波士顿房价预测demo.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.6.5"
    },
    "tianchi_metadata": {
      "competitions": [],
      "datasets": [],
      "description": "",
      "notebookId": "204137",
      "source": "dsw"
    },
    "colab": {
      "name": "波士顿房价预测demo.ipynb",
      "provenance": [],
      "include_colab_link": true
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/sujit-168/Boston_house_price/blob/main/%E6%B3%A2%E5%A3%AB%E9%A1%BF%E6%88%BF%E4%BB%B7%E9%A2%84%E6%B5%8Bdemo.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "fcb3c43a-4288-4166-a597-a54207c7f6b5",
        "id": "jpUOXa_NyKW5"
      },
      "source": [
        "# 查看当前kernel下已安装的包  list packages\n",
        "#!pip list --format=columns"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ClbAvwpXyOJO",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7703e32d-4182-4569-a0b4-25f150e5cfd4"
      },
      "source": [
        "!git clone https://github.com/sujit-168/Boston_house_price.git"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cloning into 'Boston_house_price'...\n",
            "remote: Enumerating objects: 37, done.\u001b[K\n",
            "remote: Counting objects: 100% (37/37), done.\u001b[K\n",
            "remote: Compressing objects: 100% (34/34), done.\u001b[K\n",
            "remote: Total 37 (delta 3), reused 30 (delta 2), pack-reused 0\u001b[K\n",
            "Unpacking objects: 100% (37/37), done.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "21e3b2f8-265e-4a6f-bf04-0fdcd280cdd2",
        "id": "rEuW0A5vyKXA"
      },
      "source": [
        "#!git clone https://github.com/basaraking1221/Boston_house_price.git"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "c2aefec7-4034-4437-8d03-d83b2efe21da",
        "id": "nVQPpKJFyKXB"
      },
      "source": [
        "#!git clone https://github.com/w1449550206/Boston-house-price-forecast.git"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "6ab288f9-d4e5-45a8-8077-18ef3b4488fe",
        "id": "xZvgUifCyKXC",
        "outputId": "09e22f59-91db-4ec8-9ff8-9937db641b0b"
      },
      "source": [
        "!python --version"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Python 2.7.5\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "e6561aac-ba7b-49b1-99ab-ab6fbf48a824",
        "id": "e-tRyK7ByKXC"
      },
      "source": [
        "#检查是否搭载显卡\n",
        "#!nvidia-smi"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "844c73f8-7972-4f5c-9cca-5a44d1c2b386",
        "id": "2zDDUISCyKXD",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3d044b06-23a4-4a59-f2c7-14940c148d66"
      },
      "source": [
        "# sns.histplot(train_data[col[index]],ax = ax[i][j])\n",
        "# AttributeError: module 'seaborn' has no attribute 'histplot'\n",
        "# displot() has been introduced in Seaborn v0.11.0\n",
        "\n",
        "!pip list --outdated #检查哪些包需要更新\n",
        "!pip install --upgrade seaborn    #要升级的包名\n",
        "# pip uninstall 要卸载的包名\n",
        "#!pip show --files seaborn #安装包名"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\u001b[31mERROR: Operation cancelled by user\u001b[0m\n",
            "Requirement already up-to-date: seaborn in /usr/local/lib/python3.7/dist-packages (0.11.1)\n",
            "Requirement already satisfied, skipping upgrade: matplotlib>=2.2 in /usr/local/lib/python3.7/dist-packages (from seaborn) (3.2.2)\n",
            "Requirement already satisfied, skipping upgrade: scipy>=1.0 in /usr/local/lib/python3.7/dist-packages (from seaborn) (1.4.1)\n",
            "Requirement already satisfied, skipping upgrade: pandas>=0.23 in /usr/local/lib/python3.7/dist-packages (from seaborn) (1.1.5)\n",
            "Requirement already satisfied, skipping upgrade: numpy>=1.15 in /usr/local/lib/python3.7/dist-packages (from seaborn) (1.19.5)\n",
            "Requirement already satisfied, skipping upgrade: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=2.2->seaborn) (0.10.0)\n",
            "Requirement already satisfied, skipping upgrade: python-dateutil>=2.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=2.2->seaborn) (2.8.1)\n",
            "Requirement already satisfied, skipping upgrade: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=2.2->seaborn) (2.4.7)\n",
            "Requirement already satisfied, skipping upgrade: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=2.2->seaborn) (1.3.1)\n",
            "Requirement already satisfied, skipping upgrade: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.23->seaborn) (2018.9)\n",
            "Requirement already satisfied, skipping upgrade: six in /usr/local/lib/python3.7/dist-packages (from cycler>=0.10->matplotlib>=2.2->seaborn) (1.15.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "c04cc48a-b287-4c6f-84df-e902501f61e9",
        "id": "nqxe679jyKXD"
      },
      "source": [
        "import seaborn as sns\n",
        "##用于可视化图表\n",
        "import matplotlib.pyplot as plt\n",
        "##用于做科学计算\n",
        "import numpy as np\n",
        "##用于做数据分析\n",
        "import pandas as pd\n",
        "##用于加载数据或生成数据等"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "1fa67f6b-01a6-49aa-9191-8c476f84b727",
        "id": "zwBvRVW-yKXE"
      },
      "source": [
        "from sklearn import datasets\n",
        "##加载线性模型\n",
        "from sklearn import linear_model\n",
        "###用于交叉验证以及训练集和测试集的划分\n",
        "#from sklearn.cross_validation import train_test_split\n",
        "from sklearn.model_selection import cross_val_predict\n",
        "###这个模块中含有评分函数，性能度量，距离计算等\n",
        "from sklearn import metrics\n",
        "#from sklearn.cross_validation import ShuffleSplit"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "1e53433a-6ef7-4032-82de-76332805ff62",
        "id": "JGI9p_akyKXE",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "226947f2-0a28-4871-bb95-e13cceda4fd1"
      },
      "source": [
        "!ls\n",
        "%cd Boston_house_price\n",
        "%cd /content/Boston_house_price/Demo\n",
        "!ls"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Demo  LICENSE  README.md\n",
            "[Errno 2] No such file or directory: 'Boston_house_price'\n",
            "/content/Boston_house_price\n",
            "/content/Boston_house_price/Demo\n",
            "bostonh.csv\t\tpredictions_Vs_ytest.png  visuals.py\n",
            "bostonh.txt\t\tprice.eps\t\t  波士顿房价预测1.ipynb\n",
            "data——introduction.jpg\tprice.png\t\t  波士顿房价预测2.ipynb\n",
            "model.pkl\t\ttest.pkl\t\t  波士顿房价预测.html\n",
            "output\t\t\ttrain\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "96b0acb2-4ca3-4f90-838f-5cb17d21259a",
        "id": "mSeESlqTyKXF",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "outputId": "5c00f46d-c5ad-4750-d925-1624a7d8f987"
      },
      "source": [
        "#读取txt文件，并转换为csv文件\n",
        "data_txt = pd.read_csv('bostonh.txt',sep='\\t')\n",
