6+Gz2nTiV1ChwHakny+x5/316yXdRRLWLKtYWkq4iDQpeNmrP99JUki4kVQNcNVIfbzMzM7PCkfQQqcDKUhHxySw8zvdIPTL+GBG1WgjNCqvwY+w6ieNIV2g6xYdENu7qJ8BlTurMzMys4I4mDT+Z1WEAR5G6QP5hliMy64Sc2DVARIwjFfcYIqnaxMXNdiQp0WxzJU4zMzOzziQibicVMvtVZfGRemWVQrcEfhcRUxoZn1ln4a6YZmZmZmZmBecWOzMzMzMzs4IrzHQHPdc40E2L1mHefODMvEOwLm6BuXtUnS+jPdrzefjJo2c37Pmt81l44YWjT58+eYdhZmYdbOzYsW9FxCLVthUmsTMzM7Pq+vTpw5gxY/IOw8zMOpikl2ptc2JnZlY0ci96MzMzm5ETOzOzopF7VZqZmdmMnNiZmRWNW+zMzMysghM7M7OicYudmZmZVXBiZ2ZWNG6xMzMzswpO7MzMisYtdmZmZlbBiZ2ZWdG4xc7MzMwq+OzAzKxopLYv1ilIukjSJEn/rbFdks6UNF7SE5LWbHaMZmZWTE7szMyKRj3avlhncTEwsIXtWwArZMsQ4LwmxGRmZl2Au2KamRWNW+AKKyLukdSnhV0GAZdERAAPSlpQ0hIR8XpTAjQzM0aMfpnrH3u1oY8570fvs+wKS3P8Nis39HHLObEzMysat8B1ZUsBr5T9PDFbN1NiJ2kIqVWP3r17NyU4M7N6dERi1EyjX5wCQP++vWb5seb4fCq7/vNc1h9zGxeede0sP15LnNiZmRWNW+wMiIjhwHCAfv36Rc7hmFmm6ElNIzQyMcpD/769GLT6UuzafxYvmj3zDOy8Mzz5JBx0EIftvG5jAqzBiZ2ZWdG4xa4rexVYpuznpbN1ZtYgHZ14FT2paYSGJUZFFwEffAAjR8IWW3T40zmxMzMrmg5I7CQtA1wCLAYEMDwizpDUC7gS6ANMAHaKiHckCTgD2BL4GNgjIh7JHmt34NjsoX8bEX9reMBd1w3AgZKuAPoD73l8nXVVebVsdXTi5aSmm5s0CUaMgKFDYaWV4LnnYI45mvLUTuzMzIqmR4d0xfwCOCwiHpE0PzBW0m3AHsAdEfF7SUcCRwJHMGP1xv6k6o39s0TweKAfKUEcK+mGiHinI4IuGkmXAwOAhSVNJL1XcwBExPnASFKyPJ6UMP8sn0jNWtaIpCyvli0nXtZhbr4Z9tgD3nsPttoKVlihaUkdOLEzMyueDmixy1qFXs/ufyDpGVLRjkGkRATgb8DdpMSuavXGbN/bImIKQJYcDgQub3jQBRQRg1vZHsABTQrHupFGt441IilzgmVdxqefwhFHwJlnwiqrwB13pKSuyZzYmZnZDLJy/GsAo4HFyroCvkHqqgm1qzfWWm9mHaSepK3RrWNOyswyETBwIPz733DQQfCHP8Dcc+cSihM7M7OiaUdVzPLS+JnhWVXFyv3mA64BhkbE+yp7rogISa6+aNYBZqVFrZ6kzYmYWYNFpKVHDzj00NRi14QCKS1xYmdmVjTt6IpZXhq/5sNKc5CSur9HxD+z1W+WJsjOulpOytbXqt74KtO7bpbW393mgM0KrD1J2qy0qDlpM2uySZPgZz+DTTaBQw6BbbfNOyLAiZ2ZWfF0wDx2WZXLvwDPRMRpZZtuAHYHfp/dXl+2fqbqjZJGASdLWijbbzPgqIYHbNZJVEvi2pOkOTkzK4jyAilbb513NDNwYmdmVjQdM4/d+sBuwJOSHsvWHU1K6K6StBfwErBTtq1q9caImCLpN8DD2X4nlgqpmHVGs1pUpFoS5yTNrAuqViBllVXyjmoGTuzMzIqmA1rsIuI+oNYDb1xl/5rVGyPiIuCixkVnNmtaSt5mtaiIkzizbuKRR+Dss3MvkNISJ3ZmZkXTMS12ZoU0qxUhnZiZWU0R8MAD8N3vpmXcuFymMaiXEzszs6LpgBY7s6KoTORcEdLMOkSpQMrNN8OYMbDmmp06qQMndmZmxeMWO+tGWkvknLSZWcOVF0g54wxYY428I6qLEzszs6Jxi511AfUWLXEiZ2ZNdcQR8Mc/psIot98Oq66ad0R1c2JnZlY0brGzAisldPUWLXEiZ2ZNteSS8ItfpAIpPXvmHU2bOLEzMysaJ3ZWUCNGv8zR1z4JOGEzs04iAs45B5ZYAn74Qzj44LwjajcndmZmReOumFZA5Undyduv6oTOzPJXKpAyciTsumtK7ArMiZ2ZWdG4xc46qXrmi3NSZ2adQnmBlDPPhAMPzDuiWebEzsysaNxiZ51IeTLn+eLMrBAeeQS23LKQBVJa4sTOzKxo3GJnncj1j73K06+/z0pLLODkzcw6tw8+gPnnT3PSXXIJ/OhHhSuQ0hIndmZmReMWO+skRox+mdEvTqF/315cuc96eYdjZlZdqUDKccfBvffCyivDbrvlHVXD+bKvmZmZtUupC+ag1ZfKORIzsxomTYJttklTGKy7Liy8cN4RdRi32JmZFYzcYmedQHlrnbtemlmndMstqUDKu+9OL5DShb9DndiZmRWMEzvLW/nUBW6tM7NO6/bbYZFF4LbbukyBlJa4K6aZWdGoHYtZA5W6YHrqAjPrdJ56Ch56KN0/6aR0vxskdeDEzsyscCS1eTFrFHfBNLNOqVQgpV+/NJ4uAuaaq0tVvWyNu2KamRWMEzXLkwummFmnM2kS7Lkn3HQTbLEF/PWvXXosXS1O7MzMCsaJneXFrXVm1uk8/zysv363KZDSEid2ZmYF48TO8uLWOjPrdPr2hR12gP326zZj6WrxGDszs6Jx8RTLgVvrzKzTeOop2HhjeO016NEDzj232yd14MTOzKxwXDzFms3TG5hZp1BeIOW//4UJE/KOqFNxV0wzs4JxombN5ukNzCx3kybBXnvBjTdOL5Cy2GJ5R9WpOLEzMysYJ3aWB3fBNLNcHX98mmi8mxdIaYkTOzOzgnFiZ2Zm3cKnn8Lbb8NSS8HJJ8P++3ssXQs8xs7MrGg6oHiKpIskTZL037J1V0p6LFsmSHosW99H0idl284vO2YtSU9KGi/pTDkLNTOz9njqKejfHwYNgmnTYKGFnNS1womdmVnBdJmNzNMAACAASURBVFDxlIuBgeUrImLniFg9IlYHrgH+Wbb5+dK2iNi3bP15wM+BFbJlhse04ilVwzQza4ryAilvvAEnnpgqX1qr/C6ZmRVMRyR2EXEPUPXsPWt12wm4vJW4lgAWiIgHIyKAS4Dt2vwCrVPx3HVm1jTvvAPbbpvG0G20ETzxBGy5Zd5RFYYTOzOzgmlPYidpiKQxZcuQNjzlBsCbEfG/snV9JT0q6d+SNsjWLQVMLNtnYrbOCs6FU8ysKeaeO7XSnXkm3HSTq162kYunmJl1AxExHBjezsMHM2Nr3etA74h4W9JawHWSVp7VGM3MrBv69FM45RQYOhTmnx8efBBmmy3vqArJiZ2ZWdE0sRyJpNmBHYC1SusiYiowNbs/VtLzwDeBV4Glyw5fOltnZmY2s6efhsGDU5fLb3wDdt3VSd0scFdMM7OC6aDiKbVsAoyLiK+6WEpaRNJs2f3lSEVSXoiI14H3Ja2bjcv7KXD9rDy55cuFU8ysQ0TAuefCWmulrpc33ZSSOpslTuzMzAqmIxI7SZcDDwDfkjRR0l7Zpl2YuWjKhsAT2fQHVwP7RkTp7H9/4EJgPPA8cPOsv2LLiwunmFmHOP54OOAAF0hpMHfFNDMrmI6YGi4iBtdYv0eVddeQpj+otv8YYJWGBme5cuEUM2uYL76A2WeHvfdOhVH23x883WnDuMXOzKxgmtwV08zMbNZ8+ikccghst13qhtm7d2qx8/dTQzmxMzMrGrVjsU5D0kBJz0oaL+nIKtt7S7orm07iCUnuo2RmxfX009C/P/zpT7DccvD553lH1GU5sTMzKxi32BVXVnTmHGALYCVgsKSVKnY7FrgqItYgjXE8t7lRmpk1QLUCKWeeCXPOmXdkXZYTOzOzgnFiV2jrAOMj4oWI+Ay4AhhUsU8AC2T3vwa81sT4zMwa4/334aSTXCCliVw8xcysYJyoFdpSwCtlP08E+lfscwJwq6RfAPOSppwwMyuGe++FddeFr30NHngAllnGY+maxC12ZmZF4zF2Xd1g4OKIWBrYErhU0kzf15KGSBojaczkyZObHqSZ2QxKBVI23BDOOSet693bSV0TucXOzKxg3GJXaK8Cy5T9vHS2rtxewECAiHhA0tzAwsCk8p0iYjgwHKBfv37RUQGbmbXq6adh8ODU5fLAA2GfffKOqFtyi11BLL3Ygtwy/CAeueYYxl59DAcMHgDADpuswdirj+GjsWey5kozzzO0zOILMfn+YQzdbeMZ1vfoIR64/AiuOWPfZoRvBXPiccew2YD12XmHbb5ad97ZZzD4R4PYdaftOXCfvZg8KZ1jTnjxBfbcbRe+2281Lv3bRXmF3K14jF2hPQysIKmvpDlJxVFuqNjnZWBjAEkrAnMDbpIzs87piitmLJBy1lnQs2feUXVLTuwK4osvp3Hkaf9kzR+exPd/eir77Lwh315ucZ56/jV2OezP3PfI81WP+8NhO3Dr/U/NtP7AXTfi2Rff7OiwraC2HrQdZ543fIZ1u+2xF5dffT0jrrqW7204gAsvSIX6Fljgaxx2xDH8ZPc98wi1W3JiV1wR8QVwIDAKeIZU/fIpSSdK2jbb7TDg55IeBy4H9ogIt8iZWef0rW/B5pu7QEon0LSumJK+Tar8tVS26lXghoh4plkxFNkbb73PG2+9D8CHH09l3ItvsOQiC3Ln6HE1j9lmwGpMePVtPvrksxnWL7Xoggz83sr84S+jOOgnP+jQuK2Y1lxrbV57dcbeYfPNN99X9z/59JOvusz3+vrX6fX1r3P/vf9uZojdmhO1YouIkcDIinXHld1/Gli/2XGZmdXt1lvh7rvh5JNhjTXguuvyjshoUoudpCNIJZ0FPJQtAi6vNjmrtaz3Er1Y/VtL8/B/J9TcZ96ec3LYzzblpAtGzrTtlMN/yDFnXMe0ab4AbG1z7ll/YqvNNuKWm/7FPvsflHc43ZeLp1gTjBj9MqNfnJJ3GGbWmUydmgqkbL45/Otf8OGHeUdkZZrVFXMvYO2I+H1EXJYtvyfN57NXrYPKK3598dbM3Qm7o3l7zsnlp+7N4adewwcffVpzv2P33YqzLrtzpta6LTZYhUlTPuDRZ16pcaRZbfv/Yig33XoXA7fahquu+Hve4XRb7oppzXD9Y6nVftDqS7Wyp5l1C08/DeusA3/6UyqQ8tBDUNabx/LXrK6Y04AlgZcq1i+RbauqvOJXzzUO7PbNS7PP3oPLT/05V948huvvfLzFfddeZVm232R1Thq6HV+bvyfTpgWffvY5Sy66IFt/f1UGfm9l5ppzDhaYd24u+u1P2fPYS5r0Kqwr2GLLrTn4gH3YZ/9f5B2KmXWg/n17sWv/mQtzmVk388knaaLxCLjxRthqq7wjsiqaldgNBe6Q9D+mT8zaG1ieNIjc6nD+8T/m2Rff4MzL7mx13032+tNX94/ZZ0s++ngq5195DwDHnZUKsG2w1goM/enGTuqsLi+/NIHey/YB4N933UmfvsvlG1A35hY4MzNrinfegQUXTFUu//53WGUVWHzxvKOyGpqS2EXELZK+Sep6WV485eGI+LIZMRTdd1dfjh9v3Z8nn3uVB69IwxKPP/sG5ppjdk47YkcWXmg+/nnmvjzx7Ktse8A5OUdrRXfMEYcxdsxDvPvuu2y16QCG7Hcg9993Dy9NeJEePXqw+BJLctSxJwDw1luT2X3wjnz00YeoRw+uuOwSrrz2xhmKrVhjOa8zM7MON2oU7L47/PrXaV66TTbJOyJrhYpSQdldMa0jvfnAmXmHYF3cAnP3aFg6tsLht7T58/B/pwx0OtiF9evXL8aMGdPQx9z5ggcAuHKf9Rr6uGbWyU2dCkcemcbSrbwyjBgBq62Wd1SWkTQ2IvpV29a06Q7MzKwx3GJnZmYd4umnYddd4fHHU4GUP/7Rk40XiBM7M7OC8Rg7MzPrEC+/DK+/7gIpBeXEzsysYJzXmZlZw0yeDHfdBTvtBAMHwgsvwLzz5h2VtYMTOzOzgunRuOF6ZmbWnd16ayqQ8sEHaTqDRRZxUldgzZqg3MzMGkRq+2JmZvaVqVPhkENg882hVy/4z39SUmeF5sTOzKxgJLV5MWuLEaNfZvSLU/IOw8w6whdfwPe+l6peHnAAjBnjqpddhLtimpkVjPM062jXP/YqAINWX6qVPc2sMCLSF8jss6ful8cfD1tvnXdU1kBusTMzKxi32Fkz9O/bi1379847DDNrhMmTYdCgVO0S0lQGTuq6HCd2ZmYF0xGJnaSLJE2S9N+ydSdIelXSY9myZdm2oySNl/SspM3L1g/M1o2XdGTDX3wnIGk5SUdLOif7+VuSVs47LjOzqm69NXW1HDUK3ngj72isAzmxMzMrmA4qnnIxMLDK+tMjYvVsGZmeXysBuwArZ8ecK2k2SbMB5wBbACsBg7N9uwxJmwKPA+sCu2WrFwZOzS0oM7Nqpk6FQw+dXiDl4Ydh773zjso6kBM7M7OC6YgWu4i4B6i3WsYg4IqImBoRLwLjgXWyZXxEvBARnwFXZPt2Jb8HdoyIbYEvs3WPAGvmF5KZWRXXXw+nn+4CKd2IEzszs4JpT4udpCGSxpQtQ+p8ugMlPZF11VwoW7cU8ErZPhOzdbXWdyXfiIhbsvsBEBGfAHPkF5KZWSYCxo1L93fcEUaPhrPPhp49843LmsKJnZlZwbSnxS4ihkdEv7JleB1PdR7wDWB14HVgWIe+sGJ4RdIq5SskfQeYkE84ZmaZUoGUfv3g5ZfTVb111sk7KmsiJ3ZmZgXTrAnKI+LNiPgyIqYBfyZ1tQR4FVimbNels3W11nclZwL/lPQTYDZJPwQuA07PNywz69bKC6T89rew9NJ5R2Q5cGJnZmZVSVqi7MftgVLFzBuAXSTNJakvsALwEPAwsIKkvpLmJBVYuaGZMXe0iPgzcApwBDAb8GvgjIi4NNfAzKx7ioDDDpteIOWhh2DoUOjhU/zuyBOUm5kVTEfMSyfpcmAAsLCkicDxwABJq5PGkk0A9gGIiKckXQU8DXwBHBARX2aPcyAwipT0XBQRTzU82Jxlyd2f847DzAwJPvkkFUg55RSPpevmnNiZmRVMR8w3HhGDq6z+Swv7nwScVGX9SGBkA0PrVCQ9ExErVln/ZESsmkdMZtbNRMD558Paa6fxdGef7RY6A9wV08yscDpiugOrW62BKx7QYmYdr1QgZf/94S/ZtTcndZZxi52ZWcE4T2s+SUdnd2cvu1+yPDNO82Bm1ni33gq77w5TpqT56Q46KO+IrJNxYmdmVjBugcvFptntHGX3AaYBbwB7Nj0iM+s+Ro2CgQNhpZXgllvgO9/JOyLrhJzYmZkVjPO65ouIjQAknRURv8g7HjPrJj77DOacEzbeOBVHOeAAF0ixmtwp18ysYDzGLj9O6sysKSLgvPPg299O4+pmnx1++UsnddYiJ3ZmZgXTrAnKbWaSeko6SdKDkp6X9EJpyTu2Rhkx+mVGvzgl7zDMuq/Jk2G77VKBlG9+E6ZNyzsiKwgndmZmBeMWu1ydDgwCLgUWA4YBU4GL8gyqka5/7FUABq2+VM6RmHVDt94Kq62WxtGdfjqMHAmLLZZ3VFYQTuzMzArGiV2utgG2jYhzgC+y2x8CG+UbVmP179uLXfv3zjsMs+7n3HOhVy946CEYOtRTGVibuHiKmVnBOE/L1XwRUep2+ZmkOSPiaUlr5xqVmRXX00/DPPNAnz7w17/C3HN7LJ21iy8DmJkVjFvscvWipBWz++OAPSXtAryXY0xmVkSlAilrrQUHH5zWLbSQkzprN7fYmZkVjPO0XP0O6A08A/wGuBaYE9gvz6DMrGAmT4a994YbboDNN4cLLsg7IusCnNiZmRWMW+DyofTG3wW8BRARt0laCJgzIj7KNTgzK44nnkjJ3JQpqUDKQQd5LJ01hBM7M7OCcV6XGwEvAfMD0wAi4nPg8zyDMrOCWW456N8ffv1r+M538o7GuhBfHjAzK5geUpsXm3URMQ14AVgo71jMrGCeeQZ22QU+/hjmmw+uu85JnTWcEzszM7P6DQNGSFpf0tKSliwteQdmZp1QBJx/fiqQcscd8NxzeUdkXZi7YpqZFYwb4HJ1YXa7MRDZfWX3Z8slogYaMfplRr84hf59e+UdilnxVRZIufhiWHzxvKOyLsyJnZlZwbh4Sq765h1AR7r+sVcBGLT6UjlHYtYF7L033HKLC6RY0zixMzMrmB7O63ITES/lHUNH69+3F7v27513GGbFNHVqWhZYAIYNgxNP9Fg6axpfOjAzKxhPUF5ckgZKelbSeElH1thnJ0lPS3pK0ohmx2hm7fTMM7DuurDnnunn5Zd3UmdN5cTOzKxgpLYvlj9JswHnAFsAKwGDJa1Usc8KwFHA+hGxMjC06YGaWduUF0iZOBF23z3viKybcmJnZlYwasc/6xTWAcZHxAsR8RlwBTCoYp+fA+dExDsAETGpyTGaWVu8/TZsvz3stx9ssEGafHybbfKOyropJ3ZmZgXTQ21frFNYCnil7OeJ2bpy3wS+Kel+SQ9KGljrwSQNkTRG0pjJkyd3QLhm1qrPP4exY1OBlJtvhiWWyDsi68ac2JmZFYzH2OVL0k8k3SbpieznDSXt0KCHnx1YARgADAb+LGnBajtGxPCI6BcR/RZZZJEGPb2ZtWrqVDjnHPjyyzR9wf/+B0OHuuql5c5/gWZmBdMRY+wkXSRpkqT/lq07RdI4SU9IuraUYEjqI+kTSY9ly/llx6wl6cmsOMiZ6mJZpaRDgV8DNwOl0pGTgV/VcfirwDJlPy+drSs3EbghIj6PiBeB50iJnpl1BuPGpQIpBx4It92W1s09d74xmWWc2JmZFUwPqc1LHS4GKrv93QasEhGrkRKMo8q2PR8Rq2fLvmXrzyONE1shW2p2JSyo/YAtIuI0pk9Q/hywfB3HPgysIKmvpDmBXYAbKva5jtRah6SFSV0zX2hA3GY2KyLgggtgzTVTgZQbboCBXe3jzYrOiZ2ZWcF0RItdRNwDTKlYd2tEfJH9+CCphamFuLQEsEBEPBgRAVwCbNee19iJ9YqI57L7pcROZfdryt7LA4FRwDPAVRHxlKQTJW2b7TYKeFvS08BdwOER8XZDX4GZtd3BB8O++7pAinVqnqDczKxgcurduCdwZdnPfSU9CrwPHBsR95IKgUws26dacZCie1rS1hFxY9m6gcDj9RwcESOBkRXrjiu7H8Ch2WJmeYtIV8cGD4a+fVOC57F01kk5sTMzK5j25HWShgBDylYNj4jhdR57DPAF8Pds1etA74h4W9JawHWSVm57VIV0NHCTpKuAuSSdRepSuXW+YZlZQ02dCsccA9OmwWmnwXrrpcWsE6uZ2El6kjq6lgBk4y/MzKwJ6hwzN4MsiasrkSsnaQ9S0rJx1ppEREwFpmb3x0p6njQW7FVm7K5ZrThIoUXEvZLWJY21u4s0pGFARDyVb2Rm1jDjxqUWuscegwMOmN5qZ9bJtdRid3XTojAzs7o16/Qim0PtV8D3I+LjsvWLAFMi4ktJy5GKpLwQEVMkvZ8lPqOBnwJnNSncpomIp4Ff5B2HmTVYBAwfDoccAvPOmwqkeCydFUjNxC4ift3MQMzMrD4dMcZO0uWkaowLS5oIHE+qgjkXcFv2nA9mFTA3BE6U9DkwDdg3IkqFV/YnVdjsSZoS4OaGB5sjSeOBi4CLI+K1vOMxswZ65ZWU1G2wAVx8sScbt8LxGDszMyMiBldZ/Zca+14DXFNj2xhglQaG1tn8FtgDOEHSHaQk77qI+DzXqMys/Z54AlZdFXr3hgcfhFVWcYEUK6S6/2ol/UzSrdlktS+ULx0ZoJmZzaiH2r5YY0TExRExAFgRGAOcArwu6YxcAzOztps6FX75S/jOd+Ca7FrVaqs5qbPCqusvV9LhwDBgLNCHNIHqf4FepKuVZmbWJJLavFhjRcTzEfF/QGk84YE5h2RmbTFuHKy7LgwbBvvvD1tumXdEZrOs3ksSPweGRMRRwOfA2RGxLSnZW7ajgjMzs5l1xATlVj9Js0kaJOk6YAKwILBPvlGZWd0uuQTWXBMmTkwFUs45B+aZJ++ozGZZvWPslgYeyu5/AiyQ3b88W//zBsdlZmY1uAUuP5JOBwaT5vW7FDgiIp7NNyoza5P553eBFOuS6k3s3gAWBl4GXgLWAx4DlqfOue7MzKwxPGYuV0sBPwNGRcS0vIMxszrdfju88AIMGQLbbw/bbefuDNbl1NsV805g2+z+X4DTJN0FXAn8syMCMzOz6jzGLj8RsVNE3OykzqwgSgVSNt0UzjsPvvgirffnonVB9bbYDSFLAiPifEnvAOuTyl1f0EGxmZlZFT4daS5Jv4yIU7P7R9faLyJObl5UZtaqceNg113h0UdTgZRTToHZPdOXdV11/XVnVyanlf18Jam1zszMmqyHrzQ32w+AU7P7m9bYJwAndmadxVtvwdprw9xzw/XXw7bbtn6MWcHVldhJWrOl7RHxSGPCMTOz1jiva66I2LLs/kZ5xmJmrfj005TMLbwwnHsubLKJC6RYt1HvGLsxwMPZ7Ziyn0uLmZk1icfY5UfS8Brrz2t2LGZW4fbbYfnl4Y470s+77eakzrqVehO7vsBy2W1f4JvALsCTwNYdE5qZmVXjeexytUuN9Ts1NQozm668QMoCC6TWOrNuqN4xdi9VWT1e0nvA8cDNDY3KzMxq8hi75pP03exuD0nrMWMNmxWAj5oflZlVLZDiycatm5rV0kAvAqs3IhAzM6uP87pc3JfdBnB/2foAXgeOaXpEZpa6X77yigukmFF/8ZRelauAJYATgGcbHFNV7zx8djOexrqpe/43Oe8QrIvbbMVFGvZYHjPXfBHRA0DSYxHhC5pmeXrrLXj6adhwQzjgANhlF3e/NKP+Fru3SFclywl4Bdi5oRGZmVmL6h0cbY3npM4sZ7ffDj/9KXz5JUyYAD17Oqkzy9Sb2FWWd54GTAbGR8QXjQ3JzMxa4ha75vIE5WadwNSpcMwxMGwYrLgijBiRkjoz+0q9id2LwCsRUdlqh6TeEfFyY8MyMzPrNDxBuVmePvwwdbt89FHYbz849VQXSDGroi2J3RLApPKVkr6ebZutwXGZmVkNPdxg11SeoNwsZ/PNBxttBCec4AIpZi2od6iGmHmMHcB8wKeNC8fMzFrTQ21frGNIGiDpe3nHYdblvPUW7LwzPPlk+nnYMCd1Zq1oscVO0pnZ3QB+J+njss2zAesAj3VQbGZmVoXH2OVH0q3AbyPiHkkHA78DvpR0XEScnnN4Zl1DqUDK22/DNtvAqqvmHZFZIbTWFbP0P0nAisBnZds+Ax5h+rgDMzNrArfA5Wp14IHs/s+BzYB3gesAJ3Zms2LqVDj22DSG7tvfhpEjYXUXojWrV4uJXWksgaS/AgdHxPtNicrMzGpyg12u5oyIzyUtBiwaEfcBSFo057jMiu/cc1NS5wIpZu1Sb/GUo4AFgBkSO0lLA59HxJuNDszMzKrr4cwuTy9I2h34BnAnfFVIzOPNzdojAt54A5ZYIk02vuqqsMkmeUdlVkj1Fk+5DNiiyvrNgUsbF46ZmbWmRzsWa5hfkaY1+AlwUrZuK+Dh3CIyK6q33oLtt4f+/eG992DOOZ3Umc2Cer/v+wH3VFl/b7bNzMyaRGr70vpj6iJJkyT9t2xdL0m3SfpfdrtQtl6SzpQ0XtITktYsO2b3bP//ZS1bXUpE3B4RS0XEchGRlevjcmC7POMyK5zbb4fVVkvj6IYOhfnnzzsis8KrN7GbHZiryvq5a6w3M7MO0kNq81KHi4GBFeuOBO6IiBWAO7KfIfXgWCFbhgDnQUoEgeOB/qSqyceXksGuRNJ8knaW9EtJOwNzRcTnecdlVgiffw6HHw6bbgpf+xqMHg2HHgo93LfAbFbV+79oNLBflfUH4O4nZmZN1REtdhFxDzClYvUg4G/Z/b8xvVVqEHBJJA8CC0pagtQ9/7aImBIR7wC3MXOyWGiSVgaeI1WEHpTdPidplVwDMyuK2WaDxx+HffeFsWNhjTXyjsisy6i3eMoxwJ2SViMbLA78AFgT2LgjAjMzs+raM92BpCGk1rWS4RExvJXDFouI17P7bwCLZfeXAl4p229itq7W+q7kT8AFwIkREUqTCh4LnIG/D82qi4CLLoLNN4ell4Ybb0zj6cysoepqscuuyK4HTAB2yJYXgHUB16I1M2ui9nTFjIjhEdGvbGktqZtBRAQQHfSSimRN4OTs/Si9L78nzW9nZpXeegt22AH23jtNZwBO6sw6SN0dmiPi8Yj4cUSsTOpu8xxwLTCqo4IzM7OZdURXzBrezLpYkt1Oyta/CixTtt/S2bpa67uSd4E+Fev6UDEdUBGNGP0yo1+s7I1rNgtKBVJuugmGDYPf/jbviMy6tLoTO0mzSdpB0k3Ai6SxFucDy3dUcGZmNrMeavvSTjcApcqWuwPXl63/aVYdc13gvazL5ihgM0kLZUVTNqPrXfz7G3CTpD0l/UDSnsC/SMVnCu36x1IOPmj1rtZ71nJxxRUukGLWZK2OsZP0LWBv4KfAR8AIUovdbhHxdMeGZ2ZmlUTjJyiXdDkwAFhY0kRSdcvfA1dJ2gt4Cdgp230ksCUwHvgY+BlAREyR9BumF9U6MSK6WhPQScDnwBGk1slXSEndKTnG1DD9+/Zi1/698w7DimzatJTAbbEFHHUUHHsszONRO2bN0GJiJ+leYBXgGmCniPh3tv6IJsRmZmZVzEILXE0RMbjGppkKgmTjyg6o8TgXARc1MLROJSK+BH6XLWZWEgHDh8Mll8Cdd6aWupNPzjsqs26ltTbx9YBLgNNLSZ2ZmVl3ls1jt0tpHjtJnlnZure3304FUvbdN7XOffhh3hGZdUutJXZrk1r17pP0qKRDJC3ehLjMzKyGJo6xswqS+pGqQv+BNI/dH4Hns/Vm3U95gZRTT4VRo+DrX887KrNuqcXELiIejYgDgCWA04BtSeMJegBbZYPjzcysiSS1ebGGORcYFhHLRsQGEbEsaZLy8+p9AEkDJT0rabykI1vY74eSwkmjdVrTpsHhh8MCC6QCKYcd5gIpZjmqdx67TyPi0ojYCFiRNEj8EOANSTd3ZIBmZjYjt9jlakVgWMW604Bv13OwpNmAc4AtgJWAwZJWqrLf/MDBwOhZitasIzz7LLz3XkrirrsOxo6FNdbIOyqzbq/Nl1UiYnxEHEmqBrYT8FnDozIzs5qaOI+dzewxUlGxcqtm6+uxDjA+Il6IiM+AK0hdOiv9htTd89P2BmrWcBFwwQUpiTvqqLRu2WVd9dKsk2h1uoNasspg1zN9XiMzM2uCHs7U8nQrcKOkC0lTQPQB9gSGS9q1tFNEjKhx/FKkIQ0lE4H+5TtIWhNYJiJuknR4A2M3a7+334a9904tdJtskqYxMLNOpd2JnZmZ5cNdK3O1J2keu93L1n2RrS8J0pyvbSapB6lr5x517DsEGALQu7fnnrMO9NBDsP32MHlyKpByyCEeS2fWCTmxMzMrGDfY5Sci+s7iQ7xKGspQsnS2rmR+UlfPu7OiN4sDN0jaNiLGVMQyHBgO0K9fv5jFuMxqW2KJ1OXyxhs9ls6sE/PlFjOzgumB2rxYp/EwsIKkvpLmBHYBbihtjIj3ImLhiOgTEX2AB4GZkjqzDjduXGqZmzYNllkG7r/fSZ1ZJ+fEzsysYFw8pbgi4gvgQGAU8AxwVUQ8JelESdvmG50ZqUDK8OGw5ppw6aXwwgtpvT9IzDo9d8U0MysYj7ErtogYCYysWHdcjX0HNCMmM2DmAil/+xssuWTeUZlZnZzYmZkVjKtimlnDRcBWW8Ejj7hAillBObEzMysY53X5U6pssnhEvJ53LGaz5LPP0ofKHHPAaadBz54eS2dWUL4UY2ZWZKyGwgAAIABJREFUMD2kNi/WGJLmk/QX4BNgfLZuO0nH5xuZWTs8+yystx4cn/35fve7TurMCsyJnZlZwbh4Sq6GAYsB6wOfZeseBnbOLSKztoqAP/85FUh56SVYZ528IzKzBnBXTDOzgvEVuVxtDawUEe9JCoCIeFWSK0xYMbz9Nvz853DttS6QYtbF+PzAzKxgJLV5sYbpQeqG+RVJ8wEf5hOOWRtNnAi3354KpIwa5aTOrAtxYmdmVjBqx2INcx9wVMW6XwB35RCLWX0++wyuvjrd/853UvfLww5z1UuzLsZdMc3MzOp3KHCnpJ8A80l6EpgT+EG+YZnV8OyzsOuuaRqDsWPTuLqFFso7KjPrAE7szMwKxlUu8xMRr0haBdgG6AO8BNwYEZ+0eKBZs0XAhRfC0KFpCoNrr01JnZl1WU7szMwKxmldviJiKnB13nGYtWjPPeHii2HjjeGSSzyWzqwbcGJnZlYwbrDLj6ThtbZFxJBmxmLWos02g5VXhkMP9Vg6s27CiZ2ZWcG4ymWu5qj4eUlgQ9yCZ3n77DP4v/+DPn1gv/1g8OC8IzKzJnNiZ2ZWMB1x7V3St4Ary1YtBxwHLAj8HJicrT86IkZmxxwF7AV8CRwUEaM6ILROJSJ+VrlO0vbAZjmEY5Y8+yz8+MepOMrQoXlHY2Y5cdu8mVnBdMQ8dhHxbESsHhGrA2sBHwPXZptPL20rS+pWAnYBVgYGAudKmq1DXnDndx2wc95BWDcUAX/+cyqKMmFCKpBy+ul5R2VmOXFiZ2ZWME2Yx25j4PmIeKmFfQYBV0TE1Ih4ERgPrNP2p+oStqBi0nKzpnj0URgyBNZbD554ArbbLu+IzCxH7oppZlYwTRhjtwtwednPB0r6KTAGOCwi3gGWAh4s22ditq5Lk/Q/IMpWzQssChycT0TWLb38MvTunVrq7vr/9u48vor6+v/46wAJKItoQKqsqYBfFjUCRVMtKoLFFUShqAgqQlux4voFf+6tVhGVaosWELeqRMUF9OtGBVSsG24IaNlRFgUJghggBs7vj5nES8hOcm/m5v30MY/cO5+5M+eOIXPP/Xw+Z2ZDjx4qkCIi6rETEYmaWhVYzGyEmc2LWYqs4GhmqcAZwDPhqgeAQ4AMYB1wd1W9r4i4FbgtZvkTcKi735/QqKRmyM2F0aPhkEPgvfB7leOPV1InIoB67EREIqciPXbuPgkotlR/jJOBj9392/B138YcdzLwUvh0DdAy5nUtwnVJy8zqAM2A+9x9e6LjkRomtkDK738Phx+e6IhEpJrRVzwiIhFTxXPsziFmGKaZHRTTdiawIHw8AxhkZnXNLB1oB3xQ7jcTIe6eR1AVVEmdxNfDD+9eIOWf/4R99010VCJSzajHTkQkYqpqip2Z1Qd6A7+PWX2nmWUQzCtbmd/m7gvN7GlgEZAHjHT3nVUTWbUy28yOc/c3Ex2I1CDr1wcFUh57DA4+ONHRiEg1pcRORCRialWkzmUZuPuPQFqhdeeXsH3+PLOaZCUw3cymhY935Te4+18TFJMko1mzgjl1ffrANdcEi+bSiUgJlNiJiERM1RfFlMLMbIu7NyIoIvMJQUGZQ2I2cUCJney93Fy44QYYNw6OOQZ++1sldCJSJkrsREQixqqox05KZADufkKiA5EkVrhAyj336JscESkzJXYiIhGjz3kJ4aVvIrIXli4NCqTUqxcUSNHNxkWknJTYiYiIlK6emT1U0gbuflG8gpEksnMn1K4d3JvuhhtgyBAVSBGRCtGgbRGRiKmFlXuRSrGzlEWkfGbNgo4dgyGYZjBmjJI6Eakw9diJiESMhmImxHZ3H57oICRJxBZIad8+eC4ispeU2ImIRIwSO5EIK6pAim42LiKVQImdiEjEqCpmQuikS+WYOBFWrFCBFBGpdJpjJyISMbWs/IvsHXdvmOgYJMKys2HRouDxbbfB558rqRORSqceOxGRiFGPnUiEzJoVVLps1AgWLIB99gkWEZFKph47EZGIMSv/IiJxlpsLo0dDr17QoAE8/jjU0scuEak66rETEYkY9diJVHPr18Mpp/xcIOXuu6F+/URHJSJJTomdiEjEaM6cSDWXlgYtW8L112sunYjEjRK7JHFy757sW78+tWvVonad2kx9+jn+++WX3Prnm8jJyeHgg5tz+5130aBBg0SHKhGRs/UHpk4Yy9qvlmNmnHfptaT/T2fefGkab73yHLVq1aJT11/T74JLAFizcilZD4xje86PmNXimrsmk5JaN8HvIjmpx06kGsrODoZe3norNGsWVL0UEYkjJXZJ5MGHH2X//Q8oeH7Ljddx5TWj6far7jz/3DQeeehBLr3s8gRGKFHy7JR76dDlKIaNvpW8n34id8d2Fn/+MfM/eJsxf3uElJRUfvh+EwA7d+bx2Pi/cP7l19MivR0/btlM7dr681JVNGdOpJqZPRvOP//nIZhnnpnoiESkBtIs3iS2atVKunb7FQCZmcfwxszXExyRRMW2H7eydOFnZPY6DYA6KSns26Ahc195nt5nDSYlJRWAho33B+DLTz7k4DaH0CK9HQD1G+1Hrdq1ExN8DWAVWESkCuTmwpgxcOKJQYGU995TUiciCZPwr9TN7EJ3fzjRcUSewR+GD8PMOHvA7zh74O84pG07Zs96g54n9uL1117lm2/WJTpKiYiN366jwX6Nefy+v7J25VJaHnIoZ108ivVrv2bZovm89PgkUlLr0u+CkbRu14H1a7/GMCbcfCVbt3xP12NPpFf/8xL9NpJWLXXZiVQPN9wAd94JI0bAPfeoQIqIJFR16LG7pbgGMxthZvPMbN6UyZPiGVPkPPKvqTw17Xkm/HMyT019go/mfcgtf7mNp7KeZNCA/uTk/FjQyyJSml27drJ62WJ+c3I/Ro9/mNR69Zj57OPs2rWTnB+2cNWdk+g79BIeGncj7s6uXXks+2I+Q6+8kStuv5/P3n+L/342L9FvI2mpx04q25Pvf8X7K7ITHUY0uMOWLcHja66B6dNh4kQldSKScHHpsTOz+cU1Ac2Ke527TwImAWzPw6sgtKTRrFlwGtPS0ujZqzcLPp/P0AuHMXHyQwCsXLmCt96ck8AIJUoapzWlcVpT2rTvBEBG5gnMfO5xGqc15YjM4zAz2rTvSC0ztm75nsZpB9K20xE0aNQYgE5dMvl6+WIOPaJbIt9G8lKmJpVs+qdrAOib0TzBkVRz2dkwfDisWQNvvw1NmsAZZyQ6KhERIH49ds2AIcDpRSwb4xRD0srJyeHHH7cWPH73P+/Qtm07Nm4MTu2uXbuYPPEBBvxuUCLDlAhptH8ajZscyLdrvgJg8fx5HNSyDYcf1YMln38MwPo1X5GXl0eDRo3pcGR31q5aTu6O7ezcmceShZ/wi5ZtEvgOkptV4D+R0hyVfgDnHtUq0WFUX7NmweGHw4svwllngeYRi0g1E685di8BDdz908INZjYnTjEkreyNG7nispEA5O3cySmnnsYxv+nBE/96lKypTwJwYq/e9DvzrESGKREzYPgVPHrPLezMyyOt2cEMvuxaUuvuwxP/uJ2/XnY+teukMHjUdZgZ+zZoRM8zfse4qy/GzOjYJZPO3X6d6LcgUu2YWR/gXqA28KC731Go/UrgYiAP2ABc5O6r4h6o/Cw3N5hLN24ctGsXFEjp0iXRUYmI7MHcozHCUUMxpSq9tWRDokOQJHdSh6aV1m32wfLN5f572P2X+6nbLsHMrDawGOgNrAY+BM5x90Ux25wAvO/uOWb2R+B4d/9dafvu1q2bz5tX8Xmtv5v4LgBP/T6zwvtIWj/+CF27wnHHqUCKiCScmX3k7kXOdakOxVNERKQcVDwlsroDS919ubvnAllA39gN3H22u+eET98DWsQ5RoGgQMrUqUFSV78+fPCBCqSISLWnxE5EJGqU2UVVc+DrmOerw3XFGQa8UlxjbOXoDRs06qDSZGfD2WfDuefCpLAid6NGiY1JRKQMlNiJiERMVRVPMbOVZva5mX1qZvPCdQeY2UwzWxL+3D9cb2Z2n5ktNbP5ZqZJR5XIzAYD3YBxxW3j7pPcvZu7d2vatGn8gktmsQVS7rwTRo1KdEQiImWmxE5EJGLMyr+UwwnunhEzfn8M8Ia7twPeCJ8DnAy0C5cRwAOV8+6S2hqgZczzFuG63ZhZL+A64Ax33xGn2GTiROjVKxhu+e67wT3qauljkohEh/5iiYhETJxHYvYFHg0fPwr0i1n/mAfeAxqb2UF7d6ik9yHQzszSzSwVGATMiN3AzI4EJhIkdesTEGPN1asXjBwJH38cFEsREYkYJXYiIlFTgcwudj5WuIwoYs8OvG5mH8W0N3P3deHjbwjuSwrlny9W47l7HnAp8BrwBfC0uy80sz+bWf5drscBDYBnwiGxM4rZnewtd5gyBQYPDh4fcgj8/e8qkCIikRWv+9iJiEglqcgNx919EjCplM2Odfc1ZnYgMNPMviy0Dzcz3XpmL7j7y8DLhdbdGPO4V9yDqomys2HECHj2WejZM6h+2aBBoqMSEdkr6rETEYmYqppj5+5rwp/rgecJyvN/mz/EMvyZPzywTPPFRKqd/AIpM2YEBVJmzlRSJyJJQYmdiEjEVMUcOzOrb2YN8x8DJwELCOaADQ03GwpMDx/PAIaE1TGPBjbHDNkUqZ62bYPzzlOBFBFJShqKKSISNVVzX7pmwPMWdO/VAZ5091fN7EPgaTMbBqwCBobbvwycAiwFcoALqyQqkcqwYgW0bAn77AOvvALt2mkunYgkHSV2IiIRU5E5dqVx9+XAEUWs3wicWMR6B0ZWeiAilSm/QMqoUXDjjTB6NGRkJDoqEZEqocRORCRiynlfOpGaqXCBlMGDEx2RiEiV0sByEZGIifN97ESiZ+7coEDK9OkwdmxQIKW57sYhIslNPXYiIlGjTE2kZCkp0LhxkNjpZuMiUkOox05EJGKsAv+JJL3Fi+FvfwseH3UUzJ+vpE5EahQldiIiIhJd7vDgg3DkkXDrrbBhQ7BetzEQkRpGf/VERCKmqm5QLhI52dkwYAAMHw5HHw2ffQZNmyY6KhGRhNAcOxGRiFGeJgLk5UFmJixfHhRIufpq9dKJSI2mxE5EJGqU2UlNlpcHtWtDnTrB0Mtf/lJz6URE0FBMEZHIUfEUqbEWLw6GXD76aPB8wAAldSIiISV2IiIRozl2UuPEFkhZvjy4lYGIiOxGiZ2ISMToBuVSoxQukDJ/PvTrl+ioRESqHSV2IiJRo8xOapL//AdmzAgKpMycCS1aJDoiEZFqScVTREQiRnPmJOnl5sJ770GPHnDaabB0KbRqleioRESqNfXYiYhEjObYSVJbvBiOOQZ69YJVq4J1SupEREqlxE5EJGI0ElOSkjtMmRIUSFm2DLKyoHXrREclIhIZGoopIhI1ytQk2bjDoEHw9NNwwgnw2GOaSyciUk5K7EREIkZz7CTpmMFhh0GXLnD11cENyEVEpFyU2ImIRIzmzElSyM2Fm26Cnj2hd2+4/vpERyQiEmlK7EREIkZ5nUTekiVw7rkwb17wTUXv3omOSEQk8pTYiYhEjTI7iSp3ePhhuOwySE2FZ5+F/v0THZWISFJQYiciEjGaYyeR9corMGyYCqSIiFQB3e5AREREqlZ2dvDz5JPhqadg5kwldSIilUyJnYhIxOgG5RIZublw7bXQti2sXBn8Mg4cqKqXIiJVQEMxRUQiRnmaREJsgZSLL4amTRMdkYhIUlOPnYhIxFRFj52ZtTSz2Wa2yMwWmtmocP3NZrbGzD4Nl1NiXnOtmS01s/+a2W+r7h1L5Dz0EBx5JCxbBtOmweTJUL9+oqMSEUlq6rETEYmcKumzywOucvePzawh8JGZzQzbxrv7XbtFYNYRGAR0Ag4G/m1m7d19Z1UEJxEzdy50764CKSIicaTETkQkYqpizpy7rwPWhY9/MLMvgOYlvKQvkOXuO4AVZrYU6A68W/nRSSTMmQMHHACHHw733w8pKZpLJyISRxqKKSISMVaRxWyEmc2LWUYUu3+zNsCRwPvhqkvNbL6ZPWRm+4frmgNfx7xsNSUngpKsfvopKJDSsyfceGOwrl49JXUiInGmxE5EJGIqMsfO3Se5e7eYZVLR+7YGwLPA5e6+BXgAOATIIOjRuztub1SqvyVL4Ne/hjvuCO5P98QTiY5IRKTG0lBMEZGIqaoblJtZCkFS94S7Pwfg7t/GtE8GXgqfrgFaxry8RbhOaooPPwxuNJ6aGhRIOeusREckUuV++uknVq9ezfbt2xMdiiS5evXq0aJFC1JSUsr8GiV2IiJRUwV5nZkZMAX4wt3viVl/UDj/DuBMYEH4eAbwpJndQ1A8pR3wQeVHJtWOe9ANfMQRcNFF8L//qwIpUmOsXr2ahg0b0qZNG0w3CZUq4u5s3LiR1atXk56eXubXaSimiEjEVGSOXRkcA5wP9Cx0a4M7zexzM5sPnABcAeDuC4GngUXAq8BIVcSsAebMgWOOgezsoKfuvvuU1EmNsn37dtLS0pTUSZUyM9LS0srdM6weOxGRiKmiqphzKToHfLmE19wG3Fb50Ui189NPQWGUsWOhbVtYvz6ogClSAympk3ioyO+ZEjsRkYipqjl2IkVasgTOPRfmzYOLL4a//U03GxcRqYY0FFNEJGqqaCymSJFGj4Zly4ICKZMnK6kTSTAz46qrrip4ftddd3HzzTcDcPPNN7Pvvvuyfv36gvYGDRoUuR93p2fPnmzZsgWAb7/9lnPPPZdf/vKXdO3alczMTJ5//nkAcnJyOO+88zjssMPo3Lkzxx57LFu3biU3N5cePXqQl5dXbLwvvPACZsaXX35ZsG7OnDmcdtppu213wQUXMG3aNCAoUjNmzBjatWtHly5dyMzM5JVXXinHWSra7bffTtu2bTn00EN57bXXitxm1qxZdOnShc6dOzN06NDd3tucOXPIyMigU6dOHHfccQXrx48fT6dOnejcuTPnnHNOwRDK4vY1btw4MjIyyMjIoHPnztSuXZvs7Oy9fn9K7EREIkZ5nVS5TZtgXVgz5/77Yf58Vb0UqSbq1q3Lc889x3fffVdke5MmTbj77tLvTPPyyy9zxBFH0KhRI9ydfv360aNHD5YvX85HH31EVlYWq1evBuDee++lWbNmfP755yxYsIApU6aQkpJCamoqJ554Ik899VSxx5k6dSrHHnssU6dOLfN7vOGGG1i3bh0LFizg448/5oUXXuCHH34o8+uLsmjRIrKysli4cCGvvvoql1xyCTt37j41fNeuXQwdOpSsrCwWLFhA69atefTRRwH4/vvvueSSS5gxYwYLFy7kmWeeAWDNmjXcd999zJs3jwULFrBz506ysrJK3Nc111zDp59+yqeffsrtt9/OcccdxwGVMLxdQzFFRCJG0zuizcz6APcCtYEH3f2OQu11gceArsBG4HfuvjJuAc6ZA+efDx06wOuvwy9+EbdDi0TJLS8uZNHaLZW6z44HN+Km0zuVuE2dOnUYMWIE48eP57bb9pzmfNFFF/HII48wevToEpOFJ554ghEjRgBBz1Jqaip/+MMfCtpbt27Nn/70JwDWrVtH69atC9oOPfTQgsf9+vXj2muv5bzzztvjGFu3bmXu3LnMnj2b008/nVtuuaXE9wZB7+DkyZNZsWIFdevWBaBZs2YMHDiw1NeWZPr06QwaNIi6deuSnp5O27Zt+eCDD8jMzCzYZuPGjaSmptK+fXsAevfuze23386wYcN48skn6d+/P61atQLgwAMPLHhdXl4e27ZtIyUlhZycHA4++OAS9xVr6tSpnHPOOXv13vKpx05