ML-Project-BMP_updated.ipynb

ml-project-bmp.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ML-Project-BMP_updated.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyOijvLRdRT5ki0AjhsAisV3",
      "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/gist/Sanket758/a5377ad5f817027a907ef008442deb56/ml-project-bmp.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5SxWhMplXppi",
        "colab_type": "text"
      },
      "source": [
        "# **ML Project - Bank Marketing Prediction**\n",
        "\n",
        "## Tasks to perform\n",
        "## 1 . Read in the file and get basic information about the data, including numerical summaries."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Mg5nMtEiLBEW",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import pandas as pd"
      ],
      "execution_count": 197,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i-gXny7LZBCG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "outputId": "e80a8650-5486-44b1-eddb-b22b88a1e352"
      },
      "source": [
        "df = pd.read_csv('bank-marketing.csv')\n",
        "df.head()"
      ],
      "execution_count": 198,
      "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>age</th>\n",
              "      <th>job</th>\n",
              "      <th>salary</th>\n",
              "      <th>marital</th>\n",
              "      <th>education</th>\n",
              "      <th>targeted</th>\n",
              "      <th>default</th>\n",
              "      <th>balance</th>\n",
              "      <th>housing</th>\n",
              "      <th>loan</th>\n",
              "      <th>contact</th>\n",
              "      <th>day</th>\n",
              "      <th>month</th>\n",
              "      <th>duration</th>\n",
              "      <th>campaign</th>\n",
              "      <th>pdays</th>\n",
              "      <th>previous</th>\n",
              "      <th>poutcome</th>\n",
              "      <th>response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>58</td>\n",
              "      <td>management</td>\n",
              "      <td>100000</td>\n",
              "      <td>married</td>\n",
              "      <td>tertiary</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>2143</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>unknown</td>\n",
              "      <td>5</td>\n",
              "      <td>may</td>\n",
              "      <td>261</td>\n",
              "      <td>1</td>\n",
              "      <td>-1</td>\n",
              "      <td>0</td>\n",
              "      <td>unknown</td>\n",
              "      <td>no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>44</td>\n",
              "      <td>technician</td>\n",
              "      <td>60000</td>\n",
              "      <td>single</td>\n",
              "      <td>secondary</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>29</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>unknown</td>\n",
              "      <td>5</td>\n",
              "      <td>may</td>\n",
              "      <td>151</td>\n",
              "      <td>1</td>\n",
              "      <td>-1</td>\n",
              "      <td>0</td>\n",
              "      <td>unknown</td>\n",
              "      <td>no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>33</td>\n",
              "      <td>entrepreneur</td>\n",
              "      <td>120000</td>\n",
              "      <td>married</td>\n",
              "      <td>secondary</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>2</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>unknown</td>\n",
              "      <td>5</td>\n",
              "      <td>may</td>\n",
              "      <td>76</td>\n",
              "      <td>1</td>\n",
              "      <td>-1</td>\n",
              "      <td>0</td>\n",
              "      <td>unknown</td>\n",
              "      <td>no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>47</td>\n",
              "      <td>blue-collar</td>\n",
              "      <td>20000</td>\n",
              "      <td>married</td>\n",
              "      <td>unknown</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>1506</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>unknown</td>\n",
              "      <td>5</td>\n",
              "      <td>may</td>\n",
              "      <td>92</td>\n",
              "      <td>1</td>\n",
              "      <td>-1</td>\n",
              "      <td>0</td>\n",
              "      <td>unknown</td>\n",
              "      <td>no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>33</td>\n",
              "      <td>unknown</td>\n",
              "      <td>0</td>\n",
              "      <td>single</td>\n",
              "      <td>unknown</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>1</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>unknown</td>\n",
              "      <td>5</td>\n",
              "      <td>may</td>\n",
              "      <td>198</td>\n",
              "      <td>1</td>\n",
              "      <td>-1</td>\n",
              "      <td>0</td>\n",
              "      <td>unknown</td>\n",
              "      <td>no</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   age           job  salary  marital  ... pdays previous poutcome  response\n",
              "0   58    management  100000  married  ...    -1        0  unknown        no\n",
              "1   44    technician   60000   single  ...    -1        0  unknown        no\n",
              "2   33  entrepreneur  120000  married  ...    -1        0  unknown        no\n",
              "3   47   blue-collar   20000  married  ...    -1        0  unknown        no\n",
              "4   33       unknown       0   single  ...    -1        0  unknown        no\n",
              "\n",
              "[5 rows x 19 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 198
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y3JmeDfUZJyy",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "25cb2581-2c0c-47fc-f1c9-22e2623dd5db"
      },
      "source": [
        "df.shape"
      ],
      "execution_count": 199,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(45211, 19)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 199
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QylyhZDnZU9e",
        "colab_type": "text"
      },
      "source": [
        "So, we have 45211 records with 19 collumns"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4ZmXiWPCZRpQ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "outputId": "648f2239-e8a3-4067-d96b-e10eef14d099"
      },
      "source": [
        "df.columns"
      ],
      "execution_count": 200,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['age', 'job', 'salary', 'marital', 'education', 'targeted', 'default',\n",
              "       'balance', 'housing', 'loan', 'contact', 'day', 'month', 'duration',\n",
              "       'campaign', 'pdays', 'previous', 'poutcome', 'response'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 200
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FCSuafOHZlex",
        "colab_type": "text"
      },
      "source": [
        "response column is our target variable and others are feature columns"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WxZw8ezLZdVC",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 459
        },
        "outputId": "dfb58386-97b8-4cca-85ca-9cad2d89eef5"
      },
      "source": [
        "df.info()"
      ],
      "execution_count": 201,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 45211 entries, 0 to 45210\n",
            "Data columns (total 19 columns):\n",
            " #   Column     Non-Null Count  Dtype \n",
            "---  ------     --------------  ----- \n",
            " 0   age        45211 non-null  int64 \n",
            " 1   job        45211 non-null  object\n",
            " 2   salary     45211 non-null  int64 \n",
            " 3   marital    45211 non-null  object\n",
            " 4   education  45211 non-null  object\n",
            " 5   targeted   45211 non-null  object\n",
            " 6   default    45211 non-null  object\n",
            " 7   balance    45211 non-null  int64 \n",
            " 8   housing    45211 non-null  object\n",
            " 9   loan       45211 non-null  object\n",
            " 10  contact    45211 non-null  object\n",
            " 11  day        45211 non-null  int64 \n",
            " 12  month      45211 non-null  object\n",
            " 13  duration   45211 non-null  int64 \n",
            " 14  campaign   45211 non-null  int64 \n",
            " 15  pdays      45211 non-null  int64 \n",
            " 16  previous   45211 non-null  int64 \n",
            " 17  poutcome   45211 non-null  object\n",
            " 18  response   45211 non-null  object\n",
            "dtypes: int64(8), object(11)\n",
            "memory usage: 6.6+ MB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vtNxbe_MaKZ2",
        "colab_type": "text"
      },
      "source": [
        "So, there are total 19 columns outta which 8 are numerical and 11 are categorical.\n",
        "and no null value in any column, so that's a good news!"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m70iyiOEZyvt",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "outputId": "6b3f37c8-0aa9-40fa-e95d-0d8006814cee"
      },
      "source": [
        "# numerical summary\n",
        "df.describe()"
      ],
      "execution_count": 202,
      "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>age</th>\n",
              "      <th>salary</th>\n",
              "      <th>balance</th>\n",
              "      <th>day</th>\n",
              "      <th>duration</th>\n",
              "      <th>campaign</th>\n",
              "      <th>pdays</th>\n",
              "      <th>previous</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>45211.000000</td>\n",
              "      <td>45211.000000</td>\n",
              "      <td>45211.000000</td>\n",
              "      <td>45211.000000</td>\n",
              "      <td>45211.000000</td>\n",
              "      <td>45211.000000</td>\n",
              "      <td>45211.000000</td>\n",
              "      <td>45211.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>40.936210</td>\n",
              "      <td>57006.171065</td>\n",
              "      <td>1362.272058</td>\n",
              "      <td>15.806419</td>\n",
              "      <td>258.163080</td>\n",
              "      <td>2.763841</td>\n",
              "      <td>40.197828</td>\n",
              "      <td>0.580323</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>10.618762</td>\n",
              "      <td>32085.718415</td>\n",
              "      <td>3044.765829</td>\n",
              "      <td>8.322476</td>\n",
              "      <td>257.527812</td>\n",
              "      <td>3.098021</td>\n",
              "      <td>100.128746</td>\n",
              "      <td>2.303441</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>18.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>-8019.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>33.000000</td>\n",
              "      <td>20000.000000</td>\n",
              "      <td>72.000000</td>\n",
              "      <td>8.000000</td>\n",
              "      <td>103.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>39.000000</td>\n",
              "      <td>60000.000000</td>\n",
              "      <td>448.000000</td>\n",
              "      <td>16.000000</td>\n",
              "      <td>180.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>-1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>48.000000</td>\n",
              "      <td>70000.000000</td>\n",
              "      <td>1428.000000</td>\n",
              "      <td>21.000000</td>\n",
              "      <td>319.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>-1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>95.000000</td>\n",
              "      <td>120000.000000</td>\n",
              "      <td>102127.000000</td>\n",
              "      <td>31.000000</td>\n",
              "      <td>4918.000000</td>\n",
              "      <td>63.000000</td>\n",
              "      <td>871.000000</td>\n",
              "      <td>275.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                age         salary  ...         pdays      previous\n",
              "count  45211.000000   45211.000000  ...  45211.000000  45211.000000\n",
              "mean      40.936210   57006.171065  ...     40.197828      0.580323\n",
              "std       10.618762   32085.718415  ...    100.128746      2.303441\n",
              "min       18.000000       0.000000  ...     -1.000000      0.000000\n",
              "25%       33.000000   20000.000000  ...     -1.000000      0.000000\n",
              "50%       39.000000   60000.000000  ...     -1.000000      0.000000\n",
              "75%       48.000000   70000.000000  ...     -1.000000      0.000000\n",
              "max       95.000000  120000.000000  ...    871.000000    275.000000\n",
              "\n",
              "[8 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 202
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4ANLJEL6fnZK",
        "colab_type": "text"
      },
      "source": [
        "## 2 . Describe the pdays column, make note of the mean, median and minimum values. Anything fishy in the values?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "g2WXLRc5a8UX",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 170
        },
        "outputId": "c58d7d01-a40e-4814-8bf7-5533aad67c9d"
      },
      "source": [
        "df.pdays.describe()"
      ],
      "execution_count": 203,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "count    45211.000000\n",
              "mean        40.197828\n",
              "std        100.128746\n",
              "min         -1.000000\n",
              "25%         -1.000000\n",
              "50%         -1.000000\n",
              "75%         -1.000000\n",
              "max        871.000000\n",
              "Name: pdays, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 203
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BRzRVgr2g9Y3",
        "colab_type": "text"
      },
      "source": [
        "yes, there is definitely something fishy going on here, we can see that min value is -1 and 75% of the values are -1. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qjs-g6SVft0d",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 221
        },
        "outputId": "55b10c51-f0b8-4f76-8916-e47f3f9df14f"
      },
      "source": [
        "df.pdays.value_counts()"
      ],
      "execution_count": 204,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "-1      36954\n",
              " 182      167\n",
              " 92       147\n",
              " 183      126\n",
              " 91       126\n",
              "        ...  \n",
              " 749        1\n",
              " 717        1\n",
              " 589        1\n",
              " 493        1\n",
              " 32         1\n",
              "Name: pdays, Length: 559, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 204
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mx-NYQh9hOg9",
        "colab_type": "text"
      },
      "source": [
        "As we can see there are 36,954 records having pvalue -1. this is hurting the overall mean."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zH0qF-Cgh4A2",
        "colab_type": "text"
      },
      "source": [
        "So, i did a quick search on UCI machine Learning repo site to check what does the pdays actually means:  \n",
        "**pdays:** number of days that passed by after the client was last contacted from a previous campaign (numeric; 999 means client was not previously contacted)  \n",
        "here 999 is replace by -1"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bIIV2rtgjIg8",
        "colab_type": "text"
      },
      "source": [
        "So, if pdays = -1, it indicates that this person was not previosly contacted so, we will drop these records because we want to study on people who was involved in the campaign"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TrDEEEtqf9-P",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "outputId": "a12128c7-5dbf-4f9c-d85c-1572776dc9cc"
      },
      "source": [
        "df = df.drop(df[df.pdays==-1].index)\n",
        "df.head()"
      ],
      "execution_count": 205,
      "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>age</th>\n",
              "      <th>job</th>\n",
              "      <th>salary</th>\n",
              "      <th>marital</th>\n",
              "      <th>education</th>\n",
              "      <th>targeted</th>\n",
              "      <th>default</th>\n",
              "      <th>balance</th>\n",
              "      <th>housing</th>\n",
              "      <th>loan</th>\n",
              "      <th>contact</th>\n",
              "      <th>day</th>\n",
              "      <th>month</th>\n",
              "      <th>duration</th>\n",
              "      <th>campaign</th>\n",
              "      <th>pdays</th>\n",
              "      <th>previous</th>\n",
              "      <th>poutcome</th>\n",
              "      <th>response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>24060</th>\n",
              "      <td>33</td>\n",
              "      <td>admin.</td>\n",
              "      <td>50000</td>\n",
              "      <td>married</td>\n",
              "      <td>tertiary</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>882</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>telephone</td>\n",
              "      <td>21</td>\n",
              "      <td>oct</td>\n",
              "      <td>39</td>\n",
              "      <td>1</td>\n",
              "      <td>151</td>\n",
              "      <td>3</td>\n",
              "      <td>failure</td>\n",
              "      <td>no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24062</th>\n",
              "      <td>42</td>\n",
              "      <td>admin.</td>\n",
              "      <td>50000</td>\n",
              "      <td>single</td>\n",
              "      <td>secondary</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>-247</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>telephone</td>\n",
              "      <td>21</td>\n",
              "      <td>oct</td>\n",
              "      <td>519</td>\n",
              "      <td>1</td>\n",
              "      <td>166</td>\n",
              "      <td>1</td>\n",
              "      <td>other</td>\n",
              "      <td>yes</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24064</th>\n",
              "      <td>33</td>\n",
              "      <td>services</td>\n",
              "      <td>70000</td>\n",
              "      <td>married</td>\n",
              "      <td>secondary</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>3444</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>telephone</td>\n",
              "      <td>21</td>\n",
              "      <td>oct</td>\n",
              "      <td>144</td>\n",
              "      <td>1</td>\n",
              "      <td>91</td>\n",
              "      <td>4</td>\n",
              "      <td>failure</td>\n",
              "      <td>yes</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24072</th>\n",
              "      <td>36</td>\n",
              "      <td>management</td>\n",
              "      <td>100000</td>\n",
              "      <td>married</td>\n",
              "      <td>tertiary</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>2415</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>telephone</td>\n",
              "      <td>22</td>\n",
              "      <td>oct</td>\n",
              "      <td>73</td>\n",
              "      <td>1</td>\n",
              "      <td>86</td>\n",
              "      <td>4</td>\n",
              "      <td>other</td>\n",
              "      <td>no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24077</th>\n",
              "      <td>36</td>\n",
              "      <td>management</td>\n",
              "      <td>100000</td>\n",
              "      <td>married</td>\n",
              "      <td>tertiary</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>0</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>telephone</td>\n",
              "      <td>23</td>\n",
              "      <td>oct</td>\n",
              "      <td>140</td>\n",
              "      <td>1</td>\n",
              "      <td>143</td>\n",
              "      <td>3</td>\n",
              "      <td>failure</td>\n",
              "      <td>yes</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       age         job  salary  marital  ... pdays previous poutcome  response\n",
              "24060   33      admin.   50000  married  ...   151        3  failure        no\n",
              "24062   42      admin.   50000   single  ...   166        1    other       yes\n",
              "24064   33    services   70000  married  ...    91        4  failure       yes\n",
              "24072   36  management  100000  married  ...    86        4    other        no\n",
              "24077   36  management  100000  married  ...   143        3  failure       yes\n",
              "\n",
              "[5 rows x 19 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 205
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4QqMlbMtjrTY",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "bcd654a6-77e8-42b3-c1b1-1937ded28c46"
      },
      "source": [
        "df.shape"
      ],
      "execution_count": 206,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(8257, 19)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 206
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lEeYYDrwlX_N",
        "colab_type": "text"
      },
      "source": [
        "## 3. Describe the pdays column again, this time limiting yourself to the relevant values of pdays. How different are the mean and the median values?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ILlXhgTvlHpN",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 170
        },
        "outputId": "79d7b546-935e-4597-9989-3bea32d833d7"
      },
      "source": [
        "df.pdays.describe()"
      ],
      "execution_count": 207,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "count    8257.000000\n",
              "mean      224.577692\n",
              "std       115.344035\n",
              "min         1.000000\n",
              "25%       133.000000\n",
              "50%       194.000000\n",
              "75%       327.000000\n",
              "max       871.000000\n",
              "Name: pdays, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 207
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jO7HYI49lwDS",
        "colab_type": "text"
      },
      "source": [
        "Now that we have limit ourselves to the relevant values only the mean and median looks better."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eofPftM_mcXS",
        "colab_type": "text"
      },
      "source": [
        "## 4. Plot a horizontal bar graph with the median values of balance for each education level value. Which group has the highest median?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hVaxOOeZlu6z",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import numpy as np"
      ],
      "execution_count": 208,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2LPfuMrjqVbg",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        },
        "outputId": "b44b3037-fa55-4938-e0ae-a7e4cee47cd4"
      },
      "source": [
        "df.groupby(['education'])['balance'].agg(np.median)"
      ],
      "execution_count": 209,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "education\n",
              "primary      520\n",
              "secondary    527\n",
              "tertiary     775\n",
              "unknown      782\n",
              "Name: balance, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 209
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_v6jNXETn5he",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "outputId": "d45c949e-46fc-4e3a-8295-54b59b44f6a1"
      },
      "source": [
        "df.groupby(['education'])['balance'].agg(np.median).plot(kind='barh');"
      ],
      "execution_count": 210,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VIT24jc2qYrF",
        "colab_type": "text"
      },
      "source": [
        "So the unknown group has the highest median, and then tertiary is the second highest\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bduI5y6RqwtF",
        "colab_type": "text"
      },
      "source": [
        "## 5. Make a box plot for pdays. Do you see any outliers?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R9DRvm8_po2m",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 296
        },
        "outputId": "b08adb2f-263e-48ca-c795-f515610e613d"
      },
      "source": [
        "sns.boxplot(df.pdays)"
      ],
      "execution_count": 211,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f513d20d1d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 211
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "u5zFeUmiwL83",
        "colab_type": "text"
      },
      "source": [
        "so our median lies at 200, and there are some outliers after 600."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zJfOUEmky9yL",
        "colab_type": "text"
      },
      "source": [
        "## Bivariate Analysis"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Kz9bTHTE5leY",
        "colab_type": "text"
      },
      "source": [
        "### Convert the response variable to a convenient form"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P9rJ41V35jiR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df = df.replace({'response': {\"yes\": 1,'no':0}})"
      ],
      "execution_count": 212,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NSlRdIGJzYJm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "all_cols = df.columns\n",