        "data_txt.to_csv('bostonh.csv',index=False)  \n",
        "\n",
        "# 导入数据集并查看csv数据\n",
        "train_data=pd.read_csv('bostonh.csv',sep='\\s+',encoding='utf-8')\n",
        "train_data.head()\n",
        "# 导入数据集并查看txt数据\n",
        "#train_data2=pd.read_table('bostonh.txt',encoding='utf-8')\n",
        "#train_data2.head()"
      ],
      "execution_count": null,
      "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>CRIM</th>\n",
              "      <th>ZN</th>\n",
              "      <th>INDUS</th>\n",
              "      <th>CHAS</th>\n",
              "      <th>NOX</th>\n",
              "      <th>RM</th>\n",
              "      <th>AGE</th>\n",
              "      <th>DIS</th>\n",
              "      <th>RAD</th>\n",
              "      <th>TAX</th>\n",
              "      <th>PTRATIO</th>\n",
              "      <th>B</th>\n",
              "      <th>LSTAT</th>\n",
              "      <th>MEDV</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.00632</td>\n",
              "      <td>18.0</td>\n",
              "      <td>2.31</td>\n",
              "      <td>0</td>\n",
              "      <td>0.538</td>\n",
              "      <td>6.575</td>\n",
              "      <td>65.2</td>\n",
              "      <td>4.0900</td>\n",
              "      <td>1</td>\n",
              "      <td>296.0</td>\n",
              "      <td>15.3</td>\n",
              "      <td>396.90</td>\n",
              "      <td>4.98</td>\n",
              "      <td>24.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.02731</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.07</td>\n",
              "      <td>0</td>\n",
              "      <td>0.469</td>\n",
              "      <td>6.421</td>\n",
              "      <td>78.9</td>\n",
              "      <td>4.9671</td>\n",
              "      <td>2</td>\n",
              "      <td>242.0</td>\n",
              "      <td>17.8</td>\n",
              "      <td>396.90</td>\n",
              "      <td>9.14</td>\n",
              "      <td>21.6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.02729</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.07</td>\n",
              "      <td>0</td>\n",
              "      <td>0.469</td>\n",
              "      <td>7.185</td>\n",
              "      <td>61.1</td>\n",
              "      <td>4.9671</td>\n",
              "      <td>2</td>\n",
              "      <td>242.0</td>\n",
              "      <td>17.8</td>\n",
              "      <td>392.83</td>\n",
              "      <td>4.03</td>\n",
              "      <td>34.7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.03237</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.18</td>\n",
              "      <td>0</td>\n",
              "      <td>0.458</td>\n",
              "      <td>6.998</td>\n",
              "      <td>45.8</td>\n",
              "      <td>6.0622</td>\n",
              "      <td>3</td>\n",
              "      <td>222.0</td>\n",
              "      <td>18.7</td>\n",
              "      <td>394.63</td>\n",
              "      <td>2.94</td>\n",
              "      <td>33.4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.06905</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.18</td>\n",
              "      <td>0</td>\n",
              "      <td>0.458</td>\n",
              "      <td>7.147</td>\n",
              "      <td>54.2</td>\n",
              "      <td>6.0622</td>\n",
              "      <td>3</td>\n",
              "      <td>222.0</td>\n",
              "      <td>18.7</td>\n",
              "      <td>396.90</td>\n",
              "      <td>5.33</td>\n",
              "      <td>36.2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "      CRIM    ZN  INDUS  CHAS    NOX  ...    TAX  PTRATIO       B  LSTAT  MEDV\n",
              "0  0.00632  18.0   2.31     0  0.538  ...  296.0     15.3  396.90   4.98  24.0\n",
              "1  0.02731   0.0   7.07     0  0.469  ...  242.0     17.8  396.90   9.14  21.6\n",
              "2  0.02729   0.0   7.07     0  0.469  ...  242.0     17.8  392.83   4.03  34.7\n",
              "3  0.03237   0.0   2.18     0  0.458  ...  222.0     18.7  394.63   2.94  33.4\n",
              "4  0.06905   0.0   2.18     0  0.458  ...  222.0     18.7  396.90   5.33  36.2\n",
              "\n",
              "[5 rows x 14 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "29ad2f27-2c34-4c6d-a7f3-c0a1929abd76",
        "id": "jvRT3H2ayKXF",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ba6522f2-3538-4c8e-c6e5-a4a80f1691c9"
      },
      "source": [
        "a=['CRIM:城镇的人均犯罪率','ZN:大于25,000平方英尺的地块的住宅用地比例','INDUS:每个镇的非零售业务英亩的比例','CHAS:查尔斯河虚拟变量（如果环河，则等于1；否则等于0）'\n",
        "   ,'NOX:一氧化氮的浓度','RM:每个住宅的平均房间数','AGE:1940年之前建造的自有住房的比例','DIS:到五个波士顿就业中心的加权距离','RAD:径向公路通达性的指标'\n",
        "   ,'TAX:每$ 10,000的全值财产税率','PTRATIO:各镇的师生比率','B:计算方法为1000（Bk-0.63）²，其中Bk是按城镇划分的非裔美国人的比例','LSTAT:底层人口的百分比',\n",
        "  'MEDV:自有住房数的中位数，单位为$ 1000']\n",
        "for i in a:\n",
        "    print(i)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "CRIM:城镇的人均犯罪率\n",
            "ZN:大于25,000平方英尺的地块的住宅用地比例\n",
            "INDUS:每个镇的非零售业务英亩的比例\n",
            "CHAS:查尔斯河虚拟变量（如果环河，则等于1；否则等于0）\n",
            "NOX:一氧化氮的浓度\n",
            "RM:每个住宅的平均房间数\n",
            "AGE:1940年之前建造的自有住房的比例\n",
            "DIS:到五个波士顿就业中心的加权距离\n",
            "RAD:径向公路通达性的指标\n",
            "TAX:每$ 10,000的全值财产税率\n",
            "PTRATIO:各镇的师生比率\n",
            "B:计算方法为1000（Bk-0.63）²，其中Bk是按城镇划分的非裔美国人的比例\n",
            "LSTAT:底层人口的百分比\n",
            "MEDV:自有住房数的中位数，单位为$ 1000\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "f07f996a-bdd4-43b7-bff1-35040fd6cf8e",
        "id": "Oj8B3xtkyKXG",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "47a8e91c-f5fd-44d5-eacf-3531b9f78867"
      },
      "source": [
        "#判断是否需要填补缺失值与处理类别变量\n",
        "train_data.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 506 entries, 0 to 505\n",
            "Data columns (total 14 columns):\n",
            " #   Column   Non-Null Count  Dtype  \n",
            "---  ------   --------------  -----  \n",
            " 0   CRIM     506 non-null    float64\n",
            " 1   ZN       506 non-null    float64\n",
            " 2   INDUS    506 non-null    float64\n",
            " 3   CHAS     506 non-null    int64  \n",
            " 4   NOX      506 non-null    float64\n",
            " 5   RM       506 non-null    float64\n",
            " 6   AGE      506 non-null    float64\n",
            " 7   DIS      506 non-null    float64\n",
            " 8   RAD      506 non-null    int64  \n",
            " 9   TAX      506 non-null    float64\n",
            " 10  PTRATIO  506 non-null    float64\n",
            " 11  B        506 non-null    float64\n",
            " 12  LSTAT    506 non-null    float64\n",
            " 13  MEDV     506 non-null    float64\n",
            "dtypes: float64(12), int64(2)\n",