EJGKsAv9J9WBmtYEJwMlAR+AcM+tYaLNhwCZ3bwuMB8bGI7baO/Pg2muhZ0/YZx/461/jcVgRqYCRI0fyxBNPsHnz5j3aGjRowEUXXcS9995b4j7eeecdunbtCsDChQvp0qVLsdtedNFFjB07lszMTK6//nqWLFlS0Na5c2c+/PDDIl83ffp0+vTpQ/v27UlLS+Ojjz4q9b0tXbqUVq1a0ahRo1K3veKKKwqGNMYud9xxxx7brlmzhpYtf779aosWLVizZvfbrzZp0oS8vDzmzZsHwLRp0/j6668BWLx4MZs2beL444+na9euPPbYYwA0b96cq6++mlatWnHQQQex3377cdJJJ5W4r3w5OTm8+uqrnFVJIyLUYyciEjHqsYu07sBSd18OYGZZQF+C20bk6wvcHD6eBvzDzMzdvaqCSsv+hqsm/j9Y9WVQIGX8eChmXo6IBErrWatKjRo1YsiQIdx3333ss88+e7RfdtllZGRkcPXVVxe7j+zsbBo2bFhk28iRI5k7dy6pqal8+OGHZGRksHz5cl5//XX+/e9/86tf/Yp3332XDh06ULt2bVJTU/nhhx/22N/UqVMZNWoUAIMGDWLq1Kl07dq12IqP5a0EOX78+HJtXxozIysriyuuuIIdO3Zw0kknUbt2bSDolfvoo49444032LZtG5mZmRx99NE0bdqU6dOns2LFCho3bsyAAQN4/PHHGTx4cLH7yvfiiy9yzDHHVMowTFBiJyIiEk/NgdivbFcDRxW3jbvnmdlmIA3YbUKNmY0ARgAFQ4MqqlX6QeyTUisokKK5dCKRcPnll9OlSxcuvPDCPdoaN27Mueeey4QJE4p9fZ06ddi1axe1atWiU6dOPPvsswVtEyZM4LvvvqNbt24F6xo0aED//v3p378/tWrV4uWXX6ZDhw4A7Nixg3r16u22/+zsbGbNmsXnn3+OmbFz507MjHHjxpGWlsamTZv22L5Jkya0bduWr776ii1btpTaa3fFFVcwe/bsPdYPGjSIMWPG7LauefPmu/WYrV69mubNm+/x2szMTN5++20AXn/9dRYvXgwEPXxpaWnUr1+f+vXr06NHDz777DMA0tPTadq0KQD9+/fnP//5D4MHDy52X/mysrIqbRgmaCimiIhIJLn7JHfv5u7d8j9QVNSYQUdz8OIFSupEIuSAAw5g4MCBTJkypcj2K6+8kokTJxZbsfLQQw9l+fLlAPTs2ZPt27fzwAMPFLTn5OQUPH7nnXcKErHc3FwWLVpUMOdu48aNNGnShJSUlN32P23aNM4//3xWrVrFypUr+frrr0lPT+ftt9+mXbt2rF27li+++AKAVatW8dlnn5GRkcG+++7LsGHDGDVqFLm5uQBs2LChoFhJrPHjxxcUIYldCid1AGeccQZZWVns2LGDFStWsGTJErp3777HdvkVRXfs2MHYsWML5h327duXuXPnkpeXR05ODu+//z4dOnSgVatWvPfee+Tk5ODuvPHGGwUJb3H7Ati8eTNvvvkmffv2LfL/T0UosRMRiRiz8i9SbawBWsY8bxGuK3IbM6sD7EdQRKVq6RdFJHKuuuqqEqtjnnnmmezYsaPI9lNPPZU5c+YAwRDEF154gTfffJP09HS6d+/O0KFDGTs2mOK7bNkyjjvuOA477DCOPPJIunXrVjAvbPbs2Zx66ql77H/q1KmceeaZu60766yzmDp1KnXr1uXxxx/nwgsvJCMjg7PPPpsHH3yQ/fbbD4Bbb72Vpk2b0rFjRzp37sxpp51Wpjl3JenUqRMDBw6kY8eO9OnThwkTJhQMjTzllFNYu3YtENyKoEOHDhx++OGcfvrp9OzZE4AOHTrQp08fDj/8cLp3787FF19M586dOeqoozj77LPp0qULhx12GLt27SooSlPcvgCef/55TjrpJOpX4i1krAqH7Feq7XlEI1CJpLeWbEh0CJLkTurQtNI+NW/etqvcfw/326eWPrVXA2Githg4kSCB+xA4190XxmwzEjjM3f9gZoOA/u5eYjm4bt26ef4EfRGpOl988UVBb0zUrVu3jiFDhjBz5sy92k///v254447Cqo/SuUp6vfNzD5y925Fba85diIiEaOOlegK58xdCrxGcLuDh9x9oZn9GZjn7jOAKcC/zGwpkA0MSlzEIpKsDjroIIYPH16muWzFyc3NpV+/fkrqqgkldiIiEaO8Ltrc/WXg5ULrbox5vB0YEO+4RKTm2dt7w6WmpjJkyJBKikb2lhI7EZGoUWYnIpIw7l7usvwi5VWR6XIqniIiEjG6QbmISGLUq1ePjRs3VuhDt0hZuTsbN27c4xYSpVGPnYhIxOiLYhGRxGjRogWrV69mwwYVXZOqVa9ePVq0aFGu1yixExGJGOV1IiKJkZKSQnp6eqLDECmSEjsRkahRZiciIiKFKLETEYkYzZkTERGRwpTYiYhEjObYiYiISGGmqj7JycxGuPukRMchyUm/XyLVi5ltAFbt5W6aAN9VQjjJROdkTzonRdN52ZPOyZ4q45y0dvemRTUosUtSZjbP3bslOg5JTvr9Ekk++ne9J52TPemcFE3nZU86J3uq6nOi+9iJiIiIiIhEnBI7ERERERGRiFNil7w0/0mqkn6/RJKP/l3vSedkTzonRdN52ZPOyZ6q9Jxojp2IiIiIiEjEqcdOREREREQk4pTYJSEz62Nm/zWzpWY2JtHxSPIws4fMbL2ZLUh0LCJSMaVdI8ysrpk9Fba/b2Zt4h9lfJXhnFxpZovMbL6ZvWFmrRMRZzyV9bOEmZ1lZm5mSV/9sCznxMwGhr8rC83syXjHGG9l+LfTysxmm9kn4b+fUxIRZzyV9lnJAveF52y+mXWprGMrsUsyZlYbmACcDHQEzjGzjomNSpLII0CfRAchIhVTxmvEMGCTu7cFxgNj4xtlfJXxnHwCdHP3w4FpwJ3xjTK+yvpZwswaAqOA9+MbYfyV5ZyYWTvgWuAYd+8EXB73QOOojL8n1wNPu/uRwCDg/vhGmRCPUPJnpZOBduEyAnigsg6sxC75dAeWuvtyd88FsoC+CY5JkoS7vwVkJzoOEamwslwj+gKPho+nASeamcUxxnjgzm4/AAALXUlEQVQr9Zy4+2x3zwmfvge0iHOM8VbWzxJ/IUj8t8czuAQpyzkZDkxw900A7r4+zjHGW1nOiQONwsf7AWvjGF9ClOGzUl/gMQ+8BzQ2s4Mq49hK7JJPc+DrmOerw3UiIiJluUYUbOPuecBmIC0u0SVGea+bw4BXqjSixCv1nITDx1q6+//FM7AEKsvvSXugvZm9Y2bvmVmyj3Apyzm5GRhsZquBl4E/xSe0aq3KPqvXqYydiIiIiCQ7MxsMdAOOS3QsiWRmtYB7gAsSHEp1U4dgeN3xBL26b5nZYe7+fUKjSqxzgEfc/W4zywT+ZWad3X1XogNLRuqxSz5rgJYxz1uE60RERMpyjSjYxszqEAyf2hiX6BKjTNdNM+sFXAec4e474hRbopR2ThoCnYE5ZrYSOBqYkeQFVMrye7IamOHuP7n7CmAxQaKXrMpyToYBTwO4+7tAPaBJXKKrvqrss7oSu+TzIdDOzNLNLJVgouqMBMckIiLVQ1muETOAoeHjs4FZntw3vS31nJjZkcBEgqQu2edNQSnnxN03u3sTd2/j7m0I5h2e4e7zEhNuXJTl384LBL11mFkTgqGZy+MZZJyV5Zx8BZwIYGYdCBK7DXGNsvqZAQwJq2MeDWx293WVsWMNxUwy7p5nZpcCrwG1gYfcfWGCw5IkYWZTCS5aTcLx8je5+5TERiUiZVXcNcLM/gzMc/cZwBSC4VJLCQoADEpcxFWvjOdkHNAAeCasI/OVu5+RsKCrWBnPSY1SxnPyGnCSmS0CdgLXuHvS9naX8ZxcBUw2sysICqlckORfFBX5WQlIAXD3fxLMNTwFWArkABdW2rGT/NyKiIiIiIgkPQ3FFBERERERiTgldiIiIiIiIhGnxE5ERERERCTilNiJiIiIiIhEnBI7ERERERGRiFNiJzWOmZ1tZh7z/AIz25qgWF4ys0cScWwREZF8ZjbHzK5PdBzlZWZbzSxzb7cRSQZK7KTaMLNHzMzD5SczW25md5lZ/So+9FPAL8u6sZmtNLOrqzAeERGRCgkTtB1hMpO/PJjouKqKuzdw93cBzOx4M8sraZvqIvysc2yi45DkohuUS3Xzb+B8ghs5/gZ4EKgP/DF2IzOrA+ysjJtcuvs2YNve7kdERKSa+Iu735roIGoiM0tx958SHYfUTOqxk+pmh7t/4+5fu/uTwBNAPzO72cwWhMMmlwE7gPpmtp+ZTTKz9Wb2g5m9aWbdYndoZkPMbJWZ5ZjZS0CzQu17DMU0s1PM7H0z22ZmG83sRTOrZ2ZzgNbAuPzexZjX/Do8fo6ZrTGzB8ysUUz7vmGv5FYz+9bM/l9lnzwREZGimNkgM/vMzLaY2Tozm1jciBgzqxtzbd1iZkvMbEBM+2/MbK6ZZZvZMjO7ysysmH21Ca+XF5vZYjPbbGbTzezAmG3SzOwxM/smXB41swNi2i8zsxXhdX6Nmf01ps3N7FgzOxh4Bagd01M5tNA2dcL33q9QjI+Y2cMxz4eHnzk2m9knZnZSCef1AjNbambXmNlq4NNw/V/DkUdbw3N0ecxrPgsfvh7boxp+TrgrfK/ZZvaqmbUt7tgihSmxk+puG0HvHUA6cC4wADiCILn7P6A5cBpwJPAWMMvMDgIws6OAR4BJQAbwIvDnkg5oZn2AGcBMoCtwAvAmwb+X/sDqcB8HhQtmdhjwevi6I8LtMoCHYnZ9F9AbOAs4MYy3RznPh4iISEVsJriGNiYYEfMboLg5dUOBXwEd3L0R0BNYCGBmHYGXgXFAU+BU4FKC0TYlGUJwzWsJ7AIej2l7Atgf6BAuTYB/hcdrD9wBnObuDYFOBNfa3bj7WuBkgtE8DcLl0ULb5IX7vSB/nZk1AM4mvF6b2XBgNHBeGNN1wHOlJFhtgIOBdgTnDWARcCzQEBgO3G5mvw3jOCLc5qQwzovD55OB/wGOBn4BvA+8ZGb5n4NESqShmFJtmVl3govQG+GqVOB8d/82bO9JkDw1DYdTAtxgZqcTXGDuBEYBb7j7bWH7YjP7FTCshEPfAExz99gL3vzwZ46Z7QR+cPdvYtqvAZ5y97tj4v8j8En4rWROeMyL3P21sP1CgiRRRESkMl1nu88F7+Pur8Q8X2pm9xMkW0XJBRoAHc3sXXf/OqbtEuAZd58ePv/SzP4R7uuxEmK6Jf+6aWbXAEvCXjaA3wLt3X1T2H5luN+DgDzAgE5mtsrdvwfeK/ntl+hh4DMzO9Dd1wMDgbXu/nbYPgr4s7vn96q9bGazgUFAccNbfwLGuPuO/BXuHpu4zjKz/yP4Uve1onZgZk0IPvO0jvmccwtwOXAUMLf8b1VqGvXYSXXTJxyWsB14l6AH7k9h2+r8P3ahrsC+wIaYYRdbgc7AIeE2HcL9xCptAvWR/JxMllVXYHChON4J2w4Jl9TYY7v7VuDzch5HRESkNLe5e+OY5T0z621mb5vZBjPbAowl6HEryuMEc9zHAxvNLLbHKh04x8y+z1+AmwhHsJRgZRGPWxD04AGsiGlfFv5s6e7LCXrPhgNrwyGgxQ6NLI27fwF8DAwOV11IkOzlSwcmFHp/JxCMDirOutikDgqGj35uZpvCfZxO8ec7/7gA82OOm00waqll8S8T+Zl67KS6eQsYQfDt19r8Ccjh0P0fC21bC/iWYDhJYVuqMMai1OLni2Bha4D28Q1HREQkYGapwAvA/wIPufs2M7sUKLLCczhkcSww1swaA/8gGKrYA1gV7mNkOcNow88JW5vw5+pC7UvDx/mVqr8O43mOYDhkKvAHYLqZpbl7TqFj7CpjLA8DI81sBsGwx0ExbauAm9z9mTLua4/jmtkxBOfvROB9d99pZtMIeh7zFS7+tir82c7dN5Tj2CIF1GMn1U2Ouy9191VlqCr1MUEhlF3ha2KX9eE2XxD80Y5V+HlhnxD8MS5OLlC7iFg6FRHH0nCY6DKCZLXg2BZMWu9cSiwiIiJ7KxWoC2wKk7qOBPPiimRmPc2sazi3axvBF6s7w+b7gUFmdrqZpYQFSTqa2XGlxHCDmTWzoKjYWODf7r42nBv3OnC3mTU2s/2Bu4FX3H2dmR1qZn3MbF+C6+hmgqSoqCTuG4LiKelFtMXKAtoC9wEz3X1NTNt44GYzy7DAPmHhlf8pZZ+xGhGcrw2Am9mpBPP/CsfaLv9J+LnlSeB+M2sOEJ6PM8N5gCKlUmInUfZvguGO083sZDNLN7NMM7vFzPJ78e4DepnZtWbWLpwUfWYp+70NGGBmt4YXq05mdkV4UYFgCMlvzKx5OCYegotUdzP7p5kdaWZtzew0M5sIBcMupxB8+9nbzDoRfPtZOEEUERGpVOE16I/AneFUgQkESURxmhEUGdkErCOoBj0i3NcCgoJll4dt6wmKlJU0zBCC4Z1vE/TCpbJ7sZXBwA/Af4Evge/5ef5fKnBjeKzvgcuAs9x9exHvczHwAPBBOJyxyIIu7r4ZeJ4g2XqoUNtkgjn6D4fv/yuCufflKWDyGsF8ww+A7wiKszxfaJvrgD+HQzUnhuuGE5yDOWb2A8F0jQHs2bsnUiSrhNuAiVQKM3sEaOLupxXRdjNwtrt3LrS+IcFk5rOAAwmGZr4DXOfuy8JtLgRuIbjozCGopPl3d7ew/QLgH+7eIGa/ZxDMGehMcLH5DzDQ3beb2dHAROBQoG7MfrqFsfyaIGFbDjzv7jeG7fUJLjj9CYqp/J1gQvR37n5Bxc6aiIhI9WVmbQjmz7V0dxUME6lCSuxEREREpEoosROJHw3FFBERERERiTj12ImIiIiIiESceuxEREREREQiTomdiIiIiIhIxCmxExERERERiTgldiIiIiIiIhGnxE5ERERERCTilNiJiIiIiIhEnBI7ERERERGRiPv/FBjcwCQyCtAAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 631
        },
        "id": "1xIPyLxPRHt9",
        "outputId": "2d84b076-0f95-491f-dda4-b2a6128a9aad"
      },
      "source": [
        "plt.figure(figsize=(10,10))\r\n",
        "\r\n",
        "\r\n",
        "rocauc_plot(rf_gs, \"Random Forest\", y_test, X_test)\r\n",
        "rocauc_plot(knn_gs, \"KNN\", y_test, X_test)\r\n",
        "rocauc_plot(lr_gs, \"Regresión logística\", y_test, X_test)\r\n",
        "rocauc_plot(nn_gs, \"Red Neuronal Simple\", y_test, X_test)"
      ],
      "execution_count": 260,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmQAAAJmCAYAAADy2Zy3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd5xcZdn/8c+1JdksqZCQQEISauglbFAeAZGiIUYQpIsKqBELgqIiIHk0KBawgYjyoCBIDSD4C6EXBUEJRToJoScQSEgnO5s5u/fvj/vM7uxky+zsmTlzZr7v12tf5+zM7MwVSvLNdd/nOuacQ0RERETiUxN3ASIiIiLVToFMREREJGYKZCIiIiIxUyATERERiZkCmYiIiEjMFMhEREREYqZAJiLtzGyQmX3TzO4xs3fMbL2ZrTKzJ8zsQjPbIe4ai8HM7jCzJWbWWODPTzEzZ2bfK+BnTwx/NvsrMLN3zex2M5uex3s0mtkZZvYvM1tmZi1mtsjMZpvZIXnWsaeZ/dHMXjKz1eF7vGVmt5nZ58xsQF9/bSKSP9McMhEBMLO9gFuAsUAK+A/wNjAE2CN8vA042Tn3l7jqjJqZHQzcDXzbOffrfrzP7cD/AFs755b34edOBK4A3gXuDB9uAHYCdg6/n+mcO6+bn98N+H/AFsAa4F/ASmArYApgwGzgc865li5+vg64GDglfOgV4Dn8fwNbAHsBdcArzrlt8v11iUjf1MVdgIjEz8x2AR4AGoFfAT90zq3Jet6AA4ELgS1jKbJ4fgasAP7Qz/c5H3gYODP86quXnHMnZj9gZqcBvwFmmtlVzrk3cp6fgP/3NgK4FPiuc+6DrOd3Bq4DjgJqgc908blXACfgg9jJzrl/5nzGxsB3gTMK+DWJSJ60ZClS5cKw9Vd8GPuxc+6M7DAG4Lx7gQ8Bt8dQZlGY2d7AZGC2c665P+/lnPsX8CpwclTLe8653wIv4f/yfFAXL/kjPoxd6Zz7WnYYC3/+OXyQfg84wsyOyX7ezI7Ah7GlwL65YSx8j+XOubOAAyL4JYlINxTIROQTwK7AYmBWTy90zrU45+ZlvjezK8M9T/t39Xoze93MXM5j+4c/c6WZjTGzy8P9ToGZnW5mfw+f/1Q372nh+7aFHSLMrMbMjjOza81svpmtDb+eNLPv9RCQZoTHa7r5rA+b2c3h56XM7L3wPX9tZqO7+JFrgZHA4d18XiGeD4+b5tS2E/7fXQr4Tnc/7Jx7D8gsd+Z2uTJ73v7XOfdOT0U45x7Ot2AR6TsFMhH5ZHic7ZxLl/BzRwHzws9/FLgDWIfv1gF8tpuf+wgwAXg4awmvER+GPoHvBs3B76XaEvg5MMfMuvr97pP4QPNo7hPhZvp/AUeE73lrWO8g4HRg2y7e74Gs943KkPD4Xs7j08LjXc6593t5j2sBBzSZ2UiA8LhX+Ph1EdUqIgXSHjIR2T08PlHiz50G/A043jmXyjxoZg3AauBQMxuSu3xKR1D7a9Zj6/FdqduzQ6WZDcWHjWnA8dk/Y2bb40Phv7sJot/B/6X1aOfc7Ownwu7Usi5+5nH8hQ8f7e4X3Rfh/q29gFbg3pynd8v6zB4555ab2ev4gLobcF94NPxm/ZVR1CsihVOHTERGhselJf7cFuDU7DAGEH5/M74TdUT2c2ZWj9+gvh5/5WDmZ9Y7527NDVbOudX4bhZsuIy4a3h8qZv6RoXHu3OfcM4975x7t4vHVwPvAOPNbEQ379srM2swsyZ8YB0OnJm7oR/YJDzm++8t02EbmXMs9b93EemCOmQiEpcnnXOLu3nur8BJ+G5Y9oiNqfggcqtzbkXuD4VXi34CmAhshO8AWfh07hJjZk9WdyMqngB2BK4ys/OAJ1x+c4KW40eEjMJfvZmvj+but8N3205wznW5x01EKocCmYhklt5G9fiq6L3Zw3MPAouAA8xsjHNuSfh4V8uVmc7Zn/FXDHZnSM73w8Jj7pJoxln4WWCHhl8rzexR/FWmf3HOre3m51aHx+E91NKV7Dlkw/AzzTYFfm9mzzjnns15fWbfWL7/3jIBdFnOsdT/3kWkC1qyFJH/hsc9i/DePf0ek+ruCedcG37vVy1wHICZDcEHo1X4TfvZvo0PY88D04ExwADnnAEDw9dYzs9k9k0N7aaGxfj9WwcDv8aPtPgE8DvgJTOb2E35maDX131ZLznnTgy/DsdfuHBrWN/VXVyU8HR4bOrtjcO9aJl6n846OmArM+treBSRiCmQiUhmrthR4dT2vlgfHgfnPhG+12b9qCv3asvD8fvKbupi4nxmr9mxzrnbnXPvZu0n6266fGZP1cbdFeCca3XO3euc+7Zzbk/8UuTN4fEn3fxYZu9Yv/ZmhXvpTgzr3I0Nu39zw+MnwsDVk2PxgXSec25Z+P7LgMfCx4/rT60i0n8KZCJyF/AMPmTM7OmFZjYg3GyekZldNamLl3+MfmyLcM49AzwL7Glmk+hmuTKUCUFvdfFcd2Ej0ynavg81LaFjVtvOuc+HV3VuDrzZ1R63vnLOrQJ+Gn57dnaXzDn3PP6Cgwbggu7ew8xG0fHv9Zc5T/8iPP7QzMb0VIuZfaQPpYtIHymQiVS5cKP65/AzwM41fxPx3P1WhMNfH8UvCWY8GB6/ZmabZr12G/z9EfsrE76+jZ84vwj4Rxevm5+pI/tBMzuIbm7545xbACwBdu9qcKyZfdvMuurwZeZ/dRX+MveO7KrGQv0RH3wn4a8wzfYV/NLoyWZ2seXcHN3MdsSPyxgN/M05d0P28865W/D/jDcFHjKzfXM/3MyGhxc1PJD7nIhER4FMRDLdqAPw0/rPAN41s/vN7Jpwcv6b+D+Qd8Pf8zDjQeB+/I2snzezW83sAXzH7b9A7qiGvroWf6XhDPx+smu7udLx50AAnB9O0r/OzB4B7gEu6uH95+D3mO3dxXMzgUVm9l8zu9HMrjezZ/Adq3V0fVeDj4XHyG4vFd7S6Wfht2eHt7rKPPd6+JmLgG8AS8xsbnjHgn/jbxK+K3ATfg5bV07Ch75tgH+a2ctmdkv4z/Bh/JLpD4DXovo1iciGFMhEBADn3H+A7YDTgEfwVxgejR9y+h5+eWsH59zVWT/jgE/jN7qngUPwS5+z6H7Sfl9qyu2IdbVcmbmtz//gA9h44FNAPf5m2d/v4SMuC49d1Xoq/sKCgfjN/NPD97wU2NU591gXP3Mc/urFv/XwmYW4DB+Wd8X/2to55/6L7559Fx/APoy/ifgWwC3ANOfcUbnz3rJ+PnDOnYLv7l2GD8AH4/flTcDvVTse/9+DiBSJ5TdWR0SkMpnZY/ju0Nj+3GDczPYBHgJ+4Zw7M6r6RKQ6qEMmItXubPxFAV/t5/uchd/P9fN+VyQiVUcdMhGpemZ2B7AHsJVzbl0BP9+Ev/H4mc65X/T2ehGRXApkIiIiIjHTkqWIiIhIzBTIRERERGKW6JuLjxw50k2cODHuMkRERER69cQTTyxzzo3q6rlEB7KJEyfy+OOPx12GiIiISK/MrNth2VqyFBEREYmZApmIiIhIzBTIRERERGKmQCYiIiISMwUyERERkZgpkImIiIjETIFMREREJGYKZCIiIiIxUyATERERiZkCmYiIiEjMFMhEREREYqZAJiIiIhIzBTIRERGRmCmQiYiIiMRMgUxEREQkZgpkIiIiIjFTIBMRERGJmQKZiIiISMwUyERERERiVpJAZmZ/NrP3zOy5bp43M7vIzBaa2TNmNrkUdYmIiIiUg1J1yK4Epvbw/CHAtuHXDODSEtQkIiIiUhZKEsicc/8ElvfwksOAq5z3b2C4mW1WitpERERE4lYXdwGhscBbWd8vCh97J55yRKTaOedY39pGS9BGKt1KS9qft7Y52pz/co6s7/3P+O/9eZuD1vbXOtra/PeZ59rC17vwvM1BW9b7ZT6nrS3n+/C8988Pa2jr/JltYS0bfE7mvJv3zP3MTN2trvNntrZ1/nznXNz/OiUBPvLiQzS9Mo9WW0VbzZq8fsYAcN0cwXDtx+5e29CcZvTSD3h+7605+i8PRfJrKUS5BLK8mdkM/LIm48ePj7kaESm2tjbXEYq6OLYEraTSnY8t6TZSOceWoCNU5fUeQRtJyxE1BjVm/qum49wMamvCxw3MjNqs85oawu/9a2vMqK0x/1zmPWuyzrNeU9/+vh2Pmxm1WZ/f/qej9GinJ+5nu+ceKdnntbgVpFldss8zHLiOINT+WHjc+q00AM+Hf7Q3ttWSHZw2CFqF/P9pG0ayloF1jFy+jkFr1xfwhtEpl0C2GNgi6/tx4WMbcM5dBlwG0NTUlLDfLiVpUulWln8Q7/+kxdbmHEGrI2hrY33gj+lWR9DaRtDmSLd2fJ9uCx9vdaTb2kgHmdd0fr79Z9rC17Zm3nfD9/bPt7G+1WWFpo4wlW7t3//mA+pqGFhXQ0N9badj5nxIQ92Gz9XX0hAeB2Yf62qoq6nxwaOmcwjJDixmYcCpyQ1AnQNPeyDKCkydA1FHqKk1w2o6B6Lcz5dkWHHDjayeM2eDx9fNmwfAm1sPKUkdzWnfhRpSn/15PjT1eO58OOr9PPNz+dXz5lh4cfs2nt6thmktjmmt9TBgI/9V3wgDBsOAxvzOB2wE9Rt1Pq8b0PkDL77YH089FYBP5Vdm0ZRLIPs78A0zux74ELDKOaflSimqVLqVJatSvLMqxTurmtuPS1aleHtliiWrUxUfxoqhrsaor62hrjY89vR9TQ0D6mpoHFjHgFoLg1AtA+u7DlEdYaqWhvrOx4H1Ne0/mzkOqK1RUMlDdwFBiiMTvBqnTGFp81Leb34fgDXj4eEda1g1dafuf9g5cK3Q1gqurfN5W6v/Pvu8ra3717s6pq1L84nViyC9jj61nGrqw7AzOAxC2efZIaqL8x6e+0TdIKgpwfb2n/0MzjrLn3/1q1AXfxwqSQVmdh2wPzDSzBYB/wvUAzjn/gDMBaYBC4F1wEmlqEvKy/qgjXdWNfPm8nXtX28tX8f7EbeR16SCbsPW8MZ6xgxtYPPhg9h9/HA2H9bAJoMHUsl/phtGXa1RV1vDgFqjrqZzePKPZx7r/Hx2sKqrNerCZS4pb7kBLDsgSESyw5BrZWnzct5fv8o/t2UDL+6yEU83LeHxZt97aKofAW0DmFYzjKOWrfEBaf0H/iv7vC3dtzpyu0QDBmedbwSbNnacD+jhPDt4ddVtSpIf/Qh++EN/ftppZRHGACzJmy2bmprc448/HncZkod16wMWr2hm0YpmFq9s5r3VKd5d3cK7a8JjFwGpvtbYYkQjIwcPJMo/4xsH1LLZ8EFsPqyBMcMyR//VOKA8/seUylNOXaiuAtjQ6dMZcczRcZUUj2A9pD+A9WHgSYehZ/26/M7Xrw3DUufz2Q0wd6ONOn3U44MaAGhqTm1QxrR1LRy13gpfnutuqa5U3aYkOfNM+MUv/PlZZ8H555f0483sCedcU1fP6U8fidzqVJp/zF/KAy+9x8vvrWXxyuYNwpYZjBw8kDFDGxg7vIE9xg9n9JAGNhvWwPhNGhm/cSOjhzZQW8mtKakqq+fMIfXSSzRsv33cpdA4ZUpyAlhbmw86G3SM1oYBKY/z9rCVc96nbpNlhaWOjtHs+jbmDmgDGww1w8BqeTztlyCbGseC1YDV0lRTy7QxH+aoCVM3XKqrrS/OPzvZ0AUX+OOsWXDuufHWkkOBrEK1BK28tuwDXl/2Aev7uSk6X8vWtHD/S+/x71ffJ2hzbLzRAHYZO4xdxg1j7PBBjBvhvzYfPohRgwdSV6u/uUnl6K0DlgljE66+qoRVlVCwvpuOURfLbt2dd/Vcel3f6qgd0PWG7sFjel6S6255LnNeP4jsVv3sBbOZ++pcHn/3cXDQtGlH06OJLZm21TSO2u6oqP7pSlQefBCefBJOPz3uSjagQJYQrW2OP/zjFV54p+dLlJvXt/Lq0rW8uXwdbTGsRm89aiO+uO+WHLzDaPYYP0IdLimqcl8GzNaw/fYMnT69lCVtKNNt6q5jlO+yXafgFf5MW9CHQnKX58LzgUNgyJieN4T3diVdbXH/WOsUxICm0U0KX+XuhBPguuvglVdgv/38VxlSIEuApWtaOPtvz3LPC+8ycZPGHkPOgLpadtp8GIfuPpZtNh3MViM3oqG+tiR1Ng6oZfPhg0ryWVK+ShmSymkzemTLgM5B6/oeukn9WKoLmvtWS+3ArjtGQzbrpcvUy1V1dQ1EujG0RGYvmM2sR2cBCmKJceSRcPPN/vyRR2DixFjL6YkCWRl66OWl/P2/b+OAV5au5b9vrQTgh5/akRM/smW8xYn0opR7pWLdC9XWlhV2sjpGC+/rx1Jd+F6uNf86rKaLK+kaoWFo2G3qbUN4D1fSFbnbVG4y3a/uZLpiM/eeqSCWBJ/6FGT+cnjNNXD88fHW04vq+r8tAYLWNr5/87MsXtnM5sMaGDVkIN86aDsO2XkM244uzbBAqVyl6F6V1V4p5yBoyQlC/bmqLuvn+9ptqmvoehlu6Ob5zW7q9kq6ZHabotRbkMpX9jJkV9QVS5CpU+Guu/z5zTfDEUfEW08eFMjKQLq1jTufW8ITb6xg2doWFq9s5g8n7MnUncfEXZpUgOwQVoolvoL2SrW15tFN6utSXXjsa7ep0/JcGHwahoXBKY8N4dnLdpkgVd9Ydd2mUsldRuwPBa4KsssucPfdMHeuD2cJoDlkEVqTSvPq0g+6fG7+kjVc8uBCWtJtGzz3QUvAmpaAjQbUUldbw3ajB3P9jL21IV661NcuV24IK3iJL9Nt6nZDeF+vqsv6mWDD2Uw9qmvoYkN3N92kvizV1Q2s+m5TUuRurtcyogBw9tl+pEWZDHvNpTlkJfLtG5/mnhfe7fb5rUZuxEe3G7XB4zU1xkE7bMrHJm2q27wI0HPo6r3L5TpNCG/cbUeG7t/EiIOndISoeZcXtlTnNvwLRbesNicghcGnYTgMHZvH8lwPt1mpKc2FKlJafVl61FWOsoHJk+Gpp+CGG/wVlQmjQJaHVLqVt5Z3zMI565Znef+D9Rv8RfrVpR+wxcaD+NGhG96HrLamhg9tuXHJrniUeOXVxerhPnTrnvO/mTROGrvBfegaJw5h6E5DGbHL8q43h7e2bPhZ794Lf+2mjrpBXY8QGDQi5/E+LtXVDlC3qYJEtU+rJ73t4cqmICad7LQTvPCCP589O95aCqRAlocDLnyQt1dtuKQyfdfNOn2/0+bDOPF/JrLnhBGlKk2KoHOYcvnduDfnJr7rXl0BQOP4Qd28vs0vAXajcRQMndDMiG3e9g9Ybc6G7hZoq4PGjaF+XB7Lcz2cq9tUcYoRnvoSlgqlkCUF2XZbWLjQ/wXw2Wd9OEsgBbI8ZMLY747fA4AaMz6yzUiGDdLtLsqCc5BuznP/UnfLcx0bwlffvIbU+46G4UHflujAhxuroXGzGobuMIgRe2S6TIPzW57rbqlO3SYh/6BVjPCksCRl6Te/6QhjCxbANtvEXVHBFMh6MffZdwD46v5bM33XzWOuJuFag25urVLgbVayz+nDxSk1dZ2W51a8VMPqBWnfhaqpJbWiloYthjHh9AP7dpuV+kbdyFe6FUXXKt+gpfAkVeP00/3VlL//fVkPfc2HAlkvlq31+3E+M3lszJWUSKbb1OutVQqY49S6vvfPz5bpDuV2kBo36cOVdF3MbqobAHQsTeZukm8Y5a9E5OAE3HhZEiGq0QwKWiKhMWNgu+3gn//0oy0qgAJZnkY0Doi7hM5a0wV0k/KZ49TXblP9hhO+BwyGxpEwfHzhS3V1g/rdbeptY312EItt2rskSqFdLo1mEInQJpvA8uXw/vtxVxIpBbJiSzdDy9o8lur6uGzX525TF7dWGbARbDSqwCvpGjt1m0op3zlcvY2HUBCTvuhPl0udLZEIBIEPY6tXQ20tLF0ad0WRUiArpmdmwy1fyv/1tQO67hoN3rSAoZdZ5xF0m4qtL8NO8502r8Al+cqn86Uul0iMggBGjIC1a6G+3nfIBg+Ou6pIKZAV07L5/jjtwvyW6mqr76rN7vZx9URBS6KQHcLy2SyvLpdIzIIABgyAVaugoSHuaiKnQNaLFR+kC//hIOW7U3t9ObqCEqDQbpdClkSpt65XdghT2BIpU6kUzJkDRx4JK1b4WyKV6W2R+qsyf1URuuyffmJ6QRP20ymor7wU3x11uyROuQGst66XQphImUulYOhQSKfhjjsSc5PwQimQ9WLYoHo+WN/KRgML+EcVpPxNkCtAPl0vdbuklHoLYApcIgm2dq3fMxYEPpQddFDcFRWdAlkvamuNI/YocAZZBQSyvnS9FMSkVLq64lEBTKRCrFwJI0dCaysMH+6XKquAAlkvlq/t43iJbOlmqB8UXTEltuKGG1nyv/8LKGxJ+cgOY7riUaQCbbmlD2MjR1bcaIueKJD1YE0qzQfrW1mdKnBjf9ACdQOjLaqIcpclM12xMT/6kYKYxC6zRKnxEyIV7u674eST/Y3Cq4gCWQ9SaX9j6T3GjyjsDTJXWZa57pYl1RWTcpG7RKmlSZEKs3Ah7LEH/PvfMGVK1YUxUCDLy9BBBc4HC1J+OGsJ9GXURC5txpdypa6YSBWYPx922MHfS/mUU+Chh+KuKBYKZMWUTvlbE5XA6jlzSL30Eg3bb9/nn1UQk3LS3cBWdcVEKtAzz8Duu/swtt12VRvGQIGsuEp8lWXD9tsz4eqrSvZ5IsUw99W5zF8+n0kbT1IQE6lk8+bBXnv581128eGsiimQFVORA1n2MmWh3TGRcjRp40lcMfWKuMsQkWK67z5/nDwZnngi3lrKQHnfcTrp0s1Fm9SfGUmR2f/VsP32DJ0+vSifJVIKsxfM5qQ7T2L+8vlxlyIixfTMM37g6/e/Dy+/rDAWUoesmIKWonXIMp0xjaSQpMvduJ9ZphSRCnTbbfDpT8PgwbBmDWyzTdwVlQ0FsmIKmosSyFbccCPr5s2jccoUhTFJNI2zEKki118Pxx3nz/fdN95aypACWbG0BtAWFCWQZbpjWqKUJNPEfZEqctVV8IUv+PPDD4dbbom3njKkQFYsQcofI9pDlruBX90xSYLcG4Bn02wxkSqxbFlHGDv2WLjuunjrKVMKZMWSCWR5TOrPZ6hr9vBWbeCXpMgeYZFLS5QiVWLkSGhq8qMt/vznuKspWwpkxdIeyLq+l2V2CMu9ZVFXNLxVkiK7K5YJYxphIVKFfvITuP12eOQRP3NMeqRAVizpzJJl1x2y7Mn6CltSKXI36U/aeJKumBSpRuecA+ef78/nz4dJG3bJpTMFsmLppUMGmqwvlUP3nBSRdt/9Llx4oT8/+2yFsTwpkBVLD3vIssdWiCSV7jkpIhs49VT43e/8+Y9/7DtlkhcFsmLppkOWmbAPGlshyaZ7TorIBq4KV30uvBDOOCPeWhJGgaxYutlDpgn7kkRdja/Qhn0RabdyJQwfDu+/D/feC1Onxl1R4uhelsUSNPtjF4NhNUNMkiSzUT+zLJmhDfsiAsCRR8KIEXDeeVBXpzBWIHXIiiVo8cci3ctSpNi0UV9EevXJT8LcsHu+7bbx1pJwCmTFkg47ZOGk/szcscyoC5Fy1d3NvhXGRKSTgw6C++7z57feCocdFm89CadAViztm/obOm3kz8wcEykXufvDFMREpFdHHdURxu64Q8uUEVAgK5asQKaN/FLOcm9vpCAmIr364Q9hzhy46y7Yb7+4q6kICmTFEgayFX+7vX3mmMKYlAvd3khECnLQQfCLX8DkydDcHHc1FUWBrFjSKbAaVs+9A9DMMYlfd4NcdbWkiORlxx3hxRdhn31g3bq4q6k4CmTFEqTap/SrOyZx6mqTvpYlRaRPtt0WFi4EM90ovEgUyIolSPV4H0uRYtPVkiISiQkT4M03fRhbsAC22SbuiiqSAlmxpFMbTOkXKZXMMFdQEBORfnjyyY4w9uabMG5c3BVVLAWyYlGHTGKU2SumYa4iUrAg8Jv3//hHOPRQGDMm7ooqmgJZsWTtIRMppdkLZvP4u4/TNLpJYUxECjNiBKRSsGYNzJgRdzVVQfeyLJZ0MytebGOdNj9KiWW6Y7pyUkT6LAhg2DB/s/Ag8KFMSkIdsiJZ8fhSltyzCtDICykddcdEpGBBAMOHwwcfQH09LF8OgwfHXVXVUCArktXPrwY0nV+KI/d2RxmZKyrVHRORPsuEsYEDfYesoSHuiqqKliyLYMUNN7Ju0XoaJwxWGJOiyNzuKFfT6CZt5BeRwuyxBzQ2wtq1CmMxUIcsYtk3Eh+6+6YxVyOVJtMZ0+2ORCQSa9fCV74C11wDDz0UdzVVTYEsYu03Et/XGLH3hJirkaTLXZrMHfIqIlKwtWv91ZRB4OeL/fzncVdU1RTIiqBxyhRGbPsw1KnlK/2T3Q0DDXkVkYgsW+bnirW2+lCmMBY7BbIIrbjhRtbNm0fjlCma1C/9ln3FpJYmRSQyS5bA2LHQ1gajRsF778VdkaBN/ZHKLFcOnT5dk/ql3zRPTESK4uCDfRjbbDOFsTKiQBaxxilTGHHk4eBaNalf+k3zxEQkck89BZ/9LLz9dtyVSBYFsmIIwsnG6pBJH81eMJuT7jyJk+48qcuxFiIiBXn+eairg+9+1x//+te4K5IcCmTFkA4DmfaQSR/MXjCbWY/Oar+SctLGk7RcKSL99+STsPPOfgP/3XfHXY10Q5v6i6G9Q6arLCV/mT1jGuwqIpF59FH4n//x57vtBv/9b7z1SLfUISsGBTIpkPaMiUhkHn64I4xNmaIwVuYUyHqwvrUNgPoa69sPppv9sV6BTPKTGXEhIhKZMWPADPbZBx57LO5qpBdasuzBqnVpAIYNqu/bDwYt/qgOmeQhs3cMNOJCRCJw222wySY+iK1f7zfxS9nTv6UerE4VGsjCDpkCmeRBe8dEJDLXXw/HHec7YwpjiaIlyx6sSQUADGnoPZBlpvQD6pBJn2nvmIj025VX+jAGcMQRCmMJo0DWgzbnAKjJ459Spyn92kMmedLeMRGJxCWXwEkn+fPjj4ebboq3HukzxecINU6ZwohjjoZnbvQPaFK/hGYvmN2+NC5Oo1cAACAASURBVJktE8a0d0xE+uUb3/DHL34RLr883lqkIApkxaBJ/ZIle9N+0+imTs81jW5i2lbTtFwpIv1z0UWwYAFcfHHclUiBFMiKQZP6JYs27YtIUZxzDvz2t7BsGZx6atzVSD9pD1kxqEMmObRpX0QidcYZcP758MEHcOONcVcjEVAgK4b2QKYOWbXTpn0RidzXvga/+pU///GP4fOfj7ceiYSWLIshSIHVQq3+8VarzCZ+bdoXkUidfDJccYU///Wv4fTT461HIqPEEIHMDLLGKVP8A+mU9o9VubmvzmX+8vnatC8i0Xr7bX+89FI45ZR4a5FIKZBFoNMMMvCT+jUUtmpllimbRjdxxdQr4i5HRCrBbbfBYYfBnXfCypUwfHjcFUnEFMgi0j6DDPykfgWyqqNlShEpimnT4I47oKkJ5s1TGKtQCmTFkG7WlP4qpGVKEYncAQfAAw/48x/8IN5apKgUyIpBHbKqkumMzV8+n0kbT9IypYhE4yMfgUce8ed33w0HHxxvPVJUCmT9tMGGftAesiqRu0SZ6YyJiPTbeed1hLGHHoJ99om3Hik6BbJ+2mBDP/gOma6yrHhaohSRojn3XLj6arjmGsj+C79ULAWyCHTa0A9+D1njxvEVJCWjJUoRidROO8F++/mxFgsWxF2NlJAm9RdDkNKSZQWbvWA2J915EvOXz4+7FBGpJFttBS+8AP/3f3FXIjFQh6wYFMgqWvYGfu0ZE5FIbLEFLFoEZrBwYdzVSAwUyIohndLYiwqnpUoRicyYMfDuu1BTA2+8AePGxV2RxECBrBjUIRMRkXykUvDeez6MLV7sw5lUpZLtITOzqWY238wWmtn3u3h+vJk9YGZPmdkzZpbctSAFMhER6UkQwPz50NAAr74K77+vMFblShLIzKwWuAQ4BNgROM7Mdsx52Q+AG51zewDHAr8vRW2Rcy6c1K+xF5Uoc59KEZGCBQGMGAHbb+9vhTRxom6HJCXrkO0FLHTOveqcWw9cDxyW8xoHDA3PhwFvl6i2aLWmAQd1A+OuRIpg7qtzAd2nUkQKFAQwbBisXQv19bDLLnFXJGWiVHvIxgJvZX2/CPhQzmt+CNxtZqcCGwEHlaa0iAXN/linDlmlyXTHmkY3aQisiPRdEMCQIX7f2MCBsHKlX7IUobzmkB0HXOmcGwdMA642sw3qM7MZZva4mT2+dOnSkhfZq6DFH9UhqyizF8xm1qOzAHXHRKRAo0b5MDZokO+QKYxJllIFssXAFlnfjwsfy/ZF4EYA59yjQAMwMveNnHOXOeeanHNNo0aNKlK5/ZAOO2TaQ1ZRMkuVM/eeqe6YiBTmwgt9KFu9Guo05EA6K1Ugmwdsa2ZbmtkA/Kb9v+e85k3gQAAz2wEfyMqwBdaL9g6Z/uZTabRUKSJ9tnKl37S/di188Yt+xIXCmHShJIHMORcA3wDuAl7EX035vJnNMrNDw5edAXzZzJ4GrgNOdM65UtQXqfY9ZApkIiJVbdkyGDnSD3v95CfjrkbKXMliunNuLjA357GZWecvAB8pVT1Fk075oyb1i4hUryVLYOxYaGuDTTeFf/wj7oqkzJXTpv7KEISBTB0yEZHq9PrrsPnmPoxtvrm/LZJIL7SQHbX2QKZN/Uk3e8Hs9s38mZuJi4j06he/8EPCJ0zw4UwkD+qQRa09kGnsRdLNfXUu85fPB/zNxDXuQkR6tHatP/7+9zB7tsKY9Ik6ZFFr30OmDllSZTpjma7YFVOviLskESl3Tz4Je+7Z0RU78si4K5KEUYcsarrKMvGyw5i6YiLSq0cf9WEMdE9KKZg6ZFHTHLJE6mq/mDpjItKr+++HAw/053vtBf/5T7z1SGKpQ9YPK264kXXz5nV+sH1SvwJZkmi/mIj02aJFHWFs330VxqRf1CHrh9Vz5gAwdPr0jgfVIUsU7RcTkYKNGwejR8Muu8A998RdjSScAlk/NU6Zwohjju54IGiGmnqoqY2vKMlL9g3Dm0Y3qSsmIvm59lq/VHn55X4ArEgEFMiiFrSoO5YQumG4iPTZn/4EX/qSP//hD32XTCQC2kMWtXSz9o8liG4YLiJ5u+SSjjD22c8qjEmk1CGLWpDSlP4ylX0lJWj6voj0wQUXwPe+58+//GW47LJ465GKow5Z1IKUpvSXqewrKUFXU4pIH5x7rj9+85sKY1IU6pBFLZ3SkmUZ05WUItInQQB1dX76/jXXwBlnxF2RVCh1yKIWpLSpv8zMXjCbk+48qVN3TESkV6efDvX1MGcOjBmjMCZFpQ5Z1BTIykZmz9jj7z4OaLSFiPTBV78Kf/iDP58/H7LnTYoUgQJZ1IIUNI6Mu4qq19WMMV1NKSJ5+fzn4eqr/flFF8Gpp8Zbj1QFBbKoaQ9Z7LLDmGaMiUifZIexP/4RZsyItx6pGtpDVqAu72MJflK/xl7ERmFMRPrl0EOhpgb+8heFMSkpBbICdXkfSwgn9WvsRRwUxkSkYF/7GqxcCUceCa2tvlMmUkIKZP2wwX0sIZzUrw5ZHHQrJBEpyEc/CpdeCuPHx12JVDHtIYuaOmSx0q2QRKRP9t4b/v1vf37rrfHWIlVNHbIoOac9ZDGZvWB2+3gLEZG87LlnRxh76CE44IB465Gqpg5ZlIIWf9RVliWXWa7UnDERycvDD8OTT/rzxx6DKVPirUeqnjpkUQpS/qjBsCWV6Y5puVJE8rbPPnDKKfD00wpjUhbUIYuSAlks1B0TkbxtuSW0tMDbb/uN/CJlQoEsSgpkJafumIjkbdw4WLwYzCCVggb9Xi3lQ4EsSukwkGkPWcmoOyYieRkzBt591w99XbxYYUzKjgJZlNQhK5nMjcPnL5+v7piI9GzkSHj/faithSVL/PciZUaBLEoKZEWXCWKZEReZG4eLiHRr4EAfxpYtg+HD465GpEsKZFFKN/ujJvUXTXZXbNpW09QZE5GuBQH87ndw+ul+iTIIoE5/5En50n+dBcjcWLwx91LpzBwyTeoviuwN/FdMvSLuckSkXAUBDB0Kzc3wwQdwzjkKY1L29F9oAbq/sXjYIdOk/shl3zhcS5Qi0q0ggCFDOq6iPPPMuCsSyYsGwxaoyxuLq0NWFNlhTDcOF5FupVKw0Ub+2NgIa9aoMyaJoUAWJe0hK4rMaAuFMRHp0U47wfr1MHgwrFqlMCaJokAWpfYOma6yjIoGv4pI3p54AiZPhhUrFMYkcRTIotS+h0yBLCoa/CoiPVq2zC9TXn+9H2nxxBMKY5JICmRRSmsOWZTUHRORHi1aBKNHw7p1MHNm3NWI9IsCWZSCFNQO8LfmkH7RVZUi0qPXX4fx46GtDcaOhQUL4q5IpF+UHKIUpDTyIiLayC8i3Zo/H7baCpyDiRN9p0wk4RTIohSkNPIiQlqqFJEuvfmmD2PbbQevvRZ3NSKR0M7HKKVTUK/9YyIiRfHkk7D55nDwwbB0qW4SLhVFgSxKQbOWLEVEiuHhh2Hfff1NwoNAYUwqjpYsoxS0aMlSRCRq99/vwxjAXnvFW4tIkSiQRSndrCn9/TR7wWxOuvMk5i+fH3cpIlIO5syBAw/05x/9KDzySLz1iBSJliyjpA5Zv2SPumga3aRxFyLVLgjgU5/y55/4BNx5Z7z1iBSRAlmUgmZoGBZ3FYmlURci0kldHXzhC/6+lH/7W9zViBSVliyjpA5Zv2nUhYhw2WUwYYI/v/JKhTGpCgpkUdIesoJlbpMkIlXu4ovhK1/xs8ZuuinuakRKRoEsSkFK97EskG4iLiL87GfwzW/681NOgSOPjLcekRJSIIuSAllBdBNxEeFHP4KzzvLnp50Gl14abz0iJaZAFiVN6i+IumMiwg03+ONZZ8FvfhNvLSIx0FWWUXEOWlvUISuQumMiVer11/0Nwl94AebNgylT4q5IJBbqkEUlSPmjAlmfaDO/SBX70pdgyy3hiCP89wpjUsXUIYuKAlmfZQ+C1XKlSJU54QS45hp/npnEL1LFFMiikg4DmfaQ5U2DYEWq1FFHdYy0uPxy+OIX461HpAwokEUlaPbHOs0h6wvtHROpMmef3RHGrrkGjj8+3npEyoT2kEUlaPFHTeoXEenezJkwdCjcfLPCmEgWdciikg47ZJrULyKyoYMP9kuVM2b4e1OKSCfqkEVFHTIRka596ENw773wta/FXYlI2VKHLCraQyYisqHJk+Gpp/z5Qw/FW4tIGVMgi4qushQR6WynnfzAV4AnnvDhTES6pCXLPlpxw42smzdvwyc0h0xEpMPKlT6MmcFzzymMifRCgayPVs+ZA8DQ6dM7P6FA1iea0C9SwVIpGD4c7rsPXnzRd8pEpEdasixA45QpjDjm6M4PKpD1iW4oLlKhxo6Fd96BN9+EAw6IuxqRxFAgi4r2kPWZhsKKVJhNN4WlS6GmBur0x4tIX+j/mKioQyYi1WyTTWD5cqithSVLYOTIuCsSSRQFsqgokIlItRo1yoexujrfIRs+PO6KRBJHm/qjkm72Ycws7kpERErr2GOhoQFWrFAYEymQAllUghZN6ReR6hEE/nZIABdfDM3NMHhwvDWJJJgCWVSCZk3pz5NGXogkXCrlw9e998InPhF3NSIVQXvIoqIOWV5mL5jNrEdnARp5IZJIqRQMHQrpNGy0Edx+e9wViVQEdciikm6GenXIepOZPzZz75kaeSGSNGvXwpAhPowNHeqn8Wu8hUgk9H9SVNQhy5vmj4kk1Ikn+r1jw4f7DfwiEhl1yKKiPWQiUuluugm+8x2FMZEiUCCLSjqlKf0iUnkWLfLLkgcd5L+/4IJ46xGpUApkfbDihhtZN29e108GKQ2FFZHKsnAhjB8Pra3w8stxVyNS0RTI+mD1nDkADJ0+fcMnFchEpJLMnw/bbQfOwVZbwRtvxF2RSEVTIOujxilTGHHM0Rs+oUDWK80fE0mIhQth++19GNtuO3jllbgrEql4CmRR0R6yXmVGXmj+mEiZmzgRBg6EXXbxnTIRKTqNvYhKkNJVlnnQyAuRMvbww/DUU3DqqX4ArIiUjAJZVIKU5pCJSHLdcw98/OP+/LjjYOTIeOsRqTJasoxCWyu0rtek/h5o/5hIGbvtto4wduCBCmMiMVAgi0LQ4o/qkHVL+8dEytT118OnP+3PDznE3zBcREpOgSwKQbjXQnvIupTpjmn/mEgZOuEEfzz8cJg7N95aRKqYAlkU2gOZOmRdUXdMpIw9+CB8+ctwyy1xVyJS1RTIopBu9kftIeuWumMiZeQ3v4H6ej9vbJ994LLL4q5IpOrlHcjMbCszO9vMLgm/n2RmOxWvtARp75BpDpmIlLmf/Qy+9S0IArjqqrirEZFQXoHMzA4GngY+DHwufHgkcGGR6koWBTIRSYKZM+Gss/z5t78Ns2bFW4+ItMu3Q/Yz4Cjn3KFAa/jYk8DkolSVNOkwkGlSv4iUq+9+F847z5+ffTb88pfx1iMineQbyLZ2zt0ZnjsA51wzUJ/vB5nZVDObb2YLzez73bzmaDN7wcyeN7Nr833v2KlDJiLlrib87X7WLPjJT+KtRUQ2kG8ge8vMds5+wMx2A17P54fNrBa4BDgE2BE4zsx2zHnNtsBZwEecczsBp+dZW/wUyESkXF12md8v9vOfQzoN554bd0Ui0oV8A9lFwC1mdgJQa2afAf4K/DrPn98LWOice9U5tx64Hjgs5zVfBi5xzq0AcM69l+d7x0+BTETK0Wc/C1/5Cowd67+v093yRMpVXoHMOfd/wAXAmUAt8CPgt865q/P8nLHAW1nfLwofy7YdsJ2Z/cvM/m1mU/N87/hpD1m3dMskkZgceSRcG+78+PnP461FRHqV91+XwlD2f0WuZVtgf2Ac8E8z28U5tzL7RWY2A5gBMH78+CKW0wdBOIdMk/o7mb1gNrMe9VdxaSisSAl98pMdU/evuw6OPTbeekSkV/mOvXixm8efzfNzFgNbZH0/Lnws2yLg7865tHPuNWABPqB14py7zDnX5JxrGjVqVJ4fX2S6l2WXMhP6Z+49U0NhRUrlzjs7wtittyqMiSREvnvIxvXx8VzzgG3NbEszGwAcC/w95zW34rtjmNlI/BLmq3m+f7w0qb9bmtAvUmJTp8K++8Idd8BhuVt1RaRc9bhkaWZnZ16XdZ6xDZ33hXXLOReY2TeAu/B70P7snHvezGYBjzvn/h4+93EzewE/6+y7zrn3+/BriU/QAhjUDoi7EhGpVlOm+NshPfII/POfcVcjIn3U2x6yg8NjfdY5QBuwBDg53w9yzs0F5uY8NjPr3AHfDr+SJWj2V1iaxV2JiFSjXXeFZ5/1vwcFga6mFEmgHv+vdc59DMDMLnbOnVqakhIondIVliISj+23h/nz/fnjjyuMiSRUvmMvFMZ6EqQ0gyyHxl2IlMA22/gwZgbPPQeTdTc7kaTK669SZjYI+AFwIDAKaF+bc85tVZzSEkSBbAOZKyw17kKkiBYv9mFswQIfzkQksfK9yvLX+Mn6VwOjgV8CLcCfi1RXsiiQdUlXWIoUycMP++OaNfD22wpjIhUg30D2KeBQ59wlQBAePwN8rGiVJYn2kIlIqYwa5cda/OlPfr/YmDFxVyQiEch39+dg51xmJth6MxvgnHvBzKYUq7BEUYdMREphxAhYuRJqa+Ezn4m7GhGJUL6B7DUz28E59yLwEnCyma0EVhWvtAQJUjBgcNxViEilCgLYZBNYvdp3xVasgMH6PUekkuS7ZPlTIHPjyPOAC4GrgB8WoabkSac0pT+LrrAUidi22/owVl+vMCZSoXrtkJmZAQ8AywCcc/eY2QhggHPugyLXlwxBSvexzKIrLEUidvPN/obhr70GDdoeIVKJ8umQGfBG9mvDG4ArjGUEKahThyybrrAU6adUCjbbDJ55xs8Xe+cdhTGRCtZrIHPOteFv8j2i+OUklDpkIhKlVAqGDoUlS3SDcJEqke8esl8C15rZR8xsnJltnvkqZnGJoT1k7bR/TKSf1q6FIUMgnfah7OWX465IREog36ssLw+PBwIuPLfwvDbqohJHHTLAh7FZj84CtH9MpCDLlvm5Yq2tfsTF8uVxVyQiJZJvINuyqFUkWVsrtKWrfg9ZdhibufdM7R8TKcQtt/gwNmoUvPde3NWISAnlFcicc28Uu5DESjf7Y5VP6s9cWakwJlKAJUtg5EiYMQN23BH22SfuikSkxPLdQybdCVr8sYon9Wf2jenKSpECLFwIm2/u94uBwphIlVIg668g7JBVcSDT3DGRAj3/PGy3HTjnQ5mIVC0Fsv6q8g6ZumMiBXrySdh5Zx/GdtjBd8pEpGopkPVXFe8h01WVIgUKAthzT3++227wwgvx1iMiscv3KkvM7ATgC8Bo59yuZrYfMNI5d0vRqkuCIOWPVXiVpTbyixSorg722st3xx57LO5qRKQM5NUhM7NvAz8C7qDjJuNLge8Vqa7kaA9k1TmHTEuVIn1w552w//7+/D//URgTkXb5Lll+FTjEOfcrOgbDLgC2KUpVZWjFDTeybt68DZ9Ih4FMk/pFpCe33QaHHAL/+IcPZiIiWfINZBs75xaE57mT+qvC6jlzABg6fXrnJ6q0Q6ZbJIn0wfXXw6c/7c+nTYOpU+OtR0TKTr6B7AUzy0kiTAWejriestY4ZQojjjm684NVuodMoy5E8nTllXDccf78M5+B22+PtRwRKU/5buo/G7jdzG4EBprZxcCxQG5Iqz5V2iED7R8Tycv3wq22xx8P11wTby0iUrby6pA55x4CPgw0Aw+EP7e/c+4/RawtGdrHXlRXh0xEehEE/vj223DppQpjItKjvOeQOedecM6d6pyb7pz7unPu+WIWlhjtHbLqm0MmIt047zyor4dzzvEjLk45Je6KRKTM5Tv2YqGZnW1murdHrioMZNrQL9KDc86BmTP9+fr18dYiIomRb4fsx8DHgdfN7A4zO8rM6otYV3KkU2A1UFs9/zi0oV+kG2ecAeef78/PPRcuuCDeekQkMfLdQ3alc25/YAfgceAC4B0z+20Ra0uGIOW7Y2ZxV1JS2tAvkuOcc+BXv/LnP/0pzJoVbz0ikih9upelc+4V59y5+A3+/wG+UZSqkiQTyESkun3+836/2K9/Dd//ftzViEjC5B3IzKzWzA4zs1uB14HhwFeKVVhiVFkg0/4xkRwzZsCjj8KkSZBOw+mnx12RiCRQXnPIzOzXwHFAAFwNnOmcm1/MwhIjnYL66glk2j8mkuXww+HWW+GKK3wYExEpUL6DYccCJwF3OefailhP8gSpqpvSr/1jIvhbIN1xhz+/+up4axGRxMt3U//Rzrk7FMa6EKSqZkq/litFQgcc0BHGbr0Vjj023npEJPG67ZCZ2XeccxeG52d39zrn3PnFKCwx0qmqmdKv5UoRYNkyeOABf3733XDwwfHWIyIVoaclywOAC8Pz7n7HcUB1B7IgBQ1D466i6DLdMS1XStUbOdJfSTl5Muy3X9zViEiF6DaQOeemZZ1/rDTlJFCQgrrRcVdRdOqOSdXbdVd4+WVYs0ZXUopI5PK9ddJl3Tx+abTlJFC6uWquslR3TKrWpEnw7LOQSsHrr8ddjYhUoHznkHW3Y/XoqApJrKCl4ueQaTO/VLWtt4YFC/zdOF56CbbZJu6KRKQC9Tj2wsz+JzytMbO9gez7A20LfFCswhIjaK74QKblSqlaEybAm2/6MPbqqzBxYtwViUiF6m0O2cPh0QH/ynrcAe8A5xSjqESpgg4ZaLlSqtSWW8LixX6Zcty4uKsRkQrW45Klc67GOVcDPJs5D79qnXPjnHN/KVGd5auK9pCJVI0zz/THBx+EIFAYE5Giy2tSv3Nu92IXkkitAbjWiu6QZY+7EKkKI0bAypV+iXL27LirEZEqocGw/RE0+2MFBzLtH5OqEQSw8cZ+rEVdnb8/pYhIiWgwbH+kU/5Y4ZP6tX9MKl4QwLBhsG4d1NfD6tXQULl/0RKR8qPBsP0RhIGsSu5lKVKxDjnEh7GBA/1ypcKYiJRYvnPIOjGz/c1sn6iLSZz2QFaZHTLNH5Oqcccd8LGPwdq1CmMiEot8J/XfbWb7heenAXOBO8zsW8UsruxVeIdM+8ekoq1dC42N/jZIdXVw//3+KCISg3w7ZLsDj4bnXwY+DuwNfL0YRSVGFewh0/4xqUgrV8Lw4dDcDNdcE3c1IiL5jb0ABjjn0mY2GtjUOfcwgJltWrzSEqDCO2QiFWnZMhgzBlpb/VWVS5fGXZGISN6B7FUz+wKwNXA/gJltAqSKVVgiVPgeMpGKs2QJjB0LbW2w6abw7rtxVyQiAuQfyL4H/AVoAQ4LH/skMK8YRSVGOpxDVoGT+jUQVipSEIBzsPnm/pZIIiJlIt9J/fcCY3Mevi78ql5Biz9W4GBYbeiXijJ/vu+G7befH2+hKylFpMzkfUmRmQ3Gd8W2AN4CbnfOrS1WYYlQ4ZP6taFfKsLzz8Muu/jO2Jo1MHhw3BWJiGwg37EXOwEL8JP7DwuPC8xs5yLWVv4qtEOm+WNSMZ58Enbe2YexHXdUGBORspXv2IvfAH8Exjvn9gXGA5cCvy1WYYlQoXvItFwpFeHRR2HPPf35Hnv4TpmISJnKN5BNBs53zjmA8Pgz/Hyy6lXBV1lquVISb999/XGvvXynTESkjOUbyFYCE3MemwisjrKYxAlSYLVQWznTvbVcKRXjt7+FAw+E//wn7kpERHqVbyD7C3C7mZ1sZgeY2cnA/wOuLFplSZBOVdyUfi1XSqLdcoufwJ9Kwde/DvfeG3dFIiJ5ybe18xMgDZxJx1WWVwIXFKeshAhSFTmlX8uVkkjXXguf/aw/P+88+MlP4q1HRKQP8p1D1gr8NPySjCBVkfvHRBLnT3+CL33Jnx95pMKYiCROX+eQTQfG4Ttkc51za4pVWCJUaIdMJFEuuQS+8Q1//tnPwl//Gm89IiIFyCuQmVkTMBdoBt7Ej7242MymOeeqdwd4Be4hE0mczD6xL38ZLrss3lpERAqU76b+3wO/dM5NcM7t65ybgB8Oe2nxSkuAoLnihsKKJEZmlMXf/gYvv6wwJiKJlm8g2wH4Zc5jvwK2j7achAlaFMhE4nDmmX7o6447+u+32SbeekRE+infQPZfIPc2SbuEj1evdHPFTekXKXunnw6/+IU/P+aYeGsREYlIvpv67wbmmNnlwBv4obAnA5eZ2fGZFznnro28wnKmDplIaX31q/CHP/jzn/4Uvv/9eOsREYlIvoHsZPwcsi9kPRaEj2c4oMoCmfaQiZTMTTd1hLGLLoJTT423HhGRCOU7h2zLYheSSOqQiZTOkUfCFlvAD34AM2bEXY2ISKQq5yaMcaiwPWSZ+1g2jW6KuxSRDocfDptsApdfDm++GXc1IiJFoUDWH0Gqojpkuo+llJ2pU+Guu/z5H/4AdfotS0Qqk353K5RzFRfIQPexlDKy//7wj3/48//3/xTGRKSi6Xe4QrWmwbVV1JKlSNnYe2/497/9+X33wQEHxFuPiEiR5TuHDADzNitWMYkSpPyxQjpkmf1jImXhscf88aGHFMZEpCrkFcjMbLCZ/Ql/L8uF4WOfNrP/LWZxZa3CApn2j0lZWLnSH994A554AvbZJ956RERKJN8O2S+B0cBHgPXhY/OA6hiTPedW1s2b1/mxdLM/VtDNxbV/TGI1aRKMGAH33APjxsHkyXFXJCJSMvnuIZsO7OicW2VmDsA5t9jMNi9eaeXD7vNXeQ2dPr3jwaDFHyukQyYSqy23hNdfBzMYPz7uakRESi7fQFaDX65sZ2aDgbWRV1SmGqdMYcQxR3c8EIT/OBTIRPpn3DhYvNiHsVdfhYkT465IRKTk8l2yfBg4K+exU4EHoi0nQdQhE+m/bbf1YaymBt5+W2FMRKpWvoHs28AJZvYyMNjMngVOBKr3zr7te8gUyEQKdu65fCbaUAAAIABJREFUMGgQvPsujBkTdzUiIrHJK5A5594CdsZ3yc4GZgG7O+cWF7G28qYOmUhhggCmTIG1a+Hzn4d162DkyLirEhGJVd6DYZ1zLcBNRawlWSpoD5nuYSklEwT+Ssq1a2GPPeDll+OuSESkLOQVyMzssu6ec87NiK6cBEmHc8gqYOyFZpBJSQQBDB0Kzc0wYAA8+2zcFYmIlI1895DV53xNAD4HJD+NFKp9MOzAeOuIiGaQSVGlUjBkiA9jDQ3wwQf+KCIiQJ4dMufcSbmPmdnhwMcjrygp2gNZ9WZSkbz98pc+lDU2wqpVulG4iEiOPt3LMsetVMuk/q5UWIdMpCiCwB/POQcuvFBhTESkG/0JZIeQMyy2qqQr616WIpFbudIvS2Ym759xhsKYiEg38t3U/zLgsh7aCNgUOK0YRSVCkIKaOqjVHzAiG1i2zM8Va231Yy1ERKRH+aaJH+d8vwb4r3Pu1Xw/yMymAr8FaoHLnXM/6+Z1n8GP15jinHs83/cvuSCl/WMiXVm0CCZMgLY2GD0aliyJuyIRkbLXayAzszpgNHCRcy5VyIeYWS1wCXAwsAiYZ2Z/d869kPO6Ifiu238K+ZySSjdXxJR+zSCTSKVSfonSORg71oczERHpVa97yJxzAXB2oWEstBew0Dn3qnNuPXA9cFgXrzsP+DnQn88qjaClIvaPaQaZRKqhATbd1N+TUmFMRCRv+W7qf8DMPtqPzxkLvJX1/aLwsXZmNhnYwjl3ez8+p3SC5ooIZKAZZBKBZ56BU0/150uWwGuvxVuPiEjC5LuH7HXgNjO7KTxvyzzhnDu/v0WYWQ3wK/wNy3t77QxgBsD4zNVbcaiQDplIv82bB3vt5c+POQb22SfeekREEqjHDpmZrQ5PdweeArYGDsTvBTsYOCjPz1kMbJH1/bjwsYwh+JuXP2hmrwMfBv5uZhtsbHLOXeaca3LONY0aNSrPjy+CCthDltk/JlKwhx/uCGOTJyuMiYgUqLcOmQE45z7Wz8+ZB2xrZlvig9ixwPGZJ51zq4CR7R9q9iDwnfK/yjLZgUz7x6Rf7r8fDjzQn++9NzzySLz1iIgkWG97yFwvz+clvDDgG8BdwIvAjc65581slpkdGsVnlFwFBDLQ/jHph8PC63I++lGFMRGRfuqtQ9ZgZn/u6QXOuZPz+SDn3Fxgbs5jM7t57f75vGes0qnEL1mK9Mtrr8G558Kll8ZdiYhI4uVzlWVrL1/VqUI6ZCJ9ctVVUFsLt9wCI0cqjImIRKS3DlnKOfflklSSNApkUm0uuwy+8hV/fuutcMQR8dYjIlJB+nNz8eqmQCbV5OKLO8LY5z7nO2UiIhKZ3gKZlaSKJEr4HjKNvJC8XXABfPOb/vyUUxTGRESKoMdA5pwbUqpCEsW5cFJ/Mm8uPnvBbGY9OgvQyAvJw9Zb++Npp2nPmIhIkeQ7qV+yta73x7qB8dZRoMz8sZl7z9TIC+neb34DJ5zg94ql01Cn3y5ERIpFe8gKkW72x/pkdshA88ekF6edBt/6FmwR3mBDYUxEpKgUyAoRtPhjAjtk2jsmvZoxAy66yJ/PmhVvLSIiVUJ/7S1EEHbIEriHTLdLkh6dcAJcc40//93v4Otfj7ceEZEqoUBWiHTKHxN2lWWmO6blSunS6693hLHLL4cvfjHWckREqokCWSGCMJAlbA6ZumPSo4kT4QtfgI9/HI4/Pu5qRESqigJZIRIayECb+aULBx8Mzz8Pb78NV14ZdzUiIlVJgawQCQ5kIp3stx889JA/X7QIxo2Ltx4RkSqlQFaIhO4hE+nkQx+Cxx7z5/fdpzAmIhIjBbJCqEMmSbf77vD00/78kUdg773jrUdEpMppDlkhEhjINH9MOlmxwh+feEJhTESkDCiQFSKBk/p1haUAcNtt/vjGG7BmDUyeHG89IiICKJAVpn1Sf3I6ZKArLKvehAnw6U/Dqaf67wcPjrceERFpp0BWiPZJ/ckIZFquFMaOhTffBDM444y4qxERkRza1F+IhHXItFxZ5TbdFJYuhZoaWLwYxoyJuyIREcmhQFaIdDPUDvB/wCWEliur1L77+jBWWwtLlsDIkXFXJCIiXUhOoignQUtiumNS5e64A8aPh2XLFMZERMqYAlkhgmYFMilfQeCXKa+80m/cf+MNGD487qpERKQHCmSFSKc0pV/KUxDA0KF+mfJb34q7GhERyZMCWSGClDpkUn5SKd8Ra26GhgYfykREJBEUyAqhQCblJpXynbGWFmhs9ENf63TNjohIUiiQFUKBTMrNU09BOg1DhsCqVQpjIiIJo0BWCO0hk3KxciUsWuTvR/ncc7B6tcKYiEgCKZAVImiGumTcx1JT+ivYkiWwySb+lkipFOy0U9wViYhIgRTIChG0QN3AuKvIi6b0V6hFi/ztkNra/IiLBnVsRUSSTIGsEOlmqE9Ghww0pb/iLFzoh722tcEWW8A778RdkYiI9JMCWSES1CGTCrT99uAcbLWVv2G4iIgkngJZIRK0h0wq0Iknwg47wCuvxF2JiIhERIGsEOqQSanNmweTJvnzyy+HF16Itx4REYmUAllfOefnkCVoD5kk3D//CXvtBQsWwE9+Enc1IiJSBApkfRWk/FGDYaUU7vn/7N15XFT1/vjx1wew3EVFrTSVXBCBARTMDddcS0tzNxVN7WdZaurVvl6L6lbaNTXN8lqmZgZoWZrrdcnSXBIMUXFBDROvmRsqKcrA5/fHGSYQhBmFGZD38/HgMWdmzpzznjkRb9+f93w+m6B1a2O7eXOYMsW58QghhCgQkpDZSxIy4SirVkHHjsZ227bw88/OjUcIIUSBkYTMXqmWhKwIzNQvk8IWcW+9Zdx26QJbtzo3FiGEEAVKEjJ7FaEKmUwKW0QlJxu30dHwzTewbp1z4xFCCFHgJCGzVxFKyEAmhS1y5s83Fgh/8knjfs+ezo1HCCGEQ0hCZq8ilpCJImT2bBg1ytiuVs25sQghhHAoScjsVUR6yKR/rIiZNg3GjTO2R42Czz93bjxCCCEcShIye5lvGLeFfKZ+6R8rQr74Al57zdh+9VX4+GPnxiOEEMLhJCGzl/mmcVsEZuqX/rEiok8fKF0a/u//4IMPnB2NEEIIJ3BzdgBFTqqlQiYz9Yt7NWYM1K8PL70Ef/3l7GiEEEI4kSRk9ipCFTJRiA0fDgsXglJGQiaEEKJYk4TMXkWkh0wUYgMHwldfGdvSLyaEEAJJyOxXRL5lKQqpXr2MyV4BFi2C0FCnhiOEEKJwkITMXjIPmbhbZvPfyVh4OPTr59x4hBBCFBqSkNlLEjJxN8xmcHODLVvg2jV4+mlnRySEEKIQkYTMXuYUcH3QaMYWwhYhIbBzJ5w4Ae3aOTsaIYQQhZDMQ2av1BTpHxO2Cw6GHTsgPR0SE50djRBCiEJKEjJ7mVMK/XClLJtUSPj7Q5TlOuzcCS1bOjceIYQQhZYMWdqrCCRksmxSIWAywYEDxnZ0NDRq5Nx4hBBCFGpSIbNX6o0iMUu/LJvkZO3agasrHDwoyZgQQog8SUJmL/PNQj1LvwxXOlloqPGNytmzjVsfH2dHJIQQogiQIUt7mW8U6ln6ZbjSiR5+GP74A/btg9hYZ0cjhBCiCJGEzF7mm4W+h0yGK52gShW4cAFcXOC//3V2NEIIIYoYScjslXoDSlV0dhSiMKlYEZKSjJ6xCxfA3d3ZEQkhhChiJCGzVyHvIRMONnGikYy5ucHly1C2rLMjEkIIUQRJU7+9CnkPmXCwf//bWJNSkjEhhBD3QBKyPHRJ2I3a/+vfD8hM/cJshjJloFUr4354uCRjQggh7okkZHloc3ofAOWfesp4oAg09YsClJJiJF/XrxvfphRCCCHygfSQ2UD7B1Kxbx/jjvmGJGTFVUoKlC8PqalGhSwpydkRCSGEuE9Ihcwe6emQdksSsuIoORnKlTOSsfLl/27kF0IIIfKB/EWxhznFuJUesuKnZEljWoty5eDSJWdHI4QQ4j4jFTJ7ZCRk8i3L4uOPP2DhQqMalpwsyZgQQogCIRUye1gTMpmHrFhITISaNUFrCA4Gk8nZEQkhhLhPSYXMHqk3jNsSUiG77x0//ncyVrOmJGNCCCEKlCRk9jDfNG6lQnZ/O3QI6tc3krE6deDUKWdHJIQQ4j4nCZk9zJYKWSHtIVtxbAVR56KcHUbRFxxsJGPe3kalTAghhChgkpDZo5BXyNadXAdA18e6OjmSIu7HH6FtW4iLc3YkQgghiglJyOxRBHrIgqoF0bt+b2eHUfT89BOUKgVHjxoVsq1bnR2REEKIYkQSMntYv2Up85DdVzZsgNatjZn4X3/d2dEIIYQohiQhs4ckZPefVaugSxdju317iIx0bjxCCCGKJUnI7JFaeGfql4b+uxARAc88Y2x37QqbNzs3HiGEEMWWJGT2KMQVMmnovwtHjxq3zz4La9c6NxYhhBDFmiRk9ijECRlIQ7/NNm0CsxneeAPOn4evv3Z2REIIIYo5ScjsUcgTMmGDDz6Ajh2hShXjvoeHc+MRQgghkITMPqkpgCq085CJPLz9NkyYYGw/95xzYxFCCCEykYTMHuYbRnVMKWdHIuw1ZcrfU1pMmABz5zo3HiGEECITScjsYb4p1bGi6OhRePddY3vqVPj3v50bjxBCCHEbN2cHUKSk3ijUs/SLO/DygqZN4emnYfJkZ0cjhBBCZCMJmT2kQla0DBsGhw7Bnj2wa5ezoxFCCCHuSIYs7WG+AW6Fr0Imk8LmoH9/WLQIfvkFLlxwdjRCCCFErqRCZo/UlEI3S/+KYyt4a9dbgEwKa9WjB3z3nbG9ZIlMbSGEEKLQk4TMHuaUQjcHWcYM/a83e10mhQVjCaT1643tFSugVy/nxiOEEELYQBIye5hT4IGyzo4iG5mhP5OffjJuv/vOaOIXQgghigDpIbNHIauQSe9YJgkJxm1SEkRHSzImhBCiSJGEzB6FrIdMFhS3aNwYPD2N+cXc3KBRI2dHJIQQQtjFYQmZUqqzUuqoUuq4UirbZFBKqVeVUnFKqVil1BalVC1HxWYz881CVSEDGa7Ezw/27TO227RxaihCCCHE3XJIQqaUcgXmAV2AhkB/pVTD23b7FQjSWpuAr4H3HRGbXTKWThKFg5cXHDxobO/fD8HBzo1HCCGEuEuOqpA1AY5rrU9qrW8BEUCWJh+t9Q9a6+uWu7uBGg6KzXapKYVmpv5i3z8WEgLHjhnrih45AiaTsyMSQggh7pqjErLqwOlM9xMtj93J88D6Ao3obphTCs1M/cW+f2z+fChbFk6eNCplQgghRBFW6Ka9UEo9BwQBre/w/EhgJEDNmjUdF1h6GqSnOn2m/hXHVrDu5DqOXjpaPPvH/PyMyV4bNYJr15wdjRBCCJEvHJWQnQEezXS/huWxLJRSTwBTgNZa65s5HUhrvQBYABAUFKTzP9Q7MKcYt06skGWelT+oWlDxq455eMDFi9C+PVy+7OxohBBCiHzjqIRsL1BPKeWJkYj1AwZk3kEpFQj8B+istf7TQXHZLtWSkDmxh6xYz8pfsaIxx5irK/z2m7OjEUIIIfKVQ3rItNZmYDSwETgMLNdaH1JKvaWU6m7Z7d9AWWCFUipGKbXaEbHZzMkVsowm/mI3TGk2Q/nyRjLm5mbcurs7OyohhBAiXzmsh0xrvQ5Yd9tjr2fafsJRsdwVa0LmnApZsW3i37zZ6BUrUQKuXoWSMu2IEEKI+4/M1G+r1BvGrRNm6i+W1TGz2fjp3BmWLZNkTAghxH2t0H3LstAyW75j4ISJYYtddSwlBSpUABcXozo2YEDerxFCCCGKMKmQ2cpsqZA5OCErdtWx5GSjZ+zWLaNnTAghhCgGJCGzlbWHzLEJWbGqjmU07KemGknZ5cuSlAkhhCgWJCGzlXXaC8cPWRab6liVKpCWBpUqwZUrkowJIYQoNiQhs5UTvmVZ7NarDA6GqlWNyV+FEEKIYkQSMls5YR6yYjFcmZAA3boZ2zt3wrlzTg1HCCGEcAZJyGxlnfbCsfOQ3dfDlcePw2OPwZo1MHeus6MRQgghnEYSMltZp71wTIXsvh+uPHQI6tcHraFePXj5ZWdHJIQQQjiNJGS2sk574ZgK2X09XLlvH/j6GslYw4Zw7JizIxJCCCGcShIyW5lvAgpcSzjslPftcOWzzxq3gYFGpUwIIYQo5iQhs1XqDaN/TClnR1L0xcfDP/5hVMqEEEIIIQmZzcwpTlk26b6xYYOxFNKkScb8YtOnOzsiIYQQotCQhMxWDkrIVhxbwdANQzl66WiBn8thVq6ELl2MnrH9+50djRBCCFHoSEJmq9QUh8zSv+7kOo5eOopXJa/7o6H/q6/+7hl76imjUiaEEEKILGRtGls5oEKWeSHxRZ0XFei5HOKrr2DgQGO7Vy9YscK58QghhBCFlFTIbOWAhOy+m+qiSROjb2zQIEnGhBBCiFxIhcxW5psFmpBlro4V+akuZs+GoCBo2dJYLFwIIYQQuZKEzFapN6Bk+QI7/H1THXvzTQgLM6YHSU93djRCCCFEkSBDlrYypxTYLP33TXXs//7PSMYAJk50aihCCCFEUSIVMluZUwpsHcv7ojo2dix8+KGxHRYGb7zh1HCEEEKIokQSMlulphgz9ReQIl0dM5v/Tsbeew8mT3ZuPEIIIUQRIwmZrQqoQpZ5uLLIcnODWbOM7bFjnRuLEEIIUQRJQmarAuohK9LDlX37wvffw6VLkogJIYQQ90Ca+m2VeqPAZuovksOVPXrA8uVw4wbs3OnsaIQQQogiTSpkttAadJosLp6hc2fYuNHY/uYbaNfOufEIIYQQRZwkZDZQ2jKfliRk0KYN/Pijsf3998b6lEIIIYS4J5KQ2UQbN5KQGQ38AFu2SGVMCCGEyCfSQ2aLjApZAfWQFQnz5xu3mzdDaqokY0IIIUQ+koTMFsV9yNLXF0aNgtatjftuUlgVQggh8pMkZDYoqB6yjDnICjUvLzh0yFibcu5cZ0cjhBBC3Jek1GEL65Bl/s5DVujnIPP0hIQEIxk7fNhIzoQQQgiR7yQhs4W1QpZ/M/UX+gXFR436Oxk7eRJq13Z2REIIIcR9S4YsbfD3kGX+VcgKfXXsk0/AZILff5dkTAghhChgkpDZJP8rZFBIZ+h/+GF4+WVje/9+qFHDufEIIYQQxYAkZDZQBdRDVqiYzVCxIvzxB3z6qbOjEUIIIYoV6SGzhc6YGDZ/K2SFRkYylpxsTGlx4YKzIxJCCCGKFamQ2SKfe8gK1XQXZjOUL28kYw88ANeuQdmyzo5KCCGEKFakQmYDlc8z9Reqhv6kJEhJgZIljWRMJn0VQgghHE4qZLYogIlhnd7Qn5ICe/eChwf873+SjAkhhBBOJAmZLXQ6KFdwLXHPhyoUw5XJyVCuHDRpAsePw0MPSTImhBBCOJEkZDZQpOdbdczpw5VJSeDubvSOubvLHGNCCCFEISAJmS10er71j4EThysvXDCGKNPSoHJluHxZKmNCCCFEISAJmS10er7O0u80jz5qJGMPPSRTWwghhBCFiCRkNlA6/f6Yg+zDD6F+fTh71tmRCCGEECITSchsodPzZZZ+pzT0Hz0K1aoZjfwjRxr3hRBCCFGoSEJmC63zpULm8Ib+2Fjw9oY//4Tnn3fMOYUQQghhN0nIbGB8y/LeKmQZ1TGHNfTv3Qv+/kYy6esLkZEFf04hhBBC3BVJyGyRDz1kDq2O7dhhzDEG0KgRHDhQ8OcUQgghxF2ThMwGKp96yBxWHVuwwLht1gyiowv+fEIIIYS4JzIJlS10/k0MW6ASE6FGDfjiC6OBv2VLZ0ckhChEUlNTSUxMJCUlxdmhCHFfK1myJDVq1KBECdtX+JGEzBZFISH7+mvo3RseewxOnJBkTAiRTWJiIuXKlaN27doopZwdjhD3Ja01Fy9eJDExEU9PT5tfJ0OWNlD5PFN/vvviCyMZA6OBXwghcpCSkkLlypUlGROiACmlqFy5st2VaEnIbKF14a2QLVgAQ4YY2336wKpVzo1HCFGoSTImRMG7m98zSchsUkiHLPftgxdeMLYHDZKpLYQQhZ6rqysBAQH4+vrSrVs3kpKS8uW4ixcvZvTo0flyrMzatGmDl5cXAQEBBAQE8PXXX+f7OQASEhL46quvct1n9uzZlCxZkitXrlgfy+l9t2nThqgoYxLy5ORkXnjhBerUqUPjxo1p06YNe/bsuadYtda88sor1K1bF5PJxL59+3LcLzIyEpPJhI+PD5MmTbI+furUKdq3b4/JZKJNmzYkJiZmed3Vq1epUaNGlvd169YtRo4cSf369WnQoAHffPMNAD/99BONGjXCzc2twK6No0hClgeFRt1jhazAZuhv1MhYl3LECGPYUgghCrlSpUoRExPDwYMHqVSpEvPmzXN2SHlatmwZMTExxMTE0KtXL5teYzab7TqHLQlZeHg4wcHBrFy50ubjDh8+nEqVKhEfH090dDSLFi3iwj2uZbx+/Xri4+OJj49nwYIFjBo1Kts+Fy9eZOLEiWzZsoVDhw7xxx9/sGXLFgAmTJjA4MGDiY2N5fXXX+e1117L8tqpU6fSqlWrLI+98847VK1alWPHjhEXF0fr1q0BqFmzJosXL2bAgAH39J4KA0nI8uCCNjbuoYcs3+cge/NNCA01ts+e/XuaCyGEKEKaNWvGmTNnAPjll19o1qwZgYGBNG/enKOWZd4WL15Mz5496dy5M/Xq1eMf//iH9fWLFi2ifv36NGnShJ9//tn6eEJCAu3atcNkMtG+fXt+//13AEJDQxk1ahRNmzblscceY9u2bQwbNgxvb29CM/6faoNLly7xzDPPYDKZaNq0KbGxsQCEhYUxaNAgWrRowaBBgzh//jzPPvsswcHBBAcHW2P88ccfrRW3wMBArl27xuTJk9m+fTsBAQHMmjUr2zlPnDhBcnIy//rXvwgPD7cpzhMnTrBnzx7+9a9/4eJi/Ln39PTkySeftPm95mTVqlUMHjwYpRRNmzYlKSmJs7etkXzy5Enq1atHlSpVAHjiiSesVa24uDjatWsHQNu2bVmVqdUmOjqac+fO0bFjxyzH+/zzz62Jm4uLCx4eHgDUrl0bk8lkfX9FWdF/BwXMmpDd45Blvs1BNmkShIXBkiXG+pRCCFEEpaWlsWXLFrp37w5AgwYN2L59O7/++itvvfUW//d//2fdNyYmhsjISA4cOEBkZCSnT5/m7NmzvPHGG/z888/s2LGDuLg46/4vv/wyQ4YMITY2loEDB/LKK69Yn7t8+TK7du1i1qxZdO/enXHjxnHo0CEOHDhATExMjrEOHDjQmkBdvHiRN954g8DAQGJjY3n33XcZPHiwdd+4uDg2b95MeHg4Y8aMYdy4cezdu5dvvvmG4cOHAzBjxgzmzZtHTEwM27dvp1SpUkybNo2QkBBiYmIYN25cthgiIiLo168fISEhHD16lHPnzuX5GR86dIiAgABcXV3z3Ldv377W95j554scRl/OnDnDo48+ar1fo0YNa2KdoW7duhw9epSEhATMZjPfffcdp0+fBsDf399a5fv222+5du0aFy9eJD09nfHjxzNjxowsx8oY1p46dSqNGjWid+/eNr3/okamvciDC+nGRmHoIRszBubMMbbfegvKlnVuPEKIIuvN7w8R97+r+XrMho+U541uPrnuc+PGDQICAjhz5gze3t506NABgCtXrjBkyBDi4+NRSpGammp9Tfv27alQoYJxjoYNOXXqFBcuXKBNmzbWCkzfvn05duwYALt27bL+wR80aFCWqlq3bt1QSuHn50e1atXw8/MDwMfHh4SEBAICArLFvGzZMoKCgqz3d+zYYa32tGvXjosXL3L1qvFZdu/enVKljInEN2/enCVRvHr1KsnJybRo0YJXX32VgQMH0rNnT2rUqJHnZxseHs63336Li4sLzz77LCtWrGD06NF3bB63t6k8Mp97kCtWrMgnn3xC3759cXFxoXnz5pw4cQIwEtLRo0ezePFiWrVqRfXq1XF1deXjjz+ma9eu2T4Ps9lMYmIizZs3Z+bMmcycOZMJEyawdOnSfI3Z2SQhy4OLyhiyvLuZ+jOvYXlPRo6ETz81tt9/HyZOvLfjCSGEE2T0kF2/fp1OnToxb948XnnlFaZOnUrbtm359ttvSUhIoE2bNtbXPPjg30vXubq62t2flVnGsVxcXLIc18XF5Z6Om6FMmTLW7fT0dHbv3k3Jkln/QT958mSefPJJ1q1bR4sWLdi4cWOuxzxw4ADx8fHW5PXWrVt4enoyevRoKleuzOXLl7Psf+nSJTw8PHB3d2f//v2kpaXlWSXr27evdZg4s1dffTVLBRCgevXq1moXGPPbVa9ePdtru3XrRrdu3QBYsGCBNYZHHnnEmjAnJyfzzTff4O7uzq5du9i+fTsff/wxycnJ3Lp1i7Jly/Lee+9RunRpevbsCUDv3r1ZuHBhru+nKJKELA/KWiG7u7Us861/bNEi4/ajj+Cll+7tWEKIYi+vSlZBK126NHPmzOGZZ57hxRdf5MqVK9Y/6osXL87z9Y8//jhjxozh4sWLlC9fnhUrVuDv7w9A8+bNiYiIYNCgQSxbtoyQkJB8jT0kJIRly5YxdepUtm3bhoeHB+XLl8+2X8eOHZk7dy4TLf+AjomJISAggBMnTuDn54efnx979+7lyJEjPProo1y7di3H84WHhxMWFpal+d3T05NTp04RHBzM6NGj+eOPP3jooYeIiori5s2bPProo7i4uBAUFMQbb7zB22+/jVKKhIQEDh06lK2PzJ4KWffu3fnoo4/o168fe/bsoUKFCjz88MPZ9vvzzz+pWrUqly9f5uOPP2b58uUAXLhwgUqVKuHi4sJ7773HsGHDAKMSmWHx4sVERUUxbdo0wEjutm3bRrvB3aLBAAAgAElEQVR27diyZQsNGza0Od6iQnrI8vB3D9ndr2V5T/1jGf9iO30awsMlGRNC3DcCAwMxmUyEh4fzj3/8g9dee43AwECbKlUPP/wwYWFhNGvWjBYtWuDt7W19bu7cuSxatAiTycTSpUv58MMP8zXusLAwoqOjMZlMTJ48mSVLluS435w5c4iKisJkMtGwYUPmz58PGNNX+Pr6YjKZKFGiBF26dMFkMuHq6oq/v3+2pv6IiAh69OiR5bEePXoQERFBtWrV+PDDD+natSsBAQGMHTuW8PBwa5P7Z599xrlz56hbty6+vr6EhoZStWrVe3r/Xbt25bHHHqNu3bqMGDGCjz/+2Ppc5iHfMWPG0LBhQ1q0aMHkyZOpX78+ANu2bcPLy4v69etz7tw5pkyZkuc5p0+fTlhYmPWafvDBBwDs3buXGjVqsGLFCl544QV8fJz7D417obTWzo7hrgUFBemMuVYKwsZDf5D+Qn8aqlPUWvoF1LZ/OaKhG4YCsKjzIvsDePJJWLcOvv8ennrK/tcLIUQmhw8fzpK4CCEKTk6/b0qpaK11jj1MMmSZB6c19XfoAJs3G9u3bjn23EIIIYRwKEnI8pBf017YpVUr2L7d2JbqmBBCCHHfk4QsDw6vkD3xxN/J2JYtYJk8TwghhBD3L2nqz0N+zNRvl+HDwdUVdu6UZEwIIYQoJiQhy4Ny1JBlz56QlAT9+hnfrGzWrGDPJ4QQQohCQxKyPDhkyLJePfj2W+NWCCGEEMWOJGR5KPCm/lq14PhxUAp27SqYcwghRCFRNtOSb+vWraN+/fqcOnWKsLAwSpcuzZ9//pnjvkopxo8fb70/Y8YMwsLC7nieZ555hqZNm2Z5LDQ0lK+//vqO8Rw7doyuXbtSr149GjVqRJ8+fe55zcRLly7RoUMH6tWrR4cOHbLNqp9h0qRJ+Pr64uvrm2WSVq01U6ZMoX79+nh7ezPHsnzesmXLMJlM+Pn50bx5c/bv3w9ASkoKTZo0wd/fHx8fH9544w3rsZ5//nn8/f0xmUz06tWLZFkPuVCRhCwPLqSjlQLXAvj+Q/Xq8PvvRjJ28iTUrZv/5xBCiEJoy5YtvPLKK6xfv55atWoB4OHhYZ3w83YPPvggK1eu5MKFC3keOykpiejoaK5cucLJkydtiiclJYUnn3ySUaNGER8fz759+3jxxRc5f/687W8qB9OmTaN9+/bEx8fTvn1768zzma1du5Z9+/YRExPDnj17mDFjhnVtzMWLF3P69GmOHDnC4cOH6devH2DM1P/jjz9y4MABpk6dysiRIwHjc9q6dSv79+8nJiaGDRs2sHv3bgBmzZrF/v37iY2NpWbNmnz00Uf39N5E/pKELA8uaFB39zFlrGOZo4gI+N//wMXFuK1d++6DFEKIIuSnn35ixIgRrFmzhjp16lgfHzZsGJGRkVy6dCnba9zc3Bg5cmS2WexzsnLlSrp160a/fv2IiIiwKaavvvqKZs2aWddeBGjTpg2+vr42vf5OVq1axZAhQwAYMmQI3333XbZ94uLiaNWqFW5ubpQpUwaTycSGDRsA+OSTT3j99detM+9nzLLfvHlzKlasCEDTpk1JTEwEjEpiRtUvNTWV1NRU60LjGcs7aa25ceOG3QuQi4IlCVkejCHLu/uYcl3Hsl8/mDABzp2Dhx66hwiFEKLouHnzJs888wzfffcdDRo0yPJc2bJlGTZs2B2XOnrppZdYtmwZV65cyfUc4eHh9O/fn/79+xMeHm5TXAcPHqRx48Z57nft2jUCAgJy/ImLi8u2/7lz56zrPD700EM5DoH6+/uzYcMGrl+/zoULF/jhhx+si3efOHGCyMhIgoKC6NKlC/Hx8dlev3DhQrp06WK9n5aWRkBAAFWrVqVDhw48/vjj1ueGDh3KQw89xJEjR3j55Zfz/mCEw8g8ZHlwUenou6iQZVTHsqxjaTZD5cpQtiycOQP//nc+RyuEEDZaPxn+OJC/x3zID7pkH5LLrESJEjRv3pyFCxfmmHi98sorBAQEMGHChGzPlS9fnsGDBzNnzhxKlcp5feFz584RHx9Py5YtUUpRokQJDh48iK+vb44VIXurROXKlSMmJsau12Q+V07n69ixI3v37qV58+ZUqVKFZs2a4erqChgJbMmSJYmKimLlypUMGzaM7RlzVQI//PADCxcuZMeOHdbHXF1diYmJISkpiR49eljfP8CiRYtIS0vj5ZdfJjIykqFDh97VexH5TypkeTCGLO0v62arjpnN4O4OV6/C+fN/LxouhBDFiIuLC8uXL+eXX37h3Xffzfa8u7s7AwYMYN68eTm+fuzYsSxcuJC//vorx+eXL1/O5cuX8fT0pHbt2iQkJFirZJUrV87SVH/p0iU8PDwA8PHxITo6Os/47a2QVatWjbNnzwJw9uzZOy7sPWXKFGJiYti0aRNaa+tC3DVq1KBnz56AsaB4bGys9TWxsbEMHz6cVatWUbly5WzHdHd3p23bttbhzwyurq7069ePb775Js/3KxxHKmR5cEHbXSHLVh0zm6F8ebhxAx54AK5cATf56IUQTpRHJasglS5dmrVr1xISEkK1atV4/vnnszz/6quvEhwcjDmHf7hWqlSJPn36sHDhQoYNG5bt+fDwcDZs2EAzy1yOv/32G0888QTvvPMObdq0Yfbs2QwZMoQHHniAxYsX07ZtWwAGDBjAe++9x9q1a3nyyScBo9etUqVKWfrI7K2Qde/enSVLljB58mSWLFnC008/nW2ftLQ0kpKSqFy5MrGxscTGxtKxY0fA+LboDz/8YG3iz0jUfv/9d3r27MnSpUutjwGcP3+eEiVK4O7uzo0bN9i0aROTJk1Ca82JEyeoW7cuWmtWr16dbchYOJdkBXlQpNvd1J+tOlauHKSkQMmScO2aJGNCiGKvUqVKbNiwgVatWlGlSpUsz3l4eNCjR487NvCPHz8+x28IJiQkcOrUqSzTXXh6elKhQgX27NnDU089RXR0NI0bN8bV1ZU6deowf/58AEqVKsWaNWsYO3YsY8eOpUSJEphMpjv2s9lq8uTJ1gSyVq1aLF++HICoqCjmz5/PZ599RmpqKiEhIYAxLPvll1/iZvk7MXnyZAYOHMisWbMoW7Ysn332GQBvvfUWFy9e5MUXXwSMLz1ERUVx9uxZhgwZQlpaGunp6fTp04ennnqK9PR0hgwZwtWrV9Fa4+/vzyeffHJP703kL6W1dnYMdy0oKEhHRd3hW4z5YOOhP3jk/z1BmQfdeGyz7f8iGrrBGJNf1HmR8UD16kZVLClJkjEhhNMcPnwYb29vZ4chRLGQ0++bUipaax2U0/7SQ5YHYx4y+z8ml5upMGWKcefMGUhOlmRMCCGEEDmSDCEP9k57seLYCmJP/8KGicfgitlYDik0tMDiE0IIIUTRJxWyPNg7Mezmg9+xdtIxqlwxg4eHJGNCCCGEyJNUyPJg15BlYiLTB0fi/leaMdmr5avOQgghhBC5kQpZLsr893vM511tr5AFBeH+Vxp7/CtJMiaEEEIIm0lClosyO7YA8GDAo7a94MgRlj9Th/nTuhdgVEIIIYS430hClhsNparcpGSz+nfeJzbWmOw1IoIVf27i7WdyXs5DCCFE4bZx48a7XhbpbiUkJPDVV1859JyicJKELFcaBaS7lcz56b17wd8fUlNh7tzcFxMXQgiBq6srAQEB+Pr60q1bN5KSkpwSR9euXbOce+vWrWzcuBF/f3+7jhMWFsaMGTPuKoa0tDReeumlOy5qfnuyFhUVxSuvvHJX57qd2WymSpUqTJ48OcvjtWvX5sKFC9b727Zt46mnnrLeX79+PUFBQTRs2JDAwEDGjx9/z7FER0fj5+dH3bp1eeWVV8hpftQrV67QrVs3/P398fHxYdEiY57PmJgYmjVrho+PDyaTicjISOtrPvroI+rWrYtSKst7unz5Mj169MBkMtGkSRMOHjyY5VxpaWkEBgZmed+OIAlZLpROB0C7Ppj9yZ9+giZNjO2gIPj5Z2Mz82LiQgghsihVqhQxMTEcPHiQSpUq3XHNSnvktMRSXtatW4e7u7v1frt27Zg5c6bdi43fi+PHjzNp0iS8vLxyfP72hCwoKIg5c+bky7k3bdpE/fr1WbFiRY4JUE4OHjzI6NGj+fLLL4mLiyMqKoq6devecyyjRo3i008/JT4+nvj4+GxrbwLMmzePhg0bsn//frZt28b48eO5desWpUuX5osvvuDQoUNs2LCBsWPHWhPtFi1asHnzZmrVqpXlWO+++y4BAQHExsbyxRdfMGbMmCzPf/jhh06ZQFkSslwojIQsW4Vs0yZo3drYbt7cqJQJIYSwS7NmzThz5gwAJ06coHPnzjRu3JiQkBCOHDlifbxp06b4+fnxz3/+k7JlywJG5SYkJITu3bvTsGFD0tLSmDhxIsHBwZhMJv7zn/8AxoLerVq1slbltm/fDmStBM2cORNfX198fX2ZPXs2YCRD3t7ejBgxAh8fHzp27MiNGzdyfT8xMTE0bdoUk8lEjx49rAuZ7927F5PJREBAABMnTrSujXn27Fnef/99AH788UfrIuWBgYFcu3aNyZMns337dgICApg1a1aWalVycjJDhw7Fz88Pk8lkXSh81KhRBAUF4ePjwxtvvHHHWMPDwxkzZgw1a9Zk165dNl2v999/nylTpljXwHR1dWXUqFE2vfZOzp49y9WrV2natClKKQYPHsx3332XbT+lFNeuXUNrTXJyMpUqVcLNzY369etTr149AB555BGqVq3K+fPnAQgMDKR27drZjhUXF0e7du0AaNCgAQkJCZw7dw6AxMRE1q5dy/Dhw+/pfd0NmfYiN9YK2W0J2c2bxm3btrB1q4ODEkKIezf9l+kcuXQkX4/ZoFIDJjWZZNO+aWlpbNmyxbqw+MiRI5k/fz716tVjz549vPjii2zdupUxY8YwZswY+vfvb113MsO+ffs4ePAgnp6eLFiwgAoVKrB3715u3rxJixYt6NixIytXrqRTp05MmTKFtLQ0rl+/nuUY0dHRLFq0iD179qC15vHHH6d169ZUrFiR+Ph4wsPD+fTTT+nTpw/ffPMNzz333B3f0+DBg5k7dy6tW7fm9ddf580332T27NkMHTqUTz/9lGbNmmUbIswwY8YM5s2bR4sWLUhOTqZkyZJMmzaNGTNmsGbNGsBIQjO8/fbbVKhQgQMHDgBYk7933nmHSpUqkZaWRvv27YmNjcVkMmU5V0pKCps3b+Y///kPSUlJhIeH07x58zyv2cGDB20aovzhhx8YN25ctsdLly7Nzp07szx25swZatSoYb1fo0YNa5Ke2ejRo+nevTuPPPII165dIzIyEheXrDWlX375hVu3blGnTp1c4/P392flypWEhITwyy+/cOrUKRITE6lWrRpjx47l/fff59q1a3m+z/wmCVkuMoYs0zOGLNesMYYnn3oKbtxgxe/fs86ybiXA0UtH8aqUc+lZCCEE3Lhxg4CAAM6cOYO3tzcdOnQgOTmZnTt30rv33+0eNy3/8N21a5e1YjJgwAAmTJhg3adJkyZ4enoC8N///pfY2Fi+/vprwOg5io+PJzg4mGHDhpGamsozzzxDQEBAlnh27NhBjx49KFOmDAA9e/Zk+/btdO/eHU9PT+v+jRs3JiEh4Y7v68qVKyQlJdHaMnoyZMgQevfuTVJSEteuXaNZs2bW95CRYGXWokULXn31VQYOHEjPnj2zJCk52bx5MxEREdb7FStWBGD58uUsWLAAs9nM2bNniYuLy5aQrVmzhrZt21KqVCmeffZZ3n77bWbPno2rq2uOQ7b2DuO2bds2378csXHjRgICAti6dSsnTpygQ4cOhISEUL58ecCotA0aNIglS5ZkS9RuN3nyZMaMGUNAQAB+fn4EBgbi6urKmjVrqFq1Ko0bN86S/DqKwxIypVRn4EPAFfhMaz3ttucfBL4AGgMXgb5a6wRHxZejjAqZW0lYvBiGDjXWo0xNhZIlWXdyXZYkzKuSlzT0CyGKBFsrWfkto4fs+vXrdOrUiXnz5hEaGoq7u7vdf8QzkigArTVz586lU6dO2fb76aefWLt2LaGhobz66qsMHjzYpuM/+ODf/cOurq55Dlnei8mTJ/Pkk0+ybt06WrRowcaNG+0+xm+//caMGTPYu3cvFStWJDQ0lJSUlGz7hYeHs2PHDutw3sWLF9m6dSsdOnSgcuXKXL58GQ8PDwAuXbpk3fbx8SE6OjrPLz7YUyGrXr06iYmJ1vuJiYlUr14922sXLVrE5MmTUUpRt25dPD09OXLkCE2aNOHq1as8+eSTvPPOOzRt2jT3DwkoX7689UsBWms8PT157LHHiIyMZPXq1axbt46UlBSuXr3Kc889x5dffpnnMfODQ3rIlFKuwDygC9AQ6K+Uanjbbs8Dl7XWdYFZwHRHxJYbhdHoWHrNViMZA+jVK8s+XpW8WNR5kfVHGvqFECJvpUuXZs6cOXzwwQeULl0aT09PVqxYARh/JPfv3w9A06ZNrf1RmStCt+vUqROffPIJqampABw7doy//vqLU6dOUa1aNUaMGMHw4cPZt29flteFhITw3Xffcf36df766y++/fZbQkJC7H4/FSpUoGLFitYetaVLl9K6dWvc3d0pV64ce/bsyfU9nDhxAj8/PyZNmkRwcDBHjhyhXLlydxw669ChQ5YvRFy+fJmrV69SpkwZKlSowLlz51i/fn221129epXt27fz+++/k5CQQEJCAvPmzSM8PByANm3asHTpUsAYVv7yyy9p27YtABMnTuTdd9/l2LFjAKSnp2cbRoa/K2S3/9yejAE8/PDDlC9fnt27d6O15osvvuDpp5/Otl/NmjXZssWYG/TcuXMcPXqUxx57jFu3btGjRw8GDx5Mr9v+Pt9JUlISt27dAuCzzz6jVatWlC9fnvfee4/ExEQSEhKIiIigXbt2DkvGwHFN/U2A41rrk1rrW0AEcPsn/jSwxLL9NdBeOfLrLjlQOh23K7eo/J7lP7ihQyE8nBXHVjB0w1COXjrqzPCEEKJICwwMxGQyER4ezrJly1i4cKF1WoNVq1YBMHv2bGbOnInJZOL48eNUqFAhx2MNHz6chg0b0qhRI3x9fXnhhRcwm81s27YNf39/AgMDiYyMzPaNukaNGhEaGkqTJk14/PHHGT58OIGBgXf1fpYsWcLEiRMxmUzExMTw+uuvA7Bw4UJGjBhBQEAAf/31V47vYfbs2fj6+mIymShRogRdunTBZDLh6uqKv78/s2bNyrL/P//5Ty5fvoyvry/+/v788MMP1vfZoEEDBgwYQIsWLbKd59tvv6Vdu3ZZqn9PP/0033//PTdv3mTq1KkcP37ceqy6deta++ZMJhOzZ8+mf//+eHt74+vry8mTJ+/qs8rs448/Zvjw4dStW5c6derQpUsXAObPn29N+KZOncrOnTvx8/Ojffv2TJ8+HQ8PD5YvX85PP/3E4sWLrV+KyKi0zpkzhxo1apCYmIjJZLI26h8+fBhfX1+8vLxYv349H3744T2/h/ygbP266z2dRKleQGet9XDL/UHA41rr0Zn2OWjZJ9Fy/4Rlnws5HRMgKChIR0VFFVjcWzsF0+KH/TyQmsqCkSZ292gEQNQ545xB1YLo+lhXqYoJIYqEw4cPO+Xr/Pfi+vXrlCpVCqUUERERhIeHW5O1oiI5Odn67dBp06Zx9uzZQpMEiIKT0++bUipaax2U0/5FrqlfKTUSGAlGCbMgXapent2BVYlpVJGYpxtZH5dETAghHCM6OprRo0ejtcbd3Z3PP//c2SHZbe3atbz33nuYzWZq1arF4sWLnR2SKIQcVSFrBoRprTtZ7r8GoLV+L9M+Gy377FJKuQF/AFV0LgEWdIVMCCHuJ0WxQiZEUWVvhcxRPWR7gXpKKU+l1ANAP2D1bfusBoZYtnsBW3NLxoQQQggh7hcOGbLUWpuVUqOBjRjTXnyutT6klHoLiNJarwYWAkuVUseBSxhJmxBCCCHEfc9hPWRa63XAutseez3TdgogTVlCCCGEKHZkLUshhBBCCCeThEwIIYTDuLq6Whf67tatG0lJSXa9PmP6iNsppbKsszhjxgzCwsLuJdQCERoaal3eKbPdu3fz+OOPExAQgLe3tzX21atXM23atGz75+e58xIQEEC/flm7iNq0aUPmL9UlJCRYF00HY13JVq1a4eXlRWBgIMOHD8+2jqi9fvvtNx5//HHq1q1L3759rZO7Znbr1i3rouv+/v5ZlkC6desWI0eOpH79+jRo0MA64fDvv/9O27ZtrfPirVu3LtdjXbt2zTrnWUBAAB4eHowdO/ae3htIQiaEEMKBMpZOOnjwIJUqVcoy2/y9ePDBB1m5ciUXLtxx6sq7orUmPT09X4+ZkyFDhrBgwQLrZ9OnTx8AunfvfscFyR3h8OHDpKWlsX37dv766y+bXnPu3Dl69+7N9OnTOXr0KL/++iudO3e+5wW7J02axLhx4zh+/DgVK1Zk4cKF2fb59NNPAThw4ACbNm1i/Pjx1uv3zjvvULVqVY4dO0ZcXJx13dF//etf9OnTh19//ZWIiAhefPHFXI9Vrly5LCsQ1KpVi549e97TewNJyIQQQjhJs2bNOHPmDGAsHdS5c2caN25MSEgIR44cAYyqSLNmzfDz8+Of//znHY/l5ubGyJEjs81oD3D+/HmeffZZgoODCQ4O5ueffwYgLCyMGTNmWPfz9fW1Lifk5eXF4MGD8fX15fTp00ycOBFfX1/8/PyIjIwEYNu2bbRp04ZevXrRoEEDBg4cSMbkAG+99RbBwcH4+voycuRI8po04M8//+Thhx8GjCpiw4bG6oKLFy9m9GhjDvXQ0FBGjRpF06ZNeeyxx9i2bRvDhg3D29ub0NBQ67HKli3LuHHj8PHxoX379pw/fz7b+aKjo2ndujWNGzemU6dOnD17Nse4wsPDGTRoEB07drR5Qt558+YxZMgQ64LqAL169aJatWo2vT4nWmu2bt1qXR5pyJAh1kXnM4uLi6Ndu3YAVK1aFXd3d2sl7/PPP+e1114DwMXFxbpGp1KKq1evAsYi8Y888kiex8pw7Ngx/vzzz7tabut2RW5iWCGEEPfuj3ff5ebhI/l6zAe9G/DQ//2fTfumpaWxZcsWnn/+eQBGjhzJ/PnzqVevHnv27OHFF19k69atjBkzhlGjRjF48OA8q2kvvfQSJpOJf/zjH1keHzNmDOPGjaNly5b8/vvvdOrUicOHD+d6rPj4eJYsWWJdSzMmJob9+/dz4cIFgoODadWqFQC//vorhw4d4pFHHqFFixb8/PPPtGzZktGjR1uXTho0aBBr1qyhW7dudzzfuHHj8PLyok2bNnTu3JkhQ4ZQsmTJbPtdvnyZXbt2sXr1arp3787PP//MZ599RnBwMDExMdblmYKCgpg1axZvvfUWb775Jh999JH1GKmpqbz88susWrWKKlWqEBkZyZQpU3KcdDcyMpJNmzZx5MgR5s6dy4ABA3L93AAOHjzIkCFD8tzv6NGj9O3bN8fntm3bhru7u/X+xYsXcXd3x83NSFtq1KhhTeYz8/f3Z/Xq1fTv35/Tp08THR3N6dOnqV+/PmAswbRt2zbq1KnDRx99RLVq1QgLC6Njx47MnTuXv/76i82bN+d6rCZNmljPFxERQd++fcmPlR4lIRNCCOEwN27cICAggDNnzuDt7U2HDh1ITk5m586d9O799xftb968CcDPP/9s7fUZNGgQkyZNuuOxy5cvz+DBg5kzZw6lSpWyPr5582bi4uKs969evUpycnKucdaqVYumTZsCsGPHDvr374+rqyvVqlWjdevW7N27l/Lly9OkSRNq1KgBGL1WCQkJtGzZkh9++IH333+f69evc+nSJXx8fHJNyF5//XUGDhzIf//7X7766ivCw8Oz9D9l6NatG0op/Pz8qFatGn5+fgD4+PiQkJBAQEAALi4u1kTnueeeyzacdvToUQ4ePEiHDh0AIznOqM5lFhUVhYeHBzVr1qR69eoMGzaMS5cuUalSpRwTEHuTEi8vL+u6k/ll2LBhHD58mKCgIGrVqkXz5s1xdXXFbDaTmJhI8+bNmTlzJjNnzmTChAksXbqU8PBwQkNDGT9+PLt27WLQoEEcPHjwjsfKLCIiwroY+72ShEwIIYohWytZ+S2jh+z69et06tSJefPmERoairu7+x3/ONvzh37s2LE0atSIoUOHWh9LT09n9+7d2SpObm5uWfrDUlJSrNtlypSx6XyZF+nO+MOfkpLCiy++SFRUFI8++ihhYWFZjn0nderUYdSoUYwYMYIqVapw8eLFO57PxcUly7ldXFwwm805Hvf2z09rjY+PD7t27co1nvDwcI4cOULt2rUBI5H95ptvGDFiBJUrV+by5cvWfS9dumQdAvTx8SE6Opqnn3461+PbUyGrXLkySUlJmM1m3NzcSExMpHr16tle5+bmlmXYunnz5tSvX5/KlStTunRpa3Lau3dvaw/awoUL2bBhA2AMo6ekpHDhwgWqVq2a47Ey7N+/H7PZTOPGjXN9n7aSHjIhhBAOV7p0aebMmcMHH3xA6dKl8fT0ZMWKFYCRMOzfvx+AFi1aEBERAcCyZcvyPG6lSpXo06dPlobvjOGoDBmJX+3atdm3bx8A+/bt47fffsvxmCEhIURGRpKWlsb58+f56aefsgxb3S4j+fLw8CA5OdmmbzauXbvW2mcWHx+Pq6trloTEHunp6dZzfvXVV7Rs2TLL815eXpw/f96akKWmpnLo0KFsx1i+fDkHDhyw9tWtWrWK8PBwwPiW5ZdffmmNecmSJbRt2xaA0aNHs2TJEvbs2WM93sqVKzl37ly2ODI3x2f+uf29K6Vo27at9X0tWbIkx4Tv+vXr1i8fbNq0CTc3Nxo2bIhSilSGM20AAAw4SURBVG7dulmrjlu2bLH26dWsWZMtW7YAxpcYUlJSqFKlyh2PlSE8PJz+/fvf4SrYTxIyIYQQTpExzUB4eDjLli1j4cKF+Pv74+PjY20g//DDD5k3bx5+fn459gzlZPz48Vm+bTlnzhyioqIwmUw0bNiQ+fPnA/Dss89ahxM/+uijLNWPzHr06IHJZMLf35927drx/vvv89BDD93x/O7u7owYMQJfX186depEcHBwnjEvXboULy8vAgICGDRoEMuWLcs2PGarMmXK8Msvv+Dr68vWrVutvWwZHnjgAb7++msmTZqEv78/AQEB7Ny5M8s+27dvp3r16tYGd4BWrVoRFxfH2bNnGTlyJOXKlcPf3x9/f3+Sk5OZMGECANWqVSMiIoIJEybg5eWFt7c3GzdupFy5cnf1fjJMnz6dmTNnUrduXS5evGjtP1y9erX1Pf755580atQIb29vpk+fnmU4cfr06YSFhWEymVi6dCkffPABAB988AGffvop/v7+9O/fn8WLF6OUyvVYAMuXL8/XhMwhi4sXFFlcXAghbCeLixcPZcuWzbNHThS8wrq4uBBCCCGEuANJyIQQQoj7iFTHiiZJyIQQQgghnEwSMiGEKEaKct+wEEXF3fyeSUImhBDFRMmSJbl48aIkZUIUIK01Fy9ezHGlhdzIxLBCCFFM1KhRg8TExBzXNhRC5J+SJUtaV3CwlSRkQghRTJQoUQJPT09nhyGEyIEMWQohhBBCOJkkZEIIIYQQTiYJmRBCCCGEkxXppZOUUueBUwV8Gg/gQp57CUeT61L4yDUpnOS6FD5yTQonR1yXWlrrKjk9UaQTMkdQSkXdad0p4TxyXQofuSaFk1yXwkeuSeHk7OsiQ5ZCCCGEEE4mCZkQQgghhJNJQpa3Bc4OQORIrkvhI9ekcJLrUvjINSmcnHpdpIdMCCGEEMLJpEImhBBCCOFkkpBZKKU6K6WOKqWOK6Um5/D8g0qpSMvze5RStR0fZfFjw3V5VSkVp5SKVUptUUrVckacxUle1yTTfs8qpbRSSr5NVsBsuSZKqT6W35VDSqmvHB1jcWTD/79qKqV+UEr9avl/WFdnxFmcKKU+V0r9qZQ6eIfnlVJqjuWaxSqlGjkqNknIAKWUKzAP6AI0BPorpRrettvzwGWtdV1gFjDdsVEWPzZel1+BIK21CfgaeN+xURYvNl4TlFLlgDHAHsdGWPzYck2UUvWA14AWWmsfYKzDAy1mbPxd+SewXGsdCPQDPnZslMXSYqBzLs93AepZfkYCnzggJkASsgxNgONa65Na61tABPD0bfs8DSyxbH8NtFdKKQfGWBzleV201j9ora9b7u4Gajg4xuLGlt8VgLcx/tGS4sjgiilbrskIYJ7W+jKA1vpPB8dYHNlyXTRQ3rJdAfifA+MrlrTWPwGXctnlaeALbdgNuCulHnZEbJKQGaoDpzPdT7Q8luM+WmszcAWo7JDoii9brktmzwPrCzQikec1sZT4H9Var3VkYMWYLb8n9YH6SqmflVK7lVK5VQhE/rDluoQBzymlEoF1wMuOCU3kwt6/O/nGzREnEaKgKaWeA4KA1s6OpThTSrkAM4FQJ4cisnLDGIJpg1FF/kkp5ae1TnJqVKI/sFhr/YFSqhmwVCnlq7VOd3ZgwvGkQmY4Azya6X4Ny2M57qOUcsMoL190SHTFly3XBaXUE8AUoLvW+qaDYiuu8rom5QBfYJtSKgFoCqyWxv4CZcvvSSKwWmudqrX+DTiGkaCJgmPLdXkeWA6gtd4FlMRYT1E4j01/dwqCJGSGvUA9pZSnUuoBjObK1bftsxoYYtnuBWzVMolbQcvzuiilAoH/YCRj0hdT8HK9JlrrK1prD611ba11bYy+vu5a6yjnhFss2PL/r+8wqmMopTwwhjBPOjLIYsiW6/I70B5AKeWNkZCdd2iU4nargcGWb1s2Ba5orc864sQyZInRE6aUGg1sBFyBz7XWh5RSbwFRWuvVwEKMcvJxjIbAfs6LuHiw8br8GygLrLB8x+J3rXV3pwV9n7PxmggHsvGabAQ6KqXigDRgotZaKvwFyMbrMh74VCk1DqPBP1T+oV+wlFLhGP848bD07r0BlADQWs/H6OXrChwHrgNDHRabXHshhBBCCOeSIUshhBBCCCeThEwIIYQQwskkIRNCCCGEcDJJyIQQQgghnEwSMiGEEEIIJ5OETAhR4JRS25RS/3R2HPZSSiVbZlC/p32EECIvkpAJIWxmSaxuWpKQjJ/PnB1XQdFal7XMoI5Sqo1SypzbPoWFUkorpVo6Ow4hhO1kYlghhL3e1lr/y9lBFEdKqRJa61RnxyGEyH9SIRNC3DOlVD+l1H6l1FWl1Fml1H+UUmXusO+DSqkFSqk/LfvHK6V6Z3o+RCm1Qyl1SSl1Qik1XlmWYcjhWLUt1aDhSqljSqkrSqlVSqmqmfaprJT6Qin1h+VniVKqUqbnX1FK/aaUuqaUOqOUejfTc1op1VIp9QiwHnDNVBkccts+bpb3/sxtMS5WSi3KdH+EUuqgJdZflVIdc/lcQ5VSx5VSEy2zisdYHn9XKXXSEscJpdTYTK/Zb9n8b+YKplKqtFJqhuW9XlJKbVBK1b3TuYUQjiUJmRAiP1wBBgDuQIjl5049Y0OAYMBba10eaAccAlBKNcRYuuTfQBXgSWA0MCiP8w8GWmEsCpwOfJnpuWVARcDb8uMBLLWcrz4wDXhKa10O8CH7eoNorf8HdAHSLEOUZbXWS27bx2w5bmjGY0qpshhr335uuT8CmAQMtMQ0BViZR2JUG3gEYzHwYMtjcUBLjMXcRwDvKaU6WeLwt+zT0RLncMv9T4EGGAu+PwTsAdYopUrkcm4hhINIQiaEsNcUpVRSpp+mWuv1WutDWut0rfVx4GMsiybn4BbG+qMNlVJuWuvTWus4y3MvAiu01qu01mla6yPARxgJV27e1Fr/obW+CkwEOiilHrFUtjoBr2qtL2utLwOvAl2VUg8DZkABPkqpslrrJK317nv4bBZZjp1RoesD/E9rvd1yfwzwltZ6v+WzWgf8QO5r46YCk7XWN7TW1wG01l9qrf+nDVuBtdz5885YUHwA8KLW+pzW+hbwJvAw8Pjdv10hRH6RhEwIYa93tNbumX52K6U6KKW2K6XOK6WuAtMxKlw5+RL4DJgFXFRKZa4QeQL9Myd8GIv/PpxHTAk5bNfAqJgB/Jbp+ROW20e11icxqlUjgP9ZhkrvOISYF631YWAf8JzloaEYSVoGT2Debe+vLVA9l8Oe1VrfzPyAZZj1gFLqsuUY3bjz551xXoDYTOe9hLGo8qN3fpkQwlEkIRNC3BOl1APAd0AEUNMyDDkJo/KUjdbarLWerrUOAmoB17EM6QGngM9vS/jKa6198gijdg7bicDpHJ5/zHJ72hLPSq11B4yhzOXAKqVU6RzOkZ5HDBkWAaGWJLMp8EWm504Bw257f2W11qNyOV6W8yqlWmAkvC8AHlprd+B7sn7e+rZjnLLc1rvt3KW11uE2vi8hRAGShEwIca8eAB4ELmutb1j6wEbfaWelVDulVGNL79IN4C8gzfL0x0A/pVQ3pVQJS6N8Q6VU6zximKqUqqaUKo+RrGy2DOn9D/gv8IFSyl0pVRH4AFivtT6rlPJSSnW2JGCpGL1wmpyTrz8wmvo9c3guswigLjAH2KS1PpPpuVlAmFIqQP3/9u3XJYIgCuD4d4pgsRpEsAj+L1pEbFYFi9hFg/G6ighisl62CbarGtQognAKd3IXbM8wI+p5/kDDIHw/bXeW3dmBZR/vvclGy4aAmW/u+dYYeb3ugUgpzZL72wbnOv1yEBFt4BjYTSlNAJT1mC99bpIqMyCT9CcR0QdWgUZKqQ/skH/+nxknN793gDtylmyl3OsCmAPWy1gbOOLrchzkMugZOes1wvtNAEtAD7gCLoEurz1pI8BWeVYXWAMWIuJpyHteA3tAq5T9hm40iIhHoEkOkg4Hxg6ABjmL1gFugE1y6fCnTshZtxbwQN400By4ZgPYLiXN/XJumbwGpymlHnAOLPIxmyapghThtyjpf0opTZH7wyYj4rbubCTp98yQSZIkVWZAJkmSVJklS0mSpMrMkEmSJFVmQCZJklSZAZkkSVJlBmSSJEmVGZBJkiRVZkAmSZJU2TMsRpZ6iZQpzQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "t2gcdPbvniLx"
      },
      "source": [
        "El mejor modelo es el Random Forest. Se procede a guardar el modelo y se simula la carga del mismo y su uso para predecir la probablidad de que un empleado abanode la organización sobre datos \"reales\" o no vistos (esto es, el set de pruebas que se separó antes de entrenar a los modelos)\r\n",
        "\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z0E9G7v4nvJ9"
      },
      "source": [
        "import joblib"
      ],
      "execution_count": 46,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h4CqLpgUq6lA"
      },
      "source": [
        "# Se crea una copia de la clase de nuestro pipeline de predicción\r\n",
        "model_to_save=rf_gs\r\n",
        "# Se agrega el atributo feature_names_\r\n",
        "# Este es, un array con los nombres de los features\r\n",
        "model_to_save.feature_names_=X.columns.values\r\n",
        "# Se agrega el atributo feature_df_.\r\n",
        "# Este es, un DataFrame con los Feature y su importancia en el modelo\r\n",
        "imp_df=pd.DataFrame(X.T.iloc[:,1])\r\n",
        "imp_df[1]=rf_gs.best_estimator_[1].feature_importances_\r\n",
        "imp_df=imp_df.rename(columns = {1:'Feature Importance'})\\\r\n",
        "             .sort_values(by='Feature Importance', ascending=False)\\\r\n",
        "             .reset_index().rename(columns = {'index':'Feature'})\r\n",
        "model_to_save.feature_df_=imp_df\r\n"
      ],
      "execution_count": 226,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aTy4KAx3rCCD",
        "outputId": "2abe76ea-163e-4005-a6b5-91267859b424"
      },
      "source": [
        "# Se guarda el modelo\r\n",
        "joblib.dump(model_to_save, 'clonk_rf_model.pkl')"
      ],
      "execution_count": 227,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['clonk_rf_model.pkl']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 227
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jIEZH_-vpANI"
      },
      "source": [
        "# Se carga el modelo\r\n",
        "clonk_model=joblib.load(\"clonk_rf_model.pkl\")"
      ],
      "execution_count": 228,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 456
        },
        "id": "BloUtSsWxa-B",
        "outputId": "7a8ba9cb-b063-419b-9244-520c84b0cb84"
      },
      "source": [
        "\"\"\"\r\n",
        "Usando el atributo que se creó, feature_df_, se extrae un gráfico de Feature\r\n",
        "Importance del modelo.\r\n",
        "\"\"\"\r\n",
        "plt.figure(figsize=(15, 5))\r\n",
        "plt.bar(clonk_model.feature_df_['Feature'], \\\r\n",
        "        clonk_model.feature_df_['Feature Importance'])\r\n",
        "_=plt.xticks( rotation=90)\r\n",
        "plt.title(\"Feature Importance [Modelo Preliminar]\");"
      ],
      "execution_count": 261,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 456
        },
        "id": "PrwnAyNj3f3r",
        "outputId": "0e5895e3-4606-42b4-bcb6-275f9384b7c5"
      },
      "source": [
        "\"\"\"\r\n",
        "Usando los datos a los que no ha sido expuesto el modelo, y_test y X_test\r\n",
        "para simular la exposición a datos a datos nunca vistos.\r\n",
        "\r\n",
        "Se crea un dataframe a partir de estos y realizaremos la predicción\r\n",
        "del modelo sobre estos datos y evaluaremos la calidad del modelo.\r\n",
        "\r\n",
        "\"\"\"\r\n",
        "model_test=X_test.copy()\r\n",
        "model_test[\"employee_left (Real)\"]=y_test\r\n",
        "model_test[\"employee_left (Predicted_probability)\"]=\\\r\n",
        "            clonk_model.predict_proba(model_test.iloc[:,:-1])[:,1]\r\n",
        "model_test[\"employee_left (Predicted)\"]=\\\r\n",
        "            clonk_model.predict(model_test.iloc[:,:-2])\r\n",
        "model_test.reset_index(inplace=True)\r\n",
        "model_test"
      ],
      "execution_count": 295,
      "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>index</th>\n",
              "      <th>satisfaction_level</th>\n",
              "      <th>last_evaluation</th>\n",
              "      <th>number_project</th>\n",
              "      <th>average_montly_hours</th>\n",
              "      <th>time_spend_company</th>\n",
              "      <th>Work_accident</th>\n",
              "      <th>promotion_last_5years</th>\n",
              "      <th>department_IT</th>\n",
              "      <th>department_RandD</th>\n",
              "      <th>department_accounting</th>\n",
              "      <th>department_hr</th>\n",
              "      <th>department_management</th>\n",
              "      <th>department_marketing</th>\n",
              "      <th>department_product_mng</th>\n",
              "      <th>department_sales</th>\n",
              "      <th>department_support</th>\n",
              "      <th>department_technical</th>\n",
              "      <th>salary_level_high</th>\n",
              "      <th>salary_level_low</th>\n",
              "      <th>salary_level_medium</th>\n",
              "      <th>employee_left (Real)</th>\n",
              "      <th>employee_left (Predicted_probability)</th>\n",
              "      <th>employee_left (Predicted)</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>814</td>\n",
              "      <td>0.10</td>\n",
              "      <td>0.92</td>\n",
              "      <td>7</td>\n",
              "      <td>253</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0.992</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5814</td>\n",
              "      <td>0.76</td>\n",
              "      <td>0.65</td>\n",
              "      <td>4</td>\n",
              "      <td>195</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.004</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>452</td>\n",
              "      <td>0.10</td>\n",
              "      <td>0.87</td>\n",
              "      <td>6</td>\n",
              "      <td>286</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1.000</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>6713</td>\n",
              "      <td>0.83</td>\n",
              "      <td>0.86</td>\n",
              "      <td>3</td>\n",
              "      <td>179</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.004</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>14929</td>\n",
              "      <td>0.10</td>\n",
              "      <td>0.83</td>\n",
              "      <td>7</td>\n",
              "      <td>302</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1.000</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2995</th>\n",
              "      <td>14961</td>\n",
              "      <td>0.73</td>\n",
              "      <td>0.97</td>\n",
              "      <td>6</td>\n",
              "      <td>189</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0.984</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2996</th>\n",
              "      <td>2315</td>\n",
              "      <td>0.78</td>\n",
              "      <td>0.76</td>\n",
              "      <td>5</td>\n",
              "      <td>168</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.000</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2997</th>\n",
              "      <td>3610</td>\n",
              "      <td>0.84</td>\n",
              "      <td>0.90</td>\n",
              "      <td>3</td>\n",
              "      <td>262</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.012</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2998</th>\n",
              "      <td>14924</td>\n",
              "      <td>0.10</td>\n",
              "      <td>0.81</td>\n",
              "      <td>6</td>\n",
              "      <td>245</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0.996</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2999</th>\n",
              "      <td>11279</td>\n",
              "      <td>0.22</td>\n",
              "      <td>0.91</td>\n",
              "      <td>6</td>\n",
              "      <td>222</td>\n",
              "      <td>8</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.040</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3000 rows × 24 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "      index  ...  employee_left (Predicted)\n",
              "0       814  ...                          1\n",
              "1      5814  ...                          0\n",
              "2       452  ...                          1\n",
              "3      6713  ...                          0\n",
              "4     14929  ...                          1\n",
              "...     ...  ...                        ...\n",
              "2995  14961  ...                          1\n",
              "2996   2315  ...                          0\n",
              "2997   3610  ...                          0\n",
              "2998  14924  ...                          1\n",
              "2999  11279  ...                          0\n",
              "\n",
              "[3000 rows x 24 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 295
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 509
        },
        "id": "7dahbDN_rTmv",
        "outputId": "3bedd0d1-1c99-4564-b199-8476190fddaf"
      },
      "source": [
        "# Se simula la evaluación del modelo sobre datos nunca vistos.\r\n",
        "plot_roc_conf_matrix(y_test,X_test, clonk_model, \"Modelo Preliminar - Prueba\")"
      ],
      "execution_count": 306,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      1.00      0.99      2275\n",
            "           1       0.99      0.96      0.98       725\n",
            "\n",
            "    accuracy                           0.99      3000\n",
            "   macro avg       0.99      0.98      0.99      3000\n",
            "weighted avg       0.99      0.99      0.99      3000\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}