        "numerical_cols = df._get_numeric_data().columns.to_list()\n",
        "categorical_cols = list(set(all_cols)- set(numerical_cols))"
      ],
      "execution_count": 213,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t8JWgx7tzccM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 487
        },
        "outputId": "7928ef43-c4cd-4aab-ea4f-2e2043fb492c"
      },
      "source": [
        "plt.figure(figsize=(12,8))\n",
        "sns.heatmap(df.corr(), annot=True);"
      ],
      "execution_count": 214,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NyrzuOnC88S2",
        "colab_type": "text"
      },
      "source": [
        "### Histograms"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Vr539YzCBe0L",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")"
      ],
      "execution_count": 215,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SnSx3Dtl5_k6",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "outputId": "115f5f99-6e4a-4932-edb3-97c9889260d3"
      },
      "source": [
        "for col in numerical_cols:\n",
        "    if col != 'response':\n",
        "        sns.FacetGrid(df, hue='response', height=6).map(sns.distplot, col).add_legend()"
      ],
      "execution_count": 216,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 485.25x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 485.25x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 485.25x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAd4AAAGoCAYAAADo5YQfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd5xcV33//9eZvnW2V7VV75ItYbnhisEF2xBsY5sWQmJCcBK+pADfBH7EIcUkgZDAN4GEYgwGgikxWC6Awd3qtrps1S1abe9l6vn9MSOxllfa2d2pu+/n46GHdu+cufezYO17zr2nGGstIiIikh6OTBcgIiIymyh4RURE0kjBKyIikkYKXhERkTRS8IqIiKSRK9MFnO3666+3jz/+eKbLEBGR9DKZLiBdsq7H29nZmekSREREUibrgldERGQmU/CKiIikkYJXREQkjRS8IiIiaaTgFRERSSMFr4iISBopeEVERNJIwSsiIpJGCl4REZE0UvCKiIikkYJXREQkjRS8IiIiaaTgFRERSSMFr4iISBopeEVERNJIwSsiIpJGrkwXILntoS2NCbe9e9O8FFYiIpIbFLwiWUIfYkRmB91qFhERSSMFr4iISBopeEVERNJIwSsiIpJGGlwls0KiA5c0aElEUk09XhERkTRS8IqIiKRRQsFrjLneGHPIGHPYGPPJcV73GmN+EH99izFmQfz4e4wxL4/5EzXGrE/ujyAiIpI7JgxeY4wT+ApwA7ASuMsYs/KsZh8Ceqy1i4EvAvcDWGu/a61db61dD7wPOGatfTmZP4CIiEguSaTHexFw2Fp71FobBL4P3HpWm1uBB+JfPwxca4wxZ7W5K/5eERGRWSuR4K0HmsZ83xw/Nm4ba20Y6APKz2rzbuB7413AGHOPMWa7MWZ7R0dHInWLiIjkpLQMrjLGbAKGrbV7x3vdWvs1a+1Ga+3GysrKdJQkIiKSEYkEbwswd8z3c+LHxm1jjHEBfqBrzOt3co7eroiIyGySSPBuA5YYYxqMMR5iIfrIWW0eAT4Q//o24ClrrQUwxjiAO9DzXRERkYlXrrLWho0x9wJPAE7gG9bafcaY+4Dt1tpHgK8DDxpjDgPdxML5tCuAJmvt0eSXLyIiklsSWjLSWrsZ2HzWsc+M+XoUuP0c7/0NcPHUSxQREZk5tHKViIhIGil4RURE0kjBKyIikkYKXhERkTRS8IqIiKSRgldERCSNFLwiIiJppOAVERFJIwWviIhIGil4RURE0kjBKyIikkYKXhERkTRS8IqIiKSRgldERCSNFLwiIiJppOAVERFJIwWviIhIGil4RURE0kjBKyIikkYKXhERkTRS8IqIiKSRgldERCSNFLwiIiJppOAVERFJIwWviIhIGil4RURE0kjBKyIikkYKXhERkTRS8IqIiKSRgldERCSNFLwiIiJppOAVERFJIwWviIhIGil4RURE0kjBKyIikkYKXhERkTRS8IqIiKSRgldERCSNFLwiIiJppOAVERFJIwWviIhIGiUUvMaY640xh4wxh40xnxznda8x5gfx17cYYxaMeW2tMeZFY8w+Y8weY4wveeWLiIjklgmD1xjjBL4C3ACsBO4yxqw8q9mHgB5r7WLgi8D98fe6gO8Af2itXQVcBYSSVr2IiEiOSaTHexFw2Fp71FobBL4P3HpWm1uBB+JfPwxca4wxwFuB3dbaVwCstV3W2khyShcREck9iQRvPdA05vvm+LFx21hrw0AfUA4sBawx5gljzE5jzF+OdwFjzD3GmO3GmO0dHR2T/RlERERyRqoHV7mAy4H3xP9+pzHm2rMbWWu/Zq3daK3dWFlZmeKSREREMieR4G0B5o75fk782Lht4s91/UAXsd7xM9baTmvtMLAZuHC6RYuIiOSqRIJ3G7DEGNNgjPEAdwKPnNXmEeAD8a9vA56y1lrgCWCNMSY/HshXAvuTU7qIiEjucU3UwFobNsbcSyxEncA3rLX7jDH3AduttY8AXwceNMYcBrqJhTPW2h5jzBeIhbcFNltrH03RzyIiIpL1JgxeAGvtZmK3icce+8yYr0eB28/x3u8Qm1IkIiIy62nlKhERkTRS8IqIiKSRgldERCSNFLwiIiJppOAVERFJIwWviIhIGil4RURE0kjBKyIikkYKXhERkTRS8IqIiKSRgldERCSNFLwiIiJppOAVERFJo4R2JxKR7PLQlsaE2t29aV6KKxGRyVKPV0REJI0UvCIiImmk4BUREUkjBa+IiEgaKXhFRETSSMErIiKSRgpeERGRNFLwioiIpJGCV0REJI0UvCIiImmk4BUREUkjBa+IiEgaKXhFRETSSLsTieSocDTKbw51sLOxhzy3k0Kviw3zS1k7pyTTpYnIeSh4RXLQqf5RHt7exMm+UZZWF+Iwho6BAN/f1kT3UJArl1ZijMl0mSIyDgWvSI7pHAzw1aeP4HI6eO+meays8wOxHvCPdjTz5P42+kdD3Ly2LsOVish49IxXJIeEIlG+t7URhzH80VWLzoQugMvh4PaNc7l8cQUvHe3m5abeDFYqIuei4BXJIY/uaaW1b5TbN86hNN/zhtcdxnD96hrmluaxee8p+kZCGahSRM5HwSuSI/ad7GPrsW6uWFLB8pric7ZzGMMt6+sZDoT5wpOH0lihiCRCwSuSA0KRKI/uaaXW7+O6lTUTtq8vyWPTwjIefOkEe1v60lChiCRKwSuSA1440kXvcIgb19TidCQ2Wvm6FTWU5nv4/BPq9YpkEwWvSJYbGA3xm0PtrKgpYlFlYcLvy/M4+cClC3jm1Q6OdAymsEIRmQxNJxLJcr860E4oEuX61bWTfu9dF83jy08d5sEXT/DZW1adt+1DWxoTOufdm+ZNug4R+S31eEWyWOdAgO0nutnUUE5lkXfS768s8nLT2loe3tHMYCCcggpFZLLU45VZbVHjD19/wFk2tRNt/OD0ixnHrw624XQYrl5eNeVzvP+S+fxkVws/2dnM+y5ZkLziRGRK1OMVyVJt/aPsbu7jkoUVFHqn/hn5gnmlrJvj54EXT2CtTWKFIjIVCl6RLPXLA214XA6uWFIx7XO9/5IFHG4fZPuJniRUJiLTkVDwGmOuN8YcMsYcNsZ8cpzXvcaYH8Rf32KMWRA/vsAYM2KMeTn+5z+TW77IzHSyd4R9J/u5bHEF+dPo7Z72ttU1eFwOHt3dmoTqRGQ6JgxeY4wT+ApwA7ASuMsYs/KsZh8Ceqy1i4EvAvePee2ItXZ9/M8fJqlukRntyf2nyHM7uWzR9Hu7AIVeF1ctreSxva1Eo7rdLJJJifR4LwIOW2uPWmuDwPeBW89qcyvwQPzrh4FrjfYkE5mSIx2DvNo2yFXLKsnzOJN23pvW1tLWH2Bno243i2RSIsFbDzSN+b45fmzcNtbaMNAHlMdfazDG7DLGPG2MefN4FzDG3GOM2W6M2d7R0TGpH0BkJrHW8sS+U/jz3Fy8sHziN0zCtSuqY7eb9+h2s0gmpXpwVSswz1p7AfBx4CFjzBtWd7fWfs1au9Fau7GysjLFJYlkr70n+2nuGeEtK6pxO5P7z7PQ6+LKpZU8tueUbjeLZFAi/7JbgLljvp8TPzZuG2OMC/ADXdbagLW2C8BauwM4AiydbtEiM1EoEuXJfaeoKvJywbySlFzjxjU1nOofZVeTbjeLZEoiwbsNWGKMaTDGeIA7gUfOavMI8IH417cBT1lrrTGmMj44C2PMQmAJcDQ5pYvMLL8+2E7XUJAb19TiSNEQiWtXVONxOnh096mUnF9EJjZh8Maf2d4LPAEcAP7HWrvPGHOfMeaWeLOvA+XGmMPEbimfnnJ0BbDbGPMysUFXf2it7U72DyGS6/a29PHMax1cOK+EpdVFKbtOsc/NpYvL+c2h9pRdQ0TOL6EJgtbazcDms459ZszXo8Dt47zvR8CPplmjyIwWikT5y4d3k+9xceOayW+EMFlXLKnkvkP7aeoeZm5ZfsqvJyKvp5WrRDLsn584xP7Wfm5dX0e+J/XLp785vhLWc4c7U34tEXkjBa9IBn3tmSN89ZmjvGfTPFbV+dNyzcVVhdQU+3juNQWvSCYoeEUy5PtbG/n7zQd5+9pa7rt1ddqua4zhzUsqeO5wJxFNKxJJOwWvyDRZC6f6RjnQ2s+pvlFGQ5Hztj/ZO8Iff28Xn/zxHq5aVskX7liP05Hehd7evLSSvpEQe1r60npdEdF+vCJTtrfHxb8dKGBnl4vOwK9e99qc0jyW1xSzqLKAyiIvZQUeTvaOcPDUAL860E7UWv702iV85KpFeFzp//x72aLYqljPvdbB+rmpmTMsIuNT8IpM0mDI8IV9BXzrcB6lXsuVNQHWrt1AZZGP3pEgXYNBXmsf5GBrP8+82kEwEj3z3jmledywpoaPX7eUOaWZG1FcXuhldX0xz7zWyb3XLMlYHSKzkYJXZBIGQ4Z3P13C/l4Xdy8c4S9XD+H3WNjYMG57ay19IyG6hoJUF/umtaF9sr15SSX/9cxRBgPhrKpLZKbTM16RBIWj8NGXijnY5+Lrl/XxdxcOxkL3PIwxlOR7WFRZmHXhdvniCsJRy/bjWtNGJJ0UvCIJsBY+vauIp9u8fO6CAa6pDWa6pGlbP7cEp8Ow44TWbRZJJwWvSAIebfbyvWN5fGTZEHctHM10OUlR4HWxorZIwSuSZgpekQmMRuAf9xSywh/iz1cPZbqcpNo4v4yXm3oJjxkAJiKppeAVmcA3XsunedjJp9cN4kzvdNuU2zC/lOFghIOnBjJdisiskV2jPUSyibX0th4h/9WjfM0f5tKuMAyVwJyN4J4ZmwtsmF8KwPbj3XhczgxXIzI7KHhFzmajcOwZOP4cJcOd3GXcuMMOOBaBaBgOPgrzLoZF14AvPesrp0pdSR51fh/bT/Rw6aKKTJcjMisoeEXGCgzArgeh81XCJQv5v/234Zuzlvs2xkcx9zXD0V/D8WehZQdcdA+UzIPt30zCxa9Nwjkm78L5pexU8IqkjZ7xisQVDLfAs/8M3cdg7Z18q+oT/E/4Cu5cPGbgkX8OXPA+uOIvwemGF78M7fszV3QSbJxfysm+UXqHc3+KlEguUPCKAPkjp1hx4kFwuOCyPyU692K+ezSPDeVBVpaE3/iGohq47GNQUAXb/htad6e/6CTZuKAMgBPdwxmuRGR2UPDKrOcN9rCs8SHCTh9c8sfgn8ML7W6ODbp478KRc7/R54dL7431gl/5Lgx2pK/oJFpeU0S+x0ljl4JXJB0UvDKrucJDLDvxXRw2zMH574G82E493zmaR6knyg1zAhOcwAcbPgjGCTu+CZHcu13rcjpYN6eEph4Fr0g6KHhl9rJRFjX/BG+on0Pz7mLUWwlA24iDX5z0cseCEXyJzLDJK4UL3gsDJ2Hvj1Nbc4qsmePnVN8okej5154WkenTqGaZtZYff5CSoaMcq72Jwfy5Z44/2uwlYg3vbpjE0pBVK2HxdXD4F1CzGqpXn3lpy7EENyGYl/jlkm1VXTHhqKV9YJRaf17mChGZBdTjlVmptO8A6w79K91Fy2gvvfB1rz3e4mW5P8zCosjkTrr0eiishn0/hcg4A7Ky2Or62Hzkk70zYx1qkWym4JVZxxkZ4dJX/pKAp4xjdTeD+e06kB2jhm2dbq6vn0IAOZyw6ndguBOO/SZ5BadBQ3kBHpeDk73nGUwmIkmh4JVZZ81r/4F/6Dgvrvt7wq7XL/34i5NeLIbr6ycYVHUulcugeg289iSM9Cah2vRwOAy1fp+CVyQNFLwyq5T0H2L58W9zeM7v0Fa+6Q2vP97iZUFhmGXFk7zNPNbKW2PLTh782TQqTb+6kjxa+0aJWg2wEkklBa/MHjbKRfvuI+gq5uVl/+cNLw+FHbzQ7uFtdYGxd58nr6ACGq6MLSk50DqNE6VXvT+PYCRK5+AUe/sikhAFr8wai5sepqJ3NztX/AVBT8kbXt/ZV0jYTuM281iLrganBw7/cvrnSpO6kthoZg2wEkktBa9MmrWWvuEQwXDubJ7uCfay/tC/cqp8E8fr3j5umy29RdTkRVhXloQRyZ5CmH8ZtOzEG0xwOlGGVRZ5cTmMnvOKpJjm8Waxh7Y0JtTu7k3pmQD6i/1t3P/4QZp7hhkNRako9HLFkgounF+KY1r3ZlNv9eGv4goPsWPFJxnvPnLEwt7+fN65IIgjWT/Kwqvg+LPUdTzPsfqbk3TS1HE6DDV+Hy0KXpGUUo9XJjQaivDZR/bxB9/ejstheN/F8/m/Ny5nXlkeP97Vwn/85kh272zTdYSljd/n6Nx30le0eNwmR4Z8jESdXF6VxJ/D54e5F1PR9wqeYF/yzptCdSV5nOwd0QArkRRS8Mp5hSJRPvCNrXzrheP83mUN/O+9l/FXN63knisW8aOPXModG+fSORjgu1saCUWy9Nbzr/6GiMPN7sUfPWeTPQMFGCyXVCb5A8Tia8FCbdeLyT1vitT78wiEo/QMZfEHKZEcp+CV87r/sYNsOdbNP9++js/cvBKv67eLFxtjWD+3hNs3zKWld4Sf7z6ZwUrPoXEL7P9fDjR8kFFf5Tmb7ekvoCF/lFJvknt6eaV0+VdT2fsyjkj2jxauLfEB0NqnAVYiqaLglXN6bE8r//3cMd5/yXxu2zDnnO1W1hVz5dJKth3vYceJLBpIZC08+ddQWMOBhg+cs9loxPDqUB5rilKzO09b2Uac0SAVfdm/Z29VkQ8DtA0oeEVSRYOrZFxN3cP8xcO7WTe3hL+6acWE7a9bWU1T9zCP7mllZa2fPE8i2/qk2P7/heatcMu/Ewnln7PZgcF8ItawungI8CW9jKG8egZ9ddR0b6O9dOO4g7sAFjX+MCnXOzLv9im/1+NyUFrgob0/+3vnIrlKPV4Z1z8+dpBI1PKVuy943e3lc3EYw41rahkNRXn+SGcaKpxAOAi//Gxs16D17zlv0z39BbhNlOWFKRrNawxtZRvJC3RSPHQ8NddIoqoiL2396vGKpIp6vPIGO0708OieVj72liXMKT13T/FsdSV5rKwt5oUjnVy2qCKzvd7tX4eeY/CeH8U2LziPPQP5LCscweOwiW/hB2xqKEu4bZd/FfPafkF19zb6CxsSfl8mVBf7eLVtgHA0isuhz+YiyabgzWIJ33p0ThAAGz+Y8DWttXzu0f1UFXm554qFCb/vtGtXVLH/qX6eP9LJW1ZUT/r9STHSC0/fH5tHu/ja8zbtDTlpHPFxV317SkuyDjcdpRdQ2/kinmAfQY8/pdebjupiL1ELnYNBaoqTf+tdZLbTx1l5nc17TrGrsZc/e+tS8j2T/1xW689jVV0xzx/uZCQ4jY0GpuPZf4mF73V/e87nqaftH4j16FenaGDVWG2lGwFLZe+ulF9rOqrjYduu280iKaHglTOiUcu/PHmIZdVF3LZh7pTPc83yKgLhKNszMcK55wRs+U9YfzfUrp2w+cHBfLyOKA35qQ+ZoKeE/oKFVPa+EhtxnaUqCr2xkc0KXpGUUPDKGb880MbRziHuvWYxzmmsm1jrz2NuaR67Gnux6Q6Yp/4WjBOu/quEmh8azGNJwQjONK142VG6Hm+oj+KhY+m54BS4nQ7KC720aWSzSEooeOWM/3r2KPUledywumba57pgXimn+kfTuxBDy07Y80O45KPgr5+weSAU4cSIl2WpGs08ju6iZYQdvlivN4tVF2tks0iqJBS8xpjrjTGHjDGHjTGfHOd1rzHmB/HXtxhjFpz1+jxjzKAx5s+TU7Yk267GHrYd7+H3Lm/A5Zz+57F1c0pwOgw7GnuSUF0CrIUnPw35FXDZnyb0lsaeYSyGZYWpf757mnW46fKvpqz/AM5I9gZbdbGP7qFg9i4DKpLDJhw9Y4xxAl8BrgOagW3GmEestfvHNPsQ0GOtXWyMuRO4H3j3mNe/ADyWvLJlUrZ/c8Im//1SMUVuD+92Pg3bf/PGBpMYGQ2Q53GyoraYV5p6uWF1TeqnpRx6DE48Bzf9C/iKE3rLia5hDJYlBekNwI7S9VT3bKe8bx/tZRvSeu1EVRV5sUDHQODMPr0ikhyJ/Da8CDhsrT1qrQ0C3wduPavNrcAD8a8fBq41Jjac1BjzDuAYsC85JUuyNQ05eKzZy90NIxS6k/dMdsO8EoaDEV49NZC0c44rEoJffAbKl8CF514a8mwnuoaYnxcg35neXt2Qr5ZhbxWVvS+n9bqTcXpks243iyRfIsFbDzSN+b45fmzcNtbaMNAHlBtjCoFPAH9zvgsYY+4xxmw3xmzv6OhItHZJku8djfVoPrA4uc86F1cVUeR1sbOxN6nnfYOdD0DXa3DdfeB0J/SWSNTS1D2S1ue7ZxhDR8l6Ckda8AWyYJWvcVQUenEaowFWIimQ6sFVnwW+aK0dPF8ja+3XrLUbrbUbKyvPvYOMJF84Cj884ePq2iB1+cnt+TkdhrVz/LzaNkAgnKI5vaP98Ot/gPmXw7IbEn7bqb5RgpFoWp/vjtXlX4XFUN63NyPXn4jTYago8qjHK5ICiQRvCzB2Uuec+LFx2xhjXIAf6AI2AZ83xhwHPgb8X2PMvdOsWZLo16c8dIw6eXdDanp+K2qLCUcth9vP+9lr6p7/Egx3wlsnXixjrBPdQwCZ6fECIXcR/QULYsGbpXN6K4t8dAyqxyuSbIkE7zZgiTGmwRjjAe4EHjmrzSPA6YdrtwFP2Zg3W2sXWGsXAP8K/L219stJql2S4AfH8qj0Rbi6JjUbn88vLyDP7eRAawqe8/Y2wotfhtW3Qf2Fk3rria5h/HluKjzh5NeVoC7/avKC3RSMtmashvOpLPTQMxQkrJHNIkk1YfDGn9neCzwBHAD+x1q7zxhznzHmlnizrxN7pnsY+DjwhilHkn3aRhz8+pSH2+aP4k7RQwenw7CspoiDp/qJRJPcs3vyrwED1513CMG4GruHmVeW+AYQqdBdvIKocWbt7eaKwtjI5q6h1HwoE5mtElqM11q7Gdh81rHPjPl6FDjvJqDW2s9OoT5JoYdP+IhYwx0LUvscb3lNES839bKrsYeNCxLf0ee8jj0T22/36r8C/5xJvbV/JETfSCjjwRtx+ugtXEx53z4aq98CJrvWs6ks8gLQORg4M8pZRKYvu/6lS9pYCw8f93FRRZCGotRuZrC0ugiHgV8caEvOCSNheOyTUDIPLv3jSb+9uSf2XHdOaebnp3b5V+MJD1A03JjpUt6govB08KrHK5JMCt5ZanePi2ODLt41P/WjVn1uJwsrCvnl/iQF745vQvs+eOvnwD358GzqGcZhyIqFIXqLlhJxeKjIwtvNPreTQq+LzgENsJLsYWJyOrtyuniZukeafHgcluvr0/NLdXltEUc6hjjaMc3Rzf2t8Kv7oOFKWHHLxO3H0dwzTI3fhzsJS2NOV9ThpqdoGWX9+zHRDG2jeB4VhV6NbJaMM8YsiC9b/G1gL/BpY8w2Y8xuY8zfxNsUGGMeNca8YozZa4x5d/z4cWPM540xe4wxW40xi8ec86n4OX5ljJkXP/4tY8y/GWNeMMYcNcbcFj9ea4x5xhjzcvz8b44ff6sx5kVjzE5jzA/j61ec1+Q3XJWcF7Hw8yYvV9YE8XvGH/C05djrt/Q7EpnerdAVNcX8fHcrvz7UwcLKCf+7PLfHPwHhALz9i5OaPnRa1Fqae0ZYN7dk6jUkWad/NRV9e/APHaa3aFmmy3mdyiIP+072Z7oMEYAlxGbPFBObPXMRYIBHjDFXAJXASWvtTQDGGP+Y9/ZZa9cYY95PbIbN24F/Bx6w1j5gjPk94N+Ad8Tb1wKXA8uJzdp5GLgbeMJa+3fxpZTzjTEVwF8Db7HWDhljPkFsgPF95/tBMv+RX9JuS4ebtlEnt85N3+IIpQUeGioKeOHwNFZqOvRYbEDVlX8J5YumdIrOwQCBcJS5WfB897T+woWEnHmU92XfqqoVhV6GgxGGA5mbdiUSd8Ja+xLw1vifXcBOYuG4BNgDXGeMud8Y82Zrbd+Y935vzN+XxL++BHgo/vWDxIL2tJ9aa6PxPQmq48e2AR80xnwWWGOtHQAuBlYCzxtjXib2wWD+RD+Ieryz0M+afBS4olxbm95biJctLucnO1sIRaKTv80bGIBH/xwqV8ClfzLlGpq7Tw+syuyI5rGscdJdvJKK3t04okGiDk+mSzqjsvC3I5vnefXrQjJqKP63Af7BWvvVsxsYYy4EbgQ+Z4z5lbX2dM9z7K29ROY1jv3laACstc/Ee9Y3Ad8yxnwB6AF+Ya29azI/iHq8s0wwCpubvby1Lkhemn+PXr64gqFghFeaprB285Ofhv4WuPlL4Jp6MDX1DON1Oc5MlckWXf7VOG2I0oFXM13K65we2dyhkc2SPZ4Afu/0s1RjTL0xpsoYUwcMW2u/A/wTMHZVnXeP+fvF+NcvEFsQCuA9wLPnu6gxZj7QZq39L+C/4+d/CbhszHPjAmPM0ol+AH2EnWWeOeWhL+TgljTeZj7tkoUVGAPPHe6c3HzeQ4/FRjJf+icwb9O0amjuGaG+JA/HFJ4Pp9JA/jwCrmLKe/fQ5V+d6XLOKC3w4DCxHq9INrDWPmmMWQG8GN8EbxB4L7AY+CdjTBQIAR8Z87ZSY8xuYj3Z073TPwa+aYz5C6ADmGjv06uAvzDGhOLXfL+1tsMY87vA94wxpz/N/zVw3k/QCt5Z5rEWL8XuKJdVp78H4893s7bez/OHO/nYWyb8UBgz2A7/ey9Ur4Fr/npa1w9FopzqG+XyJRXTOs9pZw9AmxZj6PKvoqZrC67wMGFXdtwKdzoMZQVeBa9klLX2OLB6zPdfAr50VrMjxHrD4/kna+0nzjrnCeCaca71u2d9Xxj/+wF+u/3t2NefAt400c8wlm41zyKhKPzypJe31AXwZOj/+csWV7CrsZfBRAbrWAuP/HHs+e67/gtc07s93No3SsTarFg4Yzxd/tU4iFLafyDTpbxOZaGHDs3lFUkaBe8ssrXDTV/IwdvqMvdL9PLFFYSjlq3HuiZu/NwX4dXHY/vsVq2Y9rWbe2JbAGbTwKqxhn01jHgqKO/PrtHNFUVeuoeCRLN0FyWR84lv1JNVG18rePQ6wHQAACAASURBVGeRx096yXNarsjAbebTLpxfitfl4LnXJgjeV5+ILZSx+jbY9OGkXLu5Z4Rinwt/njsp50u6+O3m4qHjuEMp2M1piioKvYSjlt7hUKZLEZkRFLyzRNTCEy1erqoJpH0081g+t5OLGsp47nDHuRt1vgY/+n2oWQO3/PuUFsoYT1P3cNb2dk/r8q/CAGX9+zNdyhkVY6YUicj0aXDVLLGr20X7qJPr68+xZKONQl8LtO+HQD/1w07CzgIG8ucw7KtJai2XLCrn848fonMwcOaX+hn9rfDQHeB0w53fBU9ygnI4GKZrKMjG+aVJOV+qjHorGPJVU963j7by6Y3gTpaKwtj0ra7BAFQX8dCWxFYxu3vTvFSWJZKzFLyzxBMtXtzGcnXtWbeZrYWmLXBoMwT6AQPuPOpDw5zuZ454KsgLdfPa3DsIeKe/rd/FC8sB2HqsmxvX1P72hcF2+PYtsb/f95PY7kNJcmZHogxvBZiIruJVzGt/Ck+wl6An80tbFnpdeJwO7csrkiQK3lnAxm8zX1oVpNg9ZoBMOAB7H4bmbVC2CFa8PbYylLeIrUc78YQHKRl4jfK+vax97SusOPoNXp1/NwcafndagbCm3k++x8mWo12/Dd6hLvj2rdDXDO95GOZeNM2f+vWae2IfJOqzYEeiiXT7Y8Fb3r+P1orLMl0OxhjKCjx0K3hlljPGXE9sGpMT+G9r7T9O5TwK3lngyICTE0Mu/mDp8G8PBgbhpa/AwClYej0seevrN2I3DoLuYtrLNtBetoGOsg2sPvyfrDz6DZae+D77Fn2IQwveS8Q5ySDb/k3cwIZSPy/tOwL126D/JGz/Boz2wUV/AJ2vxv6cz8aJ5rq/XlP3CJVFXnxu5+TqzYCAp5TBvHrK+7IjeAHKCz209+sZr2SHBZ989J5knu/4P970tYnaxDdG+ApwHdAMbDPGPBJfz3lSNLhqFniqNfaM7sxtZhuFXd+GoY7YiOGl178+dMfRX7iQF9Z/nkcv/wmnyi9i/av/xtufuZmFzT/BRCe/gP7FFSEO9bsYPLELnv9XiATh4j+CigQX1pgEay3NPdk/sGqsLv8qCkZP4QskMO0qDcoLPHQPa0qRzGoXAYettUettUHg+8CtUzmRgncWeOqUl+X+MPX50diBV5+I9ShXvwsql0/qXP1Fi3h2w7/xi03fZMRbxcV7PsPbn72FhuafYqKJTze5srCJ/+f+Vwr3PADF9fDmP4OyhknVkqjmnhGGghHmlmX/bebTuopXYoGyLNmxqKzASyRq6RvRlCKZteqBpjHfN8ePTZqCd4brDxm2d7q5uiZ+m7D9ALz2JMy5COZePOXzdpRt5MlLvsvTF/47IVchl+z5NO/89Vu4cP/9lPXtw9hxNnUPB2OBv/sHrNr9d1zpeIVfF98Kl3wUfP43tk+Sl+ObMuRSjzfkLmYgfz4V/XtjD+kzrDw+slnPeUWmT894Z7hn2zyEreGa2iCER+Hl70JRLay5bfrzY42hpfoqWqqupLbzORY1/4QljT9g+YnvEHZ46StazJCvDmd0FHd4CH71GgQHwOHEzLuEv+i6g6OBUq52JHHN43G80tSLy2GoKfal9DrJ1uVfRUPrZvIC7Yz4qid+QwqVF5yeUhRkUWVGSxHJlBZg7pjv58SPTZqCd4b7VasHvzvKBWUhOPw0BAfhTb8PziTu+WoMrZVvprXyzXiCfdR1PENZ/wFKBl7FP3iYsDMvtuj/6t8Bhyv2HNflZeWBPDbvc9ETMJR6U9ere6W5l7qSPJyO7NqRaCLdxStY0PoY5X37aM5w8BbnuXE6DF1DGmAls9Y2YIkxpoFY4N4J3D2VEyl4Z7CohadPebmqJogrPAhHn4KatVC6IGXXDHr8HK+/meP1N7/htbs3zYPt3zzz/cWVseeFWzrdXF+fmluYoUiUPS19bJiX3QtnjCfsKqCvoCEWvFVXJ20Fr6lwGENZvqYUyexlrQ0bY+4ltgOSE/iGtXZKgzAUvDPYK90uugIOrqkNwOFfxp6xLrsx02WdsbYshM9peanDk7LgPdg6wGgoytwcWDhjPF3+1Sw6+QgFoycZypvSOI6kKS/00DWo4JXMS2T6TypYazcDm6d7HgXvDPabU14Mliv9bbD3OZjzJihK7vKP0+FxwIbyEFs6Urdpwa6mHgDm5Wjw9hQvJ9r6c8r79mU+eAs8HOkYxFqLyWDvOyuMuXMzLZOcjy4zg0Y1z2DPt7tZWxqmpPEXgI3N180ymyqCHOxz0RtMzS/ynSd6qCryZu+ORBOIOH30FS6mvG9fxkc3lxV4CEUsA4nspSwi56TgnaEGQoZd3W6uruiH5u1QvwHyp7/OcrJdXBnCYtjamZpg3NXUy4XzSnO6h9blX40nPEDR8ImM1lEe39CiW7ebRaZFwTtDvdThJmINNzufh0gA5l2a6ZLGta4shNdheak9iaOs4zoHA5zoGuaCeZnfaGA6eoqWEXF4qOjdk9E6zkwp0gArkWlR8M5Qz7d7yHNGaeh+DorroGR+pksal9cJF5aHeCkFPd6XG2MLZ1yY5VsBTiTqcNNdvIKy/v2TWh0s2UryPTgMmlIkMk0K3hnq2TYP7y45hKO/OdbbzeJbrRdXBjnQ66Ivyc95dzX14HIY1tSnblWsdOn0r8EVDVA6MMHmESnkdBhK8jWyWWS6NKp5BmoddnBkwMVXqp6KLZRRvyHTJZ3X2Oe819Ul+Es9gVGlO/eWsNJv8L3ybRY1pnZ1rFTrL1hA0FVERe9uuv2rMlZHubYHlFnKGPMN4O1Au7V29XTOpeCdgZ5t91DEMEsHt0H9heDO7s0B1pWF8Dhi83kTDt4JRGxsHvPtC0aTcr6MMw46/aup6dqCKzxE2FWQkTLKCjw09/Rl5NoiZ3zWn9RtAflsXyLzgr8FfBn49nQvp1vNM9DzbR7e7X0RRzQI8y7JdDkT8p1+zpvE+byv9jkZjji4sHzm7KbTWbIWB9HY1KIMKSvwMBKKMBIcZxMMkRnMWvsMkJRbZwreGcba2Pzdd3legvwKKJmX6ZISckllkP29sXWbk2FHVyzELyibOcE74qtmyFtNRd/ujNVQVqBdikSmS8E7wxwZcBIKjLAsdABq12X1oKqxLq8OYjE8n6RpRS92eKjNizC3IJqU82WLztJ1FI6cJG+0LSPXLzszpUgjm0WmSs94Z5itnW6uc+7AQTQWvFnkoS2N5xzkFLGQ7/Tz4yOWt88dt0nCrIUtHR6uqA7kyueOhHX61zK37ZdU9rxMY+3b0n79snz1eEWmS8E7w2zp8HCbews2rxTjf2OCbTmWnaN7nQZWFQ2zp78Aa4PTCszX+p10Bhxndj+aScKufHqKllPRt5um6muxjvT+E/a6nRR4XQpekWnQreYZxFrY2xHiYvZicug282lri4foCLo5Nuic1nle6oj1yi6pmpnh0F56Ae7ICKUDhzJy/bJ8t4JXZh1jzPeAF4FlxphmY8yHpnou9XhnkKYhB6uDr+D2hKEmu24zJ2Jt0RAAz7V5WFg0MuXzvNjhpj5/5j3fPa2/YCEBt5+qnl0ZmdNbXujleNdQ2q8rckZi03+Sylp7V7LOpR7vDLKl08ONzq2EPH4ozc4lIs+n2huiyhPkmbapD7CK2liP9+LKGdwjM4aOkvX4h47iCfam/fKl+R76hkOEozPzg41Iqil4Z5Bd7VGudLyCq3YtmNz7v9YYWFM8xEsdbkJT/J1+qM9JT9DBJTPw+e5YHSXrsUBVz660X7u8wIMFeodm9v/GIqmS0G9nY8z1xphDxpjDxphPjvO61xjzg/jrW4wxC+LHLzLGvBz/84ox5p3JLV/GinQexmdCmJpprWaWUeuKhxkMO3ile2qLaZx+vjuje7xA0OOnt2gpVT07MNH07o97Zi7v8Mz+31gkVSYMXmOME/gKcAOwErjLGLPyrGYfAnqstYuBLwL3x4/vBTZaa9cD1wNfNcbouXIKtA47WBHcS8h4oGxRpsuZslVFQziN5anWqd1ufrHDw9yCCHNm6PPdsU6VXYQ7Mkx5f3pXsirT9oAi05JIj/ci4LC19qi1Ngh8H7j1rDa3Ag/Ev34YuNYYY6y1w9ba0x/HfYBNRtHyRls73VzpeIXRkiXgzN3PNoWuKBdXhnisxYud5H8twWhsH+JLZnhv97T+ggZGvBVUd21l0v9jTUORz4XLYehR8IpMSSLBWw80jfm+OX5s3DbxoO0DygGMMZuMMfuAPcAfjgniM4wx9xhjthtjtnd0dEz+pxCOtvXS4GijoHZZpkuZthvnjHJs0MWBvsl9gHih3UN/yMFb62bJqkrGcKrsIgpHWykcaU7jZQ1lBR71eEWmKOUjcKy1W6y1q4A3AZ8yxvjGafM1a+1Ga+3GysrKVJc0I3m6DgDgqFqe4Uqm7211ARxYNjd7J/W+x5q9FLqiXF49ewKhs2QtYYeXmq6tab1uWYGHbi0bKTIliQRvCzB2CaQ58WPjtok/w/UDXWMbWGsPAINA7o78yVLDYVga2EufqxwKcv+DS4XPcnFliM3Nid9uDkXhiZNe3lIXxDe99TdyStThoaP0Asr69+MJ9aftuqf35bVpvMUtMlMkci9vG7DEGNNALGDvBO4+q80jwAeIrepxG/CUtdbG39NkrQ0bY+YDy4HjySpeYvZ0GS5x7Ke35E34c2y1qnO5cc4of72rmIN9TlaUTLwF3ZYON71BBzfUz5D9dyehrexN1HRtoabzxbSt31xa4CEUsQwGwhT5kredYzZ7aEvjma/Pteb4aZsaylJdjuSwCXu88Wey9wJPAAeA/7HW7jPG3GeMuSXe7OtAuTHmMPBx4PSUo8uBV4wxLwM/Af7IWtuZ7B9itms92UihGcVfvzTTpSTN9fWnbze/4cnEuB5t9lHginJlzey5zXxawFNKZ8laqnt24A4NpuWa5doeUGTKEhq9Yq3dDGw+69hnxnw9Ctw+zvseBB6cZo0yAU/XAcI4KapZkulSkqbCZ9lUGeLRZi8fXzV03mWnw1F48qSXa2pn123msVoqLqeidzc1XS9ycNEHk3LORY0/POdreaMeHmAhrsbnWDR0jlvcznivb2Ny6hGZKXJveSN5HWuhYWQ/J9yLwJ1Y7zBX/M78UY4Ouvj1qfPP6d3a6aYr4ODG+tk72CfgLafLv5rq7u14A6nfgarSE8JgaQskZ/9kkdlEwZvjekfCLOMEA/7cn0Z0tnfMG6U+P8KX9hecd5DVVw/l43dHuapm9gYvQEvl5ThsiOXHU3+Tye2wlLnDtAVmx/NdkWRS8Oa4kd5WHMbir83d1arOxe2Ajy4f4pUeN8+eY+OEF9rdPN3m5aPLh8jL3XVDkmLUW0l38SqWnngIXyD1QymqvSEFr8gUKHhzXOHgCQLWzdy6s9c0mRluWzBKXV6ELx3If0Ov11q4f08hdXkR3r946tsIziRNVVfhiAZZd+hLKb9WtTdIu241i0yagjfHzQ8d4bBrES73zOx5eBzwkeXD7OjyvGH95sdbvLzS4+Zjq4Zm7aCqswW85Rxa8D4WtfyUst69Kb1WtTdEb9jFaGRmTGETSRcFbw6LhAIsso30FM6c0czjuWPBCIuKwnzkJT8/OObDWvjlSQ/3vVLI0uIw75o/++buns/eRfcw4iln44F/gBTumVvljU0lag/OzA99Iqmi4M1hwd5WnMbiqVyc6VJSyuuEH13dw6aKEJ/YUcw1T5Tx+y+U4HNaPr+xH6c6XK8TdhfyyrI/paJ3N+z5n5Rdp9ob249XI5tFJmeWD0fJbZ6BRgLWTV3dnJRf63xzOtOhxGP55uW9/NPeAh5v8fJ3F/Zzx4JR3ProOK6j9beyuPGHVDz+KWi4Eoprk36N6niPVwOsRCZHv7ZyWE3gKHtYSH3R7HjA6XLAp9YO8fQN3bxnoUL3vIyDF9f+HYRH4ScfTskt50JnlHxnRAOsRCZJPd4c5YyMMj/axFbnTWzUrVYZx0BhA1z/j/CzP4EXvgSX/5+knt+YHJxStP2bU37rROsziyRKfYYc5RtqxomlK29BpkuRbHbh+2HlrfDU56BpW9JPX+UJ5lbwimQB9XhzVV8zQevEWVyT6UokmxkDN38JTr4MD90Ov7sZqlcm7fQ13hDb+4qIWnDkyp2XcAC6DkN/C4z2QWAAnB4oqICCKqhaAe68TFcpM5iCN0cVDzey2y5ibmHqpovIDJFXCu//KXzjBnjwHfDBx6A8OSudVXmDRKyhK+ii0htOyjlTIhqBgz+Hl/4Dug/HvgdwF4CvKBbGLdtjx1xemLspNigtvzxzNcuMpeDNQSYapjbczFNcx3J3Fv+yk+xRtjAWvt+8Eb79Dnjvw1A5/fW9z0wpCnqyM3jDAdj9A3j+S7Febn45LLgCKpdDWUOsp3taJAh9LXDiOTge/7Pqd2D+ZZx3eyyRSVLw5qCC0VbchGn1LGCFfh9IoqpWwPt+DN+5Db52Fbz9i7Duzmmd8rdzed2sLkpCjckSGIgNpHrp/8FAK9Sug9sfgOEuMOcY2uL0xMK4rAGW3xybA733YehthDW3pbd+mdE0uCoH+QabABgtSP38XZlh6i6AP3wu9vdPPgw/vgd6Tkz5dOWeEE4s7dkywGqwA371t/DFVfCLT0PFUnjfT+Gep2HVO84dumfLK4E3/T4seRs0b4WX/gMTDaW2dpk11OPNQe7BkxyLVlNV6AZm91Z4MgXFtfD+R+Dp++G5L8Keh2HtHZTn3US3fxXWkfivBaeBSm8o86tX9RyHF74Mux6M3V5ecTNc/jGo3zD1cxoHLLsBimpg57dZFPlfDs95l247y7QpeHONtZSPNvKYvZCF+VqjWKbI6YJr/go2/C68+GXY/k3eFv4eIWcB7WUXMpA/n/yRk4Sd+QAYLMZGcEaDOKJBHNFQ7Gsb4vPOPIIjDha2jBJ2+Ai78gi5iqB4PhRUxraRSkVYWQstO2DLf8LeH8eCct2dcNmfQkUS1y+vuwBGeig/8AgjHRW0VF2VvHPLrKTgzTHeYDcFdpB9LOJtGlgl0+Wvh+v/Aa74C5598kfUdG2lsmcnVd07cUeGxn2LxRBxeIg6PEQdbpZaFwFrKB4axRUZwRmNLSXJyfgbnvsCVK+GmjW//btyObh9U6t5oA0O/gx2PACndoOnEC7+CFzyUSium9o5J7LwatpbG5nT8Qwj3kq6/atScx2ZFRS8OaZopBmANu98jNFUIjm/h7Y0Jt649m001b7tzLdLjj+EKzJMvL+LNU4iDg/WOF/Xg/1ZWxnfaa7i68tfpdAVxUTDeMIDrC8Lw1A7+Pxwai/sfBBC8TA3ztio6tNBXLMaqtfE5tKO7R2HgzBwEk7tgdbdcPTX0BxfCKR6Ndz0BVh7B3hTPLLLGI7X3kReoIMFrZvpL2gg7Mo/Z/MtxxJb5WrTxmQVKLlEwZtjCoea6LP5OPLKgM5MlyMzWNThJujwT9iu2hPfHjDgptAVwDpcBDylUFkWC9eNH4yfMAo9x2K91FN7Y2F64vnX76BknLEerNsXG5kcGh5zJQP+ObDsxljoFtXGQnrPw0n8qc/NOpwcq7uZ1Ue+xry2X3C0/ta0XFdmHgVvjskbamZHdAlz8zXCUrLD2O0BFxacZ7CfwxFbuKN8Eax652+PD3fHQrhtb2y6T3AIQiOxXqyvBAorY73hlu2vn3ebASO+KlorLqG+83k6StYxULAgo/VIblLw5hBnZITiUDvbo1cxTwOrJEuc3h7w1FSnFOWXwcIrY3/O59TuqZ0/yU5WXkF53z4aTj7KnkUfntQocBHQPN6cUjgce7670y5mji+Y4WpEYnxOi98Vpj04O7YHjDrcHK+9kbxgF9U92zNdjuQgBW8OKRxpJoKh3TMXj8NmuhyRM6q9s2uXor6ixfTnz6e28wVMVLMLZHIUvDmkcLiFI3YOVXmawC/Zpcoboj3Ti2ikWUvlFXjCg1T27sp0KZJjFLy5wloKRlrYHlnCvDytViXZpdobojPoIjyLZrj1FyxgIH8udR3Pq9crk6LgzRG+YBfuaIBddjHzFbySZaq9QSyG9uDsud2MMbRUXoE33E9F7yuZrkZyiII3RxSOtACwK7qYeXka0SzZ5fSUotl2u7mvYCGDeXXUdz4HdhZ192VaFLw5onC4mRF8nDTVVHh0W0uyy+kpRbNpgBUAxnCy4nK8oT5KB17NdDWSIxS8OaJwpIUDpoG5eSFtjiJZp8QVwWOimd+lKAN6ipYScBVT1a2pRZIYBW8OcERD5I+2sTW8hHn5er4r2ceY2Mjm2dDjjVp4bcjHSCT+69M46Ci9gJKho3iDia3RLLObllzJAQUjJzFYtkaWslADqyRL1cyCubzHhr18o7GaV4fyyXdGeEtFLzdWd+MuvYD6jmeo6t5BU811mS5Tspx6vDng9MCql6OLNZVIslaVN0R70IOdoWu7/KytjE8dWMCpgIf3z2ljbfHQmWM9poSeomVU9r6sqUUyIfV4c0DhSAtdjnK6KWZu3qk3vJ7oFmQiqVTtDRKIOugLOylxRzJdTlLt7nbxUHMpG/yDfGRBK4WuKDfRw2tDPj5zcD7fbamirnwDZQMHKes/QFfJmkyXLFlMPd4cUDjczEGzkCpPkHynpixIdhq7S9FMEojAn20vxu8O80fx0D1tScEoN1V381RnCS9FVzLqKaOqZ0cGq5VcoODNcp5QP57wANsius0s2e108E55l6Is9a/7C3it38WH55+iwPXGD76313VS5QnytcY6TvnXUTzciCfYl4FKJVcoeLNcQfz57tOB5QpeyWpVniAOLK2jM6fH+2qfk68eyuf2BSNc4B8at43XYfn9+adoDXj5afhSAMr796azTMkxCt4sVzjcTAQn++wCTSWSrOZyxHq9J2dQ8P7nq/l4nfCpNYPnbbeueJiF+SP8uHcRA3n1lPcqeOXcFLxZrnCkhTZ3PUHcWqNZsl6tL0jrDHnGe3LYwSONPt7dMEKZd+Kh2leV99E44uNQ/gUUBNrIG21PQ5WSixS82cxGKRg5ySGzELeJUhNflk8kW9X5ArSOeojOgClFX38tHwt8aMlwQu0vK+vHbaJ8f/QyLIbyPvV6ZXwJBa8x5npjzCFjzGFjzCfHed1rjPlB/PUtxpgF8ePXGWN2GGP2xP++Jrnlz2z5o204bZgdkcXMzQvg0FKRkuXqvEFC1kFnju9S1Bs0fO+oj5vnBphbkNhMgkJXlI0lgzzeO4fegoWx4J2pk5plWiYMXmOME/gKcAOwErjLGLPyrGYfAnqstYuBLwL3x493Ajdba9cAHwAeTFbhs8HphTOe0sAqyRG1vthdmVx/zvudI3kMRxx8eOn4A6rO5aryPgYjTra6N+IL9VI40pyiCiWXJdLjvQg4bK09aq0NAt8Hbj2rza3AA/GvHwauNcYYa+0ua+3J+PF9QJ4xxpuMwmeDwpEWgs589odrFbySE+riwZvLz3mjFh46msebq4KsKJncQiBri4cod4f41tBlRI2L8r59KapSclkiwVsPNI35vjl+bNw21tow0AeUn9XmXcBOa+0bEsQYc48xZrsxZntHR0eitc94hcMttLrnAUYDqyQn+F0R8p2RnO7xvtDu5uSIkzsaRib9XoeJPevdOlBOV8EiSgcO6nazvEFaBlcZY1YRu/384fFet9Z+zVq70Vq7sbKyMh0lZb+RXvKCnbzmaABQj1dygjGx57y5HLwPn8ij2B3lurqp/Zu7wD9IBMPLrnV4Q/0UjLYmuULJdYkEbwswd8z3c+LHxm1jjHEBfqAr/v0c4CfA+621R6Zb8KzRElt2bkd0CaXuEMUzbO1bmblqfcGcXUSjP2R4rNnLLXNH8Tmndo6lhSPkOSL8LLQRi6G0/2Byi5Scl0jwbgOWGGMajDEe4E7gkbPaPEJs8BTAbcBT1lprjCkBHgU+aa19PllFzwotO7DAM6NLmaveruSQOl+QrpCb4RzcpOfRJi+BqOH2BaNTPofLwOriYV4YqKY/fz5lAwpeeb0Jgzf+zPZe4AngAPA/1tp9xpj7jDG3xJt9HSg3xhwGPg6cnnJ0L7AY+Iwx5uX4n6qk/xQzUfM2RjwVvBoo0fNdySn18QFWRwdyb/OzHx7PY0lxmLWl0/vUsL54kM6gm2N5q8kLdOILdCapQpkJEvqXYa3dDGw+69hnxnw9Ctw+zvs+B3xumjXOPtZC83baPQsIWYee70pOqfXF/ns9Ouhk9TQDLJ2ODDjZ2e3mU2sGMdOcM7+uODYN6ZfRDazn55T1H+Rk5eVJqFJmAq1clY26j8JIN685FwIaWCW5pcYbwmA5OjDFh6QZ8qPjPpzG8s55U7/NfFqlN0y9L8AzQ/UM5tXFRjeLxCl4s1HTVgB2RRfjxJ65dSeSCzwOS6UnlFO3miMWftzo48rqIFV5ydnzen3xEAcG8ukoXEHhyEk8IW0VKDG58y9jNmneBp4itgYWUOcL4HZoHqDkljpfkKMDmV0rZ8ux7oTabWoo47k2D6dGnPx/686/C9FkrPMP8mh7GVsc62jgV5QMvEZ72caknV9yl3q82ah5K9RfyPHRPN1mlpxU7wtyZMCVM5sl/PC4jxJPlGtqk/fvbUXhCC4T5fmRBYy6SygZPJy0c0tuU/Bmm+AQtO0jULuRzqBbe/BKTpqbF2AkYmgcyv7nvH1Bw5Mnvbxj3ijeJJbrcVgWF4xyYKiA3qIlFA8ew0RzZ7CZpI6CN9u07AQb5URebB8KTSWSXHT6Ts3Bvux/mvVIk49g1HDb/OkPqjrbssIRjg356ChYgtOGKB4+kfRrSO5R8Gab5m1AbGAVaESz5KY5eQHM/9/encdHfdeJH399ZiaZSTI5SAhJyEECCeEoV7l6UajYSmtbelBtrVq1Wtdj19/+9Le/Wq2uuruu7qrVrbtae9h6tVbbirVK6UGhlLNAq0FxKAAAHgxJREFUOBIgkEAScpM7mUkyM5/94zuEXJSEzMx3MvN+Ph48Jvl+vzPz5gvJez7X+4PmaHv4t3j/cMrBnGQP81MC3xqd6+zBi+JdPQefspHcKd3NQiZXhZ+aPZA6i5IWKwlWL6kx0jUlxmdW1fNmh4Ddosl3ejnWboO9T5kdzgXVuGIpaY3h6ws7J7x2dzSzE1woNKU9ybQn5JPSVU4VHwj8G4lJRVq84URrI/HmruBoXQcz4nqD8stAiFAoTvYYiTeMbTmbjE1pbgvA2t3RJNh85Mb1crQznnZnEXF9Ldh7zwblvcTkIYk3nLSdhu4mfNnLON7QJTWaxaRWnOSlssuKK0w7bbwatp1N5rqsPqY6gjf9eq7TxfFuB2edxvCRzG4WknjDSbUxvtuYvJCuXg8z4oLzKVyIUJib4kGjKO8Iz1ZvSUcCbR4bG2aMf9/d8Sh29uD2WTnuycQVO1USr5DEG1Zq9kBMAof6pgPIUiIxqRUnGU3do2GaeLc0J5No83BdVnArw81xGom9rCuetsRCkrpPYfFJNbpoJok3nPgLZ5Q1ulAKch2SeMXklef04rBqjraFX+Lt9Fh4t93JqtQOYoP8WzAt1kN6bB/HuuJocxZi0V6SuiuD+6YirEniDRf9Lqg/BDnLOFrfwYzUeBzWSVL2R4hRWBXMTvJwLAxbvNvOJuPRFlanhaZ+8hyni6Nd8XTE5eG1xJIiy4qimiTecFF7AHweyFnB0fpO5mQmmR2REBMWjjObtYbXmlMoTHCRH6LhnGKni3aPjUZPHO0JBcY4r5YP1tFKEm+48BfOcGVczqnmboozE00OSIiJm5PsobnXQpM7fNbFHeuO44zbzvuntoXsPYsSjHHe8m4Hbc4i7P3txPU2hez9RXgJr4+i0axmN0zJp7zbgU/D3KxECN3vBSGCYo5/gtWxdhvpjn6TozG81pRCnMXLlVM6gLHvYjQRuXG92C0+yrvjaM8shDpZVhTNpMUbDrQ2lhLlLOdoXSeAdDWLiDDHX4axNEwmWHV5LOxsTWRVWkdI51BYFcyKd1PeFUdfTBLd9gxSOstD9v4ivEjiDQftNdBVDzkrKKvvIC7GSl5qvNlRCTFhaXZNTryXktYYs0MBYOvZZPq1hbUh7GY+pyjBRaXL2JChLbEQZ081uEMzuUuEl/D4GBrt/OO75Czj2MFOijMTsVjCZ0ws1MbT9beyIDWIkYhAWJzaz76z5iden4bNTSkUxoduUtVgRU4X3gbFqR47qc4ispu3Q8UWmLc+5LEIc0mLNxzU7AGbA51xGWV1HcyRiVUigixJ66fWZaXBZe6vm5KOBGp77ayb1mrK+xcOTLCKoys+B4/FAcdfNSUWYS5JvOGgZg9MX0JTj4/Wnn5JvCKiLE41JlXtbzG3g+3lhlRSY/oHJlWF2pQYL1Nj+ynvjgNlod05E068JsuKopAkXrN5eqGuBHKWU1bvn1iVJROrROSYn+IhRmn2m9jdfLrHzuHOBNZNa8Vm4m+9ogQXJ7rjAGhzFhpzO+oPmReQMIUkXrPVHQRvn39Gs/FJXFq8IpI4rDAvxcOBFvMS718ap2C3+EyZVDVYUYKLpr4YWvuttPt3K6JcupujjSRes1W9YzzmXcHR+k6ykh2kxMeaG5MQAbYkrZ+DrTF4fKF/77Z+K9tbkliT1o7TZkIAg5wrpHGiO47+GCdkLTK6m0VUkcRrttM7IHUWOKdRWtvBPOlmFhFoSWo/Lq/iuAl1m//ckIpXK26aFvxCGReTH9+LVWnKu4zuZgqvh+rd4JJqOdFEEq+ZfD6o2gEzrsLd7+VEUxfzpkviFZFncapRSCPUE6za+q282jiFa1I7yAyDylmxFk1BnJvybodxoOh60F6oeNPcwERISeI1U1MZuNtgxlWUN3Th9Wlp8YqIlJfgJTXWF/Jx3o31afRrxZ1ZzSF93/dSmODiZE8cXg1kLwNHMpRLd3M0kcRrptPnxnevpLTOqGAjLV4RiZQKfSGNtn4rm5tSWJXaQVYYtHbPKXK66fVZqHbZwWqDWe+TZUVRRhKvmap2QGIWTMmntLaDhFgruVOkVKSITMun9nOy0xayQhob69PwhFlrFwbvVDRonFeWFUUVSbxm0dqYWJV3JShFaV0Hc7OSorpUpIhsqzP7AHirPviz9qu7LWxqSmFVWntYjO0ONi22nySb5/w4b+H7jUdZVhQ1JPGapfUUdNbCjKvw+TRldZ3SzSwi2txkDxkOL1tCkHi/e9CJVcHd08OrtQtGt3tRgut8izcxQ5YVRRlJvGap2mE8zriK6tYeuno9MrFKRDSlYE1mH9saYukP4nLaXU0xvHLGwfrMs6TGeoL3RhNQlOCm1m2nvcffGpdlRVFFEq9ZTr8DjhRIn0tprVGxSlq8ItKtyeyj02MJ2iQrn4bvlDiZHufl5gzz1+1eyLlx3gM1/kQry4qiiiRes1TtgLwrwGKhtK4Dq0UxO0NKRYrIdnVGHzalg9bd/Gylg8NtMfz/BV3YLeE7S3hmghuF5kCVP/HKsqKoIonXDB11cPYEzLgKgNLaDmalJ+CIsZocmBDBlRSjWZrWz5tBSLwNLgvfPeTkyvQ+bs0N/X674xFv9ZHj6GV/tX+LQllWFFUk8ZqhcqvxWLAagNI6KRUposeazD6OtsdQH+BlRd/Y76TPq/ju0k7UJFgcUOR0s7+qDX0u0cqyoqghidcMlW8Z47uZC2np7qOu3c1cSbwiSlyXZbRG36gLXKv3rzV2NtU6+Mf53eQ7vQF73WAqSnDR7uqnsrnbOCDLiqKGJN5Q0xoq3oKCVWCxcOiMUbFqQXayyYEJERrFSV4KEz08VxkXkNdrcise3p/I/JR+Pl3UE5DXDIVzE6z2nxvnlWVFUUMSb6i1VEBHzUA388Fq44fushxJvCI6KAUfm+WipDWGkglumuDT8OU9SXT2K364vMPUTe7HK9vRR6Lddn6cF2RZUZSYRP9NI0TlW8bjzDUAlNS0MzM9gSSHeZuECxFqt89wE2/18auTE2v1PlEex9YGOw8v6qQ4eXJ0MZ9jUbAoN+V8ixeg6AZjWdHJ180LTATdmBKvUmqdUuqYUuqEUurBUc7blVLP+c/vUkrl+4+nKaXeVEp1KaUeDWzok1TFW5A4HdIKAThY08ainBSTgxIitJJiNLfPcPPnagetvZc2E+pAi43vH3KyLtvNvTPdAY4wNBbnpnC0vhNXn/9DQ84yiJ8Kx/5qbmAiqC6aeJVSVuCnwI3APOAepdS8YZfdD7RqrQuBHwHf8x93Aw8DXwlYxJOZz2fMaJ65GpSivt1NY2cvC6WbWUShj8500etTPH/KMe7n1vZY+Mw7yWTE+fj3STKLeTRL8lLw+vTAXA8sVpi9zphg5Q2vGtMicMbS4l0BnNBaV2it+4BngfXDrlkPPO3/+g/AWqWU0lp3a63fxkjAouEwuFoGxndL/FVrFkqLV0ShuSlelqf18auKeNzj6CXu9iju356M26N48uo2UmIn77rXxbnGz/7+qkHjvMU3grv9/LahIuKMJfFmA9WDvq/xHxv1Gq21B2gH0sYahFLqAaXUXqXU3qamprE+bfIZWL97LQCHatqxWpSs4RVR64tze6jutvKfh51jur7XC3+/M4lj7TYevaKD2ZNsXHe4NKedGWnxQ8d5Z10HVrt0N0ewsJhcpbV+TGu9TGu9LD093exwgqdiizG2m2x8bimpaWN2RiJxsVKxSkSn1Zl9fGxWD4+Xx7O94b0nGHb1Kz61PYU36u38y+WdA9sMTnZLclPYV9V6vpBGbIIx+fLYK1LFKkKNJfGeAXIHfZ/jPzbqNUopG5AMnA1EgBGjrwdObRtYJK+1Ma6zSMZ3RZR7aEEXMxM9fHlvEm19ow/WNrot3Ls1hZ1NMfxgeQcfmaSTqUazJG8KjZ291LUP+jsV3whtp6GxzLzARNCMJfHuAYqUUgVKqVjgbmDjsGs2Avf5v94AvKG1fFQb4tQ28LiN5QJAVUsPbT39Mr4rol6cDX68ooOzbgsffC2VV2rsAw29Xi/87Fg87/tbKmXtNn52ZTt3zoicpAvGBCtgaHfz7HXG47FXTIhIBNtFV69rrT1KqS8CmwAr8KTW+ohS6tvAXq31RuAJ4FdKqRNAC0ZyBkApdQpIAmKVUrcBN2itSwP/VwlzxzdBTALkXwMY63cBmdEsBLBgioffrm7j4f1OPr8zmVmJHrSGJreFTo+F92f18tDCLmYmTu4x3dHMyUzCbrPw7ulWPrgwyziYlAXZS43Ee60sCok0Yyobo7V+BXhl2LFvDPraDdx1gefmTyC+yKA1lG82xm1sdsCoWGW3WSjOlK0AhQBYPrWfl9e28rvKOF6tjSUpRnO13ccN03tZlRG5S2tibRaW5KWw+9Sw0bnim+CN70B7DSTnmBOcCIqwmFwV8ZqOQnsVzL5h4NC7Va1clp1MjFX+CYQ4x2Yxykn+alU7P72ig+8s6YropHvOyoI0Sms76HAP+rvOu814LB0+sicmO/mtHwrHNxmPhdcD4OrzcqimneX5qSYGJYQIFytnpuLTsPdUy/mDUwsh4zIofcm8wERQSOINhfJXIWPBwDKiA9VteHyaFQVTTA5MCBEOLs+bQqzVwq6KlqEn5t8G1bugffhCEjGZSeINNlcrVO0c0s2851QLSsHSPGnxCiHAEWNlUW4yOyuGjfPOu914LJPu5kgysT25xMWdfMPYbaToAwOH9pxqoTgjkeR42ZFIiPHYVdly8YsmqZUFafzPWyfp6vXgtPt/NZ/rbj7yElzxOXMDFAEjLd5gK3vZ2G0kZxkAHq+PfadbZXxXCDHEypmpeH166DgvGJOsqndCR605gYmAk8QbTH09cPxvMO9WY9cR4Gh9J919XpYXSOIVQpy3dMYUbBY1slU/X2Y3RxpJvMFU/ir098D82wcO7fb/UC3Pl4lVQojz4mNtLMxJZtfwcd6pRUZ38+E/mBOYCDhJvMF05EVISIcZVw8c2nu6hZwpcWQlx5kYmBAiHK2cmcbBmna6ez1DTyz8MNTsgeZycwITASWJN1j6uo31u/PWD3Qza63ZXSnju0KI0a0qnIrHp3nn5LBW78IPgbLCgd+aE5gIKEm8wXJ8E3hcQ7qZK5q7ae7qlcQrhBjVsvxU4mOtbDnWOPREYiYUroWDz4Ev8upVRxtJvMFy5EVwZkDelQOHth5vAmBV0VSzohJChLFYm4WrZk1ly7EmRmzwtuge6DgDlVvNCU4EjCTeYOjtMiZWDepmBthyrImZ6QnkpsabGJwQIpytKU7nTJuLk01dQ08U3wSOZOlujgCSeIOh9E/G3rvz7xg45O73srPiLGtmTzMxMCFEuFtTnA4YH9SHiHEYv1PK/gzuDhMiE4EilauCYd/TkFYEeVcMHNpZcZZej4/V/h8qERiRXMlIRKecKfEUTnOy5VgTn141c+jJxffCu0/BkRdg6SdMiU9MnLR4A62xzChqfvnHQamBw1uONeGIsbBSCmcIIS5izex0dle2jFxWlLMMps2H3b8w9vkWk5Ik3kDb9wxYYmDxR4Ycfut4E1fMTMMRY73AE4UQwrCmeBp9Xh87hi8rUgpWfhYaDsPp7eYEJyZMEm8g9buh5Hcw92ZIOD9z+fTZbiqbu1kzW7qZhRAXt7xgCvGxVl4/2jDy5IK7IG4K7Pp56AMTASGJN5DK/mxsA3j5fUMOv+VfRrS6WCZWCSEuzm6zsnZuBn87XE+/1zf0ZGy88Tvm6MvQVm1OgGJCJPEG0r6nIWUGFKwecnhzaQMz0uIpmJpgUmBCiMnm1kXTae3p5+3y5pEnl3/aeNzzeGiDEgEhiTdQag/AqW2w7JNgOX9bmzp72X6imZsXZpkYnBBisrl29lSSHDb+XDLKdoApuTDnZuPDfl9P6IMTEyKJN1De/iHYk2DZp4Yc/svBWnwa1i/ONikwIcRkZLdZWXdZJpuO1OPuH6VM5JVfMIa29j4Z+uDEhEjiDYSm48ZemSs+Y1SWGeSlA7XMzUpidkaiScEJISarWxdl093n5c2jjSNP5l0BM9fA2z8yquWJSUMKaATC9kfA5oCVnxty+FRzNweq2/jqjXNMCkyIyUEKoYzuyllpTHXa2VhSy40LRhmuuu5r8MT1sOcXcM0/hj5AcUmkxTtRbVXGjiFL7wPn0OVCG0tqUQpuWTTdpOCEEJOZ1aL44IJMXj/aSLurf+QFuSug8HrY/mMpIzmJSOKdqLcfARRc9fdDDmuteenAGVbkpzI9RTa9F0Jcmg8tz6XP4+O5PVWjX3DdQ8ZYr6zrnTSkq3kiGo7Au780ZjIn5ww5ta+qjYqmbj59zczRnytEhJPu48CYPz2ZlQWpPP3OaT51dQE267D2Uvblxs5F7/wElnwUkmQFRbiTFu+l0hpe+SdwJBnjLMM8tvUkyXExrF8s3cxCiIm5/5oCzrS52HRklEpWADf8C3h6YdNXQxuYuCSSeC/VkRfg9Nuw9hsQP3Tjg5NNXbxa2sDHr5xBgl06FYQQE7N2bgZ5qfE8ub1y9AvSZsG1X4EjL0L55tAGJ8ZNEu+l6O2CTV+HrEUjykMCPL6tghirhfuuyg99bEKIiGO1KD55dT7vnm7lQHXb6Bdd/SVjO9K//F8pqhHmJPFeis3fgM5auPE/wDJ0t6HGDjd/fPcMdy3NYarTblKAQohIc9eyXBLtNh59o3z0C2x2uOURY6XFm/8a2uDEuEjiHa/DL8DeJ4xZzHkrR5x+YnslHp+PzwzfwFoIISbAabfxuetm8VpZI1v9G6+MkH+NUcd5x6NGUR8RliTxjsfZk7DxHyBnOaz95ojTJxq7eOrtU6xfnE2+bIgghAiw+68pYEZaPN9+uXTkrkXnfODfIHspvPR5aL5A61iYShLvWPX1wPOfMLqWNzwF1pghp7XWfO3FQzhiLDx001xzYhRCRDS7zcrDH5zHicYuntlxevSLbHb40DNgi4XnPirlJMOQJN6x6HfB7+6GhsNw+8+NnUGGeX5vDbsqW/jqTXNJT5SxXSFEcKydO43Vs9N5ZPNxzrS5Rr8oOQc2PAnNx+G3H5LkG2Yk8V5MvxuevRcqt8L6/4bidSMuqW93829/LWN5/hQ+vGxkUhZCiEBRSvGtW+cD8MAze3H1jbJzERgbKNzxC6jaAb+5C3o7QxajeG+SeN+Lqw2euxdOvg63/gQW3zPikg53P594ajf9Hh/fvWMBFosyIVAhRDTJn5rAj+9ZTGldBw++cBCt9egXLtgAdz4B1bvg13dC1wUmZYmQksR7IXUl8NhqqNgCt/wELv/4iEt6PV4++8y7nGjs4n8+upTCabL1nxAiNN43J4Ov3FDMnw7U8shr5RdOvpfdAXc9BbUH4GfXQOW20AYqRpDEO5ynD955FB6/Hrz98Mm/GjsPDdPh7ucLv9nHjoqzfH/DQq6dnT7KiwkhRPB8fs0s7rw8hx+/Xs6Dfzx04ZnO89bDZ94AuxOeuRVe/7aM+5pI6hmeozWUbYTN34TWSij6ANz235AwdcSlh8+084Xf7qOm1cW3bp3PHZfnjPKCQggRXEop/mPDQqanOPivN05Q09bD9zcsInu0HdEyL4MH3oJX/h9s+wHs/zWseRCWfByskgpCSV2we8Iky5Yt03v37g3dG3Y1woHfwL5noKUC0ucaBccL14IaOl7b0OHm529V8Otdp0mNj+XRjyxhWX7qBV544nY9/4OgvbYQwnwr7/pywF7r+b3VfP2lwwA8cO1MPrt6Fs4L1Yqv3g2vPgzVO8GZCYs/YuxslDYrYPFcgqiZIDOmxKuUWgf8GLACj2ut/33YeTvwDLAUOAt8WGt9yn/uq8D9gBf4B631pvd6r6An3q5GqD8Ep7fDyTehdj+gIe8qWPoJuOzOIZ/+evo8bCtvZnNpAxtLavH6NLctzuahm+aQFuSSkJJ4hYhsgUy8AGfaXHzvr0fZWFJLfKyVdfMzuXXxdFYUpBIfOywJaw3lr8LeJ41H7YP0OTDzOihYBZkLIDl3RAMkiCTxDlyglBU4DlwP1AB7gHu01qWDrvk8sFBr/XdKqbuB27XWH1ZKzQN+B6wApgOvAbO11heY/x6AxOvpNXbocHeAu93YILqzFtrPGC3anmZ/0FajAtWs62D+HTQ68jhU005TZy/NXb1UNHdzrL6T8oYu+rw+Eh02bl6YxedWF5KXFn/p8Y2DJF4hIlugE+85JdVtPLunipcP1tHp9mC1KIozEpmblUT2lDiyUxwkx8WSFGfDEWPF3tNAauVG0hvfwVa9Azxu44XsSUYrOCnb+BOfahxzJIE1Fiw2o5jQnJsDkaCjJvGOpWN/BXBCa10BoJR6FlgPlA66Zj3wz/6v/wA8qpRS/uPPaq17gUql1An/6+0ITPij8Hngxc+e/z4mHpKmG3+K10HGZZAx39hZyJE8cNmOA2f40rMHBr6flmhnTlYS1xRNZXVROssLUokZvgG1EEKEoUW5KSzKTeGbt8xnx8mz7K9qZV9VG++cbKa+w83o7a35PHHffaz9SLKxqqPxCDSUGg2WsyeNWga9HaO/4TcvsGOSGNVYEm82UD3o+xpg+O4AA9dorT1KqXYgzX9857DnZg9/A6XUA8AD/m+7lFLHxhT9mHQA9cC+cT3rNEbTPsimAs3Bf5uoIPcycOReBs5F7uVXQhbIWLz/e5f4xG8FpFHyN631yApFESgsprJprR8DHjM7jlBTSu3VWi8zO45IIPcycOReBo7cSzGasXxMOQMMroOY4z826jVKKRuQjDHJaizPFUIIIaLGWBLvHqBIKVWglIoF7gaGb/S4EThXZWID8IY2Zm1tBO5WStmVUgVAEbA7MKELIYQQk89Fu5r9Y7ZfBDZhLCd6Umt9RCn1bWCv1noj8ATwK//kqRaM5Iz/ut9jTMTyAF94rxnNUSjquteDSO5l4Mi9DBy5l2KEsCugIYQQQkQyWR8jhBBChJAkXiGEECKEJPGaQCm1Til1TCl1Qin1oNnxTCZKqSeVUo1KqcODjqUqpTYrpcr9j1PMjHGyUErlKqXeVEqVKqWOKKW+5D8u93OclFIOpdRupVSJ/15+y3+8QCm1y/+z/px/gqqIcpJ4Q8xfgvOnwI3APOAef2lNMTa/BIYvsn8QeF1rXQS87v9eXJwH+LLWeh5wBfAF//9FuZ/j1wu8T2u9CFgMrFNKXQF8D/iR1roQaMWoWy+inCTe0Bsowam17gPOleAUY6C13ooxc36w9cDT/q+fBm4LaVCTlNa6Tmu9z/91J1CGUVlO7uc4acO5DW5j/H808D6MMrog91L4SeINvdFKcI4ooynGJUNrXef/uh7IMDOYyUgplQ8sAXYh9/OSKKWsSqkDQCOwGTgJtGmtPf5L5GddAJJ4RYTxF26RNXLjoJRyAn8E/o/WekgVfLmfY6e19mqtF2NU6FsBzDE5JBGmJPGGnpTRDLwGpVQWgP+x0eR4Jg2lVAxG0v2N1voF/2G5nxOgtW4D3gSuBFL8ZXRBftaFnyTe0BtLCU4xPoNLlt4H/MnEWCYN/9adTwBlWusfDjol93OclFLpSqkU/9dxGPuXl2Ek4A3+y+ReCkAqV5lCKXUT8AjnS3D+q8khTRpKqd8BazC2W2sAvgm8BPweyMPY0fFDWuvhE7DEMEqpa4BtwCHA5z/8EMY4r9zPcVBKLcSYPGXFaND8Xmv9baXUTIwJlKnAfuCj/v3JRRSTxCuEEEKEkHQ1CyGEECEkiVcIIYQIIUm8QgghRAhJ4hVCCCFCSBKvEEIIEUKSeIUwiVLqn5VSXzE7DiFEaEniFUIIIUJIEq8QIaSU+ppS6rhS6m2g2H/sM0qpPf69XP+olIpXSiUqpSr9JR1RSiUN/l4IMXlJ4hUiRJRSSzFKhC4GbgKW+0+9oLVe7t/LtQy4379N3xbgg/5r7vZf1x/aqIUQgSaJV4jQWQW8qLXu8e8CdK5G92VKqW1KqUPAvcB8//HHgU/6v/4k8FRIoxVCBIUkXiHM90vgi1rrBcC3AAeA1no7kK+UWgNYtdaHTYtQCBEwkniFCJ2twG1KqTilVCJwi/94IlDnH7+9d9hzngF+i7R2hYgYskmCECGklPoaxvZwjUAVsA/oBv4JaMLYGShRa/0J//WZQCWQ5d/nVQgxyUniFSKMKaU2AOu11h8zOxYhRGDYzA5ACDE6pdR/ATdizIAWQkQIafEKIYQQISSTq4QQQogQksQrhBBChJAkXiGEECKEJPEKIYQQISSJVwghhAih/wVvx8EAybGNdgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 485.25x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 485.25x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAd8AAAGoCAYAAAAHJ+8hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deZhcVZ3/8fe3lq7el6Q7C9khQEQSBAIBYRhEUVRGXEdERRAHZdSR+en4MI7jMI4z7s6oODooqKigghsqggjIohCIELIQIBsk3Uk63en0vtR2fn/c253eq6rTfbu76vN6nn666t5Tp85NdfrT59xzzzXnHCIiIhKc0HQ3QEREpNAofEVERAKm8BUREQmYwldERCRgCl8REZGARabrjS+66CJ39913T9fbi4jIzGLT3YAgTVvPt7m5ebreWkREZFpp2FlERCRgCl8REZGAKXxFREQCpvAVEREJmMJXREQkYApfERGRgCl8RUREAqbwFRERCZjCV0REJGAKXxERkYApfEVERAKm8BUREQmYwldERCRgCl8REZGAKXxFREQCpvAVEREJWGS6GzAjbPjuyG1rrwy+HSIiUhDU8xUREQmYwldERCRgCl8REZGAKXxFREQCpvAVEREJmMJXREQkYApfERGRgCl8RUREAqbwFRERCZjCV0REJGAKXxERkYApfEVERAKm8BUREQmYwldERCRgCl8REZGAKXxFREQCpvAVEREJmMJXREQkYApfERGRgCl8RUREAqbwFRERCZjCV0REJGAKXxERkYApfEVERAKm8BUREQmYwldERCRgGcPXzJaY2QNm9oyZbTWzj4xS5nwzazOzjf7Xp6amuSIiIrNfJIsySeCjzrknzawC+IuZ3euce2ZYuYedcxdPfhNFRETyS8aer3Nuv3PuSf9xB7ANWDTVDRMREclXOZ3zNbPlwKnA+lF2n21mT5vZ78zspWO8/moz22BmG5qamnJurIiISD7IOnzNrBz4GXCtc6592O4ngWXOuVOArwO/HK0O59yNzrm1zrm1dXV1E22ziIjIrJZV+JpZFC94f+Sc+/nw/c65dudcp//4LiBqZrWT2lIREZE8kc1sZwNuArY5574yRpkFfjnM7Ey/3kOT2VAREZF8kc1s53OAdwObzWyjv+0TwFIA59y3gLcC15hZEugBLnXOuSlor4iIyKyXMXydc48AlqHMDcANk9UoERGRfKYVrkRERAKm8BUREQmYwldERCRgCl8REZGAKXxFREQCpvAVEREJmMJXREQkYApfERGRgCl8RUREAqbwFRERCZjCV0REJGAKXxERkYApfEVERAKm8BUREQmYwldERCRgCl8REZGAKXxFREQCpvAVEREJmMJXREQkYApfERGRgCl8RUREAqbwFRERCZjCV0REJGAKXxERkYApfEVERAKm8BUREQmYwldERCRgCl8REZGAKXxFREQCpvAVEREJmMJXREQkYApfERGRgCl8RUREAqbwFRERCZjCV0REJGAKXxERkYApfEVERAKm8BUREQmYwldERCRgCl8REZGAKXxFREQCpvAVEREJmMJXREQkYApfERGRgCl8RUREAqbwFRERCZjCV0REJGAKXxERkYApfEVERAKm8BUREQmYwldERCRgCl8REZGAKXxFREQCpvAVEREJmMJXREQkYApfERGRgCl8RUREApYxfM1siZk9YGbPmNlWM/vIKGXMzL5mZjvMbJOZnTY1zRUREZn9IlmUSQIfdc49aWYVwF/M7F7n3DODyrwWON7/Wgd80/8uIiIiw2Ts+Trn9jvnnvQfdwDbgEXDil0C3OI8jwHVZrZw0lsrIiKSB3I652tmy4FTgfXDdi0C9g56Xs/IgBYRERFyCF8zKwd+BlzrnGufyJuZ2dVmtsHMNjQ1NU2kChERkVkvq/A1syhe8P7IOffzUYo0AEsGPV/sbxvCOXejc26tc25tXV3dRNorIiIy62Uz29mAm4BtzrmvjFHsTuByf9bzWUCbc27/JLZTREQkb2Qz2/kc4N3AZjPb6G/7BLAUwDn3LeAu4HXADqAbuHLymyoiIpIfMoavc+4RwDKUccAHJ6tRIiIi+UwrXImIiARM4SsiIhIwha+IiEjAFL4iIiIBU/iKiIgETOErIiISMIWviIhIwBS+IiIiAVP4ioiIBEzhKyIiEjCFr4iISMAUviIiIgFT+IqIiARM4SsiIhIwha+IiEjAFL4iIiIBU/iKiIgETOErIiISMIWviIhIwBS+IiIiAVP4ioiIBEzhKyIiEjCFr4iISMAUviIiIgFT+IqIiARM4SsiIhIwha+IiEjAFL4iIiIBU/iKiIgETOErIiISMIWviIhIwBS+IiIiAVP4ioiIBEzhKyIiEjCFr4iISMAUviIiIgFT+IqIiARM4SsiIhIwha+IiEjAFL4iIiIBU/iKiIgETOErIiISMIWviIhIwBS+IiIiAVP4ioiIBEzhKyIiEjCFr4iISMAUviIiIgFT+IqIiARM4SsiIhIwha+IiEjAFL4iIiIBU/iKiIgELDLdDZgJ1u9uGbFtZ2oPAJetWxp0c0REJM+p5ysiIhIwha+IiEjAFL4iIiIBU/iKiIgELGP4mtnNZnbQzLaMsf98M2szs43+16cmv5kiIiL5I5vZzt8DbgBuGafMw865iyelRSIiInkuY8/XOfcQMPJaHBEREZmQyTrne7aZPW1mvzOzl45VyMyuNrMNZrahqalpkt5aRERkdpmM8H0SWOacOwX4OvDLsQo65250zq11zq2tq6ubhLcWERGZfY46fJ1z7c65Tv/xXUDUzGqPumUiIiJ56qjD18wWmJn5j8/06zx0tPWKiIjkq4yznc3sNuB8oNbM6oF/A6IAzrlvAW8FrjGzJNADXOqcc1PWYhERkVkuY/g6596RYf8NeJciiYiISBa0wpWIiMwa5pn12TXrD0BERPKbmS03s+fM7BZgC/CvZvaEmW0ys3/3y5SZ2W/9y163mNnb/e0vmNkXzGyzmT1uZisH1Xm/X8d9ZrbU3/49M/uamf3ZzHaZ2Vv97QvN7CF/JcctZvZX/vZXm9mjZvakmd1uZuXZHJPCV0REZoPjgf8F/hFYBJwJvAw43czOAy4C9jnnTnHOnQzcPei1bc651XinSP/H3/Z14PvOuTXAj4CvDSq/EDgXuBj4nL/tMuAe59zLgFOAjf6VPZ8EXuWcOw3YAPy/bA5G4SsiIrPBi865x4BX+19P4a0zsQovmDcDF5rZ583sr5xzbYNee9ug72f7j88GbvUf/wAvbPv90jmXds49A8z3tz0BXGlm1wOrnXMdwFnAScCfzGwj8B5gWTYHk83aziIiItOty/9uwGedc/83vICZnQa8DviMmd3nnPu0v2vwFTjZXI3TN7ha8JZa9nvYrwe+Z2ZfAQ4D92aamDwa9XxFRGQ2uQd4b/+5VTNbZGbzzOwYoNs590Pgi8Bpg17z9kHfH/Uf/xm41H/8TuDh8d7UzJYBjc65bwPf8et/DDhn0HnkMjM7IZuDUM9XRERmDefc783sJcCj/vpOncC7gJXAF80sDSSAawa9rMbMNuH1aPt7qR8Gvmtm/wQ0AVdmeOvzgX8ys4T/npc755rM7ArgNjOL+eU+CTyf6ThsutbDWLt2rduwYcO0vPdw62//8ohtO5e+DYDL1i0NujkiIoXIpqRSsxeAtc655qmof6I07CwiIhIwDTuLiEjecs4tn+42jEY9XxERkYApfEVERAKm8BUREQmYwldERCRgCl8REZEsmNlF/g0edpjZdUdTl2Y7i4jIrLP8ut9ePZn1vfC519843n4zCwPfAC4E6oEnzOxOf/3nnKnnKyIiktmZwA7n3C7nXBz4MXDJRCtTz3cqbfju0OdrM61eJiIiM9QiYO+g5/XAuolWpp6viIhIwBS+IiIimTUASwY9X+xvmxCFr4iISGZPAMeb2QozK8K7HeGdE61M53xFREQycM4lzexDePcTDgM3O+e2TrQ+ha+IiMw6mS4NmgrOubuAuyajLg07p5IUJdqmuxUiIlJAFL4bf8Sa7d8glIpPd0tERKRAKHwP7SDskhQlO6a7JSIiUiAUvp0HAYgku6a5ISIiUigUvp2NAERTCl8REQmGwtfv+UaTndPcEBERKRQK3/6er4adRURkHGZ2s5kdNLMtR1tXYV/nm0pA9yFA4SsiMqtcXzWptxTk+rZsrhv+HnADcMvRvl1h93y7mgEHKHxFRGR8zrmHgJbJqKuww9cfcgZNuBIRkeAUePh6k636olW61EhERAJT4OHr9Xy7i+dr2FlERAKj8AW6Y/OJpPuwdHKaGyQiIoWgwMP3IMSq6CuqAjTpSkRExmZmtwGPAieaWb2ZXTXRugr7UqPORiifRyJSBkA01UmcqmlulIiIZJTdpUGTyjn3jsmqq8DD9yCUzycR8sNXPV8REQlAgQ879/d8ywGFr4iIBKPAw9fr+Sb9YWddbiQiIkEo3PCNd0G8A8rnkQ5FSYWK1PMVEZFAFG74+gtsUD4fgESkTKtciYhIIAo3fLuavO/94RsuU89XREQCUbjh69/NiNI5gN/zVfiKiEgACjd8E93e9yJvslUyUkZEw84iIhKAAg7fHu97tASAVChKOJ2YxgaJiEihUPhGSwFIW5RQOgHOTWOjRESkEBRw+PrDzn7PNx2KYjjMpaexUSIiUggKOHz9nm/kSPgChJyGnkVEZGoVcPh2Q6QYQt4/Qcr88NV5XxERmWIFHL49A0POMKjnq/AVEZEpVsDh2z0w2Qo07CwiIsEp4PAd1vP1h511uZGIiEw1ha9Pw84iIhKUAg5fDTuLiMj0KODwHX3YWT1fERGZahnD18xuNrODZrZljP1mZl8zsx1mtsnMTpv8Zk6BET3fCKDwFRGRqZdNz/d7wEXj7H8tcLz/dTXwzaNvVgCG9XxTGnYWEZGAZAxf59xDQMs4RS4BbnGex4BqM1s4WQ2cMmMOOyenq0UiIlIgJuOc7yJg76Dn9f62EczsajPbYGYbmpqaJuGtj8JYE6407CwiIlMs0AlXzrkbnXNrnXNr6+rqgnzrkYb1fLEQaQsT1rCziIhMsckI3wZgyaDni/1tM1c6DcneIT1fGHRbQRERkSk0GeF7J3C5P+v5LKDNObd/EuqdOsn+e/mWDNmcDil8RURk6kUyFTCz24DzgVozqwf+DYgCOOe+BdwFvA7YAXQDV05VYyfNsNsJ9kuHoprtLCIiUy5j+Drn3pFhvwM+OGktCkKi2/s+rOeb0rCziIgEoDBXuEqMNewcUfiKiMiUK9Dw7e/5DptwpWFnEREJQIGG7xg9X4sS1iIbIiIyxQo8fEfp+WrYWUREpljGCVd5acxzvkc/7Hzr+j0Dj4/bM3RVznVrj6pqERHJE+r5DqJFNkREJAgFGr5jXGqkYWcREQlAgYbv2JcahV0CnJuGRomISKEo0PAd41Ij/7aC5jTjWUREpk6Bhm8PYBCJDdncf1vBsIaeRURkChVo+Pr38jUbsnngnr5aaENERKZQgYZvz4jzvXBk2DmkhTZERGQKFXD4lo7YPNDz1bCziIhMoQIN3+7Re74adhYRkQAUaPiOPuycMvV8RURk6hVo+HZr2FlERKZNgYbvGBOuNOwsIiIBUPgO0j/bWdf5iojIVCrQ8NWws4iITJ8CDV8NO4uIyPQp4PAdpedr3u2N1fMVEZGpVKDh2w3R4pHbzUhZROErIiJTqvDCN5WAdGLUni94Q88h3dVIRESmUOGF7xj38u2Xtqh6viIiMqUUvsOkQ1FdaiQiIlOqAMO32/s+7rCzwldERKZOAYZvpmFnTbgSEZGpVcDhO07PV+ErIiJTqADDt3/Yeexzvhp2FhGRqVSA4Tt+zzdlUUJpXWokIiJTpwDDN4uer4adRURkChVe+CZ7ve8adhYRkWlSeOGb6VIj03W+IiIytQowfDMvshFySXDpABslIiKFpADDN/MiGwDhVG9QLRIRkQJTgOHbA6EIhKOj7k6btz2i8BURkSlSmOHr93oTqTRfvOdZDsUjA7tT/T3ftMJXRESmRgGGb/fA+d6n9rTyjQd28v298wZ29w87q+crIiJTpQDDt2cgfDfVtwKwvrWS7V3FgLe2M0A41TM97RMRkbxXgOHbPTDsvLmhjdryGFWRJLfWz8M59XxFRGTqFWD4Hun5bq5v47Sl1bxlYTPPdJbyVHuZwldERKZcgYZvKe29CXY1d7FmcRWvrGtlQSzOrfV1JE2XGomIyNQqwPD1JlxtaWgDYPXiaiIGb1xwiL29xTTEywCI6JyviIhMkQIM3x6IFLO53g/fRVUALCv1eroHEt6QtC41EhGRqVKA4etNuNrc0MbimhLmlBUBsCDmree8b6Dnq/AVEZGpUYDh60242tzQNtDrBSgNp6kIJ6n3w1fnfEVEZKoUZPj2WYwXD3WzenHVkF3zYwka+kpx6JyviIhMncIKX+cg0U1jj3fYaxZVD9k9PxanMV5EOhQlnO6bjhaKiEgBKKzwTcXBpdnXbQBDhp3B6/k2xaOkLHpUw86xvhYiye6jaqqIiOSvSOYiecS/neDeDseyuaVUlQ69s9GC4jgOI2lFEx52Loq38sYHXkXYJeiN1rBt+eXEi6oyv1BERApGYfV8E16g7m5znLxoZCDO92c891E04Z5vRfcewi7BvrpzKU4cpqx338TbKyIieanAer5e+O7vNo6rKx+xe34sDkAPRRO+1Ki05wAA25ZfzjFNj1CUaJ9gY8ew4btDn6+9cnLrFxGRKVdgPV9v2LnbFXFMVfGI3dWRFLFQmm4Xm3DPt7TXC9/WylWkLTz54SsiIrNegYWv1/PtJcbC6pIRu81gXlGcjnRswj3fst4DJEPF9EWriUcqKUp2HFWTRUQk/xRY+Ho9354xer4AC4oTdLhiwumJTbgq6W2ku2QBmBGPVqrnKyIiIxRY+HqB2kOMBWOE7/xYnNZU8VH1fLuL5wMofEVEZFQFFr5ezzdUVEpFcXTUIvNjCbpcjFByYj3f0p4DdBUvACAeraAo2e4t7iEiIuLLKnzN7CIze87MdpjZdaPsv8LMmsxso//1vslv6iTwe76VlRVjFpkfi9M70UuNUkmK+5rp7g/fSCUhlyaS6ppQc0VEJD9lvNTIzMLAN4ALgXrgCTO70zn3zLCiP3HOfWgK2jh5/PCtqhx70Yv5sQQNTHDCVcd+QqS9c75APOq9T1GinWRk5KVNIiJSmLLp+Z4J7HDO7XLOxYEfA5dMbbOmiB++c6vHDt/aogS9rogISSydyK3+9gaAgZ5vX9TrYccSmvEsIiJHZBO+i4C9g57X+9uGe4uZbTKzO8xsyWgVmdnVZrbBzDY0NTVNoLlHJ9nnDf/W1lSPWSZiEI6EAQincry5Qls9cCR849FKAE26EhGRISZrwtWvgeXOuTXAvcD3RyvknLvRObfWObe2rq5ukt46e12dHfS5KPOry8YtV+SHbySd49Cz3/Ptn3CVDJeRthBFybbcGysiInkrm/BtAAb3ZBf72wY45w455/q7id8BTp+c5k2u7q5OeijimKqRC2wMFov293xznPHc1kAiXEYy6p/fNfMW2tCws4iIDJJN+D4BHG9mK8ysCLgUuHNwATNbOOjpG4Btk9fEydPX00kPMRZWj36Nb79iv+eb6svxtoDtDXT5k6366VpfEREZLuNsZ+dc0sw+BNwDhIGbnXNbzezTwAbn3J3AP5jZG4Ak0AJcMYVtnrB4bydpl7nnWxr1/ibp7erM7Q3aGwbO9w68Z7SS8p6GMV4gIiKFKKu7Gjnn7gLuGrbtU4Me/zPwz5PbtMmX7O3GhWKUFIXHLVdaZAD09uQYvm0NdNf81ZBN8UiF1/PVQhsiIuIrqBWuXLybdHj8IWeAiiLvn6WvJ4fFMdIp6GqiJ1Y7ZHM8WknIpYjkev5YRETyVkGFL4keiJZmLNY/7BzvzaHn29MKOPqKaoa+ZcSbWR1NapUrERHxFFT4hlI9WNH453sBXNhb9zmnCVfdhwBGhG8y7IV9JJXj5C0REclbWZ3zzQc98RRF6V7SsfGv8QVIW3/45tBb7W4GoHd4+EYUviIiMlTBhO/+th6KLU6yOIvwDXnhm44ffc834fd8o0mFr4iIeApm2Hl/Wy8l9BEryXyDg/6ebyjZS28ild0bdHk9376ot3RlPJkm6TTsLCIiIxVM+O5r7aGEOCVlmcPXWYgUYUqsj32tWc5SHtTz7U2k+Nr927nxxQW4UIRUqIiIer4iIuIrmPBtbO2ixOKUlVdmLmxGMhSjhDgNWYdvCxSVkw7HuHvLAVq64vyppZKOZIhEuIyoer4iIuIrmPA91OL1TCOlY9/RaLBUuIQSemk4nG34NkPpHHYc7OTxF1pYtaCCpAvxSEsVyUiJhp1FRGRAwYRvR5sXvsQqsiqfjJZTYT3UZx2+h0iVzOXnT9VTW17EO85cynGlPdzXVEUiXKoJVyIiMqBgwrezrcV7EMti2BlIRMupCffmMOx8iBe6i2nrTvCW0xYTDYe4oLaVvb3FtLhK9XxFRGRAQYSvc47udj98i7ML33ikgupQT/bDzl2H2NVdwsp55Syb613O9PI5HcRCabbHa4hohSsREfEVxHW+7b1JL/yKgFhVVq9JRCqosoase76u+xAv9JzM8iVHriMuDac5u6adLW1zuSiShHg3FGVe3nKwW9fvGfL8uD3eHxHrVszJqR4REZk5CqLnu6+1hwr8Yd8se76JaDnldHOgvZdkKp2hcA+W6OKwq2DZ3KHhekFtGwedH/j+5UgiIlLYCid8ze/BZnvON1JOabqbVNqxv613/MJ+qLZaJYurh4bvCWU9dJvfG/aXoBQRkcJWMOFbmWPPNx6pIOZ6iJDMPPTsh2/F3PkURYb+k5pBcSw2pJyIiBS2ggjfhtZeqkM9uFAUIpnv5wuQiHqXJJWTedJVvL0JgEXHLB51f2VJEQDJjqZsmywiInmsIMJ3X2sP82NxrLjS64pmIR7xlqGssO6MPd/6hr0AHLts2aj7a8u8eW3NB/dn22QREcljBRO+tdHerM/3wpGe79KyZMae7759DQCcdNyKUfcvLDOSLkRLk8JXREQKKHxrwr1Zn+8F75wvwMqKFLubx79Gt6VpPylCzJk7f9T9C4qTtFJO1+HG7BstIiJ5K+/DN5lKc6C9lyrrya3n6w87r6p2bN3XRirtRi2XTjt62xrpCVdCaPR/zpBBl5WT7NQ5XxERKYDwbezoI+2gzHVBcXYLbADE/WHnlVVpuuIpdjd3jlpuR1MnZak20iXjL3rRFy4l0tuS+ZphERHJe3kfvv334y1Od+XY8/XCd1lZAoBN9W2jlnt8dwtzrYNo5bxx63OREqpdB7syDGGLiEj+K5jwjSY7czrnm4h4C2PURvooiYbHDN8/7WhmXriT4qq6ceuLFhUzx9rZ0jB6PSIiUjjyPnwbWnsw0oTinTn1fF0oSjJcQijewcmLKtk8Smim0o4/7zzEglArVr5g3PqisRJq6GRL/cQX2jjm4EMU92mhDhGR2S7vw3dfaw+LS5IYLqeeL/gznntbWb2omq372kacr93c0EZPTzelqQ6oGH2mc79ktJyQORr27hm33FhKew5w3pP/wIp9v5nQ60VEZOYogPDt5bhKPzRz6PmCP+O5t501i6voTaTZ0TR00tUj25uoo9V7Uj5++PbPnm5p3Et6jJnT4zl+z22EXIrK7hcp6T2Y8+tFRGTmKIDw7WF5Rcp7kmPPNxEth752Vi/2ZkkPP+/78PZm1s3zJmSRYdh5YMWs5KGcJ12FUz2s3HsHB+auI20R5rc8kdPrRURkZsn78G043MPS0qT3JMeerzfs3M6KuWVUxCJsHhS+XX1JntxzmJfP94O9fPzZzv2zp+dZK5sbWnNqx4qGXxNLtLN55d9zqOql1LZtgkR29xkWEZGZJ6/Dt703QUdfkoXFcW9DDtf5gr/EZG8boZBx8qIqNtUfCc3Hd7eQSDleVt3nbagYv+fbP3v6mHA7T+/NbcbzioY7aalYRVPNqTTOOYNwOgH7NuZUh4iIzBx5Hb79lxnNj/nhm3PP1xt2BlizuIpt+zuIJ73zxw9vbyYWCbE81gEYlNaOW5cLRaCkhlXl3Wzcm33P11yKmvbnODhnLZjRVbyQZLgY2iY2cUtERKZfQYRvbaTX25DrOV9/2Blg9eIq4qk0zzd2APDIjibOXDGHSPdBKKuDcCRzheULWBbr4Jl97QMhnvElXXuIpHtprTzB22BGd2w+tDfkdCwiIjJzZJEYs1dDqxe6NWE/fHPt+UYrINUHiV7WLKoG4Lqfb6K2PMbzjZ285bTF0HAw40znARXzmdd6mHgqzbMH2lmzuDrjS2o6ngPgcMWqgW1dxQuobH0SnJaqFBGZjfK657u3pZuiSIgy1w2hCERLcnp9/+VB9LWzZE4JF69ZiGEc6oxz5vI5vH7NQug8kPEa3wHlC6hIeItkPJ3l0HN1x/OkLUJb+XED27qL50M6AV26UYOIyGyU1z3fZw90sLKunFBfu9frNcvp9QPh29uOlc/jhstOG1mo8yDMOym7CsvnEe4+SG1ZERv3tvHuszO/pKb9OdrLlpMOFw1s6y72w759X3bvKyIiM0peh+9zB9o5Z2WtN2kqx/O9APGo/5reMWYnp9PQ2ZjDsPMCLBXnnKUhnq7Prudb0/4sjXPPGLKtJ1YHFpq88N3w3aHP1145OfWKiMio8jZ8D3fFaWzvY9WCCtjbnvP5Xhg87DxG+PYchnQy+/D1y51Rm+TO7b209yaoLI6OXb67hdK+g7RWnDhkswtFvOuKJ2nS1frdLUOe70x5M6kvW7d0UuoXEZGh8vac77MHvFnJJy6o9Hu+uV3jC0OHnUfVecD7nvU5X6/c6upenIMtY9wpacCBzQAcHha+AFQu0rCziMgslbfh+9wBLzBfssC/XGgCPd+BYee+scK30fuew7AzwMoSb43ojZmGnhu3ABy5zGiwykXQ2wrdLSP3iYjIjJa/4dvYQU1plLqKGHTsz7j842ji/pKQ9BwevUBHjuHrlyuLN7N8bmnmGc8HttATq6UzMq+UPMIAABcUSURBVIet+9rYVN/K+sPl7OuNQuUxXhk/oEVEZPbI23O+2/Z3cOKCCizeBT0tUJ37+ctkpMzrMbfVj14g155vrAKipdB5kFOWVPPYrkM457CxZmE3buZwxQnc8WT9oJs6LCZqaVafF2IlwIEtsOK8HI5KRESmW172fNNpx/ONHaxaUAlte72NEwhfzKB6GRx+cfT9nY0QLYNYefb1lc+DjgO8/Li5NLb38eSeMXq/yTg0PceW1FI21bdxwap5fOSVx/OZVS9QGk7zoY2LcUUV6vmKiMxCeRm+9Yd76I6nvJnOrf4ayBMJX4CaZXD4hdH3te/LeEOFEcoXQGcjF685hvJYhB+tHyPYm5+HVJy7Ds7lpIWVvHLVPOZXFnN8WS/XLN/Ps21RdoWWDkzKEhGR2SMvw/dZf7LViZMSvsuh9UVwbuS+5ueh9vjc6qteCi27KYtFeNOpi/jNpv20dsdHFGt/8SkA9hWv5K2nLx4yNH1qVRfvPq6beztXkD74LKQSubVBRESmVZ6Gr3eZ0Qnz/fANF0FZ7hOuAC98k71Hzu/2SyWgeTvUrRr1ZWOatwra66G3ncvWLSWeTHPHX4aeU3bO8adH/kifi/LydWdRHA2PqOYTqzs5FFtMKB0nefC5HA9KRESmU15OuHruQAfL5pZSFot44Vu1BEIT/DujZrn3/fALQ4eYW3Z56yvPe0lu9dX55Zue4yVLzuD0ZTXcun4PV527YqB3e8ujL3Js67N01hzPvKrRzyeXROAVJ9TCNvjL+odZ98aTc2vHFLl1/di3OtSiHSIinjzt+bZz4nz/MqG2vRMfcgZvwhWMnHR1cJv3fSI9X4Am7/XvXLeUXc1dPLrLu+HC9sYO/uuuZzglWs+cY0dZS3qQs5fXkCDCri2PkkjpDkciIrNF3oVvbyLF7uYub7IVeD3f6iUTr7A/uIdPump6FjCoHWUBjHHrWw6RkoHwft3qhVSXRvnwrU/xyi//kbd8888sL+qkMt2KLVg9blUWDtNbcwKL+naNGLoWEZGZK++GnXcc7CTt/GUl493ebfeOpucbLYaKY7xJV4Md3ObNhC4qza2+UAjqThgI3+JomE9dfBK/23KAokiIUxaHuGZxN9wLLDgZGsevrnzZy1jTehf/fP8O3nzaImKRkeeHx+QcdB4kkuwkGS71btYwSY7bc/vIjes+Omn1i4jMZnkXvg8+793j9mVLqwdd47vs6Cod7XKjpueOnL/NVd1LYPeDA0/ffNpi3nza4iP7H7zX+z7/pdDYMW5VtmAN1RtvJdG6jx8+5p07zopzNN/yHmp3/4rTgebYEnaueDeE8+5HQkRkxsmrYWfnHD9/sp4zl89hUXXJkcuMqo5i2Bm8SVeDwzeVgEM7jpy/zdW8Vd6Slz1jLLCx835YeAqU1GSua6l3U+CrFr3Il+55jheauzK+5M87m/nelz9O7e5f8WP3av478WZq+/ayf9uf2NFVnMuRiIjIBORVN2dLQzs7m7q46txjvQ39Q8VHM+wMXs+5fR8k+yASg0M7vZnOR9PzBe+88dKzhu7rbYO96+Hca7Ora8EaKKvj8trtfOPQWj52+9P85P1nEw6NXLLSOcdNj+zmt7+7k9uLvsOL8y7gkqtu46Gf3cCf6tt4Y/d9/NOOl1BZ28lxdVmu2jWFNHNaRPJVXvV8f/FUA0XhEK9fvdDb0LYXQtHcV6EarmY54KDVH8b2ZyofVc8XjsyYHmzXg+BSsPJV2dUVCsHKV1Gy50Gu/5tVbHjxMDc9smtEsWQqzSd/uYX//O1W/rviVkIVC1j23u9REotQHU0RWX42h4uXcF3kVu748za27R/jTk4iInLU8qbnm0ylufPpfVywah5Vpf4N6lt2Q9UiCOUwCWk0/df6HtgEtSvhxT+DhWFujqtb9ata6t1gYbTw3fEH72YOi8/Ivr6Vr4Knb+NN85u4+6T5fOme5zGMvz1jCZXFEf74XBNfvvc5tjS089WX7mL5zufgtd8aeo9jC9Gw8EJO3n0zHym9m8+sL+bUpdVDz0Vn0NWXZHtjB4e64vxp/zyqoknOrulgXmzyVuAamMgVnuN9X3vlpNUtIhKUvAnfR3Y009zZxxtPXeRtSPTCzgfgxNcefeWL10LNCnjoS14o/uX7cMo7cp/p3C8UgmUvh22/htf815FJTs7Bjvvg2L+GcDT7+o59BWDYjvv4rzdfywd/9CT/edc2vnLv8yybW8qzBzpYXFPCDX/7Ui5+6DpYsBrWvH1ENV2lizlUeRLv7ryT3825iP/306dp70lwxTnjT+Lq7Etyy6Mv8O2HdlHVs4e3hR9kdWg3v0+t5V8aXs6CMuMNC1o4c7w7OE2jsYa3NbQtIlMlb4adf/lUA5XFEV6xqs7bsP330NcGa9529JWHo/DKf4WDW+EHbwSXhr/++NHVufYq6NgHz911ZNvBbd7Sk8e9Mre6yubCotNgxx+oLY/xk/efzW8+fC4Xr1lIOGT815tWc/9Hz+fittu8SWgX/seYK37tnf9KQi7FF8p/zIUnzef6Xz/Df/72Gdp7R/ZeO/uSfOOBHZz7+fv5wt3Pcl3lvTwQ+xjXRH7DmbF6PhP9Lo+WXMvy5G6+tHMx7/v+Bva2dOd2bEDaOQ519tHSFedwIkw8PfMCXEQkF3nR832huYt7tjbyxlMHXee6+XYoq4MV50/Om5z0Jjjm67DvKTjj77zLj47GCa/xZmE/8W046Q1er/feT3kLcJxwUe71nfg6uP8/YM96WLqOkxdV8cW3nXJk//5N8PCXYM2lcNwrxqymr6iGLSs/wCnbv8633nIZnyxfybcf3s1PN9Rz9XnHsnpRFYe74+w82Mktj71Ia3eCC06Yy+cqfsq8rTexZ8GFbHjJP7Oo8QHKevexsv4XfDPxeW6tu4LP7rqQC//7Qa48ZwXvP+9YqkuLxmzHwfZeHt3ZzI6mLnY3d1KZOESFdZOgkgbqOGdfgguPifOaE3uZV6EZ2iIyu8z68N3b0s1l336MkqIwV5/nz3LubYPn74HTr5i861ZDIbjo83DPJ+C8j01CfWHvfOV9n4a9j3szn3fcC6/9AlQuzL2+dR+ADd+F3/wjvP/BocPWfZ3wy7+H0rlw0WczVvXMse9lceN9zP3dx/jsBx7hneuW8pV7n+eL9wy9gcMrTqzj2lcs45QnroOtP4d11/BIzQe8xTrM6CpZxNYVV3Lintt4V8dN/M2rX8K/1Z/Otx7cyQ8ffZF3n72MM5bPYdXCCoojYZ490MGzB9q595lGHtt1CHMpLitZz1eKH+Al4a0D73swVMevOs7nC0+9huufvp8LVs3j0jOW8Ncn1BEJjz2Y45yj/nAPT7zQwq6mLva39dLY3suhzj4qSqJUlURZUlPK0jmlFEUmNiikIWwRyUZWyWRmFwFfBcLAd5xznxu2PwbcApwOHALe7px7YXKbOtKBtl7e+Z31dPYlue3qs1hRW+bt2PZrSPXBmr+d3Ddcug7+7r7Jq+/Uy+Hhr8BNF3rPl53r9aonIlYOr/0c/ORd8PCX4byPe38wtNXDrZfCwWfg0luhdE7GqlwowmNrPsPr178LbrqQky/7KTdfcQbPHminozdJTWkRdeUxqjp3wm/eA3sehQs/DS//B3h875C6kpEyti27nOPr76D6Dx/lq+uu4YPXfJgv/XE///vHncDOEe+/ck6Ub5y8nXV7b2JO7x7aipazccVH6CpZxDEH/8jctq38XfftvLf8tzy88Ar+/cV1XPVMI+WxCGcdO4ezjp1LbXmMokiIRCrNjoOdbNvfweaGVhrb+wAIGcyvLGZ+ZTGHuxO82NJNdzwFQNiMxXNK2N/Ww1nHzuWUJdWUx0b/r5JOO5q7+jjQ1su+1l6e2N1C0jnSaUc4ZJTHIpTHItQf7mZhVcmol4CJSOHJGL5mFga+AVwI1ANPmNmdzrlnBhW7CjjsnFtpZpcCnwdGzuiZRL2JFO/8zmO0dMX54fvW8dJjBs3c7TnsXf+66PSpbMLRK6+Da/7szXBu3Arn/uPE774EsOpi7+uPn4XNd3g96PoN3szsy34Kx2d5+RLQVnE8XHkX3Pp2+M6r4OQ3s+r4V3sh37DfW6Fr8x3e87fcBKvfOmZd6XARzy99O2eWHYT13+KELT/jxlPfRecZ57A7XsVzh1JYz2FWxZpZ0bWRku2/xrY30lq+kodO+x/q510A/kStSLKTg3PWsm5OF+Fn7+L8vf/LX0e/x/4TL+TP6ZO5e/9c/ndbKX1ECeEIkaYklOT0mh7eX9vBKcvbOTbSRFXPHkId+6Gnjb50D6mSYnoqKjlAHU9HVrO1p5oHHqzhh/fX0EcR5aWlLKwpp7QoAuk+SMbp7Owg2dlMlWtnLu3MsXaWWzsVeOe140Rod6W0U8YX/lRGZ6iM4vI5lFXXUjOnjpq58ygrLaGsKEI0bMTjfSQSCfriCVKJHra92Eg42UMo1UMk1UuUBAmLkQgXs/KYOqLFZRSVVlBSXEJJcTHlpcWUFRdRHgtjZoNuQe1wDtIOkuk0qbQjkXKk0o5k2rshR8iMcMgI2eDH/lfI+4PEhpVJO0cimSaZjJNKJkkmk6RSSZLJFOlUHJdOEw6FCIXDhMIRQuEw4XCEUChMOBIhmuyiuH03EdIkFq8jEgoRCh2pO532zvenXX/7HWnnjWAMtDFkhM1vU//j0Mhj6d8uk6c3kaKtJ0FbT4I5ZUXUlsemu0mzSjY93zOBHc65XQBm9mPgEmBw+F4CXO8/vgO4wczMudHuQD85iqNhrjr3WFbOK+dlS6qH7nz5h+HsDw38wp7RapbBGVdNTl1m8LbvwzO/hMdv9IbfT30XnPE+qDsx9/qOORXedx/c/xnY+gt46gdH9pXUeMP6r/gElNVmrMpZGF7/JXjZZfCH6+FPX6XcfYXVwJDbR0RKvNneZ/wddx1aOfZ601VLYN37YfFabP3/ccyzv+WtPb/irQCjnQLu8r/Au8Rq7vHe7SCLq2jbtYlIqofiRDsnxut5afpRr1zU/wJIQ/qQEWLYj/SwSekOIxEuxVySkEsRdsmhBXqBA/4XEHdhwqQJW47/VZpG35x2RoIwaUI4DPPbe+T70NaOtm+814RybWcWNqaP443x/5j0ejPp//VgQ7YdeWajlPW2T87vFTf8Z2kidRxlFeO9fLxf327Ye//rxSdlv7StAGCZ8tHM3gpc5Jx7n//83cA659yHBpXZ4pep95/v9Ms0D6vrauBq/+mJwETuAl8LNGcsNbvomGaPfDwuHdPskO/H1Oycm8Bs09kp0AlXzrkbgRuPpg4z2+CcWztJTZoRdEyzRz4el45pdtAx5ZdsTjA2AIPvTLDY3zZqGTOLAFV4E69ERERkmGzC9wngeDNbYWZFwKXAncPK3Am8x3/8VuD+qTzfKyIiMptlHHZ2ziXN7EPAPXiXGt3snNtqZp8GNjjn7gRuAn5gZjuAFryAnipHNWw9Q+mYZo98PC4d0+ygY8ojGSdciYiIyOTKm7WdRUREZguFr4iISMBmbPia2UVm9pyZ7TCz60bZHzOzn/j715vZ8uBbmT0zW2JmD5jZM2a21cw+MkqZ882szcw2+l+fmo625sLMXjCzzX57N4yy38zsa/7ntMnMTpuOdmbLzE4c9O+/0czazezaYWVmxedkZjeb2UH/Ovz+bXPM7F4z2+5/rxnjte/xy2w3s/eMVmY6jHFMXzSzZ/2fr1+YWfUYrx33Z3W6jHFM15tZw6CfsdeN8dpxf09OlzGO6SeDjucFM9s4xmtn5Oc06ZxzM+4Lb2LXTuBYoAh4GjhpWJm/B77lP74U+Ml0tzvDMS0ETvMfVwDPj3JM5wO/me625nhcLwC14+x/HfA7vAWDzgLWT3ebczi2MN5aVMtm4+cEnAecBmwZtO0LwHX+4+uAz4/yujnALv97jf+4ZrqPZ5xjejUQ8R9/frRj8veN+7M6w47peuBjGV6X8ffkTDqmYfu/DHxqNn1Ok/01U3u+A0taOufiQP+SloNdAnzff3wH8EqbiXdq9znn9jvnnvQfdwDbgEXT26pAXALc4jyPAdVmNoHbNk2LVwI7nXMvTndDJsI59xDe1QeDDf5/833gjaO89DXAvc65FufcYeBeYEasPDTaMTnnfu/cwDqej+GtRTBrjPE5ZSOb35PTYrxj8n9P/y1wW6CNmmFmavguAgbfHqeekUE1UMb/j9cGzA2kdUfJHyI/FVg/yu6zzexpM/udmb000IZNjAN+b2Z/8ZcPHS6bz3KmupSxf0HMts+p33zn3H7/8QFg/ihlZvNn9l68kZbRZPpZnWk+5A+l3zzG6YHZ+jn9FdDonNs+xv7Z9jlNyEwN37xlZuXAz4BrnXPtw3Y/iTfEeQrwdeCXQbdvAs51zp0GvBb4oJmdN90Nmgz+gjJvAG4fZfds/JxGcN4YX95ca2hm/wIkgR+NUWQ2/ax+EzgOeBmwH2+YNl+8g/F7vbPpc5qwmRq+ebmkpZlF8YL3R865nw/f75xrd851+o/vAqJmlvmWQdPIOdfgfz8I/AJvKGywbD7Lmei1wJPOucbhO2bj5zRIY/+wv//94ChlZt1nZmZXABcD7/T/qBghi5/VGcM51+icSznn0sC3Gb2ts/FzigBvBn4yVpnZ9DkdjZkavnm3pKV/nuMmYJtz7itjlFnQf97azM7E+3xm7B8UZlZmZhX9j/EmvmwZVuxO4HJ/1vNZQNugYc+ZbMy/zmfb5zTM4P837wF+NUqZe4BXm1mNP9z5an/bjGRmFwEfB97gnOseo0w2P6szxrB5EW9i9LZm83typnkV8Kzz74A33Gz7nI7KdM/4GusLb5bs83iz+f7F3/ZpvP9g4N219XZgB/A4cOx0tznD8ZyLN8S3Cdjof70O+ADwAb/Mh4CteLMWHwNePt3tznBMx/ptfdpvd//nNPiYDPiG/zluBtZOd7uzOK4yvDCtGrRt1n1OeH887AcSeOcDr8KbF3EfsB34AzDHL7sW+M6g177X/7+1A7hyuo8lwzHtwDv32f//qv8qiGOAu8b7WZ0JX2Mc0w/8/y+b8AJ14fBj8p+P+D05E75GOyZ/+/f6/x8NKjsrPqfJ/tLykiIiIgGbqcPOIiIieUvhKyIiEjCFr4iISMAUviIiIgFT+IqIiARM4SsyC5jZG2bSXWtE5OjoUiMREZGAqecrkgUzu9xf5P5pM/uBmf2NefeRfsrM/mBm8/1y15vZ983sYTN70czebGZf8O9Pere/xGj/PUv7tz9uZiv97WPVe4WZ3eA/Ps7MHvNf+xkz6/S3n29mfzSzO8y7v+2PZvKdvkQKmcJXJAP/rkWfBC5w3s0UPgI8ApzlnDsV71ZuHx/0kuOAC/BuzPBD4AHn3GqgB3j9oHJt/vYbgP/xt41Xb7+vAl/1Xzt8mb5TgWuBk/BWCzpnQgctIlMqMt0NEJkFLgBud841AzjnWsxsNfATfw3eImD3oPK/c84lzGwz3g3P7/a3bwaWDyp326Dv/+0/XjxOvf3O5sh9eG8FvjRo3+POXzfXzDb67/dITkcrIlNOPV+Rifk6cIPf+3w/3lrj/foAnHdHmoQ7MrEizdA/eN0oj8erNxt9gx6n0B/YIjOSwlcks/uBt5nZXAAzm4N3C8v+27e9Z6wXZvD2Qd8f9R9nU+9jwFv8x5dO8L1FZBrpr2KRDJxzW83sP4EHzSwFPAVcD9xuZofxwnnFBKquMbNNeL3Vd/jbsqn3WuCH/s3j7wbaJvDeIjKNdKmRyDQwsxfwbq/YPIHXlgI9zjlnZpcC73DOXTLZbRSRqaOer8jsczpwg38ZUSvevXdFZBZRz1dERCRgmnAlIiISMIWviIhIwBS+IiIiAVP4ioiIBEzhKyIiErD/D/nb4DJGqCjmAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 485.25x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 485.25x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 485.25x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z0-mGC1IOoW4",
        "colab_type": "text"
      },
      "source": [
        "## Box Plots"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dtTJQdRbOnZ5",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "outputId": "21699bb5-34f9-4e4c-9644-94ec55c847a0"
      },
      "source": [
        "fig, axes = plt.subplots(nrows=4, ncols=2, figsize=(20, 25))\n",
        "\n",
        "for ax, col in list(zip(axes.flat, numerical_cols)):   \n",
        "    if col != 'response':    \n",
        "        sns.boxplot(col, data=df, ax=ax)"
      ],
      "execution_count": 217,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x1800 with 8 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p7D49d__BWEC",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "outputId": "2a3df7ec-a86b-4ebc-bca6-ea35ad505fea"
      },
      "source": [
        "fig, axes = plt.subplots(nrows=5, ncols=2, figsize=(20, 25))\n",
        "\n",
        "for ax, col in list(zip(axes.flat, categorical_cols)):        \n",
        "    sns.countplot(col, data=df,hue=df['response'], ax=ax)"
      ],
      "execution_count": 218,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABJgAAAV8CAYAAAB39iWQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdfbhddXnn//cHiEQNAkKkSMAgoBV8CBCepFhHKkGmFexgABUCqDgjaPnZMkAvBUGcsdLK1IcyxTGQVB6kD44MIhKpCFgQkhgIEDCpBkkMEB4EwRIl3L8/9jpxE04eyD777Jx93q/r2tde67u+a+17nXP2Ofe593d9V6oKSZIkSZIkaUNt0usAJEmSJEmSNLJZYJIkSZIkSVJHLDBJkiRJkiSpIxaYJEmSJEmS1BELTJIkSZIkSeqIBSZJkiRJkiR1ZLNeB9AN2267bU2cOLHXYUiSpC6aM2fOI1U1vtdx6HfMwSRJ6m9ry7/6ssA0ceJEZs+e3eswJElSFyW5v9cxbIySjAVuBDanlev9U1WdneQS4A+BJ5qux1fVvCQB/hY4DPh10z63OdY04JNN//OqasbaXtscTJKk/ra2/KsvC0ySJEmj2ArgHVX1VJIxwM1JvtNsO62q/mm1/u8Cdmse+wEXAvsleSVwNjAZKGBOkquq6vFhOQtJkjSiOAeTJElSH6mWp5rVMc2j1rLL4cDMZr9bga2SbA9MAWZV1WNNUWkWcGg3Y5ckSSOXBSZJkqQ+k2TTJPOAh2kViX7UbPpskjuTXJBk86ZtB+CBtt2XNG1ral/9tU5KMjvJ7OXLlw/5uUiSpJHBS+QkSeojv/3tb1myZAnPPPNMr0MZMmPHjmXChAmMGTOm16GMGFW1EpiUZCvgm0neCJwJPAi8BLgIOB04dwhe66LmeEyePHltI6UkSepb/ZaDbUj+ZYFJkqQ+smTJErbYYgsmTpxIa+7mka2qePTRR1myZAk777xzr8MZcarql0m+DxxaVX/dNK9IcjHwF836UmDHtt0mNG1Lgbev1n5DVwOWJGmE6qccbEPzLy+RkySpjzzzzDNss802Iz6xGZCEbbbZpm8+DRwOScY3I5dI8lLgncC9zbxKNHeNOwK4q9nlKuC4tOwPPFFVy4DvAock2TrJ1sAhTZskSVpNP+VgG5p/OYJJkqQ+0w+JTbt+O59hsD0wI8mmtD5MvLKqrk7yr0nGAwHmAf+16X8NcBiwCPg1cAJAVT2W5DPA7U2/c6vqsWE8D0mSRpR+ylk25FwsMEmSJPWRqroT2HOQ9nesoX8BJ69h23Rg+pAGKEmS+pKXyEmSNIpUFc8991yvw5AkSRpVRkMO5ggmSVpPe582s9chaBBzzj+u1yFs9BYvXsyUKVPYb7/9mDNnDlOnTuXqq69mxYoVvOc97+Gcc87h6aefZurUqSxZsoSVK1fyqU99iqOOOoqJEycydepUvvOd7/DSl76Uyy67jF133ZXFixdz4okn8sgjjzB+/HguvvhidtppJ44//nhe8YpXMHv2bB588EE+//nPc+SRR7Js2TKOOuoonnzySZ599lkuvPBCDjroIK677jrOPvtsVqxYwS677MLFF1/MuHHjev0lkzQM+uXvqn+HJK3JaMvBHMEkSdIosHDhQj760Y9ywQUXsHTpUm677TbmzZvHnDlzuPHGG7n22mt59atfzR133MFdd93FoYceumrfLbfckvnz53PKKadw6qmnAvCxj32MadOmceedd/L+97+fj3/846v6L1u2jJtvvpmrr76aM844A4DLLruMKVOmMG/ePO644w4mTZrEI488wnnnncf3vvc95s6dy+TJk/nCF74wvF8YSZKkLhpNOZgFJkmSRoHXvOY17L///lx33XVcd9117Lnnnuy1117ce++9LFy4kDe96U3MmjWL008/nZtuuoktt9xy1b7HHHPMqudbbrkFgFtuuYX3ve99ABx77LHcfPPNq/ofccQRbLLJJuy+++489NBDAOyzzz5cfPHFfPrTn2b+/PlsscUW3Hrrrdxzzz0ceOCBTJo0iRkzZnD//fcP15dEkiSp60ZTDuYlcpIkjQIvf/nLgdb1/2eeeSYf+chHXtBn7ty5XHPNNXzyk5/k4IMP5qyzzgKefxeR9bmjyOabb75quTV/NLztbW/jxhtv5Nvf/jbHH388n/jEJ9h666155zvfyeWXX97RuUmSJG2sRlMO5ggmSZJGkSlTpjB9+nSeeuopAJYuXcrDDz/ML37xC172spfxgQ98gNNOO425c+eu2ucb3/jGqucDDjgAgLe+9a1cccUVAFx66aUcdNBBa33d+++/n+22244Pf/jDfOhDH2Lu3Lnsv//+/PCHP2TRokUAPP300/zkJz8Z8nOWJEnqtdGQgzmCSZKkUeSQQw5hwYIFq5KUcePG8fWvf51FixZx2mmnsckmmzBmzBguvPDCVfs8/vjjvPnNb2bzzTdf9UnXl770JU444QTOP//8VRNMrs0NN9zA+eefz5gxYxg3bhwzZ85k/PjxXHLJJRxzzDGsWLECgPPOO4/Xve51XTp7SZKk3hgNOVgGhk31k8mTJ9fs2bN7HYakPtMvd7vpN9695/kWLFjAG97whiE73sSJE5k9ezbbbrvtkB1zQwx2XknmVNXkHoWkQZiD6cXol7+r/h2SBP2Zg73Y/MtL5CRJkiRJktQRL5GTJElrtHjx4l6HIEmSNOqMxBzMEUySJEmSJEnqiAUmSZIkSZIkdcQCkyRJkiRJkjpigUmSJEmSJEkdscAkSZKGzbXXXsvrX/96dt11Vz73uc/1OhxJkqS+N1z5l3eRkyRpFNr7tJlDerw55x+3zj4rV67k5JNPZtasWUyYMIF99tmHd7/73ey+++5DGoskSdLGarhzsOHMvxzBJEmShsVtt93Grrvuymtf+1pe8pKXcPTRR/Otb32r12FJkiT1reHMvywwSZKkYbF06VJ23HHHVesTJkxg6dKlPYxIkiSpvw1n/mWBSZIkSZIkSR2xwCRJkobFDjvswAMPPLBqfcmSJeywww49jEiSJKm/DWf+ZYFJkiQNi3322YeFCxfys5/9jN/85jdcccUVvPvd7+51WJIkSX1rOPMv7yInSZKGxWabbcaXv/xlpkyZwsqVKznxxBPZY489eh2WJElS3xrO/KvrBaYkmwKzgaVV9cdJdgauALYB5gDHVtVvkmwOzAT2Bh4Fjqqqxc0xzgQ+CKwEPl5V3+123JIk9bN13dK2Ww477DAOO+ywnry2JElSr/UiBxuu/Gs4LpH7M2BB2/pfARdU1a7A47QKRzTPjzftFzT9SLI7cDSwB3Ao8HdN0UqSJEmSJEkbga4WmJJMAP4z8H+a9QDvAP6p6TIDOKJZPrxZp9l+cNP/cOCKqlpRVT8DFgH7djNuSZKkkSrJ2CS3Jbkjyd1Jzmnad07yoySLknwjyUua9s2b9UXN9oltxzqzab8vyZTenJEkSRoJuj2C6X8B/x14rlnfBvhlVT3brC8BBqYv3wF4AKDZ/kTTf1X7IPuskuSkJLOTzF6+fPlQn4ckSdJIsQJ4R1W9BZgEHJpkfxxFLkmSuqhrBaYkfww8XFVzuvUa7arqoqqaXFWTx48fPxwvKUmStNGplqea1THNo3AUuSRJ6qJujmA6EHh3ksW0JvV+B/C3wFZJBiYXnwAsbZaXAjsCNNu3pDXZ96r2QfaRJEnSapJsmmQe8DAwC/h3HEUuSZK6qGsFpqo6s6omVNVEWsOr/7Wq3g98Hziy6TYN+FazfFWzTrP9X6uqmvajm/kBdgZ2A27rVtySJEkjXVWtrKpJtD6Y2xf4/S6+lqPIJUnSsNxFbnWnA59IsojWp2Nfa9q/BmzTtH8COAOgqu4GrgTuAa4FTq6qlcMetSRJ6siJJ57Iq171Kt74xjf2OpRRo6p+SevDvQNwFLkkSaPScOVgm627S+eq6gbghmb5pwxy/X5VPQO8dw37fxb4bPcilCRpdPn5uW8a0uPtdNb8dfY5/vjjOeWUUzjuuOOG9LX1fEnGA7+tql8meSnwTloTdw+MIr+CwUeR30LbKPIkVwGXJfkC8GocRS5JUsf6OQcblgKTJEnS2972NhYvXtzrMEaD7YEZzR3fNgGurKqrk9wDXJHkPODHPH8U+T80o8gfozW1AVV1d5KBUeTP4ihySZJGpOHKwSwwSZIk9ZGquhPYc5B2R5FLkqSu6cUcTJIkSZIkSeojFpgkSZIkSZLUEQtMkiRJkiRJ6ogFJkmSNCyOOeYYDjjgAO677z4mTJjA1772tXXvJEmSpI4MVw7mJN+SJI1C63NL26F2+eWXD/trSpIkbUz6OQdzBJMkSZIkSZI6YoFJkiRJkiRJHbHAJEmSJEmSpI5YYJIkqc9UVa9DGFL9dj6SJKk/9VPOsiHnYoFJkqQ+MnbsWB599NG+SXCqikcffZSxY8f2OhRJkqQ16qccbEPzL+8iJ0lSH5kwYQJLlixh+fLlvQ5lyIwdO5YJEyb0OgxJkqQ16rccbEPyLwtMkiT1kTFjxrDzzjv3OgxJkqRRxRzMS+QkSZIkSZLUIQtMkiRJkiRJ6ogFJkmSJEmSJHXEApMkSZIkSZI6YoFJkiRJkiRJHbHAJEmSJEmSpI5YYJIkSZIkSVJHLDBJkiRJkiSpIxaYJEmSJEmS1BELTJIkSZIkSeqIBSZJkiRJkiR1xAKTJElSH0myY5LvJ7knyd1J/qxp/3SSpUnmNY/D2vY5M8miJPclmdLWfmjTtijJGb04H0mSNDJs1usAJEmSNKSeBf68quYm2QKYk2RWs+2Cqvrr9s5JdgeOBvYAXg18L8nrms1fAd4JLAFuT3JVVd0zLGchSZJGFAtMkiRJfaSqlgHLmuVfJVkA7LCWXQ4HrqiqFcDPkiwC9m22LaqqnwIkuaLpa4FJkiS9gJfISZIk9akkE4E9gR81TackuTPJ9CRbN207AA+07bakaVtTuyRJ0gtYYJIkSepDScYB/wycWlVPAhcCuwCTaI1w+pshep2TksxOMnv58uVDcUhJkjQCWWCSJEnqM0nG0CouXVpV/wJQVQ9V1cqqeg74Kr+7DG4psGPb7hOatjW1P09VXVRVk6tq8vjx44f+ZCRJ0ohggUmSJKmPJAnwNWBBVX2hrX37tm7vAe5qlq8Cjk6yeZKdgd2A24Dbgd2S7JzkJbQmAr9qOM5BkiSNPE7yLUmS1F8OBI4F5ieZ17T9JXBMkklAAYuBjwBU1d1JrqQ1efezwMlVtRIgySnAd4FNgelVdfdwnogkSRo5LDBJkiT1kaq6Gcggm65Zyz6fBT47SPs1a9tPkiRpgJfISZIkSZIkqSMWmCRJkiRJktQRC0ySJEmSJEnqiAUmSZIkSZIkdcQCkyRJkiRJkjpigUmSJEmSJEkd6VqBKcnYJLcluSPJ3UnOadp3TvKjJIuSfCPJS5r2zZv1Rc32iW3HOrNpvy/JlG7FLEmSJEmSpBevmyOYVgDvqKq3AJOAQ5PsD/wVcEFV7Qo8Dnyw6f9B4PGm/YKmH0l2B44G9gAOBf4uyaZdjFuSJEmSJEkvQtcKTNXyVLM6pnkU8A7gn5r2GcARzfLhzTrN9oOTpGm/oqpWVNXPgEXAvt2KW5IkSZIkSS9OV+dgSrJpknnAw8As4N+BX1bVs02XJcAOzfIOwAMAzfYngG3a2wfZR5IkSZIkST3W1QJTVa2sqknABFqjjn6/W6+V5KQks5PMXr58ebdeRpIkSZIkSasZlrvIVdUvge8DBwBbJdms2TQBWNosLwV2BGi2bwk82t4+yD7tr3FRVU2uqsnjx4/vynlIkiRJkiTphbp5F7nxSbZqll8KvBNYQKvQdGTTbRrwrWb5qmadZvu/VlU17Uc3d5nbGdgNuK1bcUuSJEmSJOnF2WzdXTbY9sCM5o5vmwBXVtXVSe4BrkhyHvBj4GtN/68B/5BkEfAYrTvHUVV3J7kSuAd4Fji5qlZ2MW5JkiRJkiS9CF0rMFXVncCeg7T/lEHuAldVzwDvXcOxPgt8dqhjlCRJkiRJUueGZQ4mSZIkSZIk9S8LTJIkSZIkSeqIBSZJkiRJkiR1xAKTJEmSJEmSOmKBSZIkSZIkSR2xwCRJkiRJkqSOWGCSJEmSJElSRywwSZIkSZIkqSMWmCRJkiRJktQRC0ySJEl9JMmOSb6f5J4kdyf5s6b9lUlmJVnYPG/dtCfJF5MsSnJnkr3ajjWt6b8wybRenZMkSdr4WWCSJEnqL88Cf15VuwP7Aycn2R04A7i+qnYDrm/WAd4F7NY8TgIuhFZBCjgb2A/YFzh7oCglSZK0OgtMkiRJfaSqllXV3Gb5V8ACYAfgcGBG020GcESzfDgws1puBbZKsj0wBZhVVY9V1ePALODQYTwVSZI0glhgkiRJ6lNJJgJ7Aj8CtquqZc2mB4HtmuUdgAfadlvStK2pXZIk6QUsMEmSJPWhJOOAfwZOraon27dVVQE1RK9zUpLZSWYvX758KA4pSZJGIAtMkiRJfSbJGFrFpUur6l+a5oeaS99onh9u2pcCO7btPqFpW1P781TVRVU1uaomjx8/fmhPRJIkjRjrVWBKcv36tEmSJGlobGj+lSTA14AFVfWFtk1XAQN3gpsGfKut/bjmbnL7A080l9J9FzgkydbN5N6HNG2SJEkvsNnaNiYZC7wM2LZJLNJsegVegy9JkjTkhiD/OhA4FpifZF7T9pfA54Ark3wQuB+Y2my7BjgMWAT8GjgBoKoeS/IZ4Pam37lV9Vgn5yZJkvrXWgtMwEeAU4FXA3P4XYLzJPDlLsYlSZI0WnWUf1XVzW37rO7gQfoXcPIajjUdmL7ukCVJ0mi31gJTVf0t8LdJPlZVXxqmmCRJkkYt8y9JkjQSrWsEEwBV9aUkbwUmtu9TVTO7FJckSdKoZv4lSZJGkvUqMCX5B2AXYB6wsmkuwARHkiSpC8y/JEnSSLJeBSZgMrB7c42+JEmSum9U5F97n9Y/9bI55x/X6xAkSeqZTdaz313A73UzEEmSJD2P+ZckSRox1ncE07bAPUluA1YMNFbVu7sSlSRJksy/JEnSiLG+BaZPdzMISZIkvcCnex2AJEnS+lrfu8j9oNuBSJIk6XfMvyRJ0kiyvneR+xWtu5YAvAQYAzxdVa/oVmCSJEmjmfmXJEkaSdZ3BNMWA8tJAhwO7N+toCRJkkY78y9JkjSSrO9d5Faplv8LTOlCPJIkSVqN+ZckSdrYre8lcn/atroJMBl4pisRSZIkyfxLkiSNKOt7F7k/aVt+FlhMa5i2JEmSusP8S5IkjRjrOwfTCd0ORJIkSb9j/iVJkkaS9ZqDKcmEJN9M8nDz+OckE7odnCRJ0mhl/iVJkkaS9Z3k+2LgKuDVzeP/NW2SJEnqDvMvSZI0YqxvgWl8VV1cVc82j0uA8V2MS5IkabQz/5IkSSPG+haYHk3ygSSbNo8PAI92MzBJkqRRzvxLkiSNGOtbYDoRmAo8CCwDjgSO71JMkiRJMv+SJEkjyHrdRQ44F5hWVY8DJHkl8Ne0Eh9JkiQNPfMvSZI0YqzvCKY3DyQ3AFX1GLBnd0KSJEkS5l+SJGkEWd8C0yZJth5YaT5BW9/RT5IkSXrxzL8kSdKIsb5Jyt8AtyT5x2b9vcBnuxOSJEmSMP+SJEkjyHqNYKqqmcCfAg81jz+tqn9Y2z5Jdkzy/ST3JLk7yZ817a9MMivJwuZ566Y9Sb6YZFGSO5Ps1XasaU3/hUmmbejJSpIkjRQbkn9JkiT1yvpeIkdV3VNVX24e96zHLs8Cf15VuwP7Aycn2R04A7i+qnYDrm/WAd4F7NY8TgIuhFXDwc8G9gP2Bc5uHy4uSZLUrzYg/yLJ9CQPJ7mrre3TSZYmmdc8DmvbdmbzAd99Saa0tR/atC1KcsbqryNJktRuvQtML1ZVLauquc3yr4AFwA7A4cCMptsM4Ihm+XBgZrXcCmyVZHtgCjCrqh5rJrqcBRzarbglSZJGuEsYPFe6oKomNY9rAJoP/44G9mj2+bskmybZFPgKrQ8AdweOafpKkiQNalgmikwykdZdT34EbFdVy5pNDwLbNcs7AA+07bakaVtTuyRJklZTVTc2udf6OBy4oqpWAD9LsojWiHGARVX1U4AkVzR912sUlSRJGn26NoJpQJJxwD8Dp1bVk+3bqqqAGqLXOSnJ7CSzly9fPhSHlCRJ6ienNPNcTm+bbsAP+CRJ0pDoaoEpyRhaxaVLq+pfmuaHmkvfaJ4fbtqXAju27T6haVtT+/NU1UVVNbmqJo8fP35oT0SSJGlkuxDYBZgELKN1h7oh4Yd8kiQJulhgShLga8CCqvpC26argIE7wU0DvtXWflxzN7n9gSeaS+m+CxySZOvm07ZDmjZJkiSth6p6qKpWVtVzwFf53WVwHX3A1xzbD/kkSVJX52A6EDgWmJ9kXtP2l8DngCuTfBC4H5jabLsGOAxYBPwaOAGgqh5L8hng9qbfuVX1WBfjliRJ6itJtm+bA/M9wMAd5q4CLkvyBeDVtO7mexsQYLckO9MqLB0NvG94o5YkSSNJ1wpMVXUzreRkMAcP0r+Ak9dwrOnA9KGLTpIkqT8luRx4O7BtkiXA2cDbk0yiNfflYuAjAFV1d5IraU3e/SxwclWtbI5zCq1R45sC06vq7mE+FUmSNIIMy13kJEmSNDyq6phBmr+2lv6fBT47SPs1tEaYS5IkrVPX7yInSZIkSZKk/uYIpg7sfdrMXoeg1cw5/7hehyBJkiRJ0qjjCCZJkiRJkiR1xAKTJEmSJEmSOmKBSZIkSZIkSR2xwCRJkiRJkqSOWGCSJEmSJElSRywwSZIkSZIkqSMWmCRJkiRJktQRC0ySJEmSJEnqiAUmSZIkSZIkdcQCkyRJkiRJkjpigUmSJEmSJEkdscAkSZIkSZKkjlhgkiRJkiRJUkcsMEmSJEmSJKkjFpgkSZIkSZLUEQtMkiRJkiRJ6ogFJkmSJEmSJHXEApMkSZIkSZI6YoFJkiRJkiRJHbHAJEmSJEmSpI5YYJIkSZIkSVJHLDBJkiRJkiSpIxaYJEmS+kiS6UkeTnJXW9srk8xKsrB53rppT5IvJlmU5M4ke7XtM63pvzDJtF6ciyRJGjksMEmSJPWXS4BDV2s7A7i+qnYDrm/WAd4F7NY8TgIuhFZBCjgb2A/YFzh7oCglSZI0GAtMkiRJfaSqbgQeW635cGBGszwDOKKtfWa13ApslWR7YAowq6oeq6rHgVm8sGglSZK0igUmSZKk/rddVS1rlh8EtmuWdwAeaOu3pGlbU/sLJDkpyewks5cvXz60UUuSpBHDApMkSdIoUlUF1BAe76KqmlxVk8ePHz9Uh5UkSSOMBSZJkqT+91Bz6RvN88NN+1Jgx7Z+E5q2NbVLkiQNygKTJElS/7sKGLgT3DTgW23txzV3k9sfeKK5lO67wCFJtm4m9z6kaZMkSRrUZr0OQBpKPz/3Tb0OQYPY6az5vQ5BkkaNJJcDbwe2TbKE1t3gPgdcmeSDwP3A1Kb7NcBhwCLg18AJAFX1WJLPALc3/c6tqtUnDpckSVrFApMkSVIfqapj1rDp4EH6FnDyGo4zHZg+hKFJkqQ+5iVykiRJkiRJ6ogFJkmSJEmSJHXEApMkSZIkSZI6YoFJkiRJkiRJHXGSb0mSJEnaQP10F2Pv/CupExaYJEmSJEnDYu/TZvY6hCEz5/zjeh2CtFHxEjlJkiRJkiR1pGsFpiTTkzyc5K62tlcmmZVkYfO8ddOeJF9MsijJnUn2attnWtN/YZJp3YpXkiRJkiRJG6abI5guAQ5dre0M4Pqq2g24vlkHeBewW/M4CbgQWgUp4GxgP2Bf4OyBopQkSZIkSZI2Dl2bg6mqbkwycbXmw4G3N8szgBuA05v2mVVVwK1JtkqyfdN3VlU9BpBkFq2i1eXdiluSJEnaEP0y2bMTPUuSNsRwz8G0XVUta5YfBLZrlncAHmjrt6RpW1P7CyQ5KcnsJLOXL18+tFFLkiRJkiRpjXo2yXczWqmG8HgXVdXkqpo8fvz4oTqsJEmSJEmS1mG4C0wPNZe+0Tw/3LQvBXZs6zehaVtTuyRJkiRJkjYSw11gugoYuBPcNOBbbe3HNXeT2x94ormU7rvAIUm2bib3PqRpkyRJkiRJ0kaia5N8J7mc1iTd2yZZQutucJ8DrkzyQeB+YGrT/RrgMGAR8GvgBICqeizJZ4Dbm37nDkz4LUmSJEmSpI1DN+8id8waNh08SN8CTl7DcaYD04cwNEmSJEmSJA2hnk3yLUmSJEmSpP5ggUmSJEmSJEkdscAkSZIkSZKkjlhgkiRJkiRJUkcsMEmSJEmSJKkjFpgkSZIkSZLUEQtMkiRJkiRJ6ogFJkmSJEmSJHXEApMkSdIokWRxkvlJ5iWZ3bS9MsmsJAub562b9iT5YpJFSe5Msldvo5ckSRszC0ySJEmjy3+qqklVNblZPwO4vqp2A65v1gHeBezWPE4CLhz2SCVJ0ohhgUmSJGl0OxyY0SzPAI5oa59ZLbcCWyXZvhcBSpKkjZ8FJkmSpNGjgOuSzElyUtO2XVUta5YfBLZrlncAHmjbd0nT9jxJTkoyO8ns5cuXdytuSZK0kdus1wFIktSJn5/7pl6HoNXsdNb8XoegNfuDqlqa5FXArCT3tm+sqkpSL+aAVXURcBHA5MmTX9S+kiSpfziCSZIkaZSoqqXN88PAN4F9gYcGLn1rnh9uui8FdmzbfULTJkmS9AIWmCRJkkaBJC9PssXAMnAIcBdwFTCt6TYN+FazfBVwXHM3uf2BJ9oupZMkSXoeL5GTJEkaHbYDvpkEWjngZVV1bZLbgSuTfBC4H5ja9L8GOAxYBPwaOGH4Q5YkSSOFBSZJkqRRoKp+CrxlkPZHgYMHaS/g5GEITZIk9QEvkZMkSZIkSVJHLDBJkiRJkiSpIxaYJEmSJEmS1BELTJIkSZIkSeqIBSZJkiRJkiR1xAKTJEmSJEmSOmKBSZIkSZIkSR2xwCRJkiRJkqSOWGCSJEmSJElSRywwSZIkSZIkqSMWmCRJkiRJktQRC0ySJEmSJEnqyGa9DkCSJEmSJPW3vU+b2esQhjU9G90AACAASURBVMyc84/rdQgbJUcwSZIkSZIkqSMWmCRJkiRJktQRC0ySJEmSJEnqiHMwSZIkSZL0Iv383Df1OoQhs9NZ83sdgvqAI5gkSZIkSZLUEQtMkiRJkiRJ6ogFJkmSJEmSJHXEApMkSZIkSZI64iTfkiRJkiRJ68kJ3gfnCCZJkiRJkiR1xAKTJEmSJEmSOjJiCkxJDk1yX5JFSc7odTySJEmjgTmYJElaHyOiwJRkU+ArwLuA3YFjkuze26gkSZL6mzmYJElaXyOiwATsCyyqqp9W1W+AK4DDexyTJElSvzMHkyRJ6yVV1esY1inJkcChVfWhZv1YYL+qOqWtz0nASc3q64H7hj1QbQy2BR7pdRCShpXv+9HrNVU1vtdB9DNzsBfN30ejk9/30cvv/eg1mr/3a8y/NhvuSLqlqi4CLup1HOqtJLOranKv45A0fHzfS71lDvY7/j4anfy+j15+70cvv/eDGymXyC0Fdmxbn9C0SZIkqXvMwSRJ0noZKQWm24Hdkuyc5CXA0cBVPY5JkiSp35mDSZKk9TIiLpGrqmeTnAJ8F9gUmF5Vd/c4LG2cHKIvjT6+76UuMQd70fx9NDr5fR+9/N6PXn7vBzEiJvmWJEmSJEnSxmukXCInSZIkSZKkjZQFJkmSJEmSJHXEApMkSZJGrSRbJfnoBuz3l6ut/9vQRaV+keTtSa7udRzaMEnOTfJHvY5DwyfJ8Um+3Os4RioLTJIkSRrNtgLWu8CUlk2A5xWYquqtG3AM6XmSjIibMI0GSTatqrOq6ntDfFzf/+pb/mBrxEoyMcmCJF9NcneS65K8NMmkJLcmuTPJN5Ns3etYJW245tPDU9vWP5vkz5KcluT25r1+TrPt5Um+neSOJHclOap3kUsaIT4H7JJkXpLz1/C7ZWKS+5LMBO4Cvga8tNnn0qbPU83zuCTXJ5mbZH6Sw9dwjE8l+V8DQST5cJILhvfU+99gfxeS7J3kB0nmJPluku2bvrsm+V7Td26SXZpiwPnNvvMH/q40I5NuSPJPSe5NcmmSNNsObdrmAn/aFsu+SW5J8uMk/5bk9U378UmuSvKvwPVJZiY5om2/Swd+jjQ0mvfjwPdtQfN9fFmSxUn+qvnevTfJJUmObPZZnOR/Nu/72Un2an5+/j3Jf236+P7fyDRf+7va1v8iyaeb9+9fJbktyU+SHDTIvv+5ec9u2/wsfLF57/607ediTb8jvpLk3c3yN5NMb5ZPTCuXHfR/2eH5qnSPBSaNdLsBX6mqPYBfAv8FmAmcXlVvBuYDZ/cwPkmdmw4cB5DWJ35HAw/Sev/vC0wC9k7yNuBQ4BdV9ZaqeiNwbW9CljSCnAH8e1VNAmYx+O8Wmva/q6o9quoE4D+qalJVvX+14z0DvKeq9gL+E/A3A4WH9mMAfwP8SZIxzbYTaP2+09Aa7O/Cl4Ajq2pvWl/zzzZ9L6WVV74FeCuwjFaBaBLwFuCPgPMHClLAnsCpwO7Aa4EDk4wFvgr8CbA38HttsdwLHFRVewJnAf+jbdteTUx/SKuAeTxAki2bWL49JF8NtXs9rffjG4An+d1Ixkeraq+qumKQfX7e/K64CbgEOBLYHzin2e77f2TZrKr2pfU+ft7/jEneQ+vvw2FV9UjTvD3wB8Af0/pwAtb8O+ImYKBotQOt3xM0bTc2y4P9LzuiOQRTI93PqmpeszwH2AXYqqp+0LTNAP6xJ5FJGhJVtTjJo0n2BLYDfgzsAxzSLAOMo/VH+iZaydxfAVdX1U29iFnSiHUIg/9u+Tlwf1Xduh7HCPA/msLUc7T+sdiu2bbqGFX1VDNi5Y+TLADGVNX8oTsVNebT9ncBeBx4IzCr+b9/U2BZki2AHarqmwBV9QxAkj8ALq+qlcBDSX5A62/Qk8BtVbWk6TcPmAg8RSs/Xdi0fx04qYllS2BGkt2AAgaKCwCzquqx5rV/kOTvkoyn9Q/nP1fVs0P/pRn1HqiqHzbLXwc+3ix/Yy37XNU8zwfGVdWvgF8lWZFkK+BpfP+PJP/SPM+h9f4d8A5gMnBIVT3Z1v5/q+o54J4kA9/XNf2OuAk4NcnuwD3A1k3h6QBaP2vb8ML/ZdtjGJEsMGmkW9G2vJLWPAqS+s//ofVp7u/R+oTvYOB/VtXfr94xyV7AYcB5Sa6vqnOHM1BJI1oY5HdLkom0/nFcH+8HxgN7V9VvkywGxjbbVj/G/6E1l9O9wMUbFrLWpqp+0v53AfhX4O6qOqC9X1NgerFWz0PX9b/VZ4DvV9V7mp+pG9q2rf6zMRP4AK1RuydsQGxat1rD+tre6wPf8+d4/vf/OVrff9//G59nef6VW2Pblge+h6u/f/+d1qjE1wGzB+kPrb8Xa1RVS5ui46G0Riy9EpgKPFVVv0qyDS/8HeIlctJG5gng8bZraI8FfrCW/pJGhm/S+gO9D/Dd5nFiknEASXZI8qokrwZ+XVVfB86ndcmBJK3Nr4CB4sKgv1vWsN9v2y5vabcl8HDzz+V/Al6zpheuqh8BOwLvAy7f0BPQmg3yd2E/YHySA5rtY5Ls0YxEWZJm7qMkmyd5Ga1RCEcl2bQZUfQ24La1vOS9wMQkuzTrx7Rt2xJY2iwfv47QL6F12Q5Vdc96naxerJ0Gfg5ovQdvHoJj+v7f+DwEvCrJNkk2p3V527rcTzP1SpI91tF3bb8jbqX1Pr6x6fcXzXPfcgST+tE04H83ScFP8VMfacSrqt8k+T7wy2YI8nVJ3gDc0lzi8BStT3p3pXXt+3PAb4H/1quYJY0MVfVokh82k8B+B7iMF/5uWTnIrhcBdyaZu9o8TJcC/y/JfFqffN+7jhCuBCZV1eMdnooG9yZe+HfhWeCLzfxGmwH/C7ib1geTf5/k3Kbve2l9wHEAcAetES7/vaoeTPL7g71YVT2T5CTg20l+TeufyYEC5udpXSL3SdYxp1JVPdRcOvV/N/zUtQ73ASc3ky/fA1wIfKzDY/r+38g0xb5zaRV9lrLu78nAfvcmeT/wj0n+ZC1dB/0d0Wy7idZldouS3E9rFFNfF5hStfrIQEmSNi7N5N5zgfcOzGshSf0gydXABVV1fa9j0caj+aB0PrBXVT3R63j6TXOJ4tXNxO+9jMP3v/qKl8hJkjZqzeSIi4DrLS5J6hdJtkryE1p3o/OfS62S5I+ABcCXLC71J9//6leOYJIkSZIkSVJHHMEkSZIkSZKkjlhgkiRJkiRJUkcsMEmSJEmSJKkjFpgk9UySic1tobt1/H/r1rElSZL6RZLjk3x5iI95RHOjjoH1c5sJzCX1KQtMkvpWVb211zFIkiSNUkcAqwpMVXVWVX2vh/FI6jILTJJ6bdMkX01yd5Lrkrw0yaQktya5M8k3k2wNkOSGJJOb5W2TLG6W90hyW5J5zT67Ne1PNc9vb/b9pyT3Jrk0SZpthzVtc5J8McnVPfkqSJIkdUmSD7TlSn+fZNMkJyT5SZLbgAPb+l6S5Mi29afalk9PMj/JHUk+17R9OMntTds/J3lZkrcC7wbOb15zl/bjJjk4yY+bY01PsnnTvjjJOUnmNtt+f5i+RJKGgAUmSb22G/CVqtoD+CXwX4CZwOlV9WZgPnD2Oo7xX4G/rapJwGRgySB99gROpfVJ2muBA5OMBf4eeFdV7Q2MH4LzkSRJ2mgkeQNwFHBgkyutBD4AnEOrsPQHtI00Wstx3gUcDuxXVW8BPt9s+peq2qdpWwB8sKr+DbgKOK2qJlXVv7cdZyxwCXBUVb0J2Az4b20v9UhV7QVcCPzFhp+5pOFmgUlSr/2squY1y3OAXYCtquoHTdsM4G3rOMYtwF8mOR14TVX9xyB9bquqJVX1HDAPmAj8PvDTqvpZ0+fyDs5DkiRpY3QwsDdwe5J5zfr/B9xQVcur6jfAN9bjOH8EXFxVvwaoqsea9jcmuSnJfOD9wB7rOM7raeV/P2nWV8/1/qV5nkMrX5M0QlhgktRrK9qWVwJbraXvs/zu99bYgcaquozWMOz/AK5J8o71eJ3NNihaSZKkkSXAjGYk0aSqej3w6bX0X5VvJdkEeMk6jn8JcEozGukc2nK0DTSQs5mvSSOMBSZJG5sngMeTHNSsHwsMjGZaTOsTOID2uQFeS2sk0heBbwFvXs/Xug94bZKJzfpRGxy1JEnSxul64MgkrwJI8krgx8AfJtkmyRjgvW39F/O7fOvdwJhmeRZwQpKXtR0HYAtgWXOc97cd51fNttXdB0xMsmuz3p7rSRrBLDBJ2hhNozUp5J3AJODcpv2vgf+W5MfAtm39pwJ3NcO+30hrDqd1ai6l+yhwbZI5tBKhJ4bmFCRJknqvqu4BPglc1+RWs4DtaY1iugX4Ia25kwZ8lVbx6Q7gAODp5jjX0ppXaXaTcw3Mj/Qp4EfNce5tO84VwGnNZN67tMXzDHAC8I/NZXXPAf97KM9ZUm+kqnodgyT1TJJxVfVUc1e5rwALq+qCXsclSZIkSSOJI5gkjXYfbj6FuxvYktZd5SRJkiRJL4IjmCRJkiRJktQRRzBJkiRJkiSpIxaYJEmSJEmS1BELTJIkSZIkSeqIBSZJkiRJkiR1xAKTJEmSJEmSOmKBSZIkSZIkSR2xwCRJkiRJkqSOWGCSJEmSJElSRywwSZIkSZIkqSMWmCRJkiRJktQRC0ySJEmSJEnqiAUmSZIkSZIkdcQCkyRJkiRJkjpigUmSJEmSJEkd2azXAXTDtttuWxMnTux1GJIkqYvmzJnzSFWN73Uc+h1zMEmS+tva8q++LDBNnDiR2bNn9zoMSZLURUnu73UMej5zMEmS+tva8i8vkZMkSZIkSVJHLDBJkiRJkiSpIxaYJEmSJEmS1JG+nINJkqTR6re//S1LlizhmWee6XUoQ2bs2LFMmDCBMWPG9DoUSZKkQfVbDrYh+ZcFJkmS+siSJUvYYostmDhxIkl6HU7HqopHH32UJUuWsPPOO/c6HEmSpEH1Uw62ofmXl8hJktRHnnnmGbbZZpsRn9gMSMI222zTN58GSpKk/tRPOdiG5l8WmCRJ6jP9kNi067fzkSRJ/amfcpYNORcLTJIkSZIkSeqIBSZJkkaRquK5557rdRiSJEmjymjIwZzkW9pI7H3azI6PMef844YgEkn9ZvHixUyZMoX99tuPOXPmMHXqVK6++mpWrFjBe97zHs455xyefvpppk6dypIlS1i5ciWf+tSnOOqoo5g4cSJTp07lO9/5Di996Uu57LLL2HXXXVm8eDEnnngijzzyCOPHj+fiiy9mp5124vjjj+cVr3gFs2fP5sEHH+Tzn/88Rx55JMuWLeOoo47iySef5Nlnn+XCCy/koIMO4rrrruPss89mxYoV7LLLLlx88cWMGzeu118yDaOh+Pu3sfDvsCSp3WjLwRzBJEnSKLBw4UI++tGPcsEFF7B06VJuu+025s2bx5w5c7jxxhu59tprefWrX80dd9zBXXfdxaGHHrpq3y233JL58+dzyimncOqppwLwsY99jGnTpnHnnXfy/ve/n49//OOr+i9btoybb76Zq6++mjPOOAOAyy67jClTpjBv3jzuuOMOJk2axCOPPMJ5553H9773PebOncvkyZP5whe+MLxfGEmSpC4aTTlY1wpMSaYneTjJXYNs+/MklWTbZj1JvphkUZI7k+zV1ndakoXNY1q34pUkqZ+95jWvYf/99+e6667juuuuY88992Svvfbi3nvvZeHChbzpTW9i1qxZnH766dx0001sueWWq/Y95phjVj3fcsstANxyyy28733vA+DYY4/l5ptvXtX/iCOOYJNNNmH33XfnoYceAmCfffbh4osv5tOf/jTz589niy224NZbb+Wee+7hwAMPZNKkScyYMYP7779/uL4kkiRJXTeacrBuXiJ3CfBl4HnjnpPsCBwC/Lyt+V3Abs1jP+BCYL8krwTOBiYDBcxJclVVPd7FuCVJ6jsvf/nLgdb1/2eeeSYf+chHXtBn7ty5XHPNNXzyk5/k4IMP5qyzzgKefxeR9bmjyOabb75quaoAeNvb3saNN97It7/9bY4//ng+8YlPsPXWW/POd76Tyy+/vKNzkyRJ2liNphysayOYqupG4LFBNl0A/HdaBaMBhwMzq+VWYKsk2wNTgFlV9VhTVJoFHPqCI0qSpPUyZcoUpk+fzlNPPQXA0qVLefjhh/nFL37By172Mj7wgQ9w2mmnMXfu3FX7fOMb31j1fMABBwDw1re+lSuuuAKASy+9lIMOOmitr3v//fez3Xbb8eEPf5gPfehDzJ07l/33358f/vCHLFq0CICnn36an/zkJ0N+zpIkSb02GnKwYZ3kO8nhwNKqumO16tsOwANt60uatjW1D3bsk4CTAHbaaachjFqSpP5xyCGHsGDBglVJyrhx4/j617/OokWLOO2009hkk00YM2YMF1544ap9Hn/8cd785jez+eabr/qk60tf+hInnHAC559//qoJJtfmhhtu4Pzzz2fMmDGMGzeOmTNnMn78eC655BKOOeYYVqxYAcB5553H6173ui6dvSRJUm+MhhwsA8OmuiHJRODqqnpjkpcB3wcOqaonkiwGJlfVI0muBj5XVTc3+10PnA68HRhbVec17Z8C/qOq/nptrzt58uSaPXt2l85K6g7vIidpKCxYsIA3vOENQ3a8iRMnMnv2bLbddtshO+aGGOy8ksypqsk9CkmD2JAczLvISZL6QT/mYC82/xrOu8jtAuwM3NEUlyYAc5P8HrAU2LGt74SmbU3tkiRJkiRJ2kgM2yVyVTUfeNXA+mojmK4CTklyBa1Jvp+oqmVJvgv8jyRbN7sdApw5XDFLkjTaLV68uNchSJIkjTojMQfr2gimJJcDtwCvT7IkyQfX0v0a4KfAIuCrwEcBquox4DPA7c3j3KZNkiRJkiRJG4mujWD6/9m78zDLqvre/++PNAgqMrb8sBsCCprghNACRuNFUUCSCEZUjBFQcju5Fxyu0YjJvUEx5GLQEAmKIdICiYI4IB2CYsvgFBkaRUaRDqB0B6Evk+BApPn+/tir6EP1qaa6T1Wdrur363nOU/usvfbea49n7W+tvXZVvelxxu/QM1zAkWPkWwAsmNDCSZIkSZIkacJMZR9MkiRJmiJJNkjy/fYyFZLsmOTyJEuSfC7JRi39ie37kjZ+h555vL+l35Rkv+GsiSRJmg4MMEmSJM1M7wRu7Pn+YeDEqtoJuBcY6b7gCODeln5iy0eSXYBDgOcA+wOfSLLBFJVdkiRNMwaYJEnSlPnqV7/Ks5/9bHbaaSeOP/74YRdnxkoyF/hd4FPte4BXAF9oWc4ADmrDB7bvtPH7tPwHAmdX1UNVdStdX5l7TM0aSJKkiTJV9a8pe4ucJElad+z+3jMndH5XnXDo4+ZZsWIFRx55JIsWLWLu3Lm86EUv4jWveQ277LLLhJZFAPw98OfApu37VsB9VfVw+74UmNOG5wC3A1TVw0nub/nnAJf1zLN3GkmStBamug42lfUvWzBJkqQpccUVV7DTTjvxjGc8g4022ohDDjmE8847b9jFmnGS/B5wV1VdNUXLm59kcZLFy5cvn4pFSpKkcZrK+pcBJkmSNCWWLVvGdttt9+j3uXPnsmzZsiGWaMZ6CfCaJLcBZ9M9GvcxYPMkI63X5wIjG38ZsB1AG78ZcHdvep9pHlVVp1bVvKqaN3v27IlfG0mStNamsv5lgEmSJGkGqar3V9XcqtqBrpPui6vqzcAlwMEt22HAyL8vF7bvtPEXV1W19EPaW+Z2BHYGrpii1ZAkSdOMfTBJkqQpMWfOHG6//fZHvy9dupQ5c+zSZwq9Dzg7yV8D3wdOa+mnAf+cZAlwD11Qiqq6Psk5wA3Aw8CRVbVi6ostSZLW1lTWvwwwSZKkKfGiF72Im2++mVtvvZU5c+Zw9tln89nPfnbYxZrRqupS4NI2fAt93gJXVb8CXj/G9McBx01eCSVJ0mSayvqXASatlYno+X48bxySJM0cs2bN4uSTT2a//fZjxYoVvO1tb+M5z3nOsIslSZI0Y01l/csAkyRJ66FhBfkPOOAADjjggKEsW5IkadiGUQebqvqXnXxLkiRJkiRpIAaYJEmSJEmSNBADTJIkSZIkSRqIASZJkiRJkiQNxACTJEmSJEmSBmKASZIkSZIkSQMxwCRJkqbE2972Np72tKfx3Oc+d9hFkSRJWm9MVR1s1qTOXZIkrZN+cuzzJnR+2//VtY+b5/DDD+eoo47i0EMPndBlS5IkTRczuQ5mCyZJkjQlXvayl7HlllsOuxiSJEnrlamqgxlgkiRJkiRJ0kAMMEmSJEmSJGkgkxZgSrIgyV1JrutJOyHJD5Nck+TcJJv3jHt/kiVJbkqyX0/6/i1tSZKjJ6u8kiRJkiRJWjuT2YLpdGD/UWmLgOdW1fOBHwHvB0iyC3AI8Jw2zSeSbJBkA+DjwKuBXYA3tbySJEmSJElaR0xagKmqvgncMyrta1X1cPt6GTC3DR8InF1VD1XVrcASYI/2WVJVt1TVfwFnt7ySJGmaedOb3sSLX/xibrrpJubOnctpp5027CJJkiTNeFNVB5s1KXMdn7cBn2vDc+gCTiOWtjSA20el79lvZknmA/MBtt9++wktqCRJM814Xmk70c4666wpX6YkSdK6ZCbXwYbSyXeSvwQeBj4zUfOsqlOral5VzZs9e/ZEzVaSJGlaSbJxkiuS/CDJ9Uk+2NJPT3JrkqvbZ9eWniQntf4ur0myW8+8Dktyc/scNqx1kiRJ674pb8GU5HDg94B9qqpa8jJgu55sc1saq0mXJEnSqh4CXlFVDybZEPh2kq+0ce+tqi+Myv9qYOf22RM4BdgzyZbAMcA8oICrkiysqnunZC0kSdK0MqUtmJLsD/w58Jqq+kXPqIXAIUmemGRHugrOFcCVwM5JdkyyEV1H4AunssySJEnTSXUebF83bJ9azSQHAme26S4DNk+yLbAfsKiq7mlBpUWs+gIXSZIkYBIDTEnOAr4LPDvJ0iRHACcDmwKLWtPsTwJU1fXAOcANwFeBI6tqResQ/CjgQuBG4JyWV5IkjWFlA+GZYaatz1Rob+O9GriLLkh0eRt1XHsM7sQkT2xpc1i1z8s5q0mXJEl9zKQ6y9qsy6Q9IldVb+qTPGZX5VV1HHBcn/QLgAsmsGiSJM1YG2+8MXfffTdbbbUVSYZdnIFVFXfffTcbb7zxsIsyrVTVCmDXJJsD5yZ5LvB+4KfARsCpwPuAYwddli9akSRpZtXB1rb+Ncy3yEmSpAk2d+5cli5dyvLly4ddlAmz8cYbM3fu3GEXY1qqqvuSXALsX1UfackPJfk08J72fay+MJcBe49Kv7TPMk6lC1gxb968mfOvW0mS1sBMq4OtTf3LAJMkSTPIhhtuyI477jjsYmiIkswGft2CS5sArwI+nGTbqroj3b9VDwKua5MsBI5KcjZdJ9/3t3wXAn+TZIuWb1+6VlCSJGkU62AGmCRJkmaabYEzkmxA19/mOVV1fpKLW/ApwNXAn7b8FwAHAEuAXwBvBaiqe5J8iO6lKwDHVtU9U7gekiRpGjHAJEmSNINU1TXAC/ukv2KM/AUcOca4BcCCCS2gJEmakSbtLXKSJEmSJElaPxhgkiRJkiRJ0kAMMEmSJEmSJGkgBpgkSZIkSZI0EANMkiRJkiRJGogBJkmSJEmSJA3EAJMkSZIkSZIGYoBJkiRJkiRJAzHAJEmSJEmSpIEYYJIkSZIkSdJADDBJkiRJkiRpIAaYJEmSJEmSNJBZwy6ANJl2f++ZA8/jqhMOnYCSSJIkSZI0c9mCSZIkSZIkSQMxwCRJkiRJkqSBGGCSJEmSJEnSQAwwSZIkSZIkaSAGmCRJkmaQJBsnuSLJD5Jcn+SDLX3HJJcnWZLkc0k2aulPbN+XtPE79Mzr/S39piT7DWeNJEnSdDBpAaYkC5LcleS6nrQtkyxKcnP7u0VLT5KTWgXmmiS79UxzWMt/c5LDJqu8kiRJM8RDwCuq6gXArsD+SfYCPgycWFU7AfcCR7T8RwD3tvQTWz6S7AIcAjwH2B/4RJINpnRNJEnStDGZLZhOp6uM9DoauKiqdgYuat8BXg3s3D7zgVOgC0gBxwB7AnsAx4wEpSRJkrSq6jzYvm7YPgW8AvhCSz8DOKgNH9i+08bvkyQt/eyqeqiqbgWW0NXHJEmSVjFpAaaq+iZwz6jk3grM6IrNma1CdBmweZJtgf2ARVV1T1XdCyxi1aCVJEmSeiTZIMnVwF109af/AO6rqodblqXAnDY8B7gdoI2/H9iqN73PNJIkSY8x1X0wbVNVd7ThnwLbtOGxKjBWbCRJktZQVa2oql2BuXStjn5zspaVZH6SxUkWL1++fLIWI0mS1nFD6+S7qoquufaEsHIjSZL0WFV1H3AJ8GK6FuKz2qi5wLI2vAzYDqCN3wy4uze9zzS9yzi1quZV1bzZs2dPynpIkqR131QHmO5sj77R/t7V0seqwIyrYgNWbiRJkgCSzE6yeRveBHgVcCNdoOnglu0w4Lw2vLB9p42/uP0jcCFwSHvL3I50fWVeMTVrIUmSppupDjD1VmBGV2wObW+T2wu4vz1KdyGwb5ItWufe+7Y0SZIk9bctcEmSa4Ar6fqzPB94H/DuJEvo+lg6reU/Ddiqpb+b9hKWqroeOAe4AfgqcGRVrZjSNZEkSdPGrMfPsnaSnAXsDWydZCnd2+COB85JcgTwY+ANLfsFwAF0byf5BfBWgKq6J8mH6CpHAMdW1eiOwyVJktRU1TXAC/uk30Kft8BV1a+A148xr+OA4ya6jJIkaeaZtABTVb1pjFH79MlbwJFjzGcBsGACiyZJkiRJkqQJNLROviVJkiRJkjQzGGCSJEmSJEnSQAwwSZIkSZIkaSAGmCRJkiRJkjQQA0ySJEmSJEkaiAEmSZIkSZIkDcQAkyRJkiRJAphWrwAAIABJREFUkgZigEmSJEmSJEkDMcAkSZIkSZKkgRhgkiRJkiRJ0kAMMEmSJEmSJGkgBpgkSZIkSZI0EANMkiRJkiRJGogBJkmSJEmSJA3EAJMkSZIkSZIGYoBJkiRJkiRJAzHAJEmSNIMk2S7JJUluSHJ9kne29A8kWZbk6vY5oGea9ydZkuSmJPv1pO/f0pYkOXoY6yNJkqaHWcMugCRJkibUw8CfVdX3kmwKXJVkURt3YlV9pDdzkl2AQ4DnAE8Hvp7kWW30x4FXAUuBK5MsrKobpmQtJEnStGKASZIkaQapqjuAO9rwA0luBOasZpIDgbOr6iHg1iRLgD3auCVVdQtAkrNbXgNMkiRpFetNgGn395458DyuOuHQCSiJJEnS1EiyA/BC4HLgJcBRSQ4FFtO1crqXLvh0Wc9kS1kZkLp9VPqek1xkSZI0TdkHkyRJ0gyU5CnAF4F3VdXPgFOAZwK70rVw+ugELWd+ksVJFi9fvnwiZilJkqYhA0ySJEkzTJIN6YJLn6mqLwFU1Z1VtaKqHgH+iZWPwS0DtuuZfG5LGyv9Marq1KqaV1XzZs+ePfErI0mSpoVxBZiSXDSetPFK8r/aW02uS3JWko2T7Jjk8vaWks8l2ajlfWL7vqSN32FtlytJkjRdrG39K0mA04Abq+rvetK37cn2WuC6NrwQOKTVuXYEdgauAK4Edm51tI3oOgJfuLbrI0mSZrbV9sGUZGPgScDWSbYA0kY9ldV3Frm6ec4B3gHsUlW/THIOXYXlALo3m5yd5JPAEXRNuY8A7q2qnZIcAnwYeOPaLFuSJGldNwH1r5cAbwGuTXJ1S/sL4E1JdgUKuA34E4Cqur7Vx26gewPdkVW1opXlKOBCYANgQVVdP/gaSpKkmejxOvn+E+BddK+svYqVFZyfAScPuNxNkvyargJ1B/AK4A/b+DOAD9AFmA5swwBfAE5OkqqqAZYvSZK0rhqo/lVV3+6ZptcFq5nmOOC4PukXrG46SZKkEasNMFXVx4CPJXl7Vf3DRCywqpYl+QjwE+CXwNfoKk/3VdXDLVvv20vm0N5gUlUPJ7kf2Ar4fxNRHkmSpHXJZNS/JEmSJtvjtWACoKr+IclvAzv0TlNVZ67pAltT7wOBHYH7gM8D+6/pfPrMdz4wH2D77bcfdHaSJElDNZH1L0mSpMk2rgBTkn+me63t1cCKllzA2lRwXgncWlXL27y/RNdXwOZJZrVWTL1vKRl5g8nSJLOAzYC7R8+0qk4FTgWYN2+ej89JkqRpbYLrX5IkSZNqXAEmYB5dp9wTEbj5CbBXkifRPSK3D7AYuAQ4GDgbOAw4r+Vf2L5/t42/2P6XJEnSemAi61+SJEmT6gnjzHcd8P9NxAKr6nK6zrq/B1zbynAq8D7g3UmW0PWxdFqb5DRgq5b+buDoiSiHJEnSOm7C6l+SJEmTbbwtmLYGbkhyBfDQSGJVvWZtFlpVxwDHjEq+BdijT95fAa9fm+VIkiRNYxNa/5IkSZpM4w0wfWAyCyFJkqRVfGDYBZAkSRqv8b5F7huTXRBJkiStZP1LkiRNJ+N9i9wDdG8tAdgI2BD4eVU9dbIKJkmStD6z/iVJkqaT8bZg2nRkOEmAA4G9JqtQkiRJ6zvrX5IkaToZ71vkHlWdLwP7TUJ5JEmSNIr1L0mStK4b7yNyf9Dz9QnAPOBXk1IiSZIkWf+SJEnTynjfIvf7PcMPA7fRNdOWJEnS5LD+JUmSpo3x9sH01skuiCRJklay/iVJkqaTcfXBlGRuknOT3NU+X0wyd7ILJ0mStL6y/iVJkqaT8Xby/WlgIfD09vnXliZJkqTJYf1LkiRNG+MNMM2uqk9X1cPtczowexLLJUmStL6z/iVJkqaN8QaY7k7yR0k2aJ8/Au6ezIJJkiSt59aq/pVkuySXJLkhyfVJ3tnSt0yyKMnN7e8WLT1JTkqyJMk1SXbrmddhLf/NSQ6btDWVJEnT3ngDTG8D3gD8FLgDOBg4fJLKJEmSpLWvfz0M/FlV7QLsBRyZZBfgaOCiqtoZuKh9B3g1sHP7zAdOgS4gBRwD7AnsARwzEpSSJEkabbwBpmOBw6pqdlU9ja7C88HJK5YkSdJ6b63qX1V1R1V9rw0/ANwIzAEOBM5o2c4ADmrDBwJnVucyYPMk2wL7AYuq6p6quhdYBOw/casnSZJmkvEGmJ7fKhYAVNU9wAsnp0iSJEliAupfSXZo01wObFNVd7RRPwW2acNzgNt7Jlva0sZKlyRJWsV4A0xP6G0S3ZpMz5qcIkmSJIkB619JngJ8EXhXVf2sd1xVFVATUcgk85MsTrJ4+fLlEzFLSZI0DY23kvJR4LtJPt++vx44bnKKJEmSJAaofyXZkC649Jmq+lJLvjPJtlV1R3sE7q6WvgzYrmfyuS1tGbD3qPRLRy+rqk4FTgWYN2/ehAStJEnS9DOuFkxVdSbwB8Cd7fMHVfXPk1kwSZKk9dna1r+SBDgNuLGq/q5n1EJg5E1whwHn9aQf2t4mtxdwf3uU7kJg3yRbtJZU+7Y0SZKkVYy7mXVV3QDcMIllkSRJUo+1rH+9BHgLcG2Sq1vaXwDHA+ckOQL4Md0b6gAuAA4AlgC/AN7aln1Pkg8BV7Z8x7Z+oCRJklZhP0qSJEkzSFV9G8gYo/fpk7+AI8eY1wJgwcSVTpIkzVTj7eRbkiRJkiRJ6ssAkyRJkiRJkgYylABTks2TfCHJD5PcmOTFSbZMsijJze3vFi1vkpyUZEmSa5LsNowyS5IkSZIkqb9htWD6GPDVqvpN4AXAjcDRwEVVtTNwUfsO8Gpg5/aZD5wy9cWVJEmSJEnSWKY8wJRkM+BldK/Ppar+q6ruAw4EzmjZzgAOasMHAmdW5zJg8yTbTnGxJUmSJEmSNIZhtGDaEVgOfDrJ95N8KsmTgW2q6o6W56fANm14DnB7z/RLW5okSZIkSZLWAcMIMM0CdgNOqaoXAj9n5eNwwKOvy601mWmS+UkWJ1m8fPnyCSusJEmSJEmSVm8YAaalwNKqurx9/wJdwOnOkUff2t+72vhlwHY9089taY9RVadW1byqmjd79uxJK7wkSZIkSZIea9ZUL7Cqfprk9iTPrqqbgH2AG9rnMOD49ve8NslC4KgkZwN7Avf3PEo34+z+3jMHnsdVJxw6ASWRJEmSJEkanykPMDVvBz6TZCPgFuCtdK2pzklyBPBj4A0t7wXAAcAS4BctryRJkiRJktYRQwkwVdXVwLw+o/bpk7eAIye9UJIkSZIkSVorw+iDSZIkSZIkSTOIASZJkiRJkiQNxACTJEmSJEmSBmKASZIkSZIkSQMxwCRJkiRJkqSBGGCSJEmSJEnSQAwwSZIkSZIkaSCzhl0Arb9+cuzzBp7H9n917QSURJKkmSPJAuD3gLuq6rkt7QPAfweWt2x/UVUXtHHvB44AVgDvqKoLW/r+wMeADYBPVdXxU7kekiRpejHAJEmSNLOcDpwMnDkq/cSq+khvQpJdgEOA5wBPB76e5Flt9MeBVwFLgSuTLKyqGyaz4Fq/7P7e0Yfo9HTVCYcOuwiStE4wwCRJkjSDVNU3k+wwzuwHAmdX1UPArUmWAHu0cUuq6haAJGe3vAaYJElSX/bBJEmStH44Ksk1SRYk2aKlzQFu78mztKWNlb6KJPOTLE6yePny5f2ySJKk9YABJkmSpJnvFOCZwK7AHcBHJ2rGVXVqVc2rqnmzZ8+eqNlKkqRpxkfkJEmSZriqunNkOMk/Aee3r8uA7Xqyzm1prCZdkiRpFbZgkiRJmuGSbNvz9bXAdW14IXBIkicm2RHYGbgCuBLYOcmOSTai6wh84VSWWZIkTS+2YJIkSZpBkpwF7A1snWQpcAywd5JdgQJuA/4EoKquT3IOXefdDwNHVtWKNp+jgAuBDYAFVXX9FK+KJEmaRgwwSZIkzSBV9aY+yaetJv9xwHF90i8ALpjAokmSpBnMR+QkSZIkSZI0EANMkiRJkiRJGogBJkmSJEmSJA3EAJMkSZIkSZIGYoBJkiRJkiRJAzHAJEmSJEmSpIEMLcCUZIMk309yfvu+Y5LLkyxJ8rkkG7X0J7bvS9r4HYZVZkmSJEmSJK1qmC2Y3gnc2PP9w8CJVbUTcC9wREs/Ari3pZ/Y8kmSJEmSJGkdMZQAU5K5wO8Cn2rfA7wC+ELLcgZwUBs+sH2njd+n5ZckSZIkSdI6YFgtmP4e+HPgkfZ9K+C+qnq4fV8KzGnDc4DbAdr4+1t+SZIkSZIkrQOmPMCU5PeAu6rqqgme7/wki5MsXr58+UTOWpIkSZIkSasxjBZMLwFek+Q24Gy6R+M+BmyeZFbLMxdY1oaXAdsBtPGbAXePnmlVnVpV86pq3uzZsyd3DSRJkiRJkvSoKQ8wVdX7q2puVe0AHAJcXFVvBi4BDm7ZDgPOa8ML23fa+IurqqawyJIkSZIkSVqNYb5FbrT3Ae9OsoSuj6XTWvppwFYt/d3A0UMqnyRJkiRJkvqY9fhZJk9VXQpc2oZvAfbok+dXwOuntGCSJEmSJEkat3WpBZMkSZIkSZKmIQNMkiRJkiRJGogBJkmSpBkkyYIkdyW5ridtyySLktzc/m7R0pPkpCRLklyTZLeeaQ5r+W9Ocli/ZUmSJI0wwCRJkjSznA7sPyrtaOCiqtoZuIiVL015NbBz+8wHToEuIAUcA+xJ10fmMSNBKUmSpH4MMEmSJM0gVfVN4J5RyQcCZ7ThM4CDetLPrM5lwOZJtgX2AxZV1T1VdS+wiFWDVpIkSY8ywCRJkjTzbVNVd7ThnwLbtOE5wO09+Za2tLHSV5FkfpLFSRYvX758YkstSZKmjVnDLoAkSZKmTlVVkprA+Z0KnAowb968CZuvJK3rfnLs84ZdhAmz/V9dO+wiaAawBZMkSdLMd2d79I32966WvgzYriff3JY2VrokSVJfBpgkSZJmvoXAyJvgDgPO60k/tL1Nbi/g/vYo3YXAvkm2aJ1779vSJEmS+vIROUmSpBkkyVnA3sDWSZbSvQ3ueOCcJEcAPwbe0LJfABwALAF+AbwVoKruSfIh4MqW79iqGt1xuCRJ0qMMMEmSJM0gVfWmMUbt0ydvAUeOMZ8FwIIJLJokSZrBfEROkiRJkiRJAzHAJEmSJEmSpIH4iJz0OCbi9aO+9lMzwe7vPXPgeVx1wqETUBJJkiRJ6xpbMEmSJEmSJGkgBpgkSZIkSZI0EANMkiRJkiRJGogBJkmSJEmSJA3EAJMkSZIkSZIG4lvk1sCgbxPzTWKSJEmSJGkmsgWTJEmSJEmSBmKASZIkSZIkSQMxwCRJkiRJkqSBTHmAKcl2SS5JckOS65O8s6VvmWRRkpvb3y1aepKclGRJkmuS7DbVZZYkSZIkSdLYhtHJ98PAn1XV95JsClyVZBFwOHBRVR2f5GjgaOB9wKuBndtnT+CU9leSpFXs/t4zB57HVSccOgElkSRJktYfUx5gqqo7gDva8ANJbgTmAAcCe7dsZwCX0gWYDgTOrKoCLkuyeZJt23wkSdPIoG/jBN/IKWndNRHXuHWB11lJ0toYRgumRyXZAXghcDmwTU/Q6KfANm14DnB7z2RLW9pjAkxJ5gPzAbbffvtJK7Ok6cFWLJIkSZI0dYbWyXeSpwBfBN5VVT/rHddaK9WazK+qTq2qeVU1b/bs2RNYUkmSJEmSJK3OUAJMSTakCy59pqq+1JLvTLJtG78tcFdLXwZs1zP53JYmSZKkNZDktiTXJrk6yeKW5otWJEnSwKb8EbkkAU4Dbqyqv+sZtRA4DDi+/T2vJ/2oJGfTde59v/0vSf3Zv40kaRxeXlX/r+f70fiiFUmSNKBh9MH0EuAtwLVJrm5pf0EXWDonyRHAj4E3tHEXAAcAS4BfAG+d2uJKkiTNaL5oRdKUmYh+MtcV52467BJI65ZhvEXu20DGGL1Pn/wFHDmphZIkSWOy0/wZpYCvJSngH6vqVHzRiiRJmgBDfYucJEmSptRLq2pZkqcBi5L8sHdkVVULPo1bC1KdCjBv3rw1mlaSJM0cQ3uLnCRJkqZWVS1rf+8CzgX2wBetSJKkCWCASZIkaT2Q5MlJNh0ZBvYFrmPli1Zg1RetHNreJrcXvmhFkiStho/ISZIkrR+2Ac7tXujLLOCzVfXVJFfii1YkSdKADDBJkiStB6rqFuAFfdLvxhetSJKkARlgkrRGfJuUJEmSJGk0+2CSJEmSJEnSQGzBNAP95NjnDTyP7f/q2gkoiaTxGLRVmC3CNB0M+tvk75IkSdK6zRZMkiRJkiRJGogtmCRJkiRpLU3E0wPrCluLShqELZgkSZIkSZI0EANMkiRJkiRJGogBJkmSJEmSJA3EPpgkTTnfdChJkiRJM4stmCRJkiRJkjQQWzBJkjSKrewkSZKkNWMLJkmSJEmSJA3EAJMkSZIkSZIGYoBJkiRJkiRJAzHAJEmSJEmSpIHYybckTXN2SC1JkiRp2AwwSdIYDNxIkiRJ0vhMm0fkkuyf5KYkS5IcPezySJIkrQ+sg0mSpPGYFi2YkmwAfBx4FbAUuDLJwqq6YbglkyRp7ez+3jMHnse5m54w8DxsZafVsQ4mSZLGa7q0YNoDWFJVt1TVfwFnAwcOuUySJEkznXUwSZI0LqmqYZfhcSU5GNi/qv64fX8LsGdVHdWTZz4wv319NnDTJBRla+D/TcJ8J5rlnFjTpZwwfcpqOSfWdCknTJ+yWs6JNxll/Y2qmj3B81SPdagONl1Mp3NSE8f9vv5y36+/1ud9P2b9a1o8IjceVXUqcOpkLiPJ4qqaN5nLmAiWc2JNl3LC9Cmr5ZxY06WcMH3Kajkn3nQqq9bMVNTBpguP8/WT+3395b5ff7nv+5suj8gtA7br+T63pUmSJGnyWAeTJEnjMl0CTFcCOyfZMclGwCHAwiGXSZIkaaazDiZJksZlWjwiV1UPJzkKuBDYAFhQVdcPoSjTpfm35ZxY06WcMH3Kajkn1nQpJ0yfslrOiTedyqpmHaqDTRce5+sn9/v6y32//nLf9zEtOvmWJEmSJEnSumu6PCInSZIkSZKkdZQBJkmSJEmSJA3EAFOPJIcnOXkNp7kgyeaTVaa1leTYJK8cdjkmW5LNk/zPNZxmhyR/mOT0JAev4bSfSrLLasa/JsnRYyzzuj7plyaZ8tdbJtk7yflteI2P+8ksz4DzOWis/ZPktiRbD7qMNSxP3/2+Lkvy9CRfGGPcuI7X0efl6ua5BuVaZdm953CS30lyfZKrk2wyyLLWoEwPTvL8v53kK5O5jLWR5F1JnrQW06319mrXqaev7fTSmkjyjiQ3JvnMGOPnJTmpDQ/9N1RTo89v24TUXSStW7yuD8YA04Cq6oCqum8Yy04yZiftVfVXVfX1qSzPkGwOrFGACdgB+MO1WVhV/XFV3bCa8Qur6vi1mfd0sbrjbh1wEDBmAHAQw1jvYSyzqv6zqsYVeE2n3+/IY87LseY5wev3ZuD/VtWuVfXLCZyvVvUuYI0DTAM6HDDApKnyP4FXVdWb+42sqsVV9Y61mfE6/huq1VubOueYPBYkzUTrVYApyZeTXNX+yz2/pb01yY+SXAG8pCfv6UlOSXJZklvafykWtP9ond6T77YkW7eWCjcm+ac2/6+N97/oSZ6c5N+S/CDJdUnemGT3JN9o5b0wybYt76VJ/j7JYuAvk/x45Aavzef2JBuO+s/+i5L8e5v/FUk2TbJBkn9M8ssk9yS5M8lnkrwhyf1JHkryH0nmJ/luku+3eTy7zfPwJF9K8tUkNyf52571OSXJ4rYdPtiTfkCSH7Z1OikrW9A8uW3bK9pyDuxZxpeTLGrb+agk7255LkuyJXA8sFOSnyX5adtX1ya5Jl1LhpNaue9q2+oHwOeB3wFeA7yjjb+lZ3vt3bbzF1p5P5MkPdt/XhveP8n32na9qKfMJ7fh309yeZLvA/8CPLHNa3mSW5N8E9gTeGPL/2DPtjp45DhLMjvJF5Nc2T6PHqejjqN++3njJJ9u2+T7SV4+apodgL/u+b4gydIkD7Rt/r1058eCJP+c5DvAta0c1yT5k55t9o0k57VteXySN7dyXJvkmS3f6Uk+2Y6PHyX5vT7rsWXb79e0/fz+Np+HkpyR7th9MN1xe12b1/lJbgT+CDip7fvdW567kywHngq8IGOfVx9r012XZI+W/oGe9f7nsfZFy7egzeeWJL03Hlu04++X6c7PJyfZNcmKto7nJjmsbZtLk/xr20ZXAXe3+T+nbYOr2zQ7t/Q/6kn/xyQbjBxLSU5Idw5+PckePWV7zci+T/Ktto+/l+S3e9Kva8ObJDk73bXtXGATYNskNyU5E7gO+D89x8PI+X488MxWrhNGzfPwJAuTXAxclFXP/zekux5ek+S+dMfjJcBubR88ut96jpk/Bt4AfCh9WhwMuJ0OT3dcX5ruWndMn/mnzee6dMf7yDl9ZpKDevJ9Od11aIOW//Z0196bkny4lfHeJD9J8nVg055pfzfdtXjrdqyMXNt6r11jlePjPetzbpIFbfhtSY7Lan6/surv0zF0gZ5L2n5Z3bVrx1bma5M8ep1p4947+rgZqxxt/eYBn8kUtlDT+inJJ4FnAF9J8r70rwP1bbmSUS2jR86Nlv9bSRYCN/RcAx7zW6p1S7o653Xt8y5G/ba1bE9J//rieOrx7xzOmqnPb9sb09PSPV0rxUvb8FOysi59TZLXtfR+9wFj3dOsUo/rV4YhbY4ZL6OeKEjynnR190uzsv71oyS/02faodS/prWqWm8+wJbt7yZ0N0ZzgJ8As4GNgO8AJ7c8pwNnAwEOBH4GPI8uKHcVsGvLdxuwNV2rmId70s8B/mic5Xod8E893zcD/h2Y3b6/ke61wACXAp/oyXse8PKefJ/qKf/Bbb1uAV7U0p8KzALmAx9pZd4NWAxcC1wB/GVb5y/T3UjMatO+EvhiGz68zXczYGPgx8B2o7bzBq28z295bgd2bOPOAs5vw38zsq3o/jv0I+DJbRlL6G6yZgP3A3/a8p1I91/0HYAHgZ2BfYFzgYvbfroduAR4bttPt7Rpfx84v22jz7e8uwBL2vi927LmtnHfBV7as/3ntfL0rs+WPdtl5BjaAh59U+P7gKILYn4AuLOlfQe4D9gQeLBnvx4MnN6GP9uz/O2BG/scQ2Pt5z9j5bHzm3TH+8ZtHc9v229pT5n/TyvfpXSvpP4ocECb91XAkcD/bnmfSHfc7Njmdx+wbUtfBnyw5Xsn8Pc9x+VX23bduS370fK0PP8AHNOzPe9v2+cY4NvAoW1bfqfluQE4tQ1/Hljahk8CbgU+Afxum+YKxj6v/qkNvwy4rg1/oK33JqvbFy3fv7d135ouMLQhsA/wCDCv5VsCfBK4BvhFSzuWlcfjpcC/tu9PA27r2SZv7tnXmwC/1fJu2NI/ARzahgt4dRs+F/haK88LgKtb+pOAjdvwzsDiNrxDz/q/u2cbPZ/uevH7bZ32ojvnTqW7Tj6hlftlvfPoM8/D6fb7yDkz+vz/T+DTI8umu8Zc3ZY9b9R+Ox04ePTwqHNj0O10OHAHsBUrfztG9ueDPdfwRXTXvG3ozrNtgf8GfLnnun57m34+8L+B9wAfAh4A/gn4A+B7wNfprr2/AL4CvBb4FrBFz7r2u3aNVY5DgBNaniuAy9rwp4H9WM3vF/1/n24Dtu5JG+vatbBnWx/Zs71Wd9yMVY5LR7a7Hz+T/WFl3e6p9K8D7c3K36zDeWzd8eCe+TzYk//nrKwzzKfPb+mw19vPY46B3enqxU8GngJcD7yQx/627U2f+iLd78i46vF+hrZ/V/vbRlffuLQNf5hWj23ft2Ds+4Cx7mn61eNWKcOwt8tM/bBqvfQ9rLzf+WhLOwD4ehs+HDiZIda/pvNnfWua+Y4kr23D2wFvobt4LAdI8jngWT35/7WqKsm1wJ1VdW3Ldz3dAXH1qPnfWlUjaVe1PONxLfDRJB+mq2jfSxcUWdT+EbIB3Q3OiM+NGn4jXSDlELqbp17PBu6oqisBqupnbR32pQssrWDlTdwSuh/HtwL/RndjtgHw+XQtJoruR3PERVV1f5vfDcBv0F1s35CuhdgsupNrF7oT8ZaqurVNexZdBQu6m43XJHlP+74x3c07wCVV9QDwQJL76W4UR7bZ8+lukp9Ed7I/ne5i/gjdTdpmwM3Ay4HPACOtSh7oWYcvV9UjdP9R3KYn/YqqWtrW7Wq6ffntnvF7Ad8cWZ+quodVzQU+1/5r9STgV1X1nSSvotvPLwZ+Tbe/t+kz/YhXAru0YwHgqUmeUlW9/ZmMtZ9fSvejRlX9MMmPeewxPtpmdBfTbeiCJAX8Ld12PYNuWz6/5z+0m9EFJ/4LuLKq7mjL/Q+6m3Xo9lVvy6lz2ja/OcktdIGvXi+lu1BDV6l7Ct35tAHwTLqb70eA/9vyzAFel67V0fZAJXkK3Q3rncDnquobSX5Gd0yPdV6d1bbTN5M8NSv7VltYKx+56rsv2vC/VdVDwENJ7mrb8CWtrJ9q02xGt69GjlPadv0j4Js9Zfl8z3jozsu/TDIX+FJV3ZxkH7oK8JVt3psAd7X8/0UXyINu+z9UVb9u17IdWvqGwMlJdqW7DvQ7Ll5GF6ijqq5Jck1L/3FVXZbkI3Tn7/db+lPojoef9JlXr0U958zo838k7SXAx+mu1c9o6/SZ9rd3vz2eQbfTSHlHWpN9ie4YXdwz/qXAWVW1ArgzyTfogr0Lk3wiyWy6Y/orwG+39Xs+XaVzA7pr5fV02/tM4Kiq+s92XX0uXTB635Hzuul37epbDrrK0bvS9U92A12rum3prkHvoAuejfX79Zjfp6r6Vs/x/3hewspz+Z/pKum09R/ruFnb31FpMmwGnDFGHWhNXdFTB9qX/r+lt/adUsPwUuDcqvo5PHrtX6V1A/3ri/cx/nq8hmNNftuh/ce9AAAgAElEQVReSXePBUBV3Zvk9+l/HzDWPU2/etwqZZjgddT4fKn9HV3neAVdoHFY9a9pa70JMCXZm+4C8eKq+kW6Zo8/ZPX9tTzU/j7SMzzyvd+2682zgu5G5nFV1Y+S7EYXOf1ruhY411fVi8eY5Oc9wwuBv0n3uNjubdrxCF3k9s+ralfomnbTtaz4JF0rke3pWmx8vqpem+5xqkt75jF6fWcl2ZEuKvyidgE+ne7i+nhleV1V3fSYxGRPVt3uvftkFl3gakVV7Zrko8CPquofe9bna3RBrpHljNY7/4yRvoK1O1f+Afi7dpN5CF3Ue8Sv6SqrvfOvnvG92+wJwF5V9avemSe5kC6IsRj42FqUD7qoee96vw64jG6bnQW8qZXvCXTHXYC3V9WFo8qyN4+/r0b0rme/74+ZdVvuS6vqZ+k6PP5mS/9KT54LquotbZ+fX1UP9lQUes+XH1bVi8ZY1ljl6p1+rH0B/Y+ZAPf2nGPvoQuIPXPU8kYKO7I/fk7PMVBVn01yOV1LrAvSPU4R4Iyqen+fdfl1VY3M/9F9UVWPZGWfD/+LLgD3grZev1plLmMb2Sah6/foH3tHtmvFeKYfmcdjzv92Pfsa8Da6YNz1dP/ln19VvYGdVSTZjpWB6E8y+HaCNTtmRzuTLoB4CHA0XSUkwNvpKh+z6P4L/m26SsXo8+VOulacz+KxQa2xrl2rqKplLWC6P935syXdI4UPVtUDSbZijN+v0b9PaY8BjF5Ez/Do632/bbW642atfkelSfIhun909asD9fMwrfuJdN0XbNQzbvR1b5XfUk1LY/32j7ceryEY47ft0fOXx793GUvfexrgxtH1uKq6eHQZqurYtVyuVq9338Jj9+/IOTz6fu8/6P7BOZT613S2PvXBtBndjd4vkvwmXQuUTYD/lmSrJBsCrx9GwdK9GecXVfUvwAl0/fLMTvLiNn7DJM/pN21rxXIlXYDh/BY57XUTXZ8pL2rz2rTdOF1Id9MzUoZn0Z1UW9Pd0JxN17JmW7rHnaBrLvh4nkr3w3l/i+q+uqccz+i58ex9zvhC4O3Jo8+tv3AcyxlxB7AiyevbfN6WZK827kmtPBfz2H0bevo2WUuXAS9rAbWRG+LRNmPltnsdXR9MI5WNF/LYFlHQRbx/q1VKX9uT/jW6m1HasnYFqKr9quvQ+I8Zez9/i67z45F9vH3L++gy6bbRxkmeSPdY1kiU/nf7rNOFwP9o5wtJnpXkyX3yrc7rkzwhXb9MzxhVHnrL3MqyESt/CM6mC3w93HOs/4CVraAeYGXQ+Ju0lmFJXt3Wc6vVnFcjz0u/FLh/pHXeKH33xWp8h66V09Pa95EfjXvpzpHfogvmPtzSb6P7bwZ0jxqNLOcZdC0AT6J7LPb5wEXAwSPzTtd31W88Tnl6bUbX6u0RutacG/TJ801ah/hJntuW22vknHtKyzOnlecBxn+OjT7/96V7NOyzwE/prodzRpa9uushQFXd3s6LXavqkwy+nQBe1abbhK4j+e+MGv8t4I3p+lWZTdcS6Yo27nS6x3mhO+ef1v4eSfe4IXTHxcZ02/ugVs5t6VrcLae7fpy5uvUeRzkua+X4Zsv3nvZ3tfr8Pu3Gqvt3rGvXd1j5X9/ezpLHOm5WZ02OKWmi9P6OHz6O/LfR/bMPun4ex2rxNBG/pZpc3wIOSvKktm9eS3dNG8916CbGWY/XcIzx23YbK8/f1/VkX0T3mz0y7RaMfR/Q956mXz1ujDJoctwJPK3d8z8RWKUP2D5+zBDrX9PZetOCie4xiD9N1xHwTXQ7+w66VjzfpWvOOvqRtwmR5E8B2s1OP88DTkjyCF3Llv9Bd8N5UpLN6PbT39P9F7+fz9E9UrP36BFV9V/pOhr7h3Zz9Eu6llyfomu58LJ0nZ4tp+v75Dl0N+yhq1gdBRyf5H/TPTa3WlX1g3SdWv+Q7nG577T0X6Z7tetXk/ycLig24kNt/a5pNyi3Mr4Tn6q6O11LngVt3R4BzsvKR5S+UVXXJzkOWJCuk++r6SLEr6H77/oavz69qpanewzwS63MdwGvGpXtA3SPF95LF/n+Jd0P1H50x94prLzBhK51w/l0+2Ix3WMj0DWh/Hi6x5Nm0V2g/nRUecbaz58ATknXDPdh4PCqeqj97tEeBzqPLgD3LLrz4g1tOefSBWV6fYqulcX32o/nctoN8Rr4Cd1F96l0fWr9Ko9tlvwBun11DV2g4f8AX2vb+ddtu/y6J/8C4K9a/k2BzdI9CvvHdEHUz9EFGX9C1yLmw2OcV79qx+6GLV8/j7svRllC96M2Uv4t6a5Fh9E9Z301XUD2i225H6Hrt+EEusfBRrwBeEuSX9MFXf6mqu5p52XvtjmS7gdxPD4BfDHJoa1M/f6jegrw6XbdvJGu6e6jquprLUj23bYPH6R7dvw/knynXVu+Qveo21hGn/8P0gVciu4x05/RPeq6OV3Q6des/nr4GFV1w4DbCbrj9YutPP/SpxXVuXTNnX/Qyv3nVfXTtvw72/b7cjvfjqXrl2xDuuN1xza8QZvPAXSPYZxJd/yMPN76Zrrrye8ztjHLQVeZ2beqlqR7VHZLxlfB6ff79GK6a/l/VtXLGfva9U7gs0neR1ehpq1P3+OG7ro8ltOBTyb5JV1LZN8UqKnwt3SPyI2rDkTXl9p5ra4x1nUVJua3VJOoqr6XrlX0yE3ip6rqqlG/bX2PiVYnO5jx1+M19fr9tm0CnJbkQzy2teJf09X9rqP7nfpgVX1pjPuAse5pVqnH0bViHl0GTYKe+tcVdP80+OE4pxtm/WvaGumAWJp0af0GtcrUx4Gbq+rEYZdLUycrH2Fb46Bem34ecGJV9esHYZByXQq85/Eev9L6J8nhdJ1LH7WW0z+Jrq+H3cZoFSdJkiTNCOvTI3Iavv+ergPE6+laR/3j4+SXHpXkaLpWJP360pHWOUleSdfy6x8MLkmSJGmmswWTJEmSJEmSBmILJkmSJEmSJA3EAJMkSZIkSZIGYoBJkiRJkiRJAzHAJGlaSfLvqxm3d5Lzp7I8kiRJ64NWz/rtYZdD0rrLAJOkaaWqrNhIkiRNvb0B62GSxmSASdK0kuTBdE5Icl2Sa5O8sSfLU5P8W5Kbknwyidc5SZK03kqyQ5IfJvlMkhuTfCHJk5Lsk+T7rS61IMkTW/7bkmzdhucluTTJDsCfAv8rydVJfifJNknOTfKD9vntNs27Wx3tuiTvGlWG05P8qJXllUm+k+TmJHu0fE9uZbmile3AYWwzSWvHGy9J09EfALsCLwBeCZyQZNs2bg/g7cAuwDNbXkmSpPXZs4FPVNVvAT8D3g2cDryxqp4HzAL+x1gTV9VtwCeBE6tq16r6FnAS8I2qegGwG3B9kt2BtwJ7AnsB/z3JC9tsdgI+Cvxm+/wh8FLgPcBftDx/CVxcVXsAL6er4z15QraApElngEnSdPRS4KyqWlFVdwLfAF7Uxl1RVbdU1QrgrJZXkiRpfXZ7VX2nDf8LsA9wa1X9qKWdAbxsDef5CuAUgFYnu5+u3nVuVf28qh4EvgT8Tst/a1VdW1WPANcDF1VVAdcCO7Q8+wJHJ7kauBTYGNh+DcslaUhmDbsAkjTB6nG+S5IkrW9G14fuA7YaI+/DrGyIsPEEluGhnuFHer4/wsr70gCvq6qbJnC5kqaILZgkTUffAt6YZIMks+n+43ZFG7dHkh1b30tvBL49rEJKkiStI7ZP8uI2/IfAYmCHJDu1tLfQtQgHuA3YvQ2/rmceDwCb9ny/iPZYXauTbUZXRzuo9fH0ZOC1LW28LgTeniRtvi98nPyS1iEGmCRNNwWcC1wD/AC4GPjzqvppG38lcDJwI3BryytJkrQ+uwk4MsmNwBbAiXR9JX0+ybV0rYg+2fJ+EPhYksXAip55/Cvw2pFOvoF3Ai9v018F7FJV36Pr2+kK4HLgU1X1/TUo54eADYFrklzfvkuaJtI99ipJ674kWwHfq6rfGHZZJEmSpoP2Brjzq+q5Qy6KpBnOFkySpoUkTwe+C3xk2GWRJEmSJD2WLZgkSZIkSZI0EFswSZIkSZIkaSAGmCRJkiRJkjQQA0ySJEmSJEkaiAEmSZIkSZIkDcQAkyRJkiRJkgZigEmSJEmSJEkDMcAkSZIkSZKkgRhgkiRJkiRJ0kAMMEmSJEmSJGkgBpgkSZIkSZI0EANMkiRJkiRJGogBJkmSJEmSJA3EAJMkSZIkSZIGYoBJkiRJkiTp/2fv/qOsKs8E338fBCVRW1ErjKFMCqP5gahICkXtsKalBUK6xTUdczU/BMWQeyXRrEwbcSZX44/MTdq5cdJJxzt2i0J+iF57cnWMIRA7tppRsSAgCkmoVtQiRBDQjmZpIv3cP84LKbEKC06dc6hT389aZ5293/3ufZ5da1Wtp57z7vdVVYY2OoBaOOKII7Ktra3RYUiSpBpavnz5C5nZ0ug49EfmYJIkNbfd5V9NWWBqa2ujo6Oj0WFIkqQaiohnGh2D3sgcTJKk5ra7/MtH5CRJkiRJklQVC0ySJEmSJEmqigUmSZIkSZIkVaUp52CSJGmw+sMf/kBXVxevvvpqo0PpN8OHD6e1tZVhw4Y1OhRJkqQeNVsOtjf5lwUmSZKaSFdXFwcffDBtbW1ERKPDqVpmsmXLFrq6uhg9enSjw5EkSepRM+Vge5t/+YicJElN5NVXX+Xwww8f8InNDhHB4Ycf3jTfBkqSpObUTDnY3uZfFpgkSWoyzZDYdNds9yNJkppTM+Use3MvFpgkSZIkSZJUFQtMkiRJkiRJqoqTfFfhg5ctbHQI2sXy689vdAiStE/LTDKTIUP8jkkDlznYvsccTJJ2bzDkYM17Z5IkCYD169fzvve9j/PPP5+xY8dy7bXXMmHCBE444QSuuuoqAF555RU+8pGPcOKJJzJ27Fhuv/12ANra2vjiF7/I8ccfz8knn0xnZ+fOa55xxhmccMIJTJ48mWeffRaAWbNmcckll3Daaadx9NFHc+eddwKwceNGJk2axLhx4xg7diwPPvggAEuWLOHUU09l/PjxnHPOObz88sv1/vFIkiTVxGDLwSwwSZI0CKxbt46LL76YG264gQ0bNrBs2TJWrlzJ8uXLeeCBB1i8eDHvfOc7WbVqFU888QTTpk3bee4hhxzC6tWr+exnP8vnP/95AD73uc8xc+ZMHn/8cT7xiU9wySWX7Oy/ceNGHnroIe655x7mzZsHwPe//32mTp3KypUrWbVqFePGjeOFF17guuuu4yc/+QkrVqygvb2dr3/96/X9wUiSJNXQYMrBLDBJkjQIvPvd72bixIksWbKEJUuWcNJJJzF+/Hh+8YtfsG7dOo4//niWLl3K5ZdfzoMPPsghhxyy89zzzjtv5/vDDz8MwMMPP8zHP/5xAD71qU/x0EMP7ex/9tlnM2TIEMaMGcPzzz8PwIQJE7jlllv48pe/zOrVqzn44IN55JFHWLNmDaeffjrjxo1jwYIFPPPMM/X6kUiSJNXcYMrBnINJkqRB4MADDwQqz/9fccUVfOYzn3lTnxUrVnDvvffypS99icmTJ3PllVcCb1ymti9L1h5wwAE7tzMTgEmTJvHAAw/wwx/+kFmzZvGFL3yBESNGcOaZZ3LbbbdVdW+SJEn7qsGUgzmCSZKkQWTq1KnMnz9/53P2GzZsYNOmTfz617/m7W9/O5/85Ce57LLLWLFixc5zdswFcPvtt3PqqacCcNppp7Fo0SIAvve97/GhD31ot5/7zDPPMHLkSD796U9z0UUXsWLFCiZOnMjPfvaznXMKvPLKK/zqV7/q93uWJElqtMGQg9V0BFNErAd+C2wHXs/M9og4DLgdaAPWAx/LzG1RKcd9A5gO/A6YlZkrynVmAl8ql70uMxfUMm5JkprVlClTWLt27c4k5aCDDuK73/0unZ2dXHbZZQwZMoRhw4Zx44037jxn27ZtnHDCCRxwwAE7v+n65je/yQUXXMD1119PS0sLt9xyy24/9/777+f6669n2LBhHHTQQSxcuJCWlhZuvfVWzjvvPF577TUArrvuOt773vfW6O4lSZIaYzDkYLFj2FQtlAJTe2a+0K3tb4CtmfnViJgHjMjMyyNiOvA5KgWmU4BvZOYppSDVAbQDCSwHPpiZ23r73Pb29uzo6KjZfe3gErn7HpfIlTTYrV27lg984AP9dr22tjY6Ojo44ogj+u2ae6On+4qI5ZnZ3qCQ9mkRcSjwD8BYKvnThcAvqfGXfOZgg5c5mKTBrhlzsD3NvxrxiNwMYEdysgA4u1v7wqx4BDg0Io4EpgJLM3NrKSotBabtelFJkiTt9A1gcWa+HzgRWAvMA+7LzGOB+8o+wIeBY8trDnAjQPmS7yoqX/ydDFwVESPqeROSJGngqHWBKYElEbE8IuaUtpGZubFs/wYYWbZHAc91O7ertPXW/gYRMSciOiKiY/Pmzf15D5IkDVrr169v+Ogl7ZmIOASYBNwMkJm/z8wX8Us+SZIGjIGYg9W6wPSnmTmeyjdjcyNiUveDWXk+r1+e0cvMmzKzPTPbW1pa+uOSkiRJA9FoYDNwS0T8PCL+ISIOpEZf8kmSJEGNC0yZuaG8bwJ+QGV49fPlWzHK+6bSfQNwVLfTW0tbb+2SJEl6s6HAeODGzDwJeIU/Pg4H9O+XfI4ilyRJUMMCU0QcGBEH79gGpgBPAHcDM0u3mcBdZftu4PyomAi8VL5l+zEwJSJGlOf+p5Q2SZIkvVkX0JWZj5b9O6kUnGryJZ+jyCVJEtR2BNNI4KGIWAUsA36YmYuBrwJnRsQ64M/LPsC9wFNAJ/D3wMUAmbkVuBZ4rLyuKW2SJEnaRWb+BnguIt5XmiYDa/BLPkmSVENDa3XhzHyKyqolu7ZvoZLo7NqewNxerjUfmN/fMUqSNFj19zLvfV2ifPHixVx66aVs376diy66iHnz5r31SdobnwO+FxH7U/kC7wIqXyzeERGzgWeAj5W+9wLTqXzJ97vSl8zcGhE7vuQDv+STJKlqjcjB6pV/1azAJEmS1N327duZO3cuS5cupbW1lQkTJnDWWWcxZsyYRofWdDJzJdDewyG/5JMkaRCpZ/5V61XkJEmSAFi2bBnHHHMMRx99NPvvvz/nnnsud91111ufKEmSpL1Sz/zLApMkSaqLDRs2cNRRf5wzurW1lQ0bXBhWkiSpVuqZf1lgkiRJkiRJUlUsMEmSpLoYNWoUzz333M79rq4uRo0a1cCIJEmSmls98y8LTJIkqS4mTJjAunXrePrpp/n973/PokWLOOussxodliRJUtOqZ/7lKnKSJA1CfVnStr8NHTqUb33rW0ydOpXt27dz4YUXctxxx9U9DkmSpEapdw5Wz/zLApMkSaqb6dOnM3369EaHIUmSNGjUK//yETlJkiRJkiRVxQKTJEmSJEmSqmKBSZIkSZIkSVWxwCRJkiRJkqSqWGCSJEmSJElSVSwwSZIkSZIkqSpDGx2AJEmqv2evOb5fr/euK1e/ZZ8LL7yQe+65h3e84x088cQT/fr5kiRJA0Ez52COYJIkSXUxa9YsFi9e3OgwJEmSBpV65WAWmCRJUl1MmjSJww47rNFhSJIkDSr1ysEsMEmSJEmSJKkqFpgkSZIkSZJUFQtMkiRJkiRJqooFJkmSJEmSJFVlaKMDkCRJ9deXJW3723nnncf999/PCy+8QGtrK1dffTWzZ8+uexySJEmN0sw5mAUmSZJUF7fddlujQ5AkSRp06pWD+YicJEmSJEmSqmKBSZIkSZIkSVWxwCRJUpPJzEaH0K+a7X4kSVJzaqacZW/uxQKTJElNZPjw4WzZsqVpEpzMZMuWLQwfPrzRoUiSJPWqmXKwvc2/nORbkqQm0traSldXF5s3b250KP1m+PDhtLa2NjoMSZKkXjVbDrY3+ZcFJkmSmsiwYcMYPXp0o8NQg0XEeuC3wHbg9cxsj4jDgNuBNmA98LHM3BYRAXwDmA78DpiVmSvKdWYCXyqXvS4zF9TzPiRJGijMwXxETpIkqVn9WWaOy8z2sj8PuC8zjwXuK/sAHwaOLa85wI0ApSB1FXAKcDJwVUSMqGP8kiRpALHAJEmSNDjMAHaMQFoAnN2tfWFWPAIcGhFHAlOBpZm5NTO3AUuBafUOWpIkDQwWmCRJkppPAksiYnlEzCltIzNzY9n+DTCybI8Cnut2bldp6639DSJiTkR0RERHs8w7IUmS9pxzMEmSJDWfP83MDRHxDmBpRPyi+8HMzIjol2VuMvMm4CaA9vb2gb90jiRJ2iuOYJIkSWoymbmhvG8CfkBlDqXny6NvlPdNpfsG4Khup7eWtt7aJUmS3qTmBaaI2C8ifh4R95T90RHxaER0RsTtEbF/aT+g7HeW423drnFFaf9lREytdcySJEkDVUQcGBEH79gGpgBPAHcDM0u3mcBdZftu4PyomAi8VB6l+zEwJSJGlMm9p5Q2SZKkN6nHCKZLgbXd9r8G3JCZxwDbgNmlfTawrbTfUPoREWOAc4HjqEws+e2I2K8OcUuSJA1EI4GHImIVsAz4YWYuBr4KnBkR64A/L/sA9wJPAZ3A3wMXA2TmVuBa4LHyuqa0SZIkvUlN52CKiFbgI8BXgC9ERABnAB8vXRYAX6ayHO6Msg1wJ/Ct0n8GsCgzXwOejohOKsO8H65l7JIkSQNRZj4FnNhD+xZgcg/tCczt5Vrzgfn9HaMkSWo+tR7B9N+ALwL/VvYPB17MzNfLfvfVSHauVFKOv1T6u4KJJEmSJEnSPqxmBaaI+AtgU2Yur9VndJeZN2Vme2a2t7S01OMjJUmSJEmSRG0fkTsdOCsipgPDgT8BvgEcGhFDyyil7quR7FippCsihgKHAFtwBRNJkiRJkqR9Ws1GMGXmFZnZmpltVCbp/qfM/ATwU+CjpduuK5jsWNnko6V/lvZzyypzo4FjqUxYKUmSJEmSpH1ATSf57sXlwKKIuA74OXBzab8Z+E6ZxHsrlaIUmflkRNwBrAFeB+Zm5vb6hy1JkiRJkqSe1KXAlJn3A/eX7aeorAK3a59XgXN6Of8rVFaikyRJkiRJ0j6m1qvISZIkSZIkqclZYJIkSZIkSVJVLDBJkiRJkiSpKhaYJEmSJEmSVBULTJIkSZIkSaqKBSZJkiRJkiRVxQKTJEmSJEmSqmKBSZIkSZIkSVWxwCRJkiRJkqSqWGCSJEmSJElSVSwwSZIkSZIkqSoWmCRJkiRJklQVC0ySJEmSJEmqigUmSZIkSZIkVcUCkyRJkiRJkqpigUmSJEmSJElVscAkSZIkSZKkqlhgkiRJkiRJUlUsMEmSJEmSJKkqFpgkSZIkSZJUFQtMkiRJTSgi9ouIn0fEPWV/dEQ8GhGdEXF7ROxf2g8o+53leFu3a1xR2n8ZEVMbcyeSJGkgsMAkSZLUnC4F1nbb/xpwQ2YeA2wDZpf22cC20n5D6UdEjAHOBY4DpgHfjoj96hS7JEkaYCwwSZIkNZmIaAU+AvxD2Q/gDODO0mUBcHbZnlH2Kccnl/4zgEWZ+VpmPg10AifX5w4kSdJAY4FJkiSp+fw34IvAv5X9w4EXM/P1st8FjCrbo4DnAMrxl0r/ne09nCNJkvQGFpgkSZKaSET8BbApM5fX6fPmRERHRHRs3ry5Hh8pSZL2QRaYJEmSmsvpwFkRsR5YROXRuG8Ah0bE0NKnFdhQtjcARwGU44cAW7q393DOTpl5U2a2Z2Z7S0tL/9+NJEkaECwwSZIkNZHMvCIzWzOzjcok3f+UmZ8Afgp8tHSbCdxVtu8u+5Tj/5SZWdrPLavMjQaOBZbV6TYkSdIAM/Stu0iSJKkJXA4siojrgJ8DN5f2m4HvREQnsJVKUYrMfDIi7gDWAK8DczNze/3DliRJA4EFJkmSpCaVmfcD95ftp+hhFbjMfBU4p5fzvwJ8pXYRSpKkZuEjcpIkSZIkSaqKBSZJkiRJkiRVxQKTJEmSJEmSqmKBSZIkSZIkSVWpWYEpIoZHxLKIWBURT0bE1aV9dEQ8GhGdEXF7ROxf2g8o+53leFu3a11R2n8ZEVNrFbMkSZIkSZL2XC1HML0GnJGZJwLjgGkRMRH4GnBDZh4DbANml/6zgW2l/YbSj4gYQ2W53OOAacC3I2K/GsYtSZIkSZKkPVCzAlNWvFx2h5VXAmcAd5b2BcDZZXtG2accnxwRUdoXZeZrmfk00EkPS+xKkiRJkiSpMWo6B1NE7BcRK4FNwFLgX4AXM/P10qULGFW2RwHPAZTjLwGHd2/v4RxJkiRJkiQ1WJ8KTBFxX1/adpWZ2zNzHNBKZdTR+/c4wj6KiDkR0RERHZs3b67Vx0iSJNXN3uZgkiRJ9TZ0dwcjYjjwduCIiBgBRDn0J+zBKKLMfDEifgqcChwaEUPLKKVWYEPptgE4CuiKiKHAIcCWbu07dD+n+2fcBNwE0N7enn2NTZIkaV/TXzmYJElSvbzVCKbPAMupjDxa3u11F/Ct3Z0YES0RcWjZfhtwJrAW+Cnw0dJtZrkWwN1ln3L8nzIzS/u5ZZW50cCxwLK+3qAkSdIAtNc5mCRJUiPsdgRTZn4D+EZEfC4zv7mH1z4SWFBWfBsC3JGZ90TEGmBRRFwH/By4ufS/GfhORHQCW6msHEdmPhkRdwBrgNeBuZm5fQ9jkSRJGjCqzMEkSZLqbrcFph0y85sRcRrQ1v2czFy4m3MeB07qof0pelgFLjNfBc7p5VpfAb7Sl1glSZKaxd7kYJIkSY3QpwJTRHwHeA+wEtgxeigBkxtJkqQaMQeTJEkDRZ8KTEA7MKbMiSRJkqT6MAeTJEkDwltN8r3DE8C/q2UgkiRJehNzMEmSNCD0dQTTEcCaiFgGvLajMTPPqklUkiRJAnMwSZI0QPS1wPTlWgYhSZKkHn250QFIkiT1RV9XkfvnWgciSZKkNzIHkyRJA05+IkgAACAASURBVEVfV5H7LZUVSwD2B4YBr2Tmn9QqMEmSpMHOHEySJA0UfR3BdPCO7YgIYAYwsVZBSZIkyRxMkiQNHH1dRW6nrPj/gKk1iEeSJEk9MAeTJEn7sr4+Ivcfuu0OAdqBV2sSkSRJkgBzMEmSNHD0dRW5v+y2/TqwnsoQbUmSJNWOOZgkSRoQ+joH0wW1DkSSJElvZA4mSZIGij7NwRQRrRHxg4jYVF7/GBGttQ5OkiRpMDMHkyRJA0VfJ/m+BbgbeGd5/c/SJkmSpNoxB5MkSQNCXwtMLZl5S2a+Xl63Ai01jEuSJEl7kYNFxPCIWBYRqyLiyYi4urSPjohHI6IzIm6PiP1L+wFlv7Mcb+t2rStK+y8jwtXrJElSr/paYNoSEZ+MiP3K65PAlloGJkmSpL3KwV4DzsjME4FxwLSImAh8DbghM48BtgGzS//ZwLbSfkPpR0SMAc4FjgOmAd+OiP36+f4kSVKT6GuB6ULgY8BvgI3AR4FZNYpJkiRJFXucg2XFy2V3WHklcAZwZ2lfAJxdtmeUfcrxyRERpX1RZr6WmU8DncDJ/XBPkiSpCfW1wHQNMDMzWzLzHVSSnatrF5YkSZLYyxysjHZaCWwClgL/AryYma+XLl3AqLI9CngOoBx/CTi8e3sP53T/rDkR0RERHZs3b96LW5QkSc2grwWmEzJz246dzNwKnFSbkCRJklTsVQ6WmdszcxzQSmXU0ftrFWBm3pSZ7ZnZ3tLiFJ2SJA1WfS0wDYmIETt2IuIwYGhtQpIkSVJRVQ6WmS8CPwVOBQ6NiB3ntgIbyvYG4Khy/aHAIVTmedrZ3sM5kiRJb9DXAtP/DTwcEddGxLXA/wL+pnZhSZIkib3IwSKiJSIOLdtvA84E1lIpNH20dJsJ3FW27y77lOP/lJlZ2s8tq8yNBo4FlvXbnUmSpKbSp2/AMnNhRHRQmRwS4D9k5prahSVJkqS9zMGOBBaUFd+GAHdk5j0RsQZYFBHXAT8Hbi79bwa+ExGdwFYqK8eRmU9GxB3AGuB1YG5mbu/P+5MkSc1jT4ZYr6GSYEiSJKlO9jQHy8zH6WGepsx8ih5WgcvMV4FzernWV4Cv9DlYSZI0aPX1ETlJkiRJkiSpRxaYJEmSJEmSVBULTJIkSZIkSaqKBSZJkiRJkiRVxQKTJEmSJEmSqmKBSZIkSZIkSVWxwCRJkiRJkqSqWGCSJEmSJElSVSwwSZIkSZIkqSoWmCRJkiRJklQVC0ySJEmSJEmqSs0KTBFxVET8NCLWRMSTEXFpaT8sIpZGxLryPqK0R0T8bUR0RsTjETG+27Vmlv7rImJmrWKWJEmSJEnSnqvlCKbXgf+YmWOAicDciBgDzAPuy8xjgfvKPsCHgWPLaw5wI1QKUsBVwCnAycBVO4pSkiRJkiRJaryaFZgyc2NmrijbvwXWAqOAGcCC0m0BcHbZngEszIpHgEMj4khgKrA0M7dm5jZgKTCtVnFLkiRJkiRpz9RlDqaIaANOAh4FRmbmxnLoN8DIsj0KeK7baV2lrbd2SZIkSZIk7QNqXmCKiIOAfwQ+n5n/2v1YZiaQ/fQ5cyKiIyI6Nm/e3B+XlCRJkiRJUh/UtMAUEcOoFJe+l5n/ozQ/Xx59o7xvKu0bgKO6nd5a2nprf4PMvCkz2zOzvaWlpX9vRJIkSZIkSb2q5SpyAdwMrM3Mr3c7dDewYyW4mcBd3drPL6vJTQReKo/S/RiYEhEjyuTeU0qbJEmSJEmS9gFDa3jt04FPAasjYmVp+0/AV4E7ImI28AzwsXLsXmA60An8DrgAIDO3RsS1wGOl3zWZubWGcUuSJEmSJGkP1KzAlJkPAdHL4ck99E9gbi/Xmg/M77/oJEmSJEmS1F/qsoqcJEmSJEmSmpcFJkmSJEmSJFXFApMkSZIkSZKqYoFJkiRJkiRJVbHAJEmSJEmSpKpYYJIkSZIkSVJVLDBJkiQ1kYg4KiJ+GhFrIuLJiLi0tB8WEUsjYl15H1HaIyL+NiI6I+LxiBjf7VozS/91ETGzUfckSZL2fRaYJEmSmsvrwH/MzDHARGBuRIwB5gH3ZeaxwH1lH+DDwLHlNQe4ESoFKeAq4BTgZOCqHUUpSZKkXVlgkiRJaiKZuTEzV5Tt3wJrgVHADGBB6bYAOLtszwAWZsUjwKERcSQwFViamVszcxuwFJhWx1uRJEkDiAUmSZKkJhURbcBJwKPAyMzcWA79BhhZtkcBz3U7rau09dYuSZL0JhaYJEmSmlBEHAT8I/D5zPzX7scyM4Hsp8+ZExEdEdGxefPm/rikJEkagCwwSZIkNZmIGEaluPS9zPwfpfn58ugb5X1Tad8AHNXt9NbS1lv7G2TmTZnZnpntLS0t/XsjkiRpwLDAJEmS1EQiIoCbgbWZ+fVuh+4GdqwENxO4q1v7+WU1uYnAS+VRuh8DUyJiRJnce0ppkyRJepOhjQ5AkiRJ/ep04FPA6ohYWdr+E/BV4I6ImA08A3ysHLsXmA50Ar8DLgDIzK0RcS3wWOl3TWZurc8taKB59prjGx2CdvGuK1c3OgRJg4wFJkmSpCaSmQ8B0cvhyT30T2BuL9eaD8zvv+gkSVKz8hE5SZIkSZIkVcUCkyRJkiRJkqpigUmSJEmSJElVscAkSZIkSZKkqlhgkiRJkiRJUlUsMEmSJEmSJKkqQxsdgNSfnr3m+EaHoB6868rVjQ5BkiRJklRDjmCSJEmSJElSVSwwSZIkSZIkqSoWmCRJkiRJklQVC0ySJEmSJEmqigUmSZIkSZIkVcUCkyRJkiRJkqpigUmSJEmSJElVscAkSZIkSZKkqlhgkiRJkiRJUlUsMEmSJEmSJKkqQxsdgCQNFB+8bGGjQ1APll9/fqNDkCRJkgY9RzBJkiRJkiSpKjUrMEXE/IjYFBFPdGs7LCKWRsS68j6itEdE/G1EdEbE4xExvts5M0v/dRExs1bxSpIkSZIkae/UcgTTrcC0XdrmAfdl5rHAfWUf4MPAseU1B7gRKgUp4CrgFOBk4KodRSlJkiRJkiTtG2pWYMrMB4CtuzTPABaU7QXA2d3aF2bFI8ChEXEkMBVYmplbM3MbsJQ3F60kSZIkSZLUQPWeg2lkZm4s278BRpbtUcBz3fp1lbbe2iVJkiRJkrSPaNgk35mZQPbX9SJiTkR0RETH5s2b++uykiRJkiRJegv1LjA9Xx59o7xvKu0bgKO69Wstbb21v0lm3pSZ7ZnZ3tLS0u+BS5IkSZIkqWf1LjDdDexYCW4mcFe39vPLanITgZfKo3Q/BqZExIgyufeU0iZJkiRJkqR9RM0KTBFxG/Aw8L6I6IqI2cBXgTMjYh3w52Uf4F7gKaAT+HvgYoDM3ApcCzxWXteUNkmSJPUgIuZHxKaIeKJb22ERsTQi1pX3EaU9IuJvI6IzIh6PiPHdzplZ+q+LiJk9fZYkSdIOQ2t14cw8r5dDk3vom8DcXq4zH5jfj6FJkiQ1s1uBbwELu7XNA+7LzK9GxLyyfznwYeDY8joFuBE4JSIOA64C2qnMmbk8Iu4uq/pKkiS9ScMm+ZYkSVL/y8wHgF1HfM8AFpTtBcDZ3doXZsUjwKFlnsypwNLM3FqKSkuBabWPXpIkDVQWmCRJkprfyDK/JcBvgJFlexTwXLd+XaWtt3ZJkqQeWWCSJEkaRMrUBNlf14uIORHREREdmzdv7q/LSpKkAcYCkyRJUvN7vjz6RnnfVNo3AEd169da2nprf5PMvCkz2zOzvaWlpd8DlyRJA4MFJkmSpOZ3N7BjJbiZwF3d2s8vq8lNBF4qj9L9GJgSESPKinNTSpskSVKParaKnCRJkuovIm4D/j1wRER0UVkN7qvAHRExG3gG+Fjpfi8wHegEfgdcAJCZWyPiWuCx0u+azNx14nBJkqSdLDBJkiQ1kcw8r5dDk3vom8DcXq4zH5jfj6FJkqQm5iNykiRJkiRJqooFJkmSJEmSJFXFApMkSZIkSZKqYoFJkiRJkiRJVbHAJEmSJEmSpKpYYJIkSZIkSVJVLDBJkiRJkiSpKhaYJEmSJEmSVBULTJIkSZIkSarK0EYHIElSNZ695vhGh6BdvOvK1Y0OQZIkSXXmCCZJkiRJkiRVxQKTJEmSJEmSqmKBSZIkSZIkSVWxwCRJkiRJkqSqWGCSJEmSJElSVSwwSZIkSZIkqSoWmCRJkiRJklQVC0ySJEmSJEmqigUmSZIkSZIkVcUCkyRJkiRJkqpigUmSJEmSJElVscAkSZIkSZKkqlhgkiRJkiRJUlUsMEmSJEmSJKkqFpgkSZIkSZJUFQtMkiRJkiRJqooFJkmSJEmSJFVlwBSYImJaRPwyIjojYl6j45EkSRoMzMEkSVJfDIgCU0TsB/wd8GFgDHBeRIxpbFSSJEnNzRxMkiT11dBGB9BHJwOdmfkUQEQsAmYAaxoalSRJUnMzB5PUcB+8bGGjQ1APll9/fqND0D5moBSYRgHPddvvAk7p3iEi5gBzyu7LEfHLOsWmfci74QjghUbHoV1cFY2OQE3M3/t9UP1+599drw8axMzB1Cf+Ld4HmX+pxuK/zvT3fnDqNf8aKAWmt5SZNwE3NToONVZEdGRme6PjkFQ//t5LjWUOJvBvsTQY+XuvXQ2IOZiADcBR3fZbS5skSZJqxxxMkiT1yUApMD0GHBsRoyNif+Bc4O4GxyRJktTszMEkSVKfDIhH5DLz9Yj4LPBjYD9gfmY+2eCwtG9yiL40+Ph7L9WIOZj2gH+LpcHH33u9QWRmo2OQJEmSJEnSADZQHpGTJEmSJEnSPsoCkyRJkiRJkqpigUmSJEmSJElVscAkSZIkSZKkqlhg0oASEddExOe77X8lIi6NiMsi4rGIeDwiri7HDoyIH0bEqoh4IiL+t8ZFLqk/RERbRKyNiL+PiCcjYklEvC0ixkXEI+VvwA8iYkSjY5WkZuXfYqn5+X+X9oYFJg0084HzASJiCHAu8BvgWOBkYBzwwYiYBEwDfp2ZJ2bmWGBxY0KW1M+OBf4uM48DXgT+ClgIXJ6ZJwCrgasaGJ8kDQb+LZaam/93aY9ZYNKAkpnrgS0RcRIwBfg5MKHb9grg/VT+8K0GzoyIr0XEhzLzpcZELamfPZ2ZK8v2cuA9wKGZ+c+lbQEwqSGRSdLg4d9iqYn5f5f2xtBGByDthX8AZgH/jkplfTLwf2Xmf9+1Y0SMB6YD10XEfZl5TT0DlVQTr3Xb3g4c2qhAJGkQ82+x1Pz8v0t7xBFMGoh+QGUY5gTgx+V1YUQcBBARoyLiHRHxTuB3mfld4HpgfKMCllRTLwHbIuJDZf9TwD/vpr8kqf/5t1hqPv7fpT3iCCYNOJn5+4j4KfBiZm4HlkTEB4CHIwLgZeCTwDHA9RHxb8AfgP+jUTFLqrmZwP8TEW8HngIuaHA8kjQY+bdYaiL+36U9FZnZ6BikPVImmVsBnJOZ6xodjyRJkiQ1G//v0p7yETkNKBExBugE7vOPnCRJkiT1P//v0t5wBJMkSZIkSZKq4ggmSZIkSZIkVcUCkyRJkiRJkqpigUmSJEmSJElVscAkqe4i4tCIuLgOn3N2maBwT897uRbxSJIk7csi4ssR8de7Od4SEY9GxM8j4kN7cf1ZEfGtsr1XeZqkfZcFJkmNcCjQ5wJTVOzN36uzARMXSZKk/jEZWJ2ZJ2Xmg1VeyzxNajIWmCQ1wleB90TEyoi4ISLui4gVEbE6ImYARERbRPwyIhYCTwBHRcT/WdoeiojbdnzDFhHviYjFEbE8Ih6MiPdHxGnAWcD15XPe01O/cv7oiHi4fP51DfqZSJIk1V1E/OeI+FVEPAS8r7T1lFuNA/4GmFFyq7dFxI0R0RERT0bE1d2uuT4ijijb7RFx/y6f+aY8rV73K6l2hjY6AEmD0jxgbGaOi4ihwNsz819LIvJIRNxd+h0LzMzMRyJiAvBXwInAMGAFsLz0uwn43zNzXUScAnw7M88o17knM+8EiIj7du0HnAF8A7gxMxdGxNy6/AQkSZIaLCI+CJwLjKPyv+GO/Kq33OpKoD0zP1vO/8+ZuTUi9gPui4gTMvPxt/rczPxfu+ZpkgY+C0ySGi2A/xIRk4B/A0YBI8uxZzLzkbJ9OnBXZr4KvBoR/xMgIg4CTgP+34jYcc0D3vQhu+93OpXiFcB3gK/1z61JkiTt0z4E/CAzfwdQij7D6UNuVXwsIuZQ+b/ySCqPvL1lgUlSc7LAJKnRPgG0AB/MzD9ExHoqiQ3AK304fwjwYmaOq7Jf9iVYSZKkJten3CoiRgN/DUzIzG0RcSt/zOFe54/TsQzv4XRJTcg5mCQ1wm+Bg8v2IcCmUlz6M+DdvZzzM+AvI2J4GY30FwCZ+a/A0xFxDuycEPzEXT/nLfr9jMrwcKgUvCRJkgaDB4Czy3xKBwN/CfyO3nOm7v6EypeBL0XESODD3Y6tBz5Ytv+KnnXPByU1AQtMkuouM7cAP4uIJ6g8898eEauB84Ff9HLOY8DdVIZd/whYDbxUDn8CmB0Rq4AngRmlfRFwWVlK9z276XcpMLfEMKpfb1aSJGkflZkrgNuBVVTyq8fKod5ypu7nrgJ+TiV3+z6VL+x2uBr4RkR0ANt7+fhd8zRJA1xk+lSIpIEhIg7KzJcj4u1UvnGbUxIjSZIkSVIDOQeTpIHkpogYQ+VZ/gUWlyRJkiRp3+AIJkmSJEmSJFXFOZgkSZIkSZJUFQtMkiRJkiRJqooFJkmSJEmSJFXFApMkSZIkSZKqYoFJkiRJkiRJVbHAJEmSJEmSpKpYYJIkSZIkSVJVLDBJkiRJkiSpKhaYJEmSJEmSVBULTJIkSZIkSaqKBSZJkiRJkiRVxQKTJEmSJEmSqmKBSZIkSZIkSVWxwCRJkiRJkqSqDG10ALVwxBFHZFtbW6PDkCRJNbR8+fIXMrOl0XHoj8zBJElqbrvLv5qywNTW1kZHR0ejw5AkSTUUEc80Oga9kTmYJEnNbXf5l4/ISZIkSZIkqSoWmCRJkiRJklQVC0ySJEmSJEmqSlPOwSRJ0mD1hz/8ga6uLl599dVGh9Jvhg8fTmtrK8OGDWt0KJIkST1qthxsb/IvC0ySJDWRrq4uDj74YNra2oiIRodTtcxky5YtdHV1MXr06EaHI0mS1KNmysH2Nv/yETlJkprIq6++yuGHHz7gE5sdIoLDDz+8ab4NlCRJzamZcrC9zb8sMEmS1GSaIbHprtnuR5IkNadmyln25l4sMEmSJEmSJKkqFpgkSZIkSZJUFSf57sUHL1vY6BD2KcuvP7/RIUiS+kFmkpkMGeJ3TBq4zNP2PeaKkrR7gyEHa947kyRJAKxfv573ve99nH/++YwdO5Zrr72WCRMmcMIJJ3DVVVcB8Morr/CRj3yEE088kbFjx3L77bcD0NbWxhe/+EWOP/54Tj75ZDo7O3de84wzzuCEE05g8uTJPPvsswDMmjWLSy65hNNOO42jjz6aO++8E4CNGzcyadIkxo0bx9ixY3nwwQcBWLJkCaeeeirjx4/nnHPO4eWXX673j0eSJKkmBlsOZoFJkqRBYN26dVx88cXccMMNbNiwgWXLlrFy5UqWL1/OAw88wOLFi3nnO9/JqlWreOKJJ5g2bdrOcw855BBWr17NZz/7WT7/+c8D8LnPfY6ZM2fy+OOP84lPfIJLLrlkZ/+NGzfy0EMPcc899zBv3jwAvv/97zN16lRWrlzJqlWrGDduHC+88ALXXXcdP/nJT1ixYgXt7e18/etfr+8PRpIkqYYGUw5mgUmSpEHg3e9+NxMnTmTJkiUsWbKEk046ifHjx/OLX/yCdevWcfzxx7N06VIuv/xyHnzwQQ455JCd55533nk73x9++GEAHn74YT7+8Y8D8KlPfYqHHnpoZ/+zzz6bIUOGMGbMGJ5//nkAJkyYwC233MKXv/xlVq9ezcEHH8wjjzzCmjVrOP300xk3bhwLFizgmWeeqdePRJIkqeYGUw7mHEySJA0CBx54IFB5/v+KK67gM5/5zJv6rFixgnvvvZcvfelLTJ48mSuvvBJ44zK1fVmy9oADDti5nZkATJo0iQceeIAf/vCHzJo1iy984QuMGDGCM888k9tuu62qe5MkSdpXDaYczBFMkiQNIlOnTmX+/Pk7n7PfsGEDmzZt4te//jVvf/vb+eQnP8lll13GihUrdp6zYy6A22+/nVNPPRWA0047jUWLFgHwve99jw996EO7/dxnnnmGkSNH8ulPf5qLLrqIFStWMHHiRH72s5/tnFPglVde4Ve/+lW/37MkSVKjDYYczBFMkiQNIlOmTGHt2rU7k5SDDjqI7373u3R2dnLZZZcxZMgQhg0bxo033rjznG3btnHCCSdwwAEH7Pym65vf/CYXXHAB119/PS0tLdxyyy27/dz777+f66+/nmHDhnHQQQexcOFCWlpauPXWWznvvPN47bXXALjuuut473vfW6O7lyRJaozBkIPFjmFTzaS9vT07OjqquobL376RS89K0sCwdu1aPvCBD/Tb9dra2ujo6OCII47ot2vujZ7uKyKWZ2Z7g0JSD/ojB+sL87R9j7mipMGuGXOwPc2/fEROkiRJkiRJValZgSkihkfEsohYFRFPRsTVpf3WiHg6IlaW17jSHhHxtxHRGRGPR8T4bteaGRHrymtmrWKWJElvtH79+oaPXpIkSRpsBmIOVss5mF4DzsjMlyNiGPBQRPyoHLssM+/cpf+HgWPL6xTgRuCUiDgMuApoBxJYHhF3Z+a2GsYuSZIkSZKkPqrZCKaseLnsDiuv3U34NANYWM57BDg0Io4EpgJLM3NrKSotBabVKm5JkqSBLiLWR8TqMlq8o7QdFhFLy4jwpRExorQ7ilySJFWtpnMwRcR+EbES2ESlSPRoOfSVksDcEBEHlLZRwHPdTu8qbb217/pZcyKiIyI6Nm/e3O/3IkmSNMD8WWaO6zYR5zzgvsw8Friv7MMbR5HPoTKKnG6jyE8BTgau2lGUkiRJ2lVNC0yZuT0zxwGtwMkRMRa4Ang/MAE4DLi8nz7rpsxsz8z2lpaW/rikJElSM5kBLCjbC4Czu7U7ilySJFWllnMw7ZSZL0bET4FpmflfS/NrEXEL8NdlfwNwVLfTWkvbBuDf79J+f00DliSpyfX3Mu99XaJ88eLFXHrppWzfvp2LLrqIefPmvfVJ2hsJLImIBP57Zt4EjMzMjeX4b4CRZbuqUeSSJKnvGpGD1Sv/quUqci0RcWjZfhtwJvCL8o0YERFUvjl7opxyN3B+mQdgIvBSSYJ+DEyJiBFlWPaU0iZJkgaQ7du3M3fuXH70ox+xZs0abrvtNtasWdPosJrVn2bmeCqPv82NiEndD2Zmsvu5MfvMaQokSdp31TP/quUjckcCP42Ix4HHqAyxvgf4XkSsBlYDRwDXlf73Ak8BncDfAxcDZOZW4NpyjceAa0qbJEkaQJYtW8YxxxzD0Ucfzf7778+5557LXXfd1eiwmlJmbijvm4AfUJlD6fluX/QdSWWOTNj9KPKe2nf9LKcpkCRpH1XP/Ktmj8hl5uPAST20n9FL/wTm9nJsPjC/XwOUJEl1tWHDBo466o/1itbWVh599NHdnKG9EREHAkMy87dlewpwDZXR4jOBr5b3Hdnl3cBnI2IRlQm9X8rMjRHxY+C/dJvYewqVuTQlSdIAUc/8qy5zMEmSJKluRgI/qMxGwFDg+5m5OCIeA+6IiNnAM8DHSv97gelURpH/DrgAKqPII2LHKHJwFLkkSdoNC0ySJKkuRo0axXPP/XHO6K6uLkaNcs7o/paZTwEn9tC+BZjcQ7ujyCVJalL1zL9qOQeTJEnSThMmTGDdunU8/fTT/P73v2fRokWcddZZjQ5LkiSpadUz/3IEkyRJg1BflrTtb0OHDuVb3/oWU6dOZfv27Vx44YUcd9xxdY9DkiSpUeqdg9Uz/7LAJEmS6mb69OlMnz690WFIkiQNGvXKv3xETpIkSZIkSVWxwCRJkiRJkqSqWGCSJEmSJElSVSwwSZIkSZIkqSoWmCRJkiRJklQVC0ySJEmSJEmqytBGByBJkurv2WuO79frvevK1W/Z58ILL+See+7hHe94B0888US/fr4kSdJA0Mw5mCOYJElSXcyaNYvFixc3OgxJkqRBpV45mAUmSZJUF5MmTeKwww5rdBiSJEmDSr1yMAtMkiRJkiRJqooFJkmSJEmSJFXFApMkSZIkSZKqYoFJkiRJkiRJVRna6AAkSVL99WVJ2/523nnncf/99/PCCy/Q2trK1VdfzezZs+sehyRJUqM0cw5mgUmSJNXFbbfd1ugQJEmSBp165WA+IidJkiRJkqSq1KzAFBHDI2JZRKyKiCcj4urSPjoiHo2Izoi4PSL2L+0HlP3Ocryt27WuKO2/jIiptYpZkiRJkiRJe66WI5heA87IzBOBccC0iJgIfA24ITOPAbYBOx78mw1sK+03lH5ExBjgXOA4YBrw7YjYr4ZxS5I0oGVmo0PoV812P5IkqTk1U86yN/dSswJTVrxcdoeVVwJnAHeW9gXA2WV7RtmnHJ8cEVHaF2Xma5n5NNAJnFyruCVJGsiGDx/Oli1bmibByUy2bNnC8OHDGx2KJElSr5opB9vb/Kumk3yXkUbLgWOAvwP+BXgxM18vXbqAUWV7FPAcQGa+HhEvAYeX9ke6Xbb7Od0/aw4wB+Bd73pXv9+LJEkDQWtrK11dXWzevLnRofSb4f8/e/ceb1dZ3/v+84WAWKFcJLIxAZNKesGqQZeAtbqtVC5pK9iNXLwQLopnF6y+2m0NbiuK0K2lyqbVcg6VYKhIpFoOOYhARCilBSGJQLio5HCRxACBIF44oITf+WM+iYuwFiwy11oza67P+/WarzXGbzzjmb8RZiZPfut5xthmG6ZPn97rNCRJkobVb2OwTRl/jWmBqarWAbOT7ABcBPz2GL7XNyk/RQAAIABJREFU2cDZAAMDAxO/ZChJ0ibYaqutmDlzZq/TkCRJmlQcg43TU+Sq6sfAVcDrgR2SrC9sTQdWte1VwG4A7fj2wMOD40OcI0mSJEmSpB4by6fITW0zl0jyQuCtwB10Ck2HtmZzgYvb9qK2Tzv+7eosXlwEHNGeMjcTmAXcMFZ5S5IkSZIk6fkZyyVyuwIL2n2YtgAurKpLktwOLExyKvBd4JzW/hzgn5OsANbSeXIcVXVbkguB24EngRPa0jtJkiRJkiRtBsaswFRVtwB7DRG/iyGeAldVjwPvGKav04DTRjtHSZIkSZIkdW9c7sEkSZIkSZKk/mWBSZIkSZIkSV2xwCRJkiRJkqSuWGCSJEmSJElSVywwSZIkSZIkqSsWmCRJkvpQki2TfDfJJW1/ZpLvJFmR5KtJtm7xF7T9Fe34jEF9nNTi309yQG+uRJIkTQQWmCRJkvrTB4E7Bu1/BjijqvYAHgGOa/HjgEda/IzWjiR7AkcArwAOBP4xyZbjlLskSZpgLDBJkiT1mSTTgT8Cvtj2A7wF+FprsgA4pG0f3PZpx/dr7Q8GFlbVE1V1N7AC2Ht8rkCSJE00FpgkSZL6z/8G/gp4qu2/GPhxVT3Z9lcC09r2NOA+gHb80dZ+Q3yIcyRJkp7GApMkSVIfSfLHwINVtXSc3u/4JEuSLFmzZs14vKUkSdoMWWCSJEnqL28A3pbkHmAhnaVxZwI7JJnS2kwHVrXtVcBuAO349sDDg+NDnLNBVZ1dVQNVNTB16tTRvxpJkjQhWGCSJEnqI1V1UlVNr6oZdG7S/e2qehdwFXBoazYXuLhtL2r7tOPfrqpq8SPaU+ZmArOAG8bpMiRJ0gQz5bmbSJIkqQ98BFiY5FTgu8A5LX4O8M9JVgBr6RSlqKrbklwI3A48CZxQVevGP21JkjQRWGCSJEnqU1V1NXB1276LIZ4CV1WPA+8Y5vzTgNPGLkNJktQvXCInSZIkSZKkrlhgkiRJkiRJUlcsMEmSJEmSJKkrFpgkSZIkSZLUFQtMkiRJkiRJ6ooFJkmSJEmSJHXFApMkSZIkSZK6MmYFpiS7Jbkqye1JbkvywRb/RJJVSW5qrzmDzjkpyYok309ywKD4gS22Ism8scpZkiRJkiRJz9+UMez7SeAvq2pZku2ApUkWt2NnVNXfDW6cZE/gCOAVwEuBbyX5zXb4C8BbgZXAjUkWVdXtY5i7JEmSJEmSRmjMCkxVtRpY3bZ/muQOYNqznHIwsLCqngDuTrIC2LsdW1FVdwEkWdjaWmCSJEmSJEnaDIzLPZiSzAD2Ar7TQicmuSXJ/CQ7ttg04L5Bp61sseHikiRJkiRJ2gyMeYEpybbA14EPVdVPgLOAlwOz6cxw+uwovc/xSZYkWbJmzZrR6FKSJEmSJEkjMKYFpiRb0SkunV9V/wpQVQ9U1bqqegr4J361DG4VsNug06e32HDxp6mqs6tqoKoGpk6dOvoXI0mSJEmSpCGN5VPkApwD3FFVnxsU33VQs7cDt7btRcARSV6QZCYwC7gBuBGYlWRmkq3p3Ah80VjlLUmSJEmSpOdnLJ8i9wbgPcDyJDe12EeBI5PMBgq4B3g/QFXdluRCOjfvfhI4oarWASQ5Ebgc2BKYX1W3jWHekiRJkiRJeh7G8ily1wIZ4tClz3LOacBpQ8QvfbbzJEmSJEmS1Dvj8hQ5SZIkSZIk9S8LTJIkSZIkSeqKBSZJkiRJkiR1xQKTJEmSJEmSumKBSZIkSZIkSV2xwCRJkiRJkqSuWGCSJEmSJElSVywwSZIkSZIkqSsWmCRJkiRJktQVC0ySJEmSJEnqigUmSZIkSZIkdcUCkyRJkiRJkrpigUmSJEmSJEldscAkSZIkSZKkrlhgkiRJkiRJUlcsMEmSJEmSJKkrFpgkSZIkSZLUlREVmJJcOZKYJEmSRs+mjMGSbJPkhiQ3J7ktySdbfGaS7yRZkeSrSbZu8Re0/RXt+IxBfZ3U4t9PcsDoXp0kSeonz1pgagOUnYCdk+yYZKf2mgFMG48EJUmSJpsux2BPAG+pqlcDs4EDk+wLfAY4o6r2AB4BjmvtjwMeafEzWjuS7AkcAbwCOBD4xyRbjuZ1SpKk/vFcM5jeDywFfrv9XP+6GPj82KYmSZI0aW3yGKw6ftZ2t2qvAt4CfK3FFwCHtO2D2z7t+H5J0uILq+qJqrobWAHs3f2lSZKkfjTl2Q5W1ZnAmUk+UFX/ME45SZIkTWrdjsHaTKOlwB7AF4D/F/hxVT3ZmqzkVzOhpgH3tfd9MsmjwItb/PpB3Q4+Z/B7HQ8cD7D77rs/31QlSVKfeNYC03pV9Q9Jfg+YMficqjpvuHOS7AacB+xC57dmZ1fVmW2691dbX/cAh1XVI+03ZWcCc4DHgKOralnray7wsdb1qVW1AEmSpD63KWOwdnwdMDvJDsBFdGZCjVWOZwNnAwwMDNRYvY8kSdq8jajAlOSfgZcDNwHrWrjoFJCG8yTwl1W1LMl2wNIki4GjgSur6tNJ5gHzgI8ABwGz2msf4Cxgn1aQOhkYaO+5NMmiqnrkeV2pJEnSBLOJY7ANqurHSa4CXg/skGRKm8U0HVjVmq0CdgNWJpkCbA88PCi+3uBzJEmSnmZEBSY6xZ09q2rEv5WqqtXA6rb90yR30JlWfTDw5tZsAXA1nQLTwcB57T2uT7JDkl1b28VVtRagFakOBC4YaS6SJEkT1PMegyWZCvyyFZdeCLyVzo27rwIOBRYCc+nczwlgUdu/rh3/dlVVkkXAV5J8DngpnV8C3jA6lyVJkvrNSAtMtwL/hVYwer7aE0/2Ar4D7NKKTwD301lCB4PW/zfr1/kPF5ckSep3mzIG2xVY0O7DtAVwYVVdkuR2YGGSU4HvAue09ucA/5xkBbCWzpPjqKrbklwI3E5nZvoJbemdJEnSM4y0wLQzcHuSG+g8+haAqnrbc52YZFvg68CHquonnVstbTi/kozKWn1vMClJkvrQ8x6DVdUtdH6xt3H8LoZ4ClxVPQ68Y5i+TgNOe/5pS5KkyWakBaZPbErnSbaiU1w6v6r+tYUfSLJrVa1uS+AebPHh1vmv4ldL6tbHr974vbzBpCRJ6kOf6HUCkiRJIzHSp8j92/PtuD0V7hzgjqr63KBD69f5f5pnrv8/MclCOjf5frQVoS4H/ibJjq3d/sBJzzcfSZKkiWZTxmCSJEm9MNKnyP2UzhNLALYGtgJ+XlW//iynvQF4D7A8yU0t9lE6haULkxwH3Asc1o5dCswBVgCPAccAVNXaJJ8CbmztTll/w29JkqR+toljMEmSpHE30hlM263fbjOTDgb2fY5zrgUyzOH9hmhfwAnD9DUfmD+SXCVJkvrFpozBJEmSemGk92DaoBWC/u8kJwPzRj8lSSPx2g+f1+sUNitLTz+q1ylI0phyDCZJkjZnI10i96eDdrcABoDHxyQjSZIkAY7BJEnSxDHSGUx/Mmj7SeAeOlO0JUmSNHYcg0mSpAlhpPdgOmasE5EkSdLTOQaTJEkTxRYjaZRkepKLkjzYXl9PMn2sk5MkSZrMHINJkqSJYkQFJuBcYBHw0vb6f1pMkiRJY8cxmCRJmhBGWmCaWlXnVtWT7fUlYOoY5iVJkiTHYJIkaYIYaYHp4STvTrJle70beHgsE5MkSZJjMEmSNDGMtMB0LHAYcD+wGjgUOHqMcpIkSVKHYzBJkjQhjOgpcsApwNyqegQgyU7A39EZ9EiSJGlsOAaTJEkTwkhnML1q/cAGoKrWAnuNTUqSJElqHINJkqQJYaQFpi2S7Lh+p/32bKSznyRJkrRpHINJkqQJYaQDlM8C1yX5l7b/DuC0sUlJkiRJjWMwSZI0IYyowFRV5yVZArylhf60qm4fu7QkSZLkGEySJE0UI55i3QYzDmgkSZLGkWMwSZI0EYz0HkySJEmSJEnSkCwwSZIkSZIkqSsWmCRJkiRJktQVC0ySJEmSJEnqigUmSZIkSZIkdcUCkyRJkiRJkroyZgWmJPOTPJjk1kGxTyRZleSm9poz6NhJSVYk+X6SAwbFD2yxFUnmjVW+kiRJkiRJ2jRjOYPpS8CBQ8TPqKrZ7XUpQJI9gSOAV7Rz/jHJlkm2BL4AHATsCRzZ2kqSJEmSJGkzMWWsOq6qa5LMGGHzg4GFVfUEcHeSFcDe7diKqroLIMnC1vb2UU5XkiRJkiRJm6gX92A6McktbQndji02DbhvUJuVLTZcXJIkSZIkSZuJ8S4wnQW8HJgNrAY+O1odJzk+yZIkS9asWTNa3UqSJEmSJOk5jGuBqaoeqKp1VfUU8E/8ahncKmC3QU2nt9hw8aH6PruqBqpqYOrUqaOfvCRJkiRJkoY0rgWmJLsO2n07sP4Jc4uAI5K8IMlMYBZwA3AjMCvJzCRb07kR+KLxzFmSJEmSJEnPbsxu8p3kAuDNwM5JVgInA29OMhso4B7g/QBVdVuSC+ncvPtJ4ISqWtf6ORG4HNgSmF9Vt41VzpIkSZIkSXr+xvIpckcOET7nWdqfBpw2RPxS4NJRTE2SJKlvJdkNOA/Yhc4v9c6uqjOT7AR8FZhB5xd9h1XVI0kCnAnMAR4Djq6qZa2vucDHWtenVtWC8bwWSZI0cfTiKXKSJEkaO08Cf1lVewL7Aick2ROYB1xZVbOAK9s+wEF0bk8wCziezkNZaAWpk4F96Nw38+RBTwCWJEl6GgtMkiRJfaSqVq+fgVRVPwXuAKYBBwPrZyAtAA5p2wcD51XH9cAO7b6ZBwCLq2ptVT0CLAYOHMdLkSRJE4gFJkmSpD6VZAawF/AdYJeqWt0O3U9nCR10ik/3DTptZYsNF9/4PY5PsiTJkjVr1oxq/pIkaeKwwCRJktSHkmwLfB34UFX9ZPCxqio692fqWlWdXVUDVTUwderU0ehSkiRNQBaYJEmS+kySregUl86vqn9t4Qfa0jfazwdbfBWw26DTp7fYcHFJkqRnsMAkSZLUR9pT4c4B7qiqzw06tAiY27bnAhcPih+Vjn2BR9tSusuB/ZPs2G7uvX+LSZIkPcOUXicgSZKkUfUG4D3A8iQ3tdhHgU8DFyY5DrgXOKwduxSYA6wAHgOOAaiqtUk+BdzY2p1SVWvH5xIkSdJEY4FJkiSpj1TVtUCGObzfEO0LOGGYvuYD80cvO0mS1K9cIidJkiRJkqSuWGCSJEmSJElSVywwSZIkSZIkqSveg0mSJElSV354yit7nYI2svvHl/c6BUmTjDOYJEmSJEmS1BULTJIkSZIkSeqKBSZJkiRJkiR1xQKTJEmSJEmSumKBSZIkSZIkSV2xwCRJkiRJkqSuWGCSJEmSJElSVywwSZIkSZIkqSsWmCRJkiRJktQVC0ySJEmSJEnqypSx6jjJfOCPgQer6ndbbCfgq8AM4B7gsKp6JEmAM4E5wGPA0VW1rJ0zF/hY6/bUqlowVjlreD885ZW9TmGzsvvHl/c6BUmSJEmSNhtjOYPpS8CBG8XmAVdW1SzgyrYPcBAwq72OB86CDQWpk4F9gL2Bk5PsOIY5S5IkSZIk6XkaswJTVV0DrN0ofDCwfgbSAuCQQfHzquN6YIckuwIHAIuram1VPQIs5plFK0mSJEmSJPXQeN+DaZeqWt227wd2advTgPsGtVvZYsPFnyHJ8UmWJFmyZs2a0c1akiRJkiRJw+rZTb6rqoAaxf7OrqqBqhqYOnXqaHUrSZIkSZKk5zDeBaYH2tI32s8HW3wVsNugdtNbbLi4JEmSJEmSNhPjXWBaBMxt23OBiwfFj0rHvsCjbSnd5cD+SXZsN/fev8UkSZIkSZK0mZgyVh0nuQB4M7BzkpV0ngb3aeDCJMcB9wKHteaXAnOAFcBjwDEAVbU2yaeAG1u7U6pq4xuHS5IkSZIkqYfGrMBUVUcOc2i/IdoWcMIw/cwH5o9iapIkSZIkSRpFPbvJtyRJkiRJkvqDBSZJkiRJkiR1xQKTJEmSJEmSumKBSZIkSZIkSV2xwCRJkiRJkqSuWGCSJEmSJElSVywwSZIkSZIkqSsWmCRJkiRJktQVC0ySJEmSJEnqigUmSZKkPpJkfpIHk9w6KLZTksVJ7mw/d2zxJPn7JCuS3JLkNYPOmdva35lkbi+uRZIkTRxTep2AJKl/vfbD5/U6hc3G0tOP6nUKmjy+BHweGPwXcB5wZVV9Osm8tv8R4CBgVnvtA5wF7JNkJ+BkYAAoYGmSRVX1yLhdhSRJmlCcwSRJktRHquoaYO1G4YOBBW17AXDIoPh51XE9sEOSXYEDgMVVtbYVlRYDB4599pIkaaKywCRJktT/dqmq1W37fmCXtj0NuG9Qu5UtNlxckiRpSC6RkyRJmkSqqpLUaPWX5HjgeIDdd999tLqVpA1ccr95cvm/NmaBSVJf+OEpr+x1CpuV3T++vNcpSNq8PJBk16pa3ZbAPdjiq4DdBrWb3mKrgDdvFL96qI6r6mzgbICBgYFRK1xJkqSJxSVykiRJ/W8RsP5JcHOBiwfFj2pPk9sXeLQtpbsc2D/Jju2Jc/u3mCRJ0pCcwSRJktRHklxAZ/bRzklW0nka3KeBC5McB9wLHNaaXwrMAVYAjwHHAFTV2iSfAm5s7U6pqo1vHC5JkrSBBSZJkqQ+UlVHDnNovyHaFnDCMP3MB+aPYmqSJKmPuUROkiRJkiRJXbHAJEmSJEmSpK5YYJIkSZIkSVJXelJgSnJPkuVJbkqypMV2SrI4yZ3t544tniR/n2RFkluSvKYXOUuSJEmSJGlovZzB9AdVNbuqBtr+PODKqpoFXNn2AQ4CZrXX8cBZ456pJEmSJEmShrU5LZE7GFjQthcAhwyKn1cd1wM7JNm1FwlKkiRJkiTpmXpVYCrgiiRLkxzfYrtU1eq2fT+wS9ueBtw36NyVLfY0SY5PsiTJkjVr1oxV3pIkSZIkSdrIlB697+9X1aokLwEWJ/ne4INVVUnq+XRYVWcDZwMMDAw8r3MlSZIkSZK06Xoyg6mqVrWfDwIXAXsDD6xf+tZ+PtiarwJ2G3T69BaTJEmSJEnSZmDcC0xJXpRku/XbwP7ArcAiYG5rNhe4uG0vAo5qT5PbF3h00FI6SZIkSZIk9VgvlsjtAlyUZP37f6WqLktyI3BhkuOAe4HDWvtLgTnACuAx4JjxT1mSJEmSJEnDGfcCU1XdBbx6iPjDwH5DxAs4YRxSkyRJkiRJ0ibo1VPkJEmSJEmS1CcsMEmSJEmSJKkrFpgkSZIkSZLUFQtMkiRJkiRJ6ooFJkmSJEmSJHXFApMkSZIkSZK6YoFJkiRJkiRJXbHAJEmSJEmSpK5YYJIkSZIkSVJXLDBJkiRJkiSpKxaYJEmSJEmS1BULTJIkSZIkSeqKBSZJkiRJkiR1xQKTJEmSJEmSujKl1wlIkjQZ/PCUV/Y6hc3K7h9f3usUJEmSNIqcwSRJkiRJkqSuWGCSJEmSJElSVywwSZIkSZIkqSsWmCRJkiRJktQVC0ySJEmSJEnqik+RkyRJkiRJz4tPyN389PopvRNmBlOSA5N8P8mKJPN6nY8kSdJk4BhMkiSNxIQoMCXZEvgCcBCwJ3Bkkj17m5UkSVJ/cwwmSZJGakIUmIC9gRVVdVdV/QJYCBzc45wkSZL6nWMwSZI0IhOlwDQNuG/Q/soWkyRJ0thxDCZJkkakb27yneR44Pi2+7Mk3+9lPv3mZbAz8FCv89hsnJxeZ6CN+BndiJ/RzY6f0Y2Mzmf0ZaPRibrjGEzgd9xmybGAxph/7zdD4/P3ftjx10QpMK0Cdhu0P73FNqiqs4GzxzOpySTJkqoa6HUe0nD8jGpz52dUE5RjMI2I33HS5OPfe21soiyRuxGYlWRmkq2BI4BFPc5JkiSp3zkGkyRJIzIhZjBV1ZNJTgQuB7YE5lfVbT1OS5Ikqa85BpMkSSM1IQpMAFV1KXBpr/OYxJz6rs2dn1Ft7vyMakJyDKYR8jtOmnz8e6+nSVX1OgdJkiRJkiRNYBPlHkySJEmSJEnaTFlg0iZJ8tIkX3ue53wpyaFjlZP6V5IvJtlzE8+dkeTW0c5Jk0uSTyT5H0lOSfKHm0E+9yTZudd5SJIkSetNmHswqXeSTKmqJzfa/xFgsUjjoqre2+scJICq+vho9JNky6paNxp9SZIkSZsDZzD1sTZz43tt5tAPkpyf5A+T/EeSO5Ps3V7XJflukv9M8lvt3KOTLErybeDKIfY3zApJsmWS05PcmOSWJO9v8ST5fJLvJ/kW8JKe/WFowkjyoiTfSHJzkluTHJ7k6iQD7fjPkpzWjl+fZJcWf3nbX57k1CQ/G6LvIT+r0lCS/M/23XktsP678UtJDk1yYJJ/GdT2zUkuadtHts/hrUk+M6jNz5J8NsnNwOuTHNU+hzcn+efWZmqSr7fP6I1J3tDiL05yRZLbknwRyDj+UUjS07Rx4B1J/ql9L12R5IVJZrf/F9+S5KIkO/Y6V0mbps3a/tCg/dOSfDDJhweNpT/Zjj1j/N67zNVLFpj63x7AZ4Hfbq93Ar8P/A/go8D3gDdW1V7Ax4G/GXTua4BDq+q/DrO/3nHAo1X1OuB1wPuSzATeTucfZXsCRwG/N/qXpz50IPCjqnp1Vf0ucNlGx18EXF9VrwauAd7X4mcCZ1bVK4GVw/Q93GdVepokrwWOAGYDc+h8Xgb7FrBPkhe1/cOBhUleCnwGeEs793VJDmltXgR8p312HwE+Bryl7X+wtTkTOKN9Rv8b8MUWPxm4tqpeAVwE7D6a1ytJm2AW8IX2vfRjOt9Z5wEfqapXAcvpfHdJmpjm0/k3HEm2oDMuup/O3/296YxzXpvkTTz3+F2ThAWm/nd3VS2vqqeA24Arq/PowOXADGB74F/abKQzgFcMOndxVa19lv319geOSnIT8B3gxXS+eN4EXFBV69qSum+P8rWpPy0H3prkM0neWFWPbnT8F8AlbXspnc8xwOuB9TNKvjJM38N9VqWNvRG4qKoeq6qfAIsGH2zLhi8D/iTJFOCPgIvpFKKurqo1rc35dL4LAdYBX2/bbwH+paoeav2t/279Q+Dz7TO6CPj1JNu2Pr7c2n6DToFKknrp7qq6qW0vBV4O7FBV/9ZiC/jV95+kCaaq7gEeTrIXnTH0d+mMc9ZvL6MzgWEWzz1+1yThPZj63xODtp8atP8Unf/+nwKuqqq3J5kBXD2o/c836mvj/fUCfKCqLn9aMJmzaSlrMquqHyR5DZ1ZI6cmuXKjJr9sRVLo/IP9+XyPDflZlTbRQuBEYC2wpKp+mjzryrXHR3DfpS2Afavq8cHB5+hXknph8BhzHbBDrxKRNGa+CBwN/Bc6M5r2A/5XVf1fGzfcePxeVaeMZ6LaPDiDSdsDq9r20ZvYx+XAf0+yFUCS32zLRq4BDm/3vdkV+INuk1X/a0uMHquqLwOn01maORLX05meD50pvEMZ7rMqbewa4JB2T5HtgD8Zos2/0fl8vo9OsQngBuC/Jtk5yZbAka3dxr4NvCPJiwGS7NTiVwAfWN8oyexB+byzxQ4CvK+JpM3No8AjSd7Y9t/D0N9/kiaOi+gsf3sdnXH05cCxbXY1SaYleUkX43f1GWcw6W+BBUk+BnxjE/v4Ip1lSsvS+TX7GuAQOl9IbwFuB34IXNd1tpoMXgmcnuQp4JfAfwf+bgTnfQj4cpL/SWfp0lBTc4f7rEpPU1XLknwVuBl4ELhxiDbr2o29jwbmttjqJPOAq+jMmPtGVV08xLm3JTkN+Lck6+hMNT8a+HPgC0luofP/6GuA/wP4JHBBktuA/6TznSpJm5u5wP+Z5NeAu4BjepyPpC5U1S+SXAX8uM3CviLJ7wDXtdnVPwPeTee+vxuP3zUJ5VcrTSRp4mqD2f+vqirJEcCRVXVwr/OSJEmSJqJ2c+9lwDuq6s5e56PNnzOYJPWL19K5OXLoPM3m2B7nI0mSJE1ISfak82CdiywuaaScwSRJkiRJkqSueJNvSZIkSZIkdcUCkyRJkiRJkrpigUmSJEmSJEldscAkqa8keWmSr7Xt2UnmjOCcN7fHzUuSJAlI8rNe5yBpYrHAJKlvJJlSVT+qqkNbaDbwnAUmSZIkSVJ3LDBJ6rkkM5J8L8mXkvwgyflJ/jDJfyS5M8ne7XVdku8m+c8kv9XOPTrJoiTfBq5sfd2aZGvgFODwJDclOXy4PiRJkjS0dJzexlfLkxze4tsmuTLJshY/uMVnJLkjyT8luS3JFUle2NurkDQepvQ6AUlq9gDeARwL3Ai8E/h94G3AR4GjgDdW1ZNJ/hD4G+C/tXNfA7yqqtYmmQFQVb9I8nFgoKpOBEjy68/ShyRJkp7pT+nMCn81sDNwY5JrgDXA26vqJ0l2Bq5PsqidMws4sqrel+RCOuOtL/cgd0njyAKTpM3F3VW1HCDJbcCVVVVJlgMzgO2BBUlmAQVsNejcxVW1dgTv8Wx9SJIk6Zl+H7igqtYBDyT5N+B1wDeBv0nyJuApYBqwSzvn7qq6qW0vpTOWk9TnXCInaXPxxKDtpwbtP0WnGP4p4Kqq+l3gT4BtBrX/+Qjf49n6kCRJ0si9C5gKvLaqZgMP8Kux1eBx3Tqc2CBNChaYJE0U2wOr2vbRIzznp8B2XfYhSZI0mf07nXtabplkKvAm4AY646oHq+qXSf4AeFkvk5TUexaYJE0Ufwv8ryTfZeS/BbsK2HP9Tb43sQ9JkqTJ7CLgFuBm4NvAX1XV/cD5wEC7ncFRwPd6l6KkzUGqqtc5SJIkSZIkaQJzBpMkSZIkSZK6YoFJkiRJkiRJXbHAJEmSJEmSpK5YYJIkSZIkSVJXLDBJkiRJkiSpKxaYJEmSJEmS1BULTJIkSZIkSeqKBSZJkiRJkiR1xQKTJEmSJEmSumKBSZIkSZIkSV2xwCRJkiRJkqSuWGCSJEmSJElSVywwSZIkSZIkqSsWmCRJkiRJktQVC0ySJEmSJEnqigW/LqXKAAAgAElEQVQmSZIkSZIkdWVKrxMYCzvvvHPNmDGj12lIkqQxtHTp0oeqamqv89gcJdkB+CLwu0ABxwLfB74KzADuAQ6rqkeSBDgTmAM8BhxdVctaP3OBj7VuT62qBc/2vo7BJEnqb882/urLAtOMGTNYsmRJr9OQJEljKMm9vc5hM3YmcFlVHZpka+DXgI8CV1bVp5PMA+YBHwEOAma11z7AWcA+SXYCTgYG6BSpliZZVFWPDPemjsEkSepvzzb+comcJElSH0myPfAm4ByAqvpFVf0YOBhYPwNpAXBI2z4YOK86rgd2SLIrcACwuKrWtqLSYuDAcbwUSZI0gVhgkiRJ6i8zgTXAuUm+m+SLSV4E7FJVq1ub+4Fd2vY04L5B569sseHikiRJz2CBSZIkqb9MAV4DnFVVewE/p7McboOqKjrL3rqW5PgkS5IsWbNmzWh0KUmSJqC+vAeTJEmT1S9/+UtWrlzJ448/3utURs0222zD9OnT2WqrrXqdykSxElhZVd9p+1+jU2B6IMmuVbW6LYF7sB1fBew26PzpLbYKePNG8as3frOqOhs4G2BgYGBUilaSJE00/TYG25TxlwUmSZL6yMqVK9luu+2YMWMGnYeDTWxVxcMPP8zKlSuZOXNmr9OZEKrq/iT3Jfmtqvo+sB9we3vNBT7dfl7cTlkEnJhkIZ2bfD/ailCXA3+TZMfWbn/gpPG8FkmSJop+GoNt6vjLApMkSX3k8ccf74uBzXpJePGLX4xLr563DwDntyfI3QUcQ+fWCBcmOQ64Fzistb0UmAOsAB5rbamqtUk+BdzY2p1SVWvH7xIkSZo4+mkMtqnjLwtMkiT1mX4Y2AzWb9czHqrqJmBgiEP7DdG2gBOG6Wc+MH90s5MkqT/105hlU67Fm3xLkiRJkiSpKxaYJEmSJEmS1BULTJIkTSJVxVNPPdXrNCRJkiaVyTAG8x5MkvrSaz98Xtd9LD39qFHIROq9e+65hwMOOIB99tmHpUuXcthhh3HJJZfwxBNP8Pa3v51PfvKT/PznP+ewww5j5cqVrFu3jr/+67/m8MMPZ8aMGRx22GF885vf5IUvfCFf+cpX2GOPPbjnnns49thjeeihh5g6dSrnnnsuu+++O0cffTS//uu/zpIlS7j//vv527/9Ww499FBWr17N4Ycfzk9+8hOefPJJzjrrLN74xjdyxRVXcPLJJ/PEE0/w8pe/nHPPPZdtt922139kkiRNWKMxDu4nvRzTT7YxmDOYJEmaBO68807+7M/+jDPOOINVq1Zxww03cNNNN7F06VKuueYaLrvsMl760pdy8803c+utt3LggQduOHf77bdn+fLlnHjiiXzoQx8C4AMf+ABz587llltu4V3vehd//ud/vqH96tWrufbaa7nkkkuYN28eAF/5ylc44IADuOmmm7j55puZPXs2Dz30EKeeeirf+ta3WLZsGQMDA3zuc58b3z8YSZKkMTSZxmAWmCRJmgRe9rKXse+++3LFFVdwxRVXsNdee/Ga17yG733ve9x555288pWvZPHixXzkIx/h3//939l+++03nHvkkUdu+HndddcBcN111/HOd74TgPe85z1ce+21G9ofcsghbLHFFuy555488MADALzuda/j3HPP5ROf+ATLly9nu+224/rrr+f222/nDW94A7Nnz2bBggXce++94/VHIkmSNOYm0xjMJXKSJE0CL3rRi4DO+v+TTjqJ97///c9os2zZMi699FI+9rGPsd9++/Hxj38cePpjakfyyNoXvOAFG7arCoA3velNXHPNNXzjG9/g6KOP5i/+4i/Ycccdeetb38oFF1zQ1bVJkiRtribTGMwZTJIkTSIHHHAA8+fP52c/+xkAq1at4sEHH+RHP/oRv/Zrv8a73/1uPvzhD7Ns2bIN53z1q1/d8PP1r389AL/3e7/HwoULATj//PN54xvf+Kzve++997LLLrvwvve9j/e+970sW7aMfffdl//4j/9gxYoVAPz85z/nBz/4wahfsyRJUq9NhjGYM5gkSZpE9t9/f+64444Ng5Rtt92WL3/5y6xYsYIPf/jDbLHFFmy11VacddZZG8555JFHeNWrXsULXvCCDb/p+od/+AeOOeYYTj/99A03mHw2V199NaeffjpbbbUV2267Leeddx5Tp07lS1/6EkceeSRPPPEEAKeeeiq/+Zu/OUZXL0mS1BuTYQyW9dOm+snAwEAtWbKk12lI6iGfIqfJ6o477uB3fud3Rq2/GTNmsGTJEnbeeedR63NTDHVdSZZW1UCPUtIQHINJUu/5FLmnG68xfT+OwZ7v+MslcpIkSZIkSeqKS+QkSdKw7rnnnl6nIEmSNOlMxDGYM5gkSZIkSZLUFQtMkiRJkiRJ6ooFJkmSJEmSJHXFApMkSZIkSZK64k2+JUmahEb7EcYjfQTwZZddxgc/+EHWrVvHe9/7XubNmzeqeUiSJG3OejEGG6/xlzOYJEnSuFi3bh0nnHAC3/zmN7n99tu54IILuP3223udliRJUt8az/GXBSZJkjQubrjhBvbYYw9+4zd+g6233pojjjiCiy++uNdpSZIk9a3xHH9ZYJIkSeNi1apV7Lbbbhv2p0+fzqpVq3qYkSRJUn8bz/HXmBWYkuyW5Koktye5LckHW3ynJIuT3Nl+7tjiSfL3SVYkuSXJawb1Nbe1vzPJ3LHKWZIkSZIkSc/fWM5gehL4y6raE9gXOCHJnsA84MqqmgVc2fYBDgJmtdfxwFnQKUgBJwP7AHsDJ68vSkmSpIlj2rRp3HfffRv2V65cybRp03qYkSRJUn8bz/HXmBWYqmp1VS1r2z8F7gCmAQcDC1qzBcAhbftg4LzquB7YIcmuwAHA4qpaW1WPAIuBA8cqb0mSNDZe97rXceedd3L33Xfzi1/8goULF/K2t72t12lJkiT1rfEcf00Zk143kmQGsBfwHWCXqlrdDt0P7NK2pwH3DTptZYsNF5ckSZtoJI+0HW1Tpkzh85//PAcccADr1q3j2GOP5RWveMW45yFJktQr4z0GG8/x15gXmJJsC3wd+FBV/STJhmNVVUlqlN7neDpL69h9991Ho0tJkjTK5syZw5w5c3qdhiRJ0qQxXuOvMX2KXJKt6BSXzq+qf23hB9rSN9rPB1t8FbDboNOnt9hw8aepqrOraqCqBqZOnTq6FyJJkiRJkqRhjeVT5AKcA9xRVZ8bdGgRsP5JcHOBiwfFj2pPk9sXeLQtpbsc2D/Jju3m3vu3mCRJkiRJkjYDY7lE7g3Ae4DlSW5qsY8CnwYuTHIccC9wWDt2KTAHWAE8BhwDUFVrk3wKuLG1O6Wq1o5h3pIkSZIkSXoexqzAVFXXAhnm8H5DtC/ghGH6mg/MH73sJEmSJEmSNFrG9B5MkiRJkiRJ6n8WmCRJkvpMknuSLE9yU5IlLbZTksVJ7mw/d2zxJPn7JCuS3JLkNYP6mdva35lk7nDvJ0mSNJb3YJIkSZupH57yylHtb/ePL3/ONsceeyyXXHIJL3nJS7j11ltH9f01pD+oqocG7c8DrqyqTyeZ1/Y/AhwEzGqvfYCzgH2S7AScDAwABSxNsqiqHhnPi5AkqZ/08xjMGUySJGlcHH300Vx22WW9TmMyOxhY0LYXAIcMip9XHdcDOyTZFTgAWFxVa1tRaTFw4HgnLUmSujNeYzALTJIkaVy86U1vYqeddup1GpNFAVckWZrk+BbbpapWt+37gV3a9jTgvkHnrmyx4eJPk+T4JEuSLFmzZs1oXoMkSRoF4zUGc4mcJElS//n9qlqV5CXA4iTfG3ywqipJjcYbVdXZwNkAAwMDo9KnJEmaeJzBJEmS1GeqalX7+SBwEbA38EBb+kb7+WBrvgrYbdDp01tsuLgkSdIzWGCSJEnqI0lelGS79dvA/sCtwCJg/ZPg5gIXt+1FwFHtaXL7Ao+2pXSXA/sn2bE9cW7/FpMkSXoGl8hJkiT1l12Ai5JAZ6z3laq6LMmNwIVJjgPuBQ5r7S8F5gArgMeAYwCqam2STwE3tnanVNXa8bsMSZI0kVhgkiRpEhrJI21H25FHHsnVV1/NQw89xPTp0/nkJz/JcccdN+559Luqugt49RDxh4H9hogXcMIwfc0H5o92jpIkTVb9PAazwCRJksbFBRdc0OsUJEmSJp3xGoN5DyZJkiRJkiR1xQKTJEmSJEmSumKBSZKkPtO5pU7/6LfrkSRJ/amfxiybci0WmCRJ6iPbbLMNDz/8cN8McKqKhx9+mG222abXqUiSJA2rn8Zgmzr+8ibfkiT1kenTp7Ny5UrWrFnT61RGzTbbbMP06dN7nYYkSdKw+m0MtinjLwtMkiT1ka222oqZM2f2Og1JkqRJxTGYS+QkSZIkSZLUJQtMkiRJkiRJ6ooFJkmSJEmSJHXFApMkSZIkSZK6YoFJkiRJkiRJXbHAJEmSJEmSpK5YYJIkSZIkSVJXLDBJkiRJkiSpKxaYJEmSJEmS1BULTJIkSZIkSeqKBSZJkiRJkiR1xQKTJEmSJEmSumKBSZIkSZIkSV2xwCRJkiRJkqSuWGCSJEmSJElSVywwSZIkSZIkqSsWmCRJkiRJktQVC0ySJEmSJEnqigUmSZIkSZIkdcUCkyRJkiRJkrpigUmSJEmSJEldscAkSZIkSZKkrlhgkiRJ6kNJtkzy3SSXtP2ZSb6TZEWSrybZusVf0PZXtOMzBvVxUot/P8kBvbkSSZI0EVhgkiRJ6k8fBO4YtP8Z4Iyq2gN4BDiuxY8DHmnxM1o7kuwJHAG8AjgQ+MckW45T7pIkaYKxwCRJktRnkkwH/gj4YtsP8Bbga63JAuCQtn1w26cd36+1PxhYWFVPVNXdwApg7/G5AkmSNNGMWYEpyfwkDya5dVDsE0lWJbmpveYMOjbkFOwkB7bYiiTzxipfSZKkPvK/gb8Cnmr7LwZ+XFVPtv2VwLS2PQ24D6Adf7S13xAf4hxJkqSnGcsZTF+iM516Y2dU1ez2uhSGn4LdpmF/ATgI2BM4srWVJEnSEJL8MfBgVS0dp/c7PsmSJEvWrFkzHm8pSZI2Q2NWYKqqa4C1I2w+3BTsvYEVVXVXVf0CWNjaSpIkaWhvAN6W5B46Y6e3AGcCOySZ0tpMB1a17VXAbgDt+PbAw4PjQ5yzQVWdXVUDVTUwderU0b8aSZI0IfTiHkwnJrmlLaHbscWGm4I94qnZ/vZMkiQJquqkqppeVTPozBD/dlW9C7gKOLQ1mwtc3LYXtX3a8W9XVbX4Ee0pczOBWcAN43QZkiRpghnvAtNZwMuB2cBq4LOj1bG/PZMkSXpWHwH+IskKOvdYOqfFzwFe3OJ/AcwDqKrbgAuB24HLgBOqat24Zy1JkiaEKc/dZPRU1QPrt5P8E3BJ2322KdjPOTVbkiRJz1RVVwNXt+27GOIpcFX1OPCOYc4/DTht7DKUJEn9YlxnMCXZddDu24H1T5gbbgr2jcCsJDOTbE1nmvei8cxZkiRJkiRJz27MZjAluQB4M7BzkpXAycCbk8wGCrgHeD90pmAnWT8F+0kGTcFOciJwObAlML9N15YkSZIkSdJmYswKTFV15BDhc4aIrW8/5BTsqroUuHQUU5MkSZIkSdIo6sVT5CRJkiRJktRHLDBJkiRJkiSpKxaYJEmSJEmS1BULTJIkSZIkSeqKBSZJkiRJkiR1xQKTJEmSJEmSumKBSZIkSZIkSV2xwCRJkiRJkqSuWGCSJEmSJElSVywwSZIkSZIkqSsWmCRJkiRJktQVC0ySJEmSJEnqigUmSZIkSZIkdcUCkyRJkiRJkrry/7d372GSVeW9x78/wChekNuEEEDJQRKCUZCZcDGYo6CInkRQUTTqDMg5nEQiMR44IeZEEGKCIeqRGE04cWQweFeURwk4zyAXRQIzCsPFC6OiQBAGQRAJROE9f+zVUNN0zzRT3VVd3d/P89RTu9Zee++3undVrXprrbVNMEmSJEmSJKkvJpgkSZIkSZLUFxNMkiRJkiRJ6suUEkxJVkylTJIkSdPHNpgkSRoVm61vZZInAE8Etk2yFZC2agtghxmOTZIkaV6yDSZJkkbNehNMwP8E3gL8KrCKRxo39wDvn8G4JEmS5jPbYJIkaaSsN8FUVe8D3pfkzVX19wOKSZIkaV6zDSZJkkbNhnowAVBVf5/kucDOvdtU1VkzFJckSdK8ZxtMkiSNiiklmJJ8BNgFuAp4sBUXYONGkiRphtgGkyRJo2JKCSZgEbB7VdVMBiNJkqR12AaTJEkjYZMp1rsW+JWZDESSJEmPYhtMkiSNhKn2YNoWuD7JFcADY4VV9bIZiUqSJElgG0ySJI2IqSaYTprJICRJkjShk4YdgCRJ0lRM9SpyF890IJIkSVrXxrTBkjwBuAR4PF1b79NVdWKSXwM+DmwDrALeUFX/meTxdJOGLwR+DBxeVTe2ff05cBTdBOPHVtUF/T8rSZI0F01pDqYkP01yT7vdn+TBJPfMdHCSJEnz2Ua2wR4ADqiqPYA9gYOT7Au8C3hvVT0DuIsucUS7v6uVv7fVI8nuwGuAZwIHAx9Isul0P0dJkjQ3TCnBVFVPqaotqmoLYHPglcAHZjQySZKkeW5j2mDVubc9fFy7FXAA8OlWvgw4tC0f0h7T1h+YJK3841X1QFV9H1gD7D09z0ySJM01U72K3MNao+VzwItnIB5JkiRN4LG0wZJsmuQq4HZgOfBd4CdV9YtW5WZgh7a8A3BTO8YvgLvphtE9XD7BNr3HOjrJyiQr165du1HPTZIkjb4pzcGU5BU9DzcBFgH3z0hEkiRJAja+DVZVDwJ7JtkSOAfYbWYihKo6AzgDYNGiRTVTx5EkSbPbVK8i9/s9y78AbqTrNi1JkqSZ01cbrKp+kuTLwH7Alkk2a72UdgRuadVuAXYCbk6yGfBUusm+x8rH9G4jSZK0jqleRe7ImQ5EkiRJ69qYNliSBcDPW3Jpc+BFdBN3fxk4jO5KckuAz7dNzm2Pv9bWX1hVleRc4KNJ3gP8KrArcEWfT0mSJM1RU72K3I5Jzklye7t9JsmOMx2cJEnSfLaRbbDtgS8nWQ1cCSyvqi8Afwa8NckaujmWPtTqfwjYppW/FTgBoKquAz4JXA+cDxzTht5JkiQ9ylSHyH0Y+Cjwqvb49a3sRTMRlCRJkoCNaINV1WrgOROUf48JrgJXVff37H/8uncC73zMUUuSpHlnqleRW1BVH66qX7TbmcCCGYxLkiRJtsEkSdKImGqC6cdJXt8uebtpktfTTf4oSZKkmWMbTJIkjYSpJpjeCLwa+BFwK90EkEfMUEySJEnq2AaTJEkjYapzMJ0MLKmquwCSbA38HV2jR5IkSTPDNpgkSRoJU+3B9Oyxhg1AVd3JBJNHSpIkaVrZBpMkSSNhqgmmTZJsNfag/Xq23t5PSZa2y+le27tdkuVJbmj3W7XyJDk9yZokq5Ps1bPNklb/hiRLHtvTkyRJGmmPuQ0mSZI0DFNNML0b+FqSU5KcAlwG/O0GtjkTOHhc2QnAiqraFVjRHgO8BNi13Y4GPggPN6JOBPahu6zuib2NLEmSpDluY9pgkiRJAzelBFNVnQW8Arit3V5RVR/ZwDaXAHeOKz4EWNaWlwGH9pSfVZ3LgS2TbA+8GFheVXe27uHLeXTSSpIkaU7amDaYJEnSMEy5i3VVXQ9c3+fxtquqW9vyj4Dt2vIOwE099W5uZZOVS5IkzQvT1AaTJEmaUVMdIjftqqqAmq79JTk6ycokK9euXTtdu5UkSZIkSdIGDHqSyNuSbF9Vt7YhcLe38luAnXrq7djKbgGeP678ool2XFVnAGcALFq0aNoSV4O28Piz+t7HqtMWT0MkkiRJkiRJUzPoHkznAmNXglsCfL6nfHG7mty+wN1tKN0FwEFJtmqTex/UyiRJkiRJkjRLzFgPpiQfo+t9tG2Sm+muBncq8MkkRwE/AF7dqp8HvBRYA9wHHAlQVXe2K6Zc2eqdXFXjJw6XJEmSJEnSEM1YgqmqXjvJqgMnqFvAMZPsZymwdBpDkyRJkiRJ0jQa2iTfkiRJkiRJmhtMMEmSJEmSJKkvg76KnKQR55UOJUmSJEnj2YNJkiRJkiRJfTHBJEmSJEmSpL6YYJIkSZIkSVJfTDBJkiRJkiSpLyaYJEmSJEmS1BcTTJIkSZIkSeqLCSZJkiRJkiT1xQSTJEmSJEmS+mKCSZIkSZIkSX0xwSRJkiRJkqS+mGCSJEmSJElSX0wwSZIkSZIkqS8mmCRJkuaQJDsl+XKS65Ncl+RPWvnWSZYnuaHdb9XKk+T0JGuSrE6yV8++lrT6NyRZMqznJEmSZj8TTJIkSXPLL4D/VVW7A/sCxyTZHTgBWFFVuwIr2mOAlwC7ttvRwAehS0gBJwL7AHsDJ44lpSRJksYzwSRJkjSHVNWtVfX1tvxT4JvADsAhwLJWbRlwaFs+BDirOpcDWybZHngxsLyq7qyqu4DlwMEDfCqSJGmEmGCSJEmao5LsDDwH+Ddgu6q6ta36EbBdW94BuKlns5tb2WTlkiRJj2KCSZIkaQ5K8mTgM8Bbquqe3nVVVUBN03GOTrIyycq1a9dOxy4lSdIIMsEkSZI0xyR5HF1y6eyq+mwrvq0NfaPd397KbwF26tl8x1Y2Wfk6quqMqlpUVYsWLFgwvU9EkiSNjM2GHYA0kxYef1bf+1h12uJpiESSpMFIEuBDwDer6j09q84FlgCntvvP95T/cZKP003ofXdV3ZrkAuCveyb2Pgj480E8B0mSNHpMMEmSJM0tvwO8AbgmyVWt7G10iaVPJjkK+AHw6rbuPOClwBrgPuBIgKq6M8kpwJWt3slVdedgnoIkSRo1JpgkSZLmkKr6CpBJVh84Qf0CjplkX0uBpdMXnSRJmqucg0mSJEmSJEl9McEkSZIkSZKkvphgkiRJkiRJUl9MMEmSJEmSJKkvTvItSZIkSZLmpB+e/KxhhzCrPO3t18zYvu3BJEmSJEmSpL6YYJIkSZIkSVJfTDBJkiRJkiSpLyaYJEmSJEmS1BcTTJIkSZIkSeqLCSZJkiRJkiT1xQSTJEmSJEmS+mKCSZIkSZIkSX0xwSRJkiRJkqS+mGCSJEmSJElSX0wwSZIkSZIkqS8mmCRJkiRJktQXE0ySJEmSJEnqy1ASTEluTHJNkquSrGxlWydZnuSGdr9VK0+S05OsSbI6yV7DiFmSJEmSJEkTG2YPphdU1Z5Vtag9PgFYUVW7AivaY4CXALu229HABwceqSRJkiRJkiY1m4bIHQIsa8vLgEN7ys+qzuXAlkm2H0aAkiRJkiRJerRhJZgK+FKSVUmObmXbVdWtbflHwHZteQfgpp5tb25l60hydJKVSVauXbt2puKWJEmSJEnSOJsN6bj7V9UtSX4ZWJ7kW70rq6qS1GPZYVWdAZwBsGjRose0rSRJkiRJkjbeUHowVdUt7f524Bxgb+C2saFv7f72Vv0WYKeezXdsZZIkSZIkSZoFBp5gSvKkJE8ZWwYOAq4FzgWWtGpLgM+35XOBxe1qcvsCd/cMpZMkSZIkSdKQDWOI3HbAOUnGjv/Rqjo/yZXAJ5McBfwAeHWrfx7wUmANcB9w5OBDliRJkiRJ0mQGnmCqqu8Be0xQ/mPgwAnKCzhmAKFJkiRJkiRpIwzrKnKSJEmSJEmaI4Z1FTmNuIXHn9X3PladtngaIpEkSZIkScNmDyZJkiRJkiT1xQSTJEnSHJJkaZLbk1zbU7Z1kuVJbmj3W7XyJDk9yZokq5Ps1bPNklb/hiRLJjqWJEnSGBNMkiRJc8uZwMHjyk4AVlTVrsCK9hjgJcCu7XY08EHoElLAicA+wN7AiWNJKUmSpImYYJIkSZpDquoS4M5xxYcAy9ryMuDQnvKzqnM5sGWS7YEXA8ur6s6qugtYzqOTVpIkSQ8zwSRJkjT3bVdVt7blHwHbteUdgJt66t3cyiYrf5QkRydZmWTl2rVrpzdqSZI0MkwwSZIkzSNVVUBN4/7OqKpFVbVowYIF07VbSZI0YkwwSZIkzX23taFvtPvbW/ktwE499XZsZZOVS5IkTcgEkyRJ0tx3LjB2JbglwOd7yhe3q8ntC9zdhtJdAByUZKs2ufdBrUySJGlCmw07AEmSJE2fJB8Dng9sm+RmuqvBnQp8MslRwA+AV7fq5wEvBdYA9wFHAlTVnUlOAa5s9U6uqvETh0uSJD3MBJMkSdIcUlWvnWTVgRPULeCYSfazFFg6jaFJkqQ5zCFykiRJkiRJ6osJJkmSJEmSJPXFIXKSNIkfnvysvvfxtLdfMw2RSJIkSdLsZoJJkjSnLDz+rL73seq0xdMQiSRJkjR/OEROkiRJkiRJfTHBJEmSJEmSpL6YYJIkSZIkSVJfTDBJkiRJkiSpLyaYJEmSJEmS1BevIidJ0jg/PPlZfe/jaW+/ZhoikSRJkkaDPZgkSZIkSZLUFxNMkiRJkiRJ6osJJkmSJEmSJPXFBJMkSZIkSZL64iTfkgbOCZQlSZNZePxZww5hVll12uJhhyBJ0pTMmwTTdDRW/ICfn0yGSJIkSZK0fvMmwTQd+k00mGSQNJF+E+Amv+cvE+CSJEmaLUwwzUF+4ZAkSZIkSYNkgkmSRpxJ5dE0HUO3z3nKNAQiSZIkTQMTTJKkgTEZJkmSJM1Nmww7AEmSJEmSJI02ezBJkqbEIV2SJEmSJmMPJkmSJEmSJPXFBJMkSZIkSZL64hA5aQ5xAmVJkiRJ0jCYYJJmCee3kSRJkiSNKofISZIkSZIkqS/2YNLQOJxLkkbD9PSwPK2v7X2/13w1He2lucL3AUma3ezBJEmSJEmSpL6MTIIpycFJvp1kTZIThh2PJEnSfGAbTJIkTcVIJJiSbAr8A/ASYHfgtUl2H25UkiRJc5ttMEmSNFWjMgfT3sCaqvoeQJKPA4cA1w81KkmSpLnNNpg0zgHISvoAAA/ESURBVHTMSzeXrDpt8bBDkDRLjEqCaQfgpp7HNwP7DCkWSZKk+cI2mKT1ciL6RzgRvea7VNWwY9igJIcBB1fVf2+P3wDsU1V/3FPnaODo9vA3gG/PQCjbAnfMwH6nm3FOr1GJE0YnVuOcXqMSJ4xOrMY5/WYi1qdX1YJp3qd6zKI22Hw2Sq9zzU+eo5rtPEen16Ttr1HpwXQLsFPP4x1b2cOq6gzgjJkMIsnKqlo0k8eYDsY5vUYlThidWI1zeo1KnDA6sRrn9BulWLWOWdEGm8987Wi28xzVbOc5OjgjMck3cCWwa5JfS/JLwGuAc4cckyRJ0lxnG0ySJE3JSPRgqqpfJPlj4AJgU2BpVV035LAkSZLmNNtgkiRpqkYiwQRQVecB5w05jFHp/m2c02tU4oTRidU4p9eoxAmjE6txTr9RilU9ZkkbbD7ztaPZznNUs53n6ICMxCTfkiRJkiRJmr1GZQ4mSZIkSZIkzVImmDZSkkOT7D7sOEZBksuGHcNUJDk2yTeTnD3J+iOSvH/QcUm9RuX1NGZD8Sa5KIlX9ZhjkpyU5LhhxyHNtCRbJnnTFOrd28cxbkyy7cZuL20s274aJb3n61TaIUmen+QLg4lu/jDBtPEOBUwwTUFVPXfYMUzRm4AXVdXrhh3ITEoyq+dem+3xDdsIvZ6A0YtXkh6jLenaD5KkOczvKFNjgmmcJG9Ncm27vaWVLU6yOsnVST6S5LnAy4DTklyVZJcBx7hz62nz/5Jcl+RLSTZPsmeSy1us5yTZKsluSa4Yt+01A4733iRPTrIiydeTXJPkkPU9l0HG1+L4R+C/AP+a5C+SLE1yRZJvjMXa7NR6XNyQ5MQBx7hzkm8lOTPJd5KcneSFSb7a4tm73b7W4r4syW+0bY9Icm6SC4EVA4r3c0lWtf/r0a3s3iTvbWUrkixo5Rcl+b9JVgJ/MoDY+vlbXpJkz559fSXJHjMdc8/x7h3/i0uS9yc5oi3fmOQdPa+13QYV20Q2FO+wTHZ+9qw/LMmZbXmX9t56TZK/Sh89ETYizicl+WL7/Lk2yeFJFia5uMV/QZLtW92LkryvfS5dm2TvQcXZjv8X7fX0FWDs9bJLkvNbrJeOnY9JtmufU1e3m4lIjapTgV3a6+60JMcnuTJdW+wdE20wUZ2ez6Wz07WLPp3kiT2bvXn8+3qSrdt72er2HvXsVn5SunbMRUm+l+TYnmO/Pl375qok/5Rk05n702i2aefZtT2Pj2vny0VJ3tXOje8ked4E2/631i7atrWfTm/to+8lOazVSXsdXNvO1cNb+T8keVlbPifJ0rb8xiTvzCz5PqDZKY/+Lr4gyWfa++iVSX5nA9s/3Fu+nb83TlBn1nyHGnUmmHokWQgcCewD7Av8j3bC/h/ggKraA/iTqroMOBc4vqr2rKrvDiHcXYF/qKpnAj8BXgmcBfxZVT0buAY4saq+BfxSkl9r2x0OfGII8d4PvLyq9gJeALw7Sdq6iZ7LQFXVHwL/3mJ7EnBhVe3dHp+W5Emt6t4tvmcDr8rgh/Y8A3g3sFu7/QGwP3Ac8DbgW8Dzquo5wNuBv+7Zdi/gsKr6rwOK9Y1VtRBYBBybZBu6v+3K9r++GOhN0v1SVS2qqncPKL6N/Vt+CDgCIMmvA0+oqqsHFPNU3dFeax+kez56tInOz8m8D3hfVT0LuHkg0T3iYODfq2qPqvot4Hzg7+leywuBpcA7e+o/sar2pOtRsXRQQbbPz9cAewIvBX67rToDeHOL9TjgA638dODi9rm6F+Bl7zWqTgC+2153y+naNHvTvRYWJvnd3spJDlpPnd8APlBVvwncw7o9oyZ6X38H8I3W7nsbXTtwzG7Ai9txTkzyuCS/SdcO/J0W74PAnO61rcdks9b2fQvrts9I8nK6c/2lVXVHK96ert30e3SJVoBX0J3XewAvpGtDbw9cCowlrXbgkVEgzwMuactD/z6g2SfJMxn3XZyuXfbeqvptuvPkn6fhULPpO9RIs5vXuvYHzqmqnwEk+Szdl49Pjb2ZVtWdQ4yv1/er6qq2vArYBdiyqi5uZcuAT7XlT9I1KE5t94cPMtAmwF+3RtRDdB8u27V145/LzoMPbx0HAS/LI+N2nwA8rS0vr6ofw8Pnx/7AygHG9v2quqYd/zpgRVVVul5pOwNPBZYl2RUo4HE92y4f8Pl7bGuQAOxE13B4iEcSnP8CfLan/qATnxv7t/wU8JdJjgfeCJw54LinYuzvuoqusadHm+j8nMx+dMOiAT4K/N1MBjbONXQJ+XcBXwDuAn4LWN5y9JsCt/bU/xhAVV2SZIskW1bVTwYQ5/PoPj/vA0hyLt1753OBTz3yewKPb/cHAItbrA8Cdw8gRmmmHdRu32iPn0z33nLJFOr8ELipqr7ayv8FOJZH3m8mel/fn/YlvKouTLJNki3aui9W1QPAA0lup2tzHQgsBK5sr8nNgdv7fM6aO3rPsZ17yg+g+z50UFXd01P+uap6CLg+yVibfn/gY+19/bYkF9P94HAp8JZ089deD2zVEk/70Z3n2zD7vg9odjiAcd/Fk7wQ2L2nbbFFkif3eZzZ9B1qpJlgGl0P9Cw/SDcHwGQ+QdfA/yxQVXXDjEY2sdcBC4CFVfXz1jXxCW3d+Ocy7C6xAV5ZVd9epzDZh+4Np9f4xzOt92/1UM/jh+hez6cAX66qlyfZGbiop/7PBhAf0E2aR/fL1X5VdV+Si3jk/92r9+83sPiajfpbtuezHDgEeDVdY33QfsG6PVDH/23HnsuDzI73+Q3FO1DrOT97z8ehxjimqr6TZC+6XkF/BVwIXFdV+022yQYeD9ImwE9aTwlpPgjwN1X1T4+1TvucWd/r97G+r49vW23Wjr2sqv58Cttrblrf5/Fk59h36aaR+HXW/VG19xwL61FVtyTZkq5X7iXA1nRtqHur6qetF/Fs+z6g2WsTYN+qur+3sCfhNF7veT9Z+25WfIeaCxwit65LgUOTPLENiXo53Rvpq8aGTyTZutX9KfCU4YQ5obuBu3rGTL+BbggSbQjfg8BfMpzhcdBlhW9vyaUXAE8fUhxTcQHdXAcBSPKcnnUvSjfnweZ0PRq+OtEOhuipwC1t+Yghx3FX+/K+G92QU+jecw5ry38AfGUYwU3R+v6W/0w3xOfKqrprkEE1P6D75ebxrcF24BBieCxmW7yTnZ+3JfnNJJvQvf+PuZxHuuq/ZoBxkuRXgfuq6l+A0+iGcC9Isl9b/7jWfXzM2HwX+wN3V9WgegZdQvf5uXmSpwC/D9wHfD/Jq1pMySPzla0A/qiVb5rkqQOKU5puve3BC4A3jv2SnmSHJL88rv766jxt7LXN1D4jL6UNcWuJ8zvG9TAZbwVw2NjxWntmNrfHNP1uA3659XZ7PN3wtg35AW0qjnGfNxO5FDi8va8vAH4XGJsL9nK64XeXtHrHtXtpfS7k0d/FvwS8eaxCeuZGncSNPPKD8GGT1Jkt36FGngmmHlX1dbrhLlcA/wb8c+uq/E7g4iRXA+9p1T8OHJ9uIrCBTvK9Hkvoxjqvphv/fHLPuk8Ar6cbLjdoBZwNLGrDjxbTjXOdrU6h6xa5ug2dOqVn3RXAZ4DVwGeqapDD46bib4G/SfINhttz5XxgsyTfpBuaeXkr/xmwd7oJJg9g3XN0tpn0b1lVq+jmx/jwEOKqqrqJ7rV8bbv/xvo3GarZGO9k5+cJdMPQLmPdYWdvAd7a3lufwWCHcz0LuCLJVXRzYrydrnH0rvaZdBXdMLQx97dz9h+BowYVZPv8/ARwNfCvwJVt1euAo1qs19H1/INuDoUXtM+EVXhVVo2oNmz+q+1z7UV0w2i/1s7tTzPux8iq+tJ66nwbOKa9N21FN9/S+pxEN4fTarr3siUbiPV6urlMvtS2WU43j47miar6OV3b6wq6//+U2uNtTtfX0Y2IWN/3nnPo2shX0yUG/ndV/aitu5Runqc1wNfpejGZYNJ6VdV1PPq7+LF03ytXJ7ke+MMN7ObvgD9q7aNtJ6kzW75DjbxUDbP3vOa6lm3+elX5C5lIcm9V9TtGeuhar5KLgN3a/AODOu5IvZ5GLd7JpLuS03+0ObpeA7y2qg7Z0HaD1ob6HTcLE9+SNqANyfhCm8xfkqSRZHZOM6bnS/ggJ8SVZlSSxXS/pLx1wMmlkXo9jVq8G7AQeH8bNvsTusndJUmSJPWwB5MkSZIkSZL64hxMkiRJkiRJ6osJJkmSJEmSJPXFBJMkSZIkSZL6YoJJ0pyWZMskb+p5/PwkXxhmTJIkSXNBkp2T/EEf2x/RLgwiaQ4wwSRprtsSeNMGa0mSJOmx2hnY6AQTcARggkmaI0wwSZo12q9g30pyZpLvJDk7yQuTfDXJDUn2TrJ1ks8lWZ3k8iTPbtuelGRpkouSfC/JsW23pwK7JLkqyWmt7MlJPt2OdXa7/LwkSdK8kmRxa1NdneQjrS12YStbkeRprd6ZSU5PcllrZx3WdnEq8LzWzvrTtv2lSb7ebs/tOdafJbmmHevUto9FwNlt+80H/xeQNJ1SVcOOQZKALsEErAGeA1wHXAlcDRwFvAw4ErgJuKOq3pHkAOA9VbVnkpOAg4AXAE8Bvg38CrAD8IWq+q12jOcDnweeCfw78FXg+Kr6ykCepCRJ0iyQ5JnAOcBzq+qOJFsDy4BPV9WyJG8EXlZVhyY5E3gScDiwG3BuVT2jtauOq6rfa/t8IvBQVd2fZFfgY1W1KMlLgL8EXlhV9yXZuqruTHJR237lYJ+9pJmw2bADkKRxvl9V1wAkuQ5YUVWV5Bq6bthPB14JUFUXJtkmyRZt2y9W1QPAA0luB7ab5BhXVNXN7RhXtf2aYJIkSfPJAcCnquoOgJbw2Q94RVv/EeBve+p/rqoeAq5PMlkb63HA+5PsCTwI/HorfyHw4aq6b+xY0/tUJM0GJpgkzTYP9Cw/1PP4Ibr3rJ9PcdsHmfw9bqr1JEmS1OltP002vcCfArcBe9BNx3L/TAclafZwDiZJo+ZS4HXw8HC3O6rqnvXU/yndkDlJkiQ94kLgVUm2AWhD5C4DXtPWv46u3bU+49tZTwVubT2d3gBs2sqXA0e2IXRjx5poe0kjzF/tJY2ak4ClSVYD9wFL1le5qn7cJgm/FvhX4IszH6IkSdLsVlXXJXkncHGSB4FvAG8GPpzkeGAt3fyX67MaeDDJ1cCZwAeAzyRZDJwP/Kwd6/w2bG5lkv8EzgPe1rb5xyT/AexXVf8xzU9T0gA5ybckSZIkSZL64hA5SZIkSZIk9cUEkyRJkiRJkvpigkmSJEmSJEl9McEkSZIkSZKkvphgkiRJkiRJUl9MMEmSJEmSJKkvJpgkSZIkSZLUFxNMkiRJkiRJ6sv/B5teCEQ08kBOAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1440x1800 with 10 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KhRhBlzVL3Sm",
        "colab_type": "text"
      },
      "source": [
        "## 6.Are the features about the previous campaign data useful?\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "W8w1Vnk8KVSV",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 0
        },
        "outputId": "b9283eeb-bee5-4160-ffe3-e3a1f08204f1"
      },
      "source": [
        "df[numerical_cols].corr()['response']"
      ],
      "execution_count": 219,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "age         0.072577\n",
              "salary      0.036774\n",
              "balance     0.076995\n",
              "day         0.029952\n",
              "duration    0.342610\n",
              "campaign   -0.094110\n",
              "pdays      -0.152206\n",
              "previous    0.008622\n",
              "response    1.000000\n",
              "Name: response, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 219
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aDsvtpX6OdtT",
        "colab_type": "text"
      },
      "source": [
        "From the above, we can see previous collumns is not highly corelated with the response variable."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bqOgcmfmOuW0",
        "colab_type": "text"
      },
      "source": [
        "## 7. Are pdays and poutcome associated with the target? If yes, and if you plan to use them – how do you handle the pdays column with a value of -1 where the previous campaign data is missing? Explain your approach and your decision.\n",
        "\n",
        "\n",
        "yes, if the previous outcome is success there is a high chance to get successful response again, so i will use poutcome in my final model. \n",
        "\n",
        "if pdays = -1, it indicates that this person was not previosly contacted so, i have dropped these records because we want to study on people who was involved in the campaign\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jEEDr7EkQQgP",
        "colab_type": "text"
      },
      "source": [
        "## Data Preprocessing"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_8Q1GZygSHj1",
        "colab_type": "text"
      },
      "source": [
        "Dropping collumns with unknown values, because those hurts the distributions"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "G0LOTCI-C_mK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df.drop(df[df['poutcome']=='unknown'].index, inplace=True)\n",
        "df.drop(df[df['contact']=='unknown'].index, inplace=True)\n",
        "df.drop(df[df['job']=='unknown'].index, inplace=True)"
      ],
      "execution_count": 220,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wrAuWQY5SNyx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df.job = df.job.str.replace('admin.', 'admin') "
      ],
      "execution_count": 221,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JqaK53BQQqK-",
        "colab_type": "text"
      },
      "source": [
        "Convert Categorical Features into Numerical"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VkvTNvpGwh9Q",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_final = df.copy(deep=True)"
      ],
      "execution_count": 222,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Id3JKOXQOQiW",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.preprocessing import LabelEncoder\n",
        "df_final[categorical_cols] = df[categorical_cols].apply(LabelEncoder().fit_transform)"
      ],
      "execution_count": 223,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eynuvFNpORAY",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "outputId": "df5a71e2-72ed-4f80-980e-37b4b9ce32f3"
      },
      "source": [
        "df_final.head()"
      ],
      "execution_count": 224,
      "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>age</th>\n",
              "      <th>job</th>\n",
              "      <th>salary</th>\n",
              "      <th>marital</th>\n",
              "      <th>education</th>\n",
              "      <th>targeted</th>\n",
              "      <th>default</th>\n",
              "      <th>balance</th>\n",
              "      <th>housing</th>\n",
              "      <th>loan</th>\n",
              "      <th>contact</th>\n",
              "      <th>day</th>\n",
              "      <th>month</th>\n",
              "      <th>duration</th>\n",
              "      <th>campaign</th>\n",
              "      <th>pdays</th>\n",
              "      <th>previous</th>\n",
              "      <th>poutcome</th>\n",
              "      <th>response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>24060</th>\n",
              "      <td>33</td>\n",
              "      <td>0</td>\n",
              "      <td>50000</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>882</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>21</td>\n",
              "      <td>10</td>\n",
              "      <td>39</td>\n",
              "      <td>1</td>\n",
              "      <td>151</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24062</th>\n",
              "      <td>42</td>\n",
              "      <td>0</td>\n",
              "      <td>50000</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>-247</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>21</td>\n",
              "      <td>10</td>\n",
              "      <td>519</td>\n",
              "      <td>1</td>\n",
              "      <td>166</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24064</th>\n",
              "      <td>33</td>\n",
              "      <td>7</td>\n",
              "      <td>70000</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3444</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>21</td>\n",
              "      <td>10</td>\n",
              "      <td>144</td>\n",
              "      <td>1</td>\n",
              "      <td>91</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24072</th>\n",
              "      <td>36</td>\n",
              "      <td>4</td>\n",
              "      <td>100000</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>2415</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>22</td>\n",
              "      <td>10</td>\n",
              "      <td>73</td>\n",
              "      <td>1</td>\n",
              "      <td>86</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24077</th>\n",
              "      <td>36</td>\n",
              "      <td>4</td>\n",
              "      <td>100000</td>\n",
              "      <td>1</td>\n",
              "      <td>2</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>23</td>\n",
              "      <td>10</td>\n",
              "      <td>140</td>\n",
              "      <td>1</td>\n",
              "      <td>143</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       age  job  salary  marital  ...  pdays  previous  poutcome  response\n",
              "24060   33    0   50000        1  ...    151         3         0         0\n",
              "24062   42    0   50000        2  ...    166         1         1         1\n",
              "24064   33    7   70000        1  ...     91         4         0         1\n",
              "24072   36    4  100000        1  ...     86         4         1         0\n",
              "24077   36    4  100000        1  ...    143         3         0         1\n",
              "\n",
              "[5 rows x 19 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 224
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hglouAvaSv21",
        "colab_type": "text"
      },
      "source": [
        "Looks good!"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hfpvWnUIX8Q-",
        "colab_type": "text"
      },
      "source": [
        "## Train-Test Split"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "phMBvTb0X7nc",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "y = df_final.pop('response')\n",
        "X = df_final"
      ],
      "execution_count": 225,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SgMIeuq_YXGH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "78fed4c6-6211-4116-d404-0fe9cc69b227"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "X_train,X_test,y_train,y_test = train_test_split(X,y,test_size = 0.3,random_state=1)\n",
        "X_train.shape, y_train.shape"
      ],
      "execution_count": 226,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((5704, 18), (5704,))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 226
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0glaVhf9d4jr",
        "colab_type": "text"
      },
      "source": [
        "## Feature scaling"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8GhtCYA4bJNx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "sc = StandardScaler()"
      ],
      "execution_count": 227,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZKbNoCFkeHQ8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "6e675578-ab2a-4784-a96b-e4333cc0049e"
      },
      "source": [
        "sc.fit(X_train[numerical_cols[:-1]]) #Last collumn was response, but we no more have it in X_Train so -1 indicates exclusion of that"
      ],
      "execution_count": 228,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "StandardScaler(copy=True, with_mean=True, with_std=True)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 228
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UjJAiwbiehbO",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "outputId": "b02938f8-334b-449a-bcac-2836f004b640"
      },
      "source": [
        "X_train[numerical_cols[:-1]] = sc.transform(X_train[numerical_cols[:-1]])\n",
        "X_train.head()"
      ],
      "execution_count": 229,
      "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>age</th>\n",
              "      <th>job</th>\n",
              "      <th>salary</th>\n",
              "      <th>marital</th>\n",
              "      <th>education</th>\n",
              "      <th>targeted</th>\n",
              "      <th>default</th>\n",
              "      <th>balance</th>\n",
              "      <th>housing</th>\n",
              "      <th>loan</th>\n",
              "      <th>contact</th>\n",
              "      <th>day</th>\n",
              "      <th>month</th>\n",
              "      <th>duration</th>\n",
              "      <th>campaign</th>\n",
              "      <th>pdays</th>\n",
              "      <th>previous</th>\n",
              "      <th>poutcome</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>40491</th>\n",
              "      <td>1.049063</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.250633</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.120390</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>-0.924492</td>\n",
              "      <td>5</td>\n",
              "      <td>-0.763877</td>\n",
              "      <td>-0.053701</td>\n",
              "      <td>-1.328660</td>\n",
              "      <td>0.163412</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>43777</th>\n",
              "      <td>0.961920</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.250633</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.265456</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1.224110</td>\n",
              "      <td>8</td>\n",
              "      <td>1.884625</td>\n",
              "      <td>-0.686580</td>\n",
              "      <td>1.395901</td>\n",
              "      <td>1.171091</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32747</th>\n",
              "      <td>0.526205</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.250633</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.278923</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.339392</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.802510</td>\n",
              "      <td>-0.686580</td>\n",
              "      <td>-0.672247</td>\n",
              "      <td>-0.441195</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44680</th>\n",
              "      <td>-1.216658</td>\n",
              "      <td>6</td>\n",
              "      <td>0.065422</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.317445</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>-1.430045</td>\n",
              "      <td>11</td>\n",
              "      <td>1.214987</td>\n",
              "      <td>4.376450</td>\n",
              "      <td>2.178200</td>\n",
              "      <td>-0.239659</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32910</th>\n",
              "      <td>0.177632</td>\n",
              "      <td>4</td>\n",
              "      <td>1.329641</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.438022</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.339392</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.562127</td>\n",
              "      <td>1.212056</td>\n",
              "      <td>1.027231</td>\n",
              "      <td>-0.441195</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            age  job    salary  marital  ...  campaign     pdays  previous  poutcome\n",
              "40491  1.049063    0 -0.250633        2  ... -0.053701 -1.328660  0.163412         0\n",
              "43777  0.961920    0 -0.250633        2  ... -0.686580  1.395901  1.171091         0\n",
              "32747  0.526205    0 -0.250633        2  ... -0.686580 -0.672247 -0.441195         0\n",
              "44680 -1.216658    6  0.065422        2  ...  4.376450  2.178200 -0.239659         0\n",
              "32910  0.177632    4  1.329641        1  ...  1.212056  1.027231 -0.441195         1\n",
              "\n",
              "[5 rows x 18 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 229
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kMHK3-2hfI-M",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "outputId": "c3380a2f-af08-4856-8ddd-e99d9d982fc0"
      },
      "source": [
        "X_test[numerical_cols[:-1]] = sc.transform(X_test[numerical_cols[:-1]])\n",
        "X_test.head()"
      ],
      "execution_count": 230,
      "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>age</th>\n",
              "      <th>job</th>\n",
              "      <th>salary</th>\n",
              "      <th>marital</th>\n",
              "      <th>education</th>\n",
              "      <th>targeted</th>\n",
              "      <th>default</th>\n",
              "      <th>balance</th>\n",
              "      <th>housing</th>\n",
              "      <th>loan</th>\n",
              "      <th>contact</th>\n",
              "      <th>day</th>\n",
              "      <th>month</th>\n",
              "      <th>duration</th>\n",
              "      <th>campaign</th>\n",
              "      <th>pdays</th>\n",
              "      <th>previous</th>\n",
              "      <th>poutcome</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>25359</th>\n",
              "      <td>0.003346</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.250633</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.077231</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.465780</td>\n",
              "      <td>9</td>\n",
              "      <td>0.708467</td>\n",
              "      <td>-0.053701</td>\n",
              "      <td>-0.420473</td>\n",
              "      <td>-0.239659</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36458</th>\n",
              "      <td>0.526205</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.250633</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.746450</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>-0.418938</td>\n",
              "      <td>8</td>\n",
              "      <td>-0.866898</td>\n",
              "      <td>0.579177</td>\n",
              "      <td>1.000255</td>\n",
              "      <td>-0.038124</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35041</th>\n",
              "      <td>-1.216658</td>\n",
              "      <td>8</td>\n",
              "      <td>-1.704486</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.476796</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>-1.050880</td>\n",
              "      <td>8</td>\n",
              "      <td>-0.849728</td>\n",
              "      <td>1.212056</td>\n",
              "      <td>1.207070</td>\n",
              "      <td>-0.441195</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32655</th>\n",
              "      <td>-0.345227</td>\n",
              "      <td>9</td>\n",
              "      <td>0.065422</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.521017</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.339392</td>\n",
              "      <td>0</td>\n",
              "      <td>2.837571</td>\n",
              "      <td>-0.053701</td>\n",
              "      <td>0.361827</td>\n",
              "      <td>-0.038124</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35572</th>\n",
              "      <td>0.177632</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.250633</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.447105</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.924492</td>\n",
              "      <td>8</td>\n",
              "      <td>0.060292</td>\n",
              "      <td>0.579177</td>\n",
              "      <td>1.216062</td>\n",
              "      <td>-0.239659</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            age  job    salary  marital  ...  campaign     pdays  previous  poutcome\n",
              "25359  0.003346    0 -0.250633        1  ... -0.053701 -0.420473 -0.239659         0\n",
              "36458  0.526205    0 -0.250633        2  ...  0.579177  1.000255 -0.038124         1\n",
              "35041 -1.216658    8 -1.704486        2  ...  1.212056  1.207070 -0.441195         0\n",
              "32655 -0.345227    9  0.065422        1  ... -0.053701  0.361827 -0.038124         0\n",
              "35572  0.177632    0 -0.250633        0  ...  0.579177  1.216062 -0.239659         0\n",
              "\n",
              "[5 rows x 18 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 230
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "m58wvaR-gmjf",
        "colab_type": "text"
      },
      "source": [
        "## Predictive model 1: Logistic regression"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "opU5otGlfRg2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.metrics import confusion_matrix, accuracy_score,precision_score,recall_score, classification_report\n",
        "from sklearn.model_selection import cross_val_score"
      ],
      "execution_count": 231,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9vkNo75ihMmX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "clf = LogisticRegression()"
      ],
      "execution_count": 232,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zUKEe_bthPgs",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "outputId": "cf6ee58e-be4f-4cf4-cfa1-6854266972a9"
      },
      "source": [
        "clf.fit(X_train, y_train)"
      ],
      "execution_count": 233,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
              "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
              "                   multi_class='auto', n_jobs=None, penalty='l2',\n",
              "                   random_state=None, solver='lbfgs', tol=0.0001, verbose=0,\n",
              "                   warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 233
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1rdU5yW4hiWm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "y_pred = clf.predict(X_test)"
      ],
      "execution_count": 234,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WJ8zo48ODNHw",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def show_report(clf, y_true, y_pred): \n",
        "    print('Accuracy: ',accuracy_score(y_true, y_pred))\n",
        "    print('================================')\n",
        "    print('Precision: ',precision_score(y_true, y_pred))\n",
        "    print('================================')\n",
        "    print('Recall: ',recall_score(y_true, y_pred))\n",
        "    print('================================')\n",
        "    print('\\nClassification Report\\n')\n",
        "    print(classification_report(y_true, y_pred))\n",
        "    print('================================')\n",
        "    print('\\nconfusion_matrix\\n')\n",
        "    print(confusion_matrix(y_true, y_pred))\n",
        "    print('================================')\n",
        "    print('\\ncross validation:',cross_val_score(clf, X, y, cv=5))"
      ],
      "execution_count": 235,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Pj80f01GhsW0",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 476
        },
        "outputId": "8075df4c-93c4-4aae-dbec-5e67b49f2335"
      },
      "source": [
        "show_report(clf, y_pred, y_test)"
      ],
      "execution_count": 236,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy:  0.8253578732106339\n",
            "================================\n",
            "Precision:  0.5051369863013698\n",
            "================================\n",
            "Recall:  0.6812933025404158\n",
            "================================\n",
            "\n",
            "Classification Report\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.93      0.86      0.89      2012\n",
            "           1       0.51      0.68      0.58       433\n",
            "\n",
            "    accuracy                           0.83      2445\n",
            "   macro avg       0.72      0.77      0.73      2445\n",
            "weighted avg       0.85      0.83      0.83      2445\n",
            "\n",
            "================================\n",
            "\n",
            "confusion_matrix\n",
            "\n",
            "[[1723  289]\n",
            " [ 138  295]]\n",
            "================================\n",
            "\n",
            "cross validation: [0.79509202 0.72331288 0.77852761 0.77361963 0.73173726]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ASJWlGuTpwun",
        "colab_type": "text"
      },
      "source": [
        "## RFE Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FIZNDfCDokLD",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.feature_selection import RFE\n",
        "rfe = RFE(clf, 15)"
      ],
      "execution_count": 237,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2swmjv_sokUX",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "93ff6b0f-f86f-49f1-a26a-4e7441a32683"
      },
      "source": [
        "rfe = rfe.fit(X_train, y_train)\n",
        "ranks = rfe.ranking_\n",
        "ranks"
      ],
      "execution_count": 238,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1, 3, 1, 1, 1, 4, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 238
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5Os7jcUusWGM",
        "colab_type": "text"
      },
      "source": [
        "The higher values mean that they were dropped at the early stage and the smaller values mean that they survive till the end and prove themselves fittest for the model. In a nutshell, columns with lower ranks are good for us.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r1HQJk_MokOA",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "outputId": "5b1f9cdf-25d1-4b37-f9f5-7398584bbae9"
      },
      "source": [
        "print(X_train.columns[rfe.support_])\n",
        "cols = X_train.columns[rfe.support_]\n",
        "clf.fit(X_train[cols],y_train)\n",
        "y_pred2 = clf.predict(X_test[cols])"
      ],
      "execution_count": 239,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Index(['age', 'salary', 'marital', 'education', 'default', 'balance',\n",
            "       'housing', 'loan', 'contact', 'day', 'duration', 'campaign', 'pdays',\n",
            "       'previous', 'poutcome'],\n",
            "      dtype='object')\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pMMQY2iCokHh",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 476
        },
        "outputId": "e81df05b-cdb9-4f85-f313-d2fe2fd3dee9"
      },
      "source": [
        "show_report(clf, y_pred2, y_test)"
      ],
      "execution_count": 240,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy:  0.8233128834355828\n",
            "================================\n",
            "Precision:  0.4982876712328767\n",
            "================================\n",
            "Recall:  0.6767441860465117\n",
            "================================\n",
            "\n",
            "Classification Report\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.93      0.85      0.89      2015\n",
            "           1       0.50      0.68      0.57       430\n",
            "\n",
            "    accuracy                           0.82      2445\n",
            "   macro avg       0.71      0.77      0.73      2445\n",
            "weighted avg       0.85      0.82      0.83      2445\n",
            "\n",
            "================================\n",
            "\n",
            "confusion_matrix\n",
            "\n",
            "[[1722  293]\n",
            " [ 139  291]]\n",
            "================================\n",
            "\n",
            "cross validation: [0.79509202 0.72331288 0.77852761 0.77361963 0.73173726]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1CvXkO5stRB-",
        "colab_type": "text"
      },
      "source": [
        "## VIF"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sqx8lTkDokFb",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from statsmodels.stats.outliers_influence import variance_inflation_factor"
      ],
      "execution_count": 241,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5tNT-6DookC2",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 514
        },
        "outputId": "25b7080d-9ea1-4274-ea58-50a86aa4d116"
      },
      "source": [
        "vif = pd.DataFrame()\n",
        "vif['features'] = X_train[cols].columns\n",
        "vif['VIF'] = [variance_inflation_factor(X_train[cols].values, i).round(2) for i in range(X_train[cols].shape[1])]\n",
        "vif = vif.sort_values(by='VIF', ascending=True)\n",
        "vif"
      ],
      "execution_count": 242,
      "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>features</th>\n",
              "      <th>VIF</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>default</td>\n",
              "      <td>1.01</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>day</td>\n",
              "      <td>1.01</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>previous</td>\n",
              "      <td>1.02</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>duration</td>\n",
              "      <td>1.03</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>balance</td>\n",
              "      <td>1.04</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>campaign</td>\n",
              "      <td>1.04</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>contact</td>\n",
              "      <td>1.15</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>loan</td>\n",
              "      <td>1.17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>age</td>\n",
              "      <td>1.20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>salary</td>\n",
              "      <td>1.21</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>pdays</td>\n",
              "      <td>1.21</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>poutcome</td>\n",
              "      <td>1.65</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>housing</td>\n",
              "      <td>2.40</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>education</td>\n",
              "      <td>3.89</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>marital</td>\n",
              "      <td>3.98</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     features   VIF\n",
              "4     default  1.01\n",
              "9         day  1.01\n",
              "13   previous  1.02\n",
              "10   duration  1.03\n",
              "5     balance  1.04\n",
              "11   campaign  1.04\n",
              "8     contact  1.15\n",
              "7        loan  1.17\n",
              "0         age  1.20\n",
              "1      salary  1.21\n",
              "12      pdays  1.21\n",
              "14   poutcome  1.65\n",
              "6     housing  2.40\n",
              "3   education  3.89\n",
              "2     marital  3.98"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 242
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HB3JHixZvhgX",
        "colab_type": "text"
      },
      "source": [
        "## P-value"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h9FzUiQfvhKO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import statsmodels.api as sm\n",
        "from statsmodels.sandbox.regression.predstd import wls_prediction_std\n",
        "a = sm.add_constant(X_train[cols])\n",
        "b = y_train"
      ],
      "execution_count": 243,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BxibeeLvvhIl",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 680
        },
        "outputId": "b6e9d344-6e48-4c58-877b-d872321dc1c4"
      },
      "source": [
        "model = sm.OLS(b, a)\n",
        "results = model.fit()\n",
        "print(results.summary())"
      ],
      "execution_count": 244,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:               response   R-squared:                       0.314\n",
            "Model:                            OLS   Adj. R-squared:                  0.313\n",
            "Method:                 Least Squares   F-statistic:                     173.9\n",
            "Date:                Fri, 28 Aug 2020   Prob (F-statistic):               0.00\n",
            "Time:                        16:04:26   Log-Likelihood:                -2047.9\n",
            "No. Observations:                5704   AIC:                             4128.\n",
            "Df Residuals:                    5688   BIC:                             4234.\n",
            "Df Model:                          15                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const          0.1922      0.017     11.184      0.000       0.158       0.226\n",
            "age            0.0084      0.005      1.550      0.121      -0.002       0.019\n",
            "salary         0.0062      0.005      1.203      0.229      -0.004       0.016\n",
            "marital        0.0134      0.008      1.605      0.109      -0.003       0.030\n",
            "education      0.0250      0.007      3.454      0.001       0.011       0.039\n",
            "default       -0.0313      0.058     -0.538      0.591      -0.145       0.083\n",
            "balance        0.0105      0.005      2.240      0.025       0.001       0.020\n",
            "housing       -0.1664      0.011    -15.459      0.000      -0.187      -0.145\n",
            "loan          -0.0622      0.014     -4.541      0.000      -0.089      -0.035\n",
            "contact       -0.0337      0.018     -1.891      0.059      -0.069       0.001\n",
            "day            0.0053      0.005      1.145      0.252      -0.004       0.014\n",
            "duration       0.1203      0.005     25.816      0.000       0.111       0.129\n",
            "campaign      -0.0221      0.005     -4.709      0.000      -0.031      -0.013\n",
            "pdays          0.0046      0.005      0.907      0.364      -0.005       0.015\n",
            "previous       0.0034      0.005      0.728      0.467      -0.006       0.012\n",
            "poutcome       0.1731      0.006     27.380      0.000       0.161       0.185\n",
            "==============================================================================\n",
            "Omnibus:                      422.816   Durbin-Watson:                   2.014\n",
            "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              521.711\n",
            "Skew:                           0.710   Prob(JB):                    5.15e-114\n",
            "Kurtosis:                       3.424   Cond. No.                         29.0\n",
            "==============================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-8QS35skvhGa",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 514
        },
        "outputId": "1f53afc3-69d9-4807-e841-c3440196773a"
      },
      "source": [
        "pvals = []\n",
        "for i in cols:\n",
        "    a = round(results.pvalues[i],3)\n",
        "    pvals.append(a)\n",
        "\n",
        "a = zip(cols, pvals)\n",
        "df_pvals = pd.DataFrame(a, columns=['Column', 'pvalues']).sort_values('pvalues')\n",
        "df_pvals"
      ],
      "execution_count": 245,
      "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>Column</th>\n",
              "      <th>pvalues</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>housing</td>\n",
              "      <td>0.000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>loan</td>\n",
              "      <td>0.000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>duration</td>\n",
              "      <td>0.000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>campaign</td>\n",
              "      <td>0.000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>poutcome</td>\n",
              "      <td>0.000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>education</td>\n",
              "      <td>0.001</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>balance</td>\n",
              "      <td>0.025</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>contact</td>\n",
              "      <td>0.059</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>marital</td>\n",
              "      <td>0.109</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>age</td>\n",
              "      <td>0.121</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>salary</td>\n",
              "      <td>0.229</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>day</td>\n",
              "      <td>0.252</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>pdays</td>\n",
              "      <td>0.364</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>previous</td>\n",
              "      <td>0.467</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>default</td>\n",
              "      <td>0.591</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       Column  pvalues\n",
              "6     housing    0.000\n",
              "7        loan    0.000\n",
              "10   duration    0.000\n",
              "11   campaign    0.000\n",
              "14   poutcome    0.000\n",
              "3   education    0.001\n",
              "5     balance    0.025\n",
              "8     contact    0.059\n",
              "2     marital    0.109\n",
              "0         age    0.121\n",
              "1      salary    0.229\n",
              "9         day    0.252\n",
              "12      pdays    0.364\n",
              "13   previous    0.467\n",
              "4     default    0.591"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 245
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bQOPqGuYvhEG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "46735bd4-c53a-4704-f27d-a2371b85f604"
      },
      "source": [
        "final_features = df_pvals.nlargest(8, 'pvalues')['Column'].to_list()\n",
        "final_features"
      ],
      "execution_count": 246,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['default', 'previous', 'pdays', 'day', 'salary', 'age', 'marital', 'contact']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 246
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UrcM3yQ34g4d",
        "colab_type": "text"
      },
      "source": [
        "## USing selected features"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MukICXBI1w-L",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "clf.fit(X_train[final_features], y_train)\n",
        "y_pred3 = clf.predict(X_test[final_features])"
      ],
      "execution_count": 247,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "w8MjFaDE3SB9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 476
        },
        "outputId": "4cf49fab-24ad-46e1-bd11-34662deb8c09"
      },
      "source": [
        "show_report(clf, y_pred3, y_test)"
      ],
      "execution_count": 248,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy:  0.7603271983640082\n",
            "================================\n",
            "Precision:  0.0\n",
            "================================\n",
            "Recall:  0.0\n",
            "================================\n",
            "\n",
            "Classification Report\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       1.00      0.76      0.86      2443\n",
            "           1       0.00      0.00      0.00         2\n",
            "\n",
            "    accuracy                           0.76      2445\n",
            "   macro avg       0.50      0.38      0.43      2445\n",
            "weighted avg       1.00      0.76      0.86      2445\n",
            "\n",
            "================================\n",
            "\n",
            "confusion_matrix\n",
            "\n",
            "[[1859  584]\n",
            " [   2    0]]\n",
            "================================\n",
            "\n",
            "cross validation: [0.79509202 0.72331288 0.77852761 0.77361963 0.73173726]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p3H0mk0SBVAj",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "426b8bf6-efc9-4c8b-c048-3165c4ad6772"
      },
      "source": [
        "from sklearn.feature_selection import SelectFromModel\n",
        "smf = SelectFromModel(clf)\n",
        "smf.fit(X_train,y_train)\n",
        "features = smf.get_support()\n",
        "feature_name = X_train.columns[features]\n",
        "feature_name"
      ],
      "execution_count": 249,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['housing', 'loan', 'contact', 'duration', 'poutcome'], dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 249
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-U6s_1f2Bm-H",
        "colab_type": "text"
      },
      "source": [
        "So these are the most imposrtant features"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jhZfAUKf45po",
        "colab_type": "text"
      },
      "source": [
        "## Predictive model 2: Random Forest"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CmQGeKdoFLaF",
        "colab_type": "text"
      },
      "source": [
        "Using all features"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UOzeNDPh3Zuj",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ca223992-aaa9-47ae-f24e-9f3d26d4278e"
      },
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "clf2 = RandomForestClassifier(n_estimators=48,max_depth=5)\n",
        "clf2.fit(X_train,y_train)\n"
      ],
      "execution_count": 269,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight=None,\n",
              "                       criterion='gini', max_depth=5, max_features='auto',\n",
              "                       max_leaf_nodes=None, max_samples=None,\n",
              "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
              "                       min_samples_leaf=1, min_samples_split=2,\n",
              "                       min_weight_fraction_leaf=0.0, n_estimators=48,\n",
              "                       n_jobs=None, oob_score=False, random_state=None,\n",
              "                       verbose=0, warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 269
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n-S_HReCChdd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "y_pred_rf = clf2.predict(X_test)"
      ],
      "execution_count": 270,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nM4vbkA7C7iZ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "733f843f-b959-4abd-b07e-02f632550524"
      },
      "source": [
        "show_report(clf2, y_pred_rf, y_test)"
      ],
      "execution_count": 271,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy:  0.8347648261758691\n",
            "================================\n",
            "Precision:  0.4160958904109589\n",
            "================================\n",
            "Recall:  0.7941176470588235\n",
            "================================\n",
            "\n",
            "Classification Report\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.97      0.84      0.90      2139\n",
            "           1       0.42      0.79      0.55       306\n",
            "\n",
            "    accuracy                           0.83      2445\n",
            "   macro avg       0.69      0.82      0.72      2445\n",
            "weighted avg       0.90      0.83      0.85      2445\n",
            "\n",
            "================================\n",
            "\n",
            "confusion_matrix\n",
            "\n",
            "[[1798  341]\n",
            " [  63  243]]\n",
            "================================\n",
            "\n",
            "cross validation: [0.7797546  0.82883436 0.84846626 0.86196319 0.59975445]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gRFuFx0uHXGU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 303
        },
        "outputId": "a31e7bd6-a793-45ad-b5f2-0ada786abb69"
      },
      "source": [
        "importances=clf2.feature_importances_\n",
        "feature_importances=pd.Series(importances, index=X_train.columns).sort_values(ascending=False)\n",
        "selected_features = feature_importances.index.to_list()[:5]\n",
        "print(selected_features)\n",
        "sns.barplot(x=feature_importances[0:10], y=feature_importances.index[0:10])\n",
        "plt.title('Feature Importance',size=20)\n",
        "plt.ylabel(\"Features\")\n",
        "plt.show()"
      ],
      "execution_count": 272,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['poutcome', 'duration', 'housing', 'pdays', 'month']\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mJnHhBM3FOno",
        "colab_type": "text"
      },
      "source": [
        "Using Selected Features"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YEfJGTpMDAaq",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "clf2.fit(X_train[selected_features], y_train)\n",
        "y_pred_rf2 = clf2.predict(X_test[selected_features])"
      ],
      "execution_count": 275,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_V05XAa8Fcqd",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 476
        },
        "outputId": "7fc09c4e-1f0c-45d2-8e75-23beaa0f55ec"
      },
      "source": [
        "show_report(clf2, y_pred_rf2, y_test)"
      ],
      "execution_count": 276,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy:  0.8441717791411043\n",
            "================================\n",
            "Precision:  0.5291095890410958\n",
            "================================\n",
            "Recall:  0.744578313253012\n",
            "================================\n",
            "\n",
            "Classification Report\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.94      0.86      0.90      2030\n",
            "           1       0.53      0.74      0.62       415\n",
            "\n",
            "    accuracy                           0.84      2445\n",
            "   macro avg       0.74      0.80      0.76      2445\n",
            "weighted avg       0.87      0.84      0.85      2445\n",
            "\n",
            "================================\n",
            "\n",
            "confusion_matrix\n",
            "\n",
            "[[1755  275]\n",
            " [ 106  309]]\n",
            "================================\n",
            "\n",
            "cross validation: [0.78588957 0.84110429 0.85030675 0.86257669 0.59116022]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bbW9x_nEH5Bq",
        "colab_type": "text"
      },
      "source": [
        "## Compare the performance of the Random Forest and the logistic model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QjL3bcWPH-p9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "4fdcc6ee-665a-4975-d021-e60eb40ef40b"
      },
      "source": [
        "print('Logistic Regression\\n')\n",
        "show_report(clf , y_pred, y_test)\n",
        "print('\\n')\n",
        "print('='*100)\n",
        "print('='*100)\n",
        "print('\\nRandom Forest Classifier\\n')\n",
        "show_report(clf2, y_pred_rf2, y_test)"
      ],
      "execution_count": 277,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Logistic Regression\n",
            "\n",
            "Accuracy:  0.8253578732106339\n",
            "================================\n",
            "Precision:  0.5051369863013698\n",
            "================================\n",
            "Recall:  0.6812933025404158\n",
            "================================\n",
            "\n",
            "Classification Report\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.93      0.86      0.89      2012\n",
            "           1       0.51      0.68      0.58       433\n",
            "\n",
            "    accuracy                           0.83      2445\n",
            "   macro avg       0.72      0.77      0.73      2445\n",
            "weighted avg       0.85      0.83      0.83      2445\n",
            "\n",
            "================================\n",
            "\n",
            "confusion_matrix\n",
            "\n",
            "[[1723  289]\n",
            " [ 138  295]]\n",
            "================================\n",
            "\n",
            "cross validation: [0.79509202 0.72331288 0.77852761 0.77361963 0.73173726]\n",
            "\n",
            "\n",
            "====================================================================================================\n",
            "====================================================================================================\n",
            "\n",
            "Random Forest Classifier\n",
            "\n",
            "Accuracy:  0.8441717791411043\n",
            "================================\n",
            "Precision:  0.5291095890410958\n",
            "================================\n",
            "Recall:  0.744578313253012\n",
            "================================\n",
            "\n",
            "Classification Report\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.94      0.86      0.90      2030\n",
            "           1       0.53      0.74      0.62       415\n",
            "\n",
            "    accuracy                           0.84      2445\n",
            "   macro avg       0.74      0.80      0.76      2445\n",
            "weighted avg       0.87      0.84      0.85      2445\n",
            "\n",
            "================================\n",
            "\n",
            "confusion_matrix\n",
            "\n",
            "[[1755  275]\n",
            " [ 106  309]]\n",
            "================================\n",
            "\n",
            "cross validation: [0.7803681  0.81595092 0.85092025 0.86503067 0.57704113]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kcWv-0VMJLL6",
        "colab_type": "text"
      },
      "source": [
        "So Random Forest has less number of false negatives and slightly better accuracy."
      ]
    }
  ]
}