            "memory usage: 55.5 KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "7fbd50a1-a825-4cdb-a90a-9b1df6f0e82e",
        "id": "BIcVZoE6yKXG",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "fb981c5a-588c-48e9-b1de-1a3b771fe546"
      },
      "source": [
        "#结果显示：数据点完备，无需填补；数据皆为浮点型，无需处理类别变量。\n",
        "train_data.isnull().sum() "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CRIM       0\n",
              "ZN         0\n",
              "INDUS      0\n",
              "CHAS       0\n",
              "NOX        0\n",
              "RM         0\n",
              "AGE        0\n",
              "DIS        0\n",
              "RAD        0\n",
              "TAX        0\n",
              "PTRATIO    0\n",
              "B          0\n",
              "LSTAT      0\n",
              "MEDV       0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "f78007f6-91eb-4100-8ecb-ee4269eb7f79",
        "id": "H2k5b3IsyKXG",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ea1e306e-b90e-4ead-a94f-7dfd3cf6ccf0"
      },
      "source": [
        "#计算每一个特征和price的相关系数\n",
        "train_data.corr()['MEDV']"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CRIM      -0.388305\n",
              "ZN         0.360445\n",
              "INDUS     -0.483725\n",
              "CHAS       0.175260\n",
              "NOX       -0.427321\n",
              "RM         0.695360\n",
              "AGE       -0.376955\n",
              "DIS        0.249929\n",
              "RAD       -0.381626\n",
              "TAX       -0.468536\n",
              "PTRATIO   -0.507787\n",
              "B          0.333461\n",
              "LSTAT     -0.737663\n",
              "MEDV       1.000000\n",
              "Name: MEDV, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Brcnm2HzyKXH",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f6a2029e-2652-4acb-b43c-499ba5585579"
      },
      "source": [
        "# 载入波士顿房屋的数据集\n",
        "MEDV = train_data['MEDV']\n",
        "features = train_data.drop('MEDV', axis = 1)\n",
        "# TODO: Minimum price of the data\n",
        "#目标：计算价值的最小值\n",
        "minimum_price = np.min(MEDV)\n",
        "# TODO: Maximum price of the data\n",
        "#目标：计算价值的最大值\n",
        "maximum_price = np.max(MEDV)\n",
        "# TODO: Mean price of the data\n",
        "#目标：计算价值的平均值\n",
        "mean_price = np.mean(MEDV)\n",
        "# TODO: Median price of the data\n",
        "#目标：计算价值的中值\n",
        "median_price = np.median(MEDV)\n",
        "# TODO: Standard deviation of prices of the data\n",
        "#目标：计算价值的标准差\n",
        "std_price = np.std(MEDV)\n",
        "# Show the calculated statistics\n",
        "#目标：输出计算的结果\n",
        "print(\"Statistics for Boston housing dataset:\\n\")\n",
        "print(\"Minimum price: ${:,.2f}\".format(minimum_price))\n",
        "print(\"Maximum price: ${:,.2f}\".format(maximum_price)) \n",
        "print(\"Mean price: ${:,.2f}\".format(mean_price)) \n",
        "print(\"Median price ${:,.2f}\".format(median_price))\n",
        "print(\"Standard deviation of prices: ${:,.2f}\".format(std_price))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Statistics for Boston housing dataset:\n",
            "\n",
            "Minimum price: $5.00\n",
            "Maximum price: $50.00\n",
            "Mean price: $22.53\n",
            "Median price $21.20\n",
            "Standard deviation of prices: $9.19\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "026df8a9-fdbe-451e-90f7-fb43b65ce9c7",
        "id": "D8k9RjMPyKXH"
      },
      "source": [
        "## 找出挖掘数据的统计特征并处理异常值"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "87b95f48-1ec6-430f-b7f2-2737cd5538cb",
        "id": "5M0mTrdiyKXI",
        "outputId": "3ed88c2d-93e7-4090-adb0-ad86e57fb432"
      },
      "source": [
        "train_data.describe() "
      ],
      "execution_count": null,
      "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>CRIM</th>\n",
              "      <th>ZN</th>\n",
              "      <th>INDUS</th>\n",
              "      <th>CHAS</th>\n",
              "      <th>NOX</th>\n",
              "      <th>RM</th>\n",
              "      <th>AGE</th>\n",
              "      <th>DIS</th>\n",
              "      <th>RAD</th>\n",
              "      <th>TAX</th>\n",
              "      <th>PTRATIO</th>\n",
              "      <th>B</th>\n",
              "      <th>LSTAT</th>\n",
              "      <th>MEDV</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "      <td>506.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>3.613524</td>\n",
              "      <td>11.363636</td>\n",
              "      <td>11.136779</td>\n",
              "      <td>0.069170</td>\n",
              "      <td>0.554695</td>\n",
              "      <td>6.284634</td>\n",
              "      <td>68.574901</td>\n",
              "      <td>3.795043</td>\n",
              "      <td>9.549407</td>\n",
              "      <td>408.237154</td>\n",
              "      <td>18.455534</td>\n",
              "      <td>356.674032</td>\n",
              "      <td>12.653063</td>\n",
              "      <td>22.532806</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>8.601545</td>\n",
              "      <td>23.322453</td>\n",
              "      <td>6.860353</td>\n",
              "      <td>0.253994</td>\n",
              "      <td>0.115878</td>\n",
              "      <td>0.702617</td>\n",
              "      <td>28.148861</td>\n",
              "      <td>2.105710</td>\n",
              "      <td>8.707259</td>\n",
              "      <td>168.537116</td>\n",
              "      <td>2.164946</td>\n",
              "      <td>91.294864</td>\n",
              "      <td>7.141062</td>\n",
              "      <td>9.197104</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.006320</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.460000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.385000</td>\n",
              "      <td>3.561000</td>\n",
              "      <td>2.900000</td>\n",
              "      <td>1.129600</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>187.000000</td>\n",
              "      <td>12.600000</td>\n",
              "      <td>0.320000</td>\n",
              "      <td>1.730000</td>\n",
              "      <td>5.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.082045</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>5.190000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.449000</td>\n",
              "      <td>5.885500</td>\n",
              "      <td>45.025000</td>\n",
              "      <td>2.100175</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>279.000000</td>\n",
              "      <td>17.400000</td>\n",
              "      <td>375.377500</td>\n",
              "      <td>6.950000</td>\n",
              "      <td>17.025000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.256510</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>9.690000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.538000</td>\n",
              "      <td>6.208500</td>\n",
              "      <td>77.500000</td>\n",
              "      <td>3.207450</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>330.000000</td>\n",
              "      <td>19.050000</td>\n",
              "      <td>391.440000</td>\n",
              "      <td>11.360000</td>\n",
              "      <td>21.200000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>3.677082</td>\n",
              "      <td>12.500000</td>\n",
              "      <td>18.100000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.624000</td>\n",
              "      <td>6.623500</td>\n",
              "      <td>94.075000</td>\n",
              "      <td>5.188425</td>\n",
              "      <td>24.000000</td>\n",
              "      <td>666.000000</td>\n",
              "      <td>20.200000</td>\n",
              "      <td>396.225000</td>\n",
              "      <td>16.955000</td>\n",
              "      <td>25.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>88.976200</td>\n",
              "      <td>100.000000</td>\n",
              "      <td>27.740000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.871000</td>\n",
              "      <td>8.780000</td>\n",
              "      <td>100.000000</td>\n",
              "      <td>12.126500</td>\n",
              "      <td>24.000000</td>\n",
              "      <td>711.000000</td>\n",
              "      <td>22.000000</td>\n",
              "      <td>396.900000</td>\n",
              "      <td>37.970000</td>\n",
              "      <td>50.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             CRIM          ZN       INDUS        CHAS         NOX          RM  \\\n",
              "count  506.000000  506.000000  506.000000  506.000000  506.000000  506.000000   \n",
              "mean     3.613524   11.363636   11.136779    0.069170    0.554695    6.284634   \n",
              "std      8.601545   23.322453    6.860353    0.253994    0.115878    0.702617   \n",
              "min      0.006320    0.000000    0.460000    0.000000    0.385000    3.561000   \n",
              "25%      0.082045    0.000000    5.190000    0.000000    0.449000    5.885500   \n",
              "50%      0.256510    0.000000    9.690000    0.000000    0.538000    6.208500   \n",
              "75%      3.677082   12.500000   18.100000    0.000000    0.624000    6.623500   \n",
              "max     88.976200  100.000000   27.740000    1.000000    0.871000    8.780000   \n",
              "\n",
              "              AGE         DIS         RAD         TAX     PTRATIO           B  \\\n",
              "count  506.000000  506.000000  506.000000  506.000000  506.000000  506.000000   \n",
              "mean    68.574901    3.795043    9.549407  408.237154   18.455534  356.674032   \n",
              "std     28.148861    2.105710    8.707259  168.537116    2.164946   91.294864   \n",
              "min      2.900000    1.129600    1.000000  187.000000   12.600000    0.320000   \n",
              "25%     45.025000    2.100175    4.000000  279.000000   17.400000  375.377500   \n",
              "50%     77.500000    3.207450    5.000000  330.000000   19.050000  391.440000   \n",
              "75%     94.075000    5.188425   24.000000  666.000000   20.200000  396.225000   \n",
              "max    100.000000   12.126500   24.000000  711.000000   22.000000  396.900000   \n",
              "\n",
              "            LSTAT        MEDV  \n",
              "count  506.000000  506.000000  \n",
              "mean    12.653063   22.532806  \n",
              "std      7.141062    9.197104  \n",
              "min      1.730000    5.000000  \n",
              "25%      6.950000   17.025000  \n",
              "50%     11.360000   21.200000  \n",
              "75%     16.955000   25.000000  \n",
              "max     37.970000   50.000000  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "97cb063d-9295-419d-8d3d-53ad38cf9627",
        "id": "J7D_faZ3yKXI"
      },
      "source": [
        "以上数据显示：房价最大值是最小值的10倍。由于表格太大，其余特征与房价的关系及特征间交叉关系并不直观，故而进入一下可视化的探索。"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "8a51f174-af49-4b48-abc0-87f4f3a878a2",
        "id": "T3Eu2SuhyKXI"
      },
      "source": [
        "## 我们将尝试了解因变量和自变量、自变量和自变量之间的关系。\n",
        "sns.pairplot(train_data) \n",
        "#最后一行和一列为price与其他特征的相关关系图。不难发现：其中NOX,RM,PTRATIO,LSTAT与price有较为明显的线性相关性。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "06853ba2-3cc3-49a8-9e63-526c2e76f89f",
        "id": "bpjBpBblyKXJ"
      },
      "source": [
        "#sns.set() \n",
        "#sns.distplot(train_data.CRIM,hist=True,bins=100,kde=False)\n",
        "#案例2：绘制直方图和核函数密度估计图\n",
        "#sns.distplot(train_data.CRIM)\n",
        "#plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "9a587aeb-0ab2-4bb1-a403-406b48af92fa",
        "id": "FRUx1H13yKXJ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 901
        },
        "outputId": "f3287f9c-adfe-44ac-b9bf-c65ef32c8311"
      },
      "source": [
        "rows = 2\n",
        "cols = 7\n",
        "\n",
        "fig, ax = plt.subplots(nrows=rows,ncols=cols,figsize = (24,8))\n",
        "\n",
        "col = train_data.columns\n",
        "index = 0\n",
        "\n",
        "for i in range(rows):\n",
        "    for j in range(cols):\n",
        "        sns.distplot(train_data[col[index]],ax = ax[i][j])\n",
        "        index = index + 1\n",
        "plt.tight_layout()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABrgAAAI4CAYAAAAxqel1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXhdd33v+/dX8yxLsjxJnhI7AWcgCcKhlDCFgKGQtGVoArThnHBSbsmlbTinhNvewE3Lc4GeC6fnNG0JkIbhhCSE0rrFkIapIYEYK4mJYyeOZcW2JNuarXna0vf+sZecHUWytqS99vh5PY8er/Vbv7X2dz+Pf2vvvb6/wdwdERERERERERERERERkUyRl+oARERERERERERERERERJZCCS4RERERERERERERERHJKEpwiYiIiIiIiIiIiIiISEZRgktEREREREREREREREQyihJcIiIiIiIiIiIiIiIiklEKUh1Aoqxevdq3bNmS6jBEEuKJJ57ocff6VMeRDtS2JZuobb9IbVuyidp2lNq1ZBO16xepbUs2UduOUruWbKJ2/SK1bckm8bbtrElwbdmyhebm5lSHIZIQZnY81TGkC7VtySZq2y9S25ZsorYdpXYt2UTt+kVq25JN1Laj1K4lm6hdv0htW7JJvG1bUxSKiIiIiIiIiIiIiIhIRlGCS0RERERERERERERERDKKElwiIiIiIiIiIiIiIiKSUUJNcJnZLjM7bGYtZnbbOeq9x8zczJpiyj4VnHfYzN4eZpwiIiIiIiIiIiIiIiKSOQrCurCZ5QN3AtcA7cA+M9vt7ofm1KsE/hjYG1O2A7geuAjYAPzIzC5w9+mw4hUREREREREREREREZHMEOYIrp1Ai7u3uvskcB9w3Tz1/hL4PDAeU3YdcJ+7T7j7C0BLcD0RERERERERERERERHJcWEmuBqAtpj99qDsLDO7Atjo7t9f6rnB+TebWbOZNXd3dycmahEREREREZEsZWbHzOyAme03s+ZUxyMiIiIislyhTVG4GDPLA74IfHi513D3u4C7AJqamjwxkYmIiIiIiIhktTe7e0+qgxARERERWYkwE1wdwMaY/cagbFYlcDHwMzMDWAfsNrNr4zhXREREREREREREREREclSYUxTuA7ab2VYzKwKuB3bPHnT3AXdf7e5b3H0L8Dhwrbs3B/WuN7NiM9sKbAd+FWKsIiIiIiIiIrnAgX83syfM7Ob5Kmg5ABERERHJBKGN4HL3iJndAjwE5AN3u/tBM7sDaHb33ec496CZPQAcAiLAx9x9OqxY53Pv3hNntz9w5aZkvrSIpBHdC0QkG8Xe2+Kle6Bkgvn+b+v/rsjLvN7dO8xsDfCwmT3n7o/EVsi05QDi/VzT/UBEkmHuPUn3HhFJtlz6zR/qGlzuvgfYM6fs9gXqvmnO/meBz4YWnIiIiIiIiEiOcfeO4N8uM/sesBN45NxniYiIiIiknzCnKBQRERERkRQxs41m9lMzO2RmB83sj4PyWjN72MyOBP/WLHD+jUGdI2Z2Y3KjF5EwmFm5mVXObgNvA55JbVQiIiIiIsujBJeIiIiISHaKAJ9w9x3Aa4GPmdkO4Dbgx+6+HfhxsP8SZlYLfBq4kujojk8vlAgTkYyyFnjUzH5NdJ3r77v7D1Mck4icg5ntMrPDZtZiZvN9Zm8KOrQ8ZWZPm9k7UxGniIhIKoQ6RaGIiIiIiKSGu58CTgXbQ2b2LNAAXAe8Kaj2deBnwCfnnP524GF37wMws4eBXcC3Qw9cRELj7q3Aq1Idh4jEx8zygTuBa4B2YJ+Z7Xb3QzHV/gJ4wN3/PujIsgfYkvRgRUREUkAjuEREREREspyZbQEuB/YCa4PkF8BpoiM65moA2mL224Oyude92cyazay5u7s7oTGLiIgIO4EWd29190ngPqIdVWI5UBVsVwMnkxifiIhISinBJSIiIiKSxcysAvgu8CfuPhh7zN2d6IOxZXH3u9y9yd2b6uvrVxipiIiIzBFPh5PPAB8ys3aio7f+z/kupE4pIiKSjZTgEhERERHJUmZWSDS59b/d/Z+C4k4zWx8cXw90zXNqB7AxZr8xKBMREZH0cgNwj7s3Au8EvmlmL3vep04pIuGJY628YjO7Pzi+N5hdATPbaWb7g79fm9nvxHtNEYlSgktEREREJAuZmQFfA5519y/GHNoN3Bhs3wj8yzynPwS8zcxqzKwGeFtQJiIiIskTT4eTm4AHANz9l0AJsDop0YlI7Fp57wB2ADcE6+HFugnod/dtwJeAzwflzwBN7n4Z0fVuv2xmBXFeU0RQgktEREREJFv9JvD7wFtieoa+E/gccI2ZHQHeGuxjZk1m9lUAd+8D/hLYF/zdEZSJiIhI8uwDtpvZVjMrAq4n2lEl1gngagAzeyXRBJfmIBRJnnjWyrsO+Hqw/SBwtZmZu4+6eyQoL+HFqcPjuaaIAAWpDkBERERERBLP3R8FbIHDV89Tvxn4SMz+3cDd4UQnIiIii3H3iJndQnQUdT5wt7sfNLM7gGZ33w18AviKmf0p0YfjHw7W2BSR5JhvrbwrF6oTtOsBoA7oMbMriX7n3gz8fnA8nmsC0fX1gJsBNm3atPJ3I5JhlOASERERERERERFJQ+6+B9gzp+z2mO1DREdti0gGcve9wEXBCMyvm9kPlnj+XcBdAE1NTUpuS87RFIUiIiIiIiIiIiIiIksXz1p5Z+uYWQFQDfTGVnD3Z4Fh4OI4rykiKMElIiIigJntMrPDZtZiZrfNc7zYzO4Pju81sy1BeaGZfd3MDpjZs2b2qWTHLiIiIiIiIpIi8ayVtxu4Mdh+L/ATd/fgnAIAM9sMvAI4Fuc1RQRNUSgiIpLzzCwfuBO4hujc3vvMbHcw3cmsm4B+d99mZtcDnwd+D3gfUOzul5hZGXDIzL7t7seS+y5EREREREREkivOtfK+BnzTzFqAPqIJK4DXA7eZ2RQwA/yRu/cAzHfNpL4xkQyhBJeIiIjsBFrcvRXAzO4DrgNiE1zXAZ8Jth8E/tbMjOhC1uVBr7NSYBIYTFLcIiIiIiIiIikVx1p540Q7h84975vAN+O9poi8nKYoFBERkQagLWa/PSibt467R4ABoI5osmsEOAWcAP67u/fN9yJmdrOZNZtZc3d3d2LfgYiIiIiIiIiI5BQluERERGQldgLTwAZgK/AJMztvvorufpe7N7l7U319fTJjFBERERERERGRLKMEl4iIiHQAG2P2G4OyeesE0xFWA73AB4AfuvuUu3cBjwFNoUcsIiIiIiIiIiI5TQkuERER2QdsN7OtZlZEdMHb3XPq7AZuDLbfC/zE3Z3otIRvATCzcuC1wHNJiVpERERERERERHKWElwiIiI5LlhT6xbgIeBZ4AF3P2hmd5jZtUG1rwF1ZtYC3ArcFpTfCVSY2UGiibJ/dPenk/sOREREREREREQk1xSkOgARERFJPXffA+yZU3Z7zPY48L55zhuer1xERERERERERCRMoY7gMrNdZnbYzFrM7LZ5jn/UzA6Y2X4ze9TMdgTlW8xsLCjfb2b/EGacIiIiIiIiIiIiIiIikjlCG8FlZvlEpy26BmgH9pnZbnc/FFPtXnf/h6D+tcAXgV3BsaPufllY8YmIiIiIiIiIiIiIiEhmCnME106gxd1b3X0SuA+4LraCuw/G7JYDHmI8IiIiIiIiIiIiIiIikgXCTHA1AG0x++1B2UuY2cfM7CjwBeDjMYe2mtlTZvYfZnbVfC9gZjebWbOZNXd3dycydhEREREREREREREREUlToa7BFQ93v9Pdzwc+CfxFUHwK2OTulwO3AveaWdU8597l7k3u3lRfX5+8oEVERERERERERERERCRlwkxwdQAbY/Ybg7KF3Af8NoC7T7h7b7D9BHAUuCCkOEVERERERERERERERCSDhJng2gdsN7OtZlYEXA/sjq1gZttjdn8LOBKU15tZfrB9HrAdaA0xVhEREREREREREREREckQBWFd2N0jZnYL8BCQD9zt7gfN7A6g2d13A7eY2VuBKaAfuDE4/Q3AHWY2BcwAH3X3vrBiFREREREREREREVmJrqFxjveO0FhTRn6epTocEZGsF1qCC8Dd9wB75pTdHrP9xwuc913gu2HGJiIiIiKSzczsbuBdQJe7XxyU3Q9cGFRZBZxx98vmOfcYMARMAxF3b0pK0CIiIiIZ5N69J85unxoY42uPvsDo5DTb11Rw4+u2kGdKcomIhCnMKQpFJAuZ2S4zO2xmLWZ22zzHbzWzQ2b2tJn92Mw2xxy70cyOBH83zj1XREREEuoeYFdsgbv/nrtfFiS1vgv80znOf3NQV8ktERGRFInjN/iXzGx/8Pe8mZ1JRZwCP3jmNAa8+cI1HOka5vHW3lSHJCKS9ZTgEpG4BWvj3Qm8A9gB3GBmO+ZUewpocvdLgQeBLwTn1gKfBq4EdgKfNrOaZMUuIiKSa9z9EWDeab7NzID3A99OalAiIiISt3h+g7v7n8Z0XvlfnLvzioTk1MAYLV3DvH57PW995Rq21JXzaEsP0zOe6tBERLKaElwishQ7gRZ3b3X3SeA+4LrYCu7+U3cfDXYfBxqD7bcDD7t7n7v3Aw8zp1e5iIiIJM1VQKe7H1nguAP/bmZPmNnNC13EzG42s2Yza+7u7g4lUBERkRy26G/wOW5AnVdS4uDJQQxo2lyDmfH6bXWcGZ3i+c6hVIcmIpLVlOASkaVoANpi9tuDsoXcBPxgqefqYZmIiEjoFnsA9np3v4Joj/GPmdkb5qvk7ne5e5O7N9XX14cRp4iISC5byu/ozcBW4CcLHNfv7BA93zlEY00p5cUFAFy4rorSwnwOdAykODIRkeymBJeIhMLMPgQ0AX+91HP1sExERCQ8ZlYA/C5w/0J13L0j+LcL+B7RHuQiIiKSvq4HHnT36fkO6nd2eEYmInT0j3HBusqzZfl5xivXV/Hc6UEmIzMpjE5EJLsVpDoAEckoHcDGmP3GoOwlzOytwJ8Db3T3iZhz3zTn3J+FEqWISAju3Xtiyed84MpNIUQismJvBZ5z9/b5DppZOZDn7kPB9tuAO5IZoIiIiABx/gYPXA98LPSI5GXa+8dwYGtd+UvKd6yv4skT/TxxvJ/fOL8uNcGJiGQ5jeASkaXYB2w3s61mVkT0C/Tu2ApmdjnwZeDaoNf3rIeAt5lZjZnVEH1Y9lCS4hYREck5ZvZt4JfAhWbWbmY3BYeuZ870hGa2wcz2BLtrgUfN7NfAr4Dvu/sPkxW3iIiInLXob3AAM3sFUEP0c1+SrOPMKAZsWFX6kvLz6ssx4LGWnpTEJSKSCzSCS0Ti5u4RM7uFaGIqH7jb3Q+a2R1As7vvJjolYQXwHTMDOOHu17p7n5n9JdEv6AB3uHtfCt6GiIhITnD3GxYo//A8ZSeBdwbbrcCrQg1OREREFhXnb3CIJr7uc3dPVay5rKN/jNUVxZQU5r+kvKQwn8aaUh472sN/5cIURScikt2U4BKRJXH3PcCeOWW3x2y/9Rzn3g3cHV50IiIiIiIiItljsd/gwf5nkhmTvFTHmTHOr6+Y99h59RX8/Eg39zx2jKKCFyfS0lTmIiKJoQSXiIiIiIiIiEggnnU39XBaRADGJqcZHI+wrrpk3uObasuYcTg1MMbmOWt0iYjIymkNLhEREREREZEcYmb5ZvaUmf1bqmMREclkXUPjANRXFM97vKEmui5XW/9Y0mISEcklSnCJiIiIiIiI5JY/Bp5NdRAiIpmue2gCgPrK+RNcVSWFVJcW0t4/msywRERyhhJcIiIiIiIiIjnCzBqB3wK+mupYREQyXffQBAV5Rk150YJ1GmtKadcIrqxmZrvM7LCZtZjZbfMcLzaz+4Pje81sS1B+jZk9YWYHgn/fEnPOz4Jr7g/+1iTvHYlkDiW4RERERERERHLH/wD+DJhJdSAiIpmua2iC1RXF5JktWKexpoy+kUlGJiJJjEySxczygTuBdwA7gBvMbMecajcB/e6+DfgS8PmgvAd4t7tfAtwIfHPOeR9098uCv67Q3oRIBlOCS0RERERERCQHmNm7gC53f2KRejebWbOZNXd3dycpOhGRzNM7MkldxcKjtyA6ggug44xGcWWpnUCLu7e6+yRwH3DdnDrXAV8Pth8ErjYzc/en3P1kUH4QKDWz+ee7FJF5KcElIiIiIiIikht+E7jWzI4RfQD3FjP71txK7n6Xuze5e1N9fX2yYxQRyQgzM07/6CS155ieEKBhVSkGtPVpHa4s1QC0xey3B2Xz1nH3CDAA1M2p8x7gSXefiCn7x2B6wv/bbP5hguqUIrlOCS4RERERERGRHODun3L3RnffAlwP/MTdP5TisEREMlLX0ATTM75ogqukMJ+6imJODYwnKTLJNGZ2EdFpC/8wpviDwdSFVwV/vz/fueqUIrlOCS4RERERERERERGRJTgRjMiqLTt3ggtgbVUxnYNKcGWpDmBjzH5jUDZvHTMrAKqB3mC/Efge8AfufnT2BHfvCP4dAu4lOhWiiMyhBJeIiIiIiIhIjnH3n7n7u1Idh4hIpppNcNUsMoILYG1VCX0jk0xGZsIOS5JvH7DdzLaaWRHREdK759TZDdwYbL+X6AhqN7NVwPeB29z9sdnKZlZgZquD7ULgXcAzIb8PkYykBJeIiIiIiIiIiIjIEpzoG8WAVWWFi9ZdW1WCA91DE4vWlcwSrKl1C/AQ8CzwgLsfNLM7zOzaoNrXgDozawFuBW4Lym8BtgG3B2tt7TezNUAx8JCZPQ3sJzoC7CvJe1cimaMgzIub2S7gb4B84Kvu/rk5xz8KfAyYBoaBm939UHDsU8BNwbGPu/tDYcYqIiIiIiIiIiIiEo+O/jGqSgspyFt8/MDaqmIAOgfHaagpDTs0STJ33wPsmVN2e8z2OPC+ec77K+CvFrjsqxMZo0i2Cm0El5nlA3cC7wB2ADeY2Y451e5190vc/TLgC8AXg3N3EB3OeRGwC/i74HoiIiIiIiIiIiIiKdU5OE5VSXxjB+rKi8nPMzqHtA6XiEgihTlF4U6gxd1b3X0SuA+4LraCuw/G7JYDHmxfB9zn7hPu/gLQghbSExERERERERERkTRwaiA6gise+XnGmspiOgeV4BIRSaQwE1wNQFvMfntQ9hJm9jEzO0p0BNfHl3juzWbWbGbN3d3dCQtcREREREREREREZCGnB8apjjPBBdF1uDoHtQaXiEgihZngiou73+nu5wOfBP5iiefe5e5N7t5UX18fToAiIiIiIiIiIiIigaHxKUYmp5eW4KosZmBsivGp6RAjExHJLWEmuDqAjTH7jUHZQu4DfnuZ54qIiIikzOhEhIGxKdx98coiIiIiIpLRTg9EpxqsKok/wbWmqgRA0xSKiCRQfCshLs8+YLuZbSWanLoe+EBsBTPb7u5Hgt3fAma3dwP3mtkXgQ3AduBXIcYqIiIismTTM873D5zkVy/0MeNw3upy3t+0Me65+NNd5+A4x3pHuHxjTapDERERERFJuXv3ngDgSNcQwJK+99dXFgPQMzyZ+MBERHJUaCO43D0C3AI8BDwLPODuB83sDjO7Nqh2i5kdNLP9wK3AjcG5B4EHgEPAD4GPubvG74qIiEha+denT/J4ax+v2VLLrovW0X5mjLsfe4GJLJh2ZGBsiq/+vJV/2X+Se35xTKPTMpCZ3W1mXWb2TEzZZ8ysw8z2B3/vXODcXWZ22MxazOy25EUtIiIiseL5TDaz95vZoeAZ273JjjEXDY5NASxpisKasiLyDHqGtQ6XiEiihDmCC3ffA+yZU3Z7zPYfn+PczwKfDS86ERERkeU70jnEr17o46ptq3nHJesB2LCqlH987AV+cPA0v31ZQ4ojXJlfHu1lbGqaps01NB/vZ9+xfnZurU11WLI09wB/C3xjTvmX3P2/L3SSmeUDdwLXAO3APjPb7e6HwgpUREREXi6ez2Qz2w58CvhNd+83szWpiTa3DAQJrsqS+B+t5ucZteVFSnCJiCRQmGtwiYiIiGSlGXd+ePA0NWWFXHPR2rPl29ZU8Lrz69j3Qt/Zefkz0Yw7v24/w/Y1lbz7VRsoLczn3r3HUx2WLJG7PwL0LePUnUCLu7e6+yTRtXKvS2hwIiIiEo94PpP/C3Cnu/cDuHtXkmPMSYNjEcqK8inMX9qj1dUVxfRqikIRkYRRgktERERkiVq6hjk1MM7Vr1hLQd5Lv069+RVrKCrI48fPdaYoupXr6B9jYGyKSxurKczPY/vaCh5t6dU0hdnjFjN7OpjCcL4F1hqAtpj99qDsZczsZjNrNrPm7u7uMGIVERHJZfF8Jl8AXGBmj5nZ42a2K2nR5bCBsaklTU84a3VFMT3DE8zM6Hu1iEgiKMElIiIiskSPt/ZSXlzApY3VLztWVlTAb5xXx6GTg7T1jaYgupU73jsCwPn1FQBsq6+gZ3iC5zuHUxmWJMbfA+cDlwGngP9vJRdz97vcvcndm+rr6xMRn4iIiCxNAbAdeBNwA/AVM1s1t5I6pSTW4PgUVSVLT3DVVRQRmXFODWbubA8iIulECS4RERGRJRieiHD49BBNm2soWGBKkivPq8MMvpWh0/od7xulpqyQqqBX6vlroomuXx7tSWVYkgDu3unu0+4+A3yF6NRHc3UAG2P2G4MyERERSa54PpPbgd3uPuXuLwDPE014vYQ6pSTWSkZwARzrGUl0SCIiOUkJLhEREZEleKZjAAde1fiyjrFnVZcWcsHaSnbvP5lx04+4Oyf6RtlcV362bFVpIasrinjm5GAKI5NEMLP1Mbu/AzwzT7V9wHYz22pmRcD1wO5kxCciIiIvEc9n8j8THb2Fma0mOmVhazKDzDVT0zOMTk5TVVqw5HNnE1ytSnCJiCSEElwiIiIiS3CgY4D6ymLWVhWfs96ljas4NTBO8/H+JEWWGEMTEYbGIzTWlJ4tMzN2bKjmoBJcGcXMvg38ErjQzNrN7CbgC2Z2wMyeBt4M/GlQd4OZ7QFw9whwC/AQ8CzwgLsfTMmbEBERyWELfSab2R1mdm1Q7SGg18wOAT8F/pu796Ym4twwNB4BWNYIrqqSAgrzjRe6leASEUmEpXc1EBERkawTLEb9N0A+8FV3/9yc48XAN4BXA73A77n7seDYpcCXgSpgBniNu2flpPLjU9Mc7x3hqu31mNk5675yfSUlhXn8669PsnNrbZIiXLmuwQkA1laVvKT8og1VfOWRViYi0xQX5KciNFkid79hnuKvLVD3JPDOmP09wJ6QQhMREZE4zfeZ7O63x2w7cGvwJ0kwMDYFQOUy1uAyM1ZXFPNCj9a2FRFJBI3gEhERyXFmlg/cCbwD2AHcYGY75lS7Ceh3923Al4DPB+cWAN8CPuruFxGdHmUqSaEn3ZGuYWYcLlhbuWjd4oJ8rn7FWvYcOEVkeiYJ0SVGZ7Dg9XwJrsiM8/xp/RgXEZH05e48cbyPux5p5VuPH+dEr0ZJiEhiDU9ER3BVlixv3EA0waV7k4hIIijBJSIiIjuBFndvdfdJ4D7gujl1rgO+Hmw/CFxt0SFMbwOedvdfA7h7r7tPJynupDvSOURJYR6basviqv/uV62nd2SSvS/0hRxZ4nQOjlNelE9F8Ut/sF8YJPVauodSEZaIiEhcfn6kh+8+2cHoZIQTfaPc9fNW9h3LnM9hEUl/swmuud+X41VbXkR7/xjTGbZWr4hIOlKCS0RERBqAtpj99qBs3jrBWgADQB3RRazdzB4ysyfN7M8WehEzu9nMms2subu7O6FvIFmO9Y6yubac/LxzT0846w0X1FOUn8fPDneFHFnidA6Os2bO6C2ATXVl5Bm0ar0AERFJUz1DE/z7odNc3FDNx6/ezieuuYBtayr456c6eKZjINXhiUiWGJmIYEBZ0fITXJEZ59TAWGIDExHJQUpwiYiIyEoUAK8HPhj8+ztmdvV8Fd39Lndvcvem+vr6ZMaYEMMTEXqGJ9iyujzuc8qKCti5tZb/eD5zEno9w5PUVxa/rLy4IJ9NtWVKcImISNr60XOdFOTlce2rNpBnRnFhPh/YuZmNtWU80NymkVwikhDD4xFKi/Lj7vQ2V215EQAn+kYTGZaISE5SgktEREQ6gI0x+41B2bx1gnW3qoFeoqO9HnH3HncfJboA9hWhR5wCx4M1PLbUxTc94aw3XlDP853DnDyT/j00RycjjE1NUxf86J7rvPoKjnZrDS4REUk/IxMRDnYM8potNS+ZNqyoII8/eO1mqksLufkbzRzTujciskLDE5FlT08IUFsW/a7dpgSXiMiKKcElIiIi+4DtZrbVzIqA64Hdc+rsBm4Mtt8L/MTdHXgIuMTMyoLE1xuBQ0mKO6mO9YxQkGc0rCpd0nlvujA6Wi0TRnH1jUwCUFf+8hFcAOetLudY7wgzWi9ARETSzK/bzzDtzhWba152rKy4gBtftwUH/vM9+zgzOpn8AEUkawxPRKgoWX6Cq6q0kII80wguEZEEUIJLREQkxwVrat1CNFn1LPCAux80szvM7Nqg2teAOjNrAW4FbgvO7Qe+SDRJth940t2/n+z3kAzHekdprCmjIH9pX5+2ralgQ3UJ/3E4/RNcvcNBgqti/hFcW+vLGZ+aoXNoPJlhiYiILOrgyUHWVhWzvnr+jiirK4q56/ebaO8f48P/uI/e4YkkRygi2WKlI7jy84yGmlJO9KX/DA8iIulOCS4RERHB3fe4+wXufr67fzYou93ddwfb4+7+Pnff5u473b015txvuftF7n6xu/9Zqt5DmEYmIpwaGFvy9IQAZsYbLqjnsZYeptN85FPvSPRhX+0CUxTOjl7r6NePcRERSR/jU9Mc7x3hFeuqzllv59Za/tcHLufZU4P8zt/9ggPtA0mKUESyyUoTXAAba8o0gktEJAGU4BIRERFZxNPtA8w4bF5GggvgN86vY2giwrOnBhMcWWL1Dk9SVVJA4QKj1Bprou+/XQkuERFJI0e7h5lxuGBt5aJ1337ROu67+bWMT03z7r99lPd/+Zfc89gLPHWin/Gp6SREKyKZbDIyw2RkZuUJrtoyrcElIpIAK7sbi4iIiOSAg/AAuUIAACAASURBVCejPbw3LHH9rVlXbq0D4PHWXi5uqE5YXIl2ZmyKmrL5R29BzAiuM0pwiYhI+mjtGaEw39hUG19HlMs31fCjT7yRb/ziGP/0ZAef+dfo8qH5ecYr11eyubacnVtrF+zwISK5a3giArDiBNem2jL6RiYZGp+isqQwEaGJiOQkJbhEREREFnHw5CCVJQXL/vG5rrqEzXVl7H2hj49cdV6Co0ucgbEpGmsWTuKVFuVTV15Ee796m4qISPo40TvKxpoy8vMs7nOqSgq55S3b+dibt3FyYJwD7QM80zHAY0d7+P6BU+x9oY8PXbmJNVUlIUYuIplmJIEJLoC2vjF2bFCCS0RkudQdSURERGQRB08OsGGBRevjdeXWWvYd62MmTdfhmnFnYGyKVaXn/oHdWFOqKQpFRCRtDE9EOHlmbNnTCJsZDatK2XXxOv7r2y/ke3/0m/yn121hfGqar/y8lZ6hiQRHLCKZ7OwIrpIEJbjUcUxEZEWU4BKRJTGzXWZ22MxazOy2eY6/wcyeNLOImb13zrFpM9sf/O1OXtQiIss3PjXN0e4R1q9aWQ/unVvrODM6xeHOoQRFllgjExGmZ5zqRRJcDTWlmqJQRETSxtNtZ3BgU215wq65fW0lN191Hg58a+9xJiMzCbu2iGS24fFEj+BSgisbxPGsrNjM7g+O7zWzLUH5NWb2hJkdCP59S8w5rw7KW8zsf5pZ/MOURXJIqAmuOBr3rWZ2yMyeNrMfm9nmmGN6EC6SZswsH7gTeAewA7jBzHbMqXYC+DBw7zyXGHP3y4K/a0MNVkQkQQ6fHmJ6xhMyggtgb2tvIsJKuIGxKQBWnWMNLoiuw9XRP4Z7eo5EExGR3PJMsE5mwzmm2F2O1ZXFXP+aTXQNTfCjZzsTem0RyVxDwQiu8hUmuKrLCqkqKeCEElwZL85nZTcB/e6+DfgS8PmgvAd4t7tfAtwIfDPmnL8H/guwPfjbFdqbEMlgoSW44mzcTwFN7n4p8CDwhZhjehAukn52Ai3u3uruk8B9wHWxFdz9mLs/Daibo4hkhYMnBwHYsGplD8421paxobqEJ06cSURYCXdmNJrgWmwEV2NNGRORGXqGJ5MRloiIyDkdPDlIdWnhikdTzGfbmgpes6WWx1p66BoaT/j1RSTzjExEKC7IozB/5Y9UN9WVKcGVHRZ9Vhbsfz3YfhC42szM3Z9y95NB+UGgNBjttR6ocvfHPdqz8BvAb4f/VkQyT5gjuOJ5EP5Td5+9kz8ONIYYj4isXAPQFrPfHpTFq8TMms3scTNb8IPZzG4O6jV3d3cvN1YRkYQ4eHKAypICaspWvvjz5ZtqeOpEfwKiSrzZEVyLTlEYJPratV6AiIikgWc6BthQvbJphM/lmh1rKczP4yfPdYX2GiKSOYYnIglLqG+sUYIrS8TzrOxsHXePAANA3Zw67wGedPeJoH77ItcUEcJNcC31QfhNwA9i9hd9EK6H4CIZZ7O7NwEfAP6HmZ0/XyV3v8vdm9y9qb6+PrkRiojMcfDkIDvWV5GIKc8v37SK9v6xtOwFPjA2RWG+UVaUf856s1NAaR0uERFJtdHJCK09I6xf4Sjrc6koLuC159VxoH2AzsH0+/wWkeQanohQUZKYBNem2jLa+8aYmdHU37nOzC4iOm3hHy7jXD0fl5wW6hpc8TKzDwFNwF/HFC/6IFwPwUWSrgPYGLPfGJTFxd07gn9bgZ8BlycyOBGRRHN3nu8c4pXrqxJyvcs3rQJgfxpOU3hmbIrq0sJFE3lnE1z9SnCJiEhqHe0awR3WVYU3ggvgqu2rKSzI46eHNYpLJNcNjydwBFdtGZPTM3SmYec3WZJ4npWdrWNmBUA10BvsNwLfA/7A3Y/G1I+d6WzB5296Pi65LswEV1wPws3srcCfA9cGQzABPQgXSVP7gO1mttXMioDrgd3xnGhmNWZWHGyvBn4TOBRapCIiCXByYJzRyWm2ralIyPUu2lBNYb7xZBomuAZGJxednhCgqiS6IHa7Elxpz8zuNrMuM3smpuyvzew5M3vazL5nZqsWOPeYmR0ws/1m1py8qEVE4ne0exiA+sriUF+nvLiAnVtqeaZjgMHxqVBfS0TSWyKnKNxUWwbAiV5NU5jh4nlWthu4Mdh+L/ATd/fgu/j3gdvc/bHZyu5+Chg0s9datAfiHwD/EvYbEclEYSa4Fm3cZnY58GWiya2umHI9CBdJQ8E8wbcADwHPAg+4+0Ezu8PMrgUws9eYWTvwPuDLZnYwOP2VQLOZ/Rr4KfA5d1e7FpG01tIVfXC2PUEJrpLCfHasr0rLdbgGxqZYVVoUV92GmjJNUZgZ7gF2zSl7GLjY3S8Fngc+dY7z3+zulwWzKoiIpJ2WrmHy84y68vg+v1Zi59ZaZhyaj/WF/loiscxsl5kdNrMWM7ttnuMfNrPuoFPKfjP7SCrizAVT0zOMTU0nPsGldbgyWjzPyoCvAXVm1gLcCsy25VuAbcDtMW14TXDsj4CvAi3AUV66tI+IBBJzR56Hu0fMbLZx5wN3zzZuoNnddxOdkrAC+E4wHc4Jd7+W6IPwL5vZDNEknB6Ei6QJd98D7JlTdnvM9j5eOox6tvwXwCWhBygikkBHOocA2LamgqPdIwm55uWbarh/XxuR6RkK8tNitmimZ5yh8QjVZYuP4AJYX13C6QFNpZLu3P0RM9syp+zfY3YfJ9qDVERyhJmVAI8AxUSfBzzo7p9ObVTLd7R7mE21ZUn5PF1dUcz2NRX86oU+3njBmsVPEEkAM8sH7gSuIbq2/T4z2z3PM7L73f2WpAeYY3qHJwEStgbXhlWlmEGbZkbIeHE8Kxsn2hF87nl/BfzVAtdsBi5ObKSSq06eGeMXR3soKcznjRfUU1kS32//TBBaggviatxvXeA8PQgXERGRlDvaPUxteRF1FYmb+ujyTau45xfHONw5xEUbqhN23ZUYHJvCIa4pCgHWVpXw67b0m2ZRluw/A/cvcMyBfzczB77s7nclLywRCdEE8BZ3HzazQuBRM/uBuz+e6sCWo6VrmPPrEzPKOh5Xbq3jW3uPc/j0YNJeU3LeTqAlWL4DM7sPuA7NcpQSPcPRlVUSNYKrqCCPDdWltGkEl4iEqKN/jK882kqewVTEae0e4Q/feB7FBfmpDi0h0qPbsIiIiEgaOtI5zLYEPzi7YlMNAE+l0TpcZ8ai64msijvBVUzvyCSTkZkww5IQmdmfAxHgfy9Q5fXufgXwDuBjZvaGBa5zs5k1m1lzd3d3SNGKSKJ41HCwWxj8eQpDWrbI9AzHekcStk5mPC5cV0lVSQHNx9NvqmHJWg1AW8x+e1A213uC9TUfNLON811In9krl+gEF0BjjRJcIhKe6Rnnu0+2U1KQx59cfQF/8LrNdA6O8x+Hs+dzQAkuERERkXm4O0e6htm2NrEPzhprSlldUZRWCa7B8WiCqyrOBNe6qhIAuoY0TWEmMrMPA+8CPuju8z7YdveO4N8u4HtEe5DPV+8ud29y96b6+vqQIhaRRDKzfDPbD3QBD7v73nnqpP2D8BN9o0xNO+fXlyftNfPzjMs31XD49BCdg/oMlLTxr8CWYH3Nh4Gvz1dJn9kr1zM7RWECE1wba8to61eCS0TCsb/tDKcHx3nXpRuoKi1k+5pKLm6o5petvYxNTqc6vIRQgktERERkHj3DkwyMTSV8BJeZcdnGGp5qS5/e30PjEQCq4pyHe22Q4OocnAgtJgmHme0C/gy41t3nfZpiZuVmVjm7DbwNeCZ5UYpImNx92t0vI7pu7k4ze9n6HpnwILylKzoQLZkjuABevbkGB/7pyY6kvq7krA4gdkRWY1B2lrv3uvvsl7KvAq9OUmw5J4wRXJtqy+gcnGB8KjseNItI+nB3fn6km3VVJVy0oeps+RsuqGciMsPTHenT6XYl4kpwmdk/mdlvmZkSYiJZRG1bJPuoXSfO7IOz7QkewQXRdbhau0cYGJ1K+LWXY2hsioI8o6Qwvv82Lya41Hs9WZbTts3s28AvgQvNrN3MbgL+FqgEHjaz/Wb2D0HdDWY2u3buWqLr8vwa+BXwfXf/YULfkIis2Eo/8939DPBTYFdiI0uOo90jAJyf5ATX6opittSV8Z3mNhYYBCsyr2W22X3AdjPbamZFwPXA7jnXXR+zey3w7Mqjlfn0DE1QmG8UFSTup9bG2lIA2vvHEnZNWT79npZs0tY3StfQBK87vw4zO1u+obqENZXF7E+jWWVWIt7G+nfAB4AjZvY5M7swxJhEJHnUtkWyj9p1grR0DQHh9Ay/pKEagIMnBxJ+7eUYmohQWVLwki+957KuOprgOj2gBFcSLbltu/sN7r7e3QvdvdHdv+bu29x9o7tfFvx9NKh70t3fGWy3uvurgr+L3P2z4b41EVmmJd8XzKzezFYF26XANcBz4YYZjpauYdZUFsc9+jiRXr25ltaeEZ7QWlyyNMv5LI8AtwAPEU1cPeDuB83sDjO7Nqj2cTM7GHRM+Tjw4XDCl96RScqL4//OHI9NtWUAWocrfej3tGSNpzsGyM8zLg6eP8wyM161cRXH+0bPLleQyeJKcLn7j9z9g8AVwDHgR2b2CzP7T2aW/G+TIpIQatsi2UftOnGOdo9QVpR/dr2pRJr9gnmgIz0SXIPjU1Qu4QFhTVkhRfl5dGoNrqRR2xaRuZZ5X1gP/NTMniY6MuRhd/+35EScWMd6R9i6Onnrb8W6uKGK8qJ8HmhuS8nrS2Za7me5u+9x9wvc/fzZTifufru77w62PxV0SHmVu7/Z3TMyaZ0JeoYnEjo9IcDGmiDBpXW40oK+c0u2mHHnQMcAF66tpKQw/2XHL1xbCUBL53CyQ0u4pUxxUke0F8hHgKeAvyHa2B8OJTIRSQq1bZHso3adGMd6R9hSV57QHpqzasuLaFhVmjYJrqHx6AiueJkZa6qK6dQIrqRS2xaRuZZ6X3D3p939cne/1N0vdvc7khZsgrX1jZ4d+ZBsxQX5vOvSDfzb06cYmYikJAbJTPosz2zdQ4lPcNVXFlNckKcRXGlE7VSywbHeEYbGI1zSWD3v8XXVJVQUF/B8MHNNJovrrmxm3wMuBL4JvNvdTwWH7jez5rCCE5FwqW2LZB+168Q53jvKjvVVi1dcposbqngmbRJcU5xfv7SpGNdWldA5OLF4RUkItW0RmSuX7wvjU9N0DU2kLMEF8P7XNHJ/cxvfP3CK9zdtTFkckjlyuc1mi57hSbbUJfa+Y2ZsrC3jhBJcaUHtVLLFgfYBCvONV66b/5lGnhnb1lRwpHMIdw+lY2+yxNvt4Cvuvie2wMyK3X3C3ZtCiEtEkkNtWyT7qF0nQGR6hra+Ud5x8brQXuOShmoeOtjJ4PhUStYPmTU1PcP41AxVSxjBBbCuqoRnTw2GFJXMQ21bRObK2ftCezCV18YUJriu2FTDefXlfKe5TQkuiVfOttlsMDPj9I1McPGGxHeA21hTSlvfWMKvK8uidioZz905fHqI7WsqKSpYeAK/zXVl7G87Q+/IJKsripMYYWLF+yTjr4A9c8p+SXR4pohkLrVtkeyjdh2He/eeOOfx3uEJIjNO5+D4onWXa3YdroMdg/zG+XWhvEY8hsajUystZYpCiI7g+tnhrjBCkvmpbYvIXDl7X5h9ELyxtjRlMZgZ72/ayOd+8Byt3cOct8SR0JKTcrbNZoP+0UlmHCqW+J05Hptqy2g+1p/xoyiyhNqpZLzWnhHOjE3xxgvrz1lvS110LdPjvaPZm+Ays3VAA1BqZpcDs3fZKiB1XaVEZEXUtkWyj9p1YvWOTAJQVx7el7zZBNczHQMpTnBNAVC5xFFka6uKGZmcZmh8asnnSvzUtkVkLt0XODuVVypHcAH87hUN/PVDh/nOE+18ctcrUhqLpC+12ezQMxz9fZDoNbggei8bmogwMDbFqrKihF9fFqd2Ktnk5893A7B9TeU569VXFlNamM/x3hFevbkmGaGFYrG78tuJLqrXCHwxpnwI+L9CiklEwqe2LZJ91K4TqGc4urZUXUV4PzBXVxSzvrqEAyleh2twmSO41lWXANA5OK4EV7jUtkVkrpy/L7T1jVJSmEd9insbr6ks4c0XruG7T7TziWsuoCB/4WmAJKflfJvNBrO/DxKV4IqdJeJYzwgQHZ2qBFfKqJ1K1vj5kR5qy4uoLT/3/STPjIaaUk4OZPYUqee8K7v714Gvm9l73P27SYpJREKmti2SfdSuE6t3ZJKigrxQemjGurihmmdSnOBa/giu2QTXBNsW6Rkmy6e2LSJz6b4Abf2jbKwpS4upvN7f1MiPnu3kP57v5upXrk11OJKG1Gazw2yCqzyE3wc1wUPoE32jXNJYnfDry+LUTiVbTEZmeLy19+yMMYtZX1XCL1t7mZ7xkCMLz2JTFH7I3b8FbDGzW+ced/cvznOaiKQ5tW2R7KN2nVi9wxOsLi8K/cHZJQ3VPHyoM6XT/A2NR8gzKCvKX9J5swmu0wPjYYQlAbVtEZlL9wU40TeW8ukJZ735FWtYXVHEA81tSnDJvNRms8PsFIWVYSS4glFbbf2jCb+2xEftVLLFUyf6GZmcZvua+NYGXb+qhMiM0x0k8TPRYnfl8uBfrZYqkl3UtkWyj9p1AvUOT7JhVfgL118S9Ko6dHKQK89LzTpcs8m1vCUm89ZURqeF6hrK3C/CGUJtW0Tmyun7grvT3jfKzi3psVZEYX4ev3tFI3c/+gI9wxMZvUi7hCan22y26BmeoCDPKFlip7B4lBTmU1aUf3Z9QUkJtVPJCntf6MMMtq6OM8FVHX3ucTqDpylcbIrCLwf//j/JCUdEkkFtWyT7qF0nzvSM0z86eTb5FKbZaQMOdAykMMEVWfL6WxCdnqWiuICuIY3gCpPatojMlev3hTOjUwxNRNJmBBfA+17dyF2PtPLPT3XwkavOS3U4kmZyvc1mi56hCeoqipbcKSxeNWVFtCnBlTJqp5It9h3r48K1lZTGmYxfXVFMQZ5x6kzm/q6PawVUM/uCmVWZWaGZ/djMus3sQ2EHJyLhUtsWyT5q1yt3ZnSSGYe6ivAXeK6vLGZdVUlK1+EaGo8se6qVNZXFdA1qBFcyqG2LyFy5el+YncIrnRJc29dWcvmmVdy/rw33zF3DQsKVq202W4Q9QrO2vIj2/swdQZEt1E4lk0WmZ3jyeD+v2VIb9zn5ecbaqhJOZfDSA3EluIC3ufsg8C7gGLAN+G9hBSUiSaO2LZJ91K5XqH90CnhxseewXdxQxYEUJrgGx6eoLF3e+l/1lcUawZU8atsiMldO3hfa+qIPgDfWpE+CC+ADOzdxpGuYR470pDoUSV852WazRe/IZKgJrpqyItr7R5meUZI8xdROJWM9d3qIkclpmpY4jfP66hJODYxlbCedeBNcs916fwv4jrun7imMiCSS2rZI9lG7XqH+0egC0rOLPYft4oZqWntGGJ6IJOX1YkVmZhidnF7WFIUAa6tKtAZX8qhti8hcOXlfOBWsEdFQE/5amUtx3WUNrK8u4e9+2pLqUCR95WSbzRazUxSGpba8iKlpp3NQncdSTO1UMta+Y30ASxrBBdEE18jkdMb+to83wfVvZvYc8Grgx2ZWD+iOK5L51LZFso/a9Qr1j06SZ1BVsrxRTUt1SUM17nDo5GBSXi/W8Hg0qVZVvLz3uqaymM7B8Yzt6ZVh1LZFZK6cvC+cPDNOWVE+VcvsnBGWooI8PnLVeex9oY8njvenOhxJTznZZrOBu9MzPEl9mCO4yqPfx09oHa5UUzuVjNV8rJ+GVaVsWLW0TkDrqqP1D51K/jOJRIgrweXutwGvA5rcfQoYAa5b7Dwz22Vmh82sxcxum+f4rWZ2yMyeDuY13Rxz7EYzOxL83Rj/WxKReC23bYtI+lK7Xrkzo1NUlxaSnxfOAtJzXdxQDcDBk8nvHDgUJLiWO4JrTVUx41MzDKVg9FmuUdsWkbly9b5wenCM9dUlmCXnc3oprn/NRmrKCvmfPz6S6lAkDeVqm80Gg+MRJqdnwl2DK5g9ok0JrpRSO5VM5e7sO9a35OkJIdpxFeBo13Ciw0qKpTzNeAWwxcxiz/nGQpXNLB+4E7gGaAf2mdludz8UU+0pojeMUTP7P4AvAL9nZrXAp4EmwIEngnPVDUok8ZbUtkUkI6hdr0D/yGTSpieE6JfJ1RXFPNOR/N5SQ+PR9cYqlzlabW1VCQBdgxNJG/GW49S2RWSunLsvnDwzzvrq9JqecFZ5cQEffeP5/L8/eI7HW3t57Xl1qQ5J0k/Otdls0DMcnbZrdWURY5MzobzGqrIi8gza+sdCub4sidqpZJy2vjG6hiZoWuL0hBD9/lJWlE9rz0gIkYUvrgSXmX0TOB/YD0wHxc65G/dOoMXdW4Nr3Ec04302weXuP42p/zjwoWD77cDD7t4XnPswsAv4djzxikh8ltm2RSSNqV2vXP/oJNvXVCbt9cyMizZUpWQE1+DsCK7S5Y3gqg96enUNjrNtTUXC4pKXU9sWkbly9b5wemCcq7avTnUYC7rxdVu45xfH+NwPnuN7f/S6tBxpJqmRq202G/QE69KsriimrS+cBFR+nrG+ulQjuFJM7VQy1Yvrby19BBdE72/ZPoKrCdjhS1tgoQFoi9lvB648R/2bgB+c49yGuSeY2c3AzQCbNm1aQmgiElhO2xaR9KZ2vQKR6RmGxiOsKk/uaKSLNlTxWEsP41PTlBTmJ+11h8YjGFBRvMwpCiuDEVwZuhhthlHbFpG5cu6+EJmeoWtonPXVJakOZUElhfn8yVu388nvHuChg6fZdfH6VIck6SPn2my26BmeBMJNcAFsrFWCKw2onUpGaj7eR1VJARcss7NufUVxxo7gimsNLuAZYF1YQZjZh4jeQP56Kee5+13u3uTuTfX19eEEJ5LdQm3bIpISatcrMDA2hUNSpyiE6DpckRnn+c6hpL7u0PgUFcUF5C2zd/naqmAE15DWXU6CJbdtM7vbzLrM7JmYslozezhY5/ZhM5u3i5/WwxXJCDn3md81NMGMw/olLp6ebO+5opHz68v5wkOHiUyHM52ZZKSca7PZ4uwUhSGuwQWwsaaME0pwpZraqWSkfcf6adpSS94y1xJfXVlM99DE2WUMMkm83XVXA4fM7FfA2S667n7tOc7pADbG7DcGZS9hZm8F/hx4o7tPxJz7pjnn/izOWEUkfstp2yKS3tSuV6BvNNo7M9kJros2VAFw8OQglzauStrrDo1HqCxZ3ugtiI78Ki3Mp3NQI7iSYDlt+x7gb3nplCq3AT9298+Z2W3B/idjT9J6uCIZI+c+808NREdOrFvmCK57955IWCyLXes3zqvjW3tPcH9zGx+8cnPCXlcy2rLarJntAv4GyAe+6u6fW6Dee4AHgde4e3PCohZ6hyfIM6gtD/c3wqbaMrqGJpI+q4O8RM59tkrm6xuZpKVrmN+94mUT4MWtviJ6f2vtHuFVG5P3TCIR4n2i8ZllXHsfsN3+f/buPL7Ruzr0/+dItmXLtmx5HXtsz+KZzGQmG2SyE9YkhAJJbxtKoED4/XKbH73QW8rtkva2KU1vW2j7gi7QQlrS0l5SSKDAtAyE0GwkJJNZss7u8Xi8jvddlm1J5/eHpInj8Yw3Sc8j+bxfL78iS4+k4ySPnkfP+Z5zRDYRT1jdCXx47gYi8ibgq8Ctqto356FHgT+ds6L0FuB3VxCDMebCPut0AMaYlPus0wFks5HJ+GqloD+zLQqbKvyUFuZlfA7XeHiW0sKV/60iQk3AZy0KM+Ozy32Cqj4tIhvn3X07ry8k+zrxRWS/M28bm4drTHb4rNMBZFr3SLxiuL7M3RVcABfXBdhQ4eeLj53g9ivWr7gdsMkpn13uE0TEC3wZuJn4+I59iUUnh+dtVwr8OrA3BXGaefonZqgoLsC7wsqIpWqs8APQORxiSwZnAps3+KzTARizXAdOx9chXrWxYsWvkaxQbR2YyLoE15JaFKrqU0AbkJ+4vQ84uMhzIsCniCerjgAPq+ohEblfRJJZ778ASoBHROQlEdmdeO4Q8MeJ99kH3J/8gm2MSZ2V7NvGGHez/Xp1hkMzeIRVJX1WQkTYURfgta6xjL7vaiu4AGpLC+kbsxaF6ZbCfbtWVXsSt88AtQtss6R5uBCfiSsi+0Vkf39//wrCMcas1Fo85p8ZjR9vVlrBlUkiwnsurWNgYpoHnm51OhzjAivcZ68GWlS1VVVngG8SX6wy3x8DnwfspCwNBiam096eEF5PcKVzzpe5sJUeW0XkVhE5JiItiS4J8x/3ici3Eo/vTS5CE5FKEXlCRCZE5EvznvNk4jVfSvzUpOBPNDlof9sQBV4Pl64vW/FrVJTEk/it/dk3h2tJVzRE5FeAe4AKoJn4F9yvAO+60PNUdQ+wZ9599825fdMFnvsg8OBS4jPGrMxK921jjHvZfr06w6EZyoryU7Y6czmtkPI8wsHuUaIxTfvqUIBoTJmYjqw6mVcd8HGkO7OJubUoHfu2qqqIrGqAtqo+ADwAsGvXLhvGbUwGrcVjfvfoFMUFXgKrXJyRKU0Vft57WR3/8HQrv3xNE7UB9yfmTPqscJ9daNHJNfNe981Ao6r+QER+6wLvf0/i/WlqalrJn7BmZS7BFa9OtTlczlnJfrrESsu7gWFV3SIidxJPSH+QeFL6D4BLEj/z/bK1HDWL2dc2xGUNZatqbZrn8dAYLMrKBNeSKriATwI3AGMAqnoCsKyxMdnP9m1jco/t16swHJrN+PytpPryImajSmv/REbeb3I6ggKBotVdJKwp9dFrFVyZkKp9u1dE6gAS/+xbYJslzdI1xjhuzR3zz4yGqSsvQiT9C0FS5XfevZ1ILMYXfnzc6VCM81K+z4qIB/gC8L8W21ZVH1DVXaq6q7q6ejVvu+YMahQcLwAAIABJREFUTExTWZL+7wjVJT78BV5OD1qCy0Er2U+XUml5O/H24BCflfcuERFVnVTVZ7DqS7NC4dkor3aNsmsV7QmTNleXcDJD1yNSaakJrunEDgqAiOQRHzhtjMlutm8bk3tWtF+vtKXCnMebEm0VfjMFf4NjRkIzjiW46srjKzZfy9AcrvFwBIBS3+oquGoDhUzORJmcjqQiLHN+qTpm7wbuSty+C/j+Ats8CtwiIsHETNxbEvcZY9xlzZ3Ld4+GqcuC9oRzNVX6+ei1G3nkQAfHzow7HY5x1kr22cUWnZQSr/p4UkTagGuB3SKyKyURGwAGxmcyUsElImyoLObUQPZdYM4hK9lPl9Le++w2ibE+o0DlEuL5p0R7wj+Q86zusJbha9tLHSPMRpWrNgZX/Vqbq4o5NTCJanadTi41wfWUiPweUCQiNwOPAP+RvrCMMRli+7YxuWfZ+/WclgrvAXYAHxKRHfM2O9tSAfgi8ZYKc30B+GEK4ndMJBpjLByhvDiz87eSqkt85HmEQxmawzUWngVY9QyumtL4l/2+8elVx2QuaCX79r8BzwHbRKRTRO4GPgfcLCIngJsSvyMiu0TkH8Hm4RqTRdbcufyZ0amsS3AB/No7t+AvyONvHj/hdCjGWSvZZ/cBW0Vkk4gUAHcSX6wCgKqOqmqVqm5U1Y3A88Bt1tIsdSanI0zNRjOS4IL4BeY2q+BykpuOrb+sqpcCNyZ+PrrQRladubbtb4t/Tbtyw+oTXBsq/UxHYvRn2Xf7pSa47gX6gVeB/4/4XK3fT1dQxpiMsX3bmNyzkv16xS0VAETk54FTwKGU/AUOGZmKJ3wqHKrg8nqEdWWFGa/gChStLqFXUxq/0GhtCtNu2fu2qn5IVetUNV9VG1T1a6o6qKrvUtWtqnpTMnGlqvtV9b/Pee6Dqrol8fNPafy7jDErt6bO5WejMfrGp1lXVuR0KMsWLC7gY9dtYM+rPbT0WWXGGraSY3kE+BTxSuojwMOqekhE7heR29IcryHenhCgKgMtCh/a287kdITTg5P863On0/5+ZkErObYupb332W0SVWFlwOCFXlRVuxL/HAceIv693Zg32Nc2zEW1JZSn4DpGY4UfyL45gEtasquqMRH5HvA9VbVaR2NyhO3bxuSeFe7Xiw6vZl5LBREZBSpFJAz8DvGBuhdsT+j2wdbDoXgnilScGK5UfVkRh7vHUNW0zxcZC88iQIlvlRVcAavgygQ7Zhtj5ltrnwt949OoQn0WVnAB3P2WTfzTs2383RMtfOGDVzgdjnHASvdZVd1D/CL73PvuO8+2b19VkOYcAxPx7whVpZmp4Koq8RHT17+bmMxa4X56ttKSeCLrTuDD87ZJtgl/DrgDeFwv0AcukQQrV9UBEckH3gf8ZHl/jcl10Zhy8PQw77+iPiWv1zQnwZWKmV6ZcsErGomV2X9IfLWIJ3FfFPhbVb0//eEZY9LB9m1jco+D+/VngS+q6sRiCRlVfQB4AGDXrl2ua+o8Mhmv4Ar6nWlRCFBXXsgLbUN0Dk+dXT2VLuPhCH5fHl7P6hJptYkKrj6r4EoLO2YbY+Zba58LD+1tB+D04CQAR8+Mn70vG8yN9coNQb73UhebqoqpnNfu7MPXuG/xj0mNtbbP5ppkBVd1hloUViYqxZLvazJjNftpYgFostLSCzyYrLQE9qvqbuBrwL+KSAswRDwJlnzvNiAAFCS6o9wCnAYeTSS3vMSTW/+Qwj/Z5IBjZ8YZn46kZP4WwPpgESLZV8G1WIvC3wBuAK5S1QpVrSC+ovsGEfmNtEdnjEkX27eNyT2r2a9X01LhGuDPEyflnwZ+L3Fyn3WGQjN4ZPUt+1ZjfXm87dJrXelvUzgeniWwyvlbAIGiPAryPFbBlT52zDbGzLcmPxdGE62Eyxw8Tq/WW7ZW4RHhqeM5X3Bn3mhN7rO54KG97ex5tQeAn54YyEhyPZn8HpywCq4MW9V+qqp7VPUiVW1W1T9J3HdfIrmFqoZV9QOJ9t9Xq2rrnOduTLxnSaKt+GFVnVTVK1X1MlXdqaq/rqrR9PzpJlu9cCre5XLXhtRUW/nyvNQFCnMuwfVR4EOqeip5R2IH/AjwsXQGZoxJK9u3jck9q9mvLzi8OiHZUgHmtFRQ1RvnDLX+K+BPVfVLq/9zMm84NEO5vwBPmlsDXkhtoBCvRzjUPZb29xoPRyhNQYJLRKgp9VkFV/rYMdsYM9+a/FzIhQRXoDCfKzcEebFjhInpiNPhmMxZk/tsrkjuq6tt671UxQVeCvM9VsGVebafmqzzQtsQ68uLUtr9pbHCT0eOJbjyVXVg/p2JHqTZe1ZpjLF925jcs+L9eonDq79GfOZWC/AZ4sN3c8pIaJZyB9sTAuR7PWypLuFQd2YquEoLU/P31gYKrYIrfeyYbYyZb01+LoxOzeLL81CY73U6lFW5vrmKaEzPrro2a8Ka3GdzxUQ4QlG+d9VtvZdKRKgq8VkFV+bZfmqyiqrywqkhrtmU2llZTRX+rKvgWmz5wYU+Te2T1pjsZfu2MblnVfv1YsOrVTUMfGCR1/jsYu/jZsOhGS6qLXU6DHauD/DTE+d8t0qpmGrKKrgAakp9HO8dT8lrmXPYMdsYM9+a/FwYnZp1tI1wqlSX+riotoS9p4Z460XV5HkWW3dscsCa3GdzxcR0hJIUnTMvVWVxAaez7AJzDrD91GSVk/2TDEzMcHUaEly9Y9OEZ6NZs6hosTOpy0VkbIGfceDSTARojEmLFe/bInKriBwTkRYROaeCQ0TeKiIHRSQiInfMe+wuETmR+Llr/nONMatix+xVmI3GGA9HCDpcwQWws76M/vFp+sbT1/JvcjqCQsoquGpKfVbBlT62bxtj5luTnwujU7OU50CCC+C6zVWMhyO81pX+lsTGFdbkPpsrJqYjGWtPmFRV4mM0NEt41kYuZZDtpyar7E1Ugl+zuTKlr9tUGW93mE1tCi/4Ca2q2ZGmM8Ysy0r3bRHxAl8GbgY6gX0isltVD8/ZrB34OPCb855bAfwhsAtQ4EDiucMricUY80Z2zF6d0VB8rkfQX+BwJLCzPgDAoe4xarYVpuU9xsPxWQKBVFVwBQoZD0eYmolSVGD/K6aS7dvGmPnW6ufC6NQstYH0HBczbWttCVUlBfzs5ABXNJY7HY5Js7W6z+aKiXCE+vKijL5nVYkPBdoGJ9m+LpDR916rbD812WZv6xDVpT42VqZu/hZwdp5X+1CIrS7ocLMUVgtvjFmOq4EWVW1V1Rngm8DtczdQ1TZVfQWIzXvuu4HHVHUokdR6DLg1E0EbY8xihkPxrhPlLkhw7UgmuLrSN4drLBxP6KWyggtIa9WZMcaYtSsaUybCEcpypILLI8J1zVV0Dk9l3ZwLY9aayZnMV3BVJ86tW/omMvq+xpjsMHf+lkhq5wM2zUlwZQtLcBljlmM90DHn987EfSl9rojcIyL7RWR/f3//igI1xpjlGD5bweX8hbNAYT4bKv0c6k5f26J0VHAB1qbQGGNMWoyFZ1HImQQXwJubyvHlefjZyfTO3TTGrNxsNEZ4NpayubVLVV3qQ4ATvZbgMsacq30oxJmxMNekeP4WxGcA+gu8luAyxpjVUNUHVHWXqu6qrq52OhxjzBowHJrBK+Ka4fWX1JelOcEVT+ilamB2bSBRwTVmCS5jjDGpNzYVP27lUoLLl+dl14Ygr3WNMpr4+4wx7pJcFJaqrgdLle/1ECwuoKXfElzGmHPtPTUEpH7+FoCI0FThz6oZXJbgMsYsRxfQOOf3hsR96X6uMcak1XBohjJ/Pp4Ul/ev1I76AO1DobRd8BoLR/AXeMnzpOZUsKY0XsHVO2YtCo0xxs1EpFFEnhCRwyJySER+3emYlmIkBxNcANc1V6H6+qB4Y4y7JJPrqep6sBw1pT5OWotCY8wC9rYOUVFcwNaakrS8fmOF3yq4jDE5ax+wVUQ2iUgBcCewe4nPfRS4RUSCIhIEbkncZ4wxjhuenHFFe8KknYk5XIfTVMU1Ho4QSOFK1KA/n3yvWItCY4xxvwjwv1R1B3At8EkR2eFwTIsaDeVmgquiuIDtdQFeODVEeDbqdDjGmHnOzq114LOnutRHa/8kkej88ebGmLVMVXnu5ABXb0z9/K2kpkSCS1XT8vqpZgkuY8ySqWoE+BTxxNQR4GFVPSQi94vIbQAicpWIdAIfAL4qIocSzx0C/ph4kmwfcH/iPmOMcdxIaJagv8DpMM7aWV8GwKHu0bS8/nh4NqWzBESEmtJC+satgssYY9xMVXtU9WDi9jjxc/qlztR1zGh4Fl+eh8J8r9OhpNz1zZWEZqLsfrnb6VCMMfOkem7tctSUFjITjdExPJXx9zbGuFfrwCTdo2FuvKgqbe/RVOEnPBujfyI7FrBm/hPaGJPVVHUPsGfefffNub2PePvBhZ77IPBgWgM0xphlmo3GGJ+OUO6iBFd1qY/agC9tc7jGwxFqSn0pfc3qUh/9VsFljDFZQ0Q2Am8C9i7w2D3APQBNTU0ZjWsho6HZnKveStpcVUxtwMc/PdvGB65sSNtqbGPM8o2HZ8nzCEUOJNeT5+onesfZVFWc8fc3xrjTT4/3A3Djluq0vUdThR+AjqHQ2XEEbpbWCi4RuVVEjolIi4jcu8DjbxWRgyISEZE75j0WFZGXEj9LbYFmjDHGGLMsI4m2R25qUQjxKq50VHDFYpqo4Ert31tT6rMZXMYYkyVEpAT4DvBpVT1nNYWqPqCqu1R1V3V1+i6gLNVYOHcTXCLC9c1VHOkZOzs03hjjDmPhCKWFeY4knqsTCa6WfpvDZYx53TMtAzRV+Gmq9KftPRoTCa7Tg9kxhyttFVwi4gW+DNwMdAL7RGS3qh6es1k78HHgNxd4iSlVvSJd8RljjDHGAAyHZgBc1aIQ4JL6AE8e62NqJkpRQepWjQ6FZogpKW1RCFAT8PFCm12YM8YYtxORfOLJrW+o6r87Hc9SjIZmWbfO/SuIV+qKxnKePNbHV586ybWbK8+73UN725f0eh++xvmqO2NywVgaFoUtVWG+l3WBQlr6LMFljImbjcZ4vnWI26+oT+v7NASLAOjMkhap6azguhpoUdVWVZ0BvgncPncDVW1T1VcAm5hojDHGGEecTXAVuyvBtaO+jJjC0TOpbVPYNxZvIxhI8Zf12tJCRkKzTEeiKX1dY4wxqSPxMoSvAUdU9QtOx7MUkViMielIzlZwAeR7PfzKWzfzxLF+DpwedjocY0zC+FTEkflbSVtrSzjeO+7Y+xtj3OWljhEmpiPcuDV987cgnmCvKfXRMZQdFVzpTHCtBzrm/N7J8obXForIfhF5XkR+PrWhGWOMMcbEjYRm8YqkvKJptXbWBwBSPofrzFh8FVaqv6zXBOJtVJIJNGOMMa50A/BR4J1zRgL8nNNBXcj4VASFnE5wAdx13UYqiwv44mPHnQ7FGJPgZAUXwPZ1pZzonSAStboAYww8eawPr0e4rjm9CS6ItynsGLYE12ptUNVdwIeBvxKR5vkbiMg9iSTY/v7+/sxHaIwxxpisNxyaocyfj8dlQ90bgkWUFeWnfA5X90h8TlZZilsyJofP9o1bgsvtRGTbnAvbL4nImIh8et42bxeR0Tnb3OdUvMaY1FHVZ1RVVPUyVb0i8bPH6bguZHQqPisz1xNcxb48fvXtzTzTMsBTx+36hnndEubbf0JEXk0cr58RkR1OxJlrQjMRpiMxRyu4tq8LMB2J0TY46VgMxhj3+K8jfVy1MZiRc6LGYBEdQ9aisAtonPN7Q+K+JVHVrsQ/W4EngTctsI2rBt8aY4wxJvsMT84Q9LvvopmIsLM+kPIKrp7RKTySnhlcAP3j4ZS+rkk9VT2WvLANXAmEgO8usOlP51wAvz+zURpjTFwywRXI8QQXwEeu3cDm6mJ+799fZWI64nQ4xgXmzLd/D7AD+NACCayHVPXSxHH9z4GsaD/qdsmuBKUOfvZsrysF4EiPtSk0Zq3rHA5x9Mw4N11cm5H3a6zw0zM6xWwWVJCmcxnCPmCriGwinti6k3g11qJEJAiEVHVaRKqIt1H487RFaowxxpg1azg0y/Z1pU6HcdbcAfIeEQ53j/Gvz53G61m4wmy5g+S7R8IEClNfsZas4Oq1FoXZ5l3ASVU97XQgxhizkGSCq3wNJLgK8738xR2XccdXnuNPfnCEP/uFS50OyTjv7Hx7ABFJzrc/nNxAVeeuhioGNKMR5qjesfiirVTPrV2OLTUl5HmEo2fGeP/l9Y7FYYxx3n8d6QPgXRlKcDUEi4gpnBkN01jhz8h7rlTaKrhUNQJ8CngUOAI8rKqHROR+EbkNQESuEpFO4APAV0XkUOLpFwP7ReRl4Angc6p6+Nx3McYYY4xZudlofHB9eYrb9aVKfXkRkZjSl8KqqO6RKcrSULFWWVxAnkc4M2YVXFnmTuDfzvPYdSLysoj8UER2LrSBtQw3xqTb6NQsvjwPvnyv06FkxJUbKrjnxs382wvt/ONPW8+7XUyVyelIVqysNquypPn2IvJJETlJfHH4/8xQbDmtN9F228k5vb48L83VJRy1Ci5j1ryfHOllc3Uxm6qKM/J+jcF4UqtjyP1zuNL6KZ3o5b1n3n33zbm9j3jrwvnP+xlgS5WMMcYYk1bDoRkAKorduSq8vixeFdUzEqaurCglr9kzGqY8DQkuj0eoDRRyZtQSXNlCRAqA24DfXeDhg8Rn4k6IyM8B3wO2zt9IVR8AHgDYtWuXrRg3xqTc6NRszs/fmu+3b91O+1CI//ODIxzpGedX3rqJpkSroLaBSY73TnBqcJKZSAwBmqtLuPWSddSXx88V5laDX8hyq8CNe6nql4Evi8iHgd8H7pq/jYjcA9wD0NRk/+0X0+eCCi6Ityncd2rI0RiMMc4aC8+yt3WIj9+wMWPvmaza6hhe4wkuY4wxxhg3GwnF2x4FXVrBVVXqI98rdI9O8WaCq369WEzpGZ1iY2V6WgzUlxfSPZIdg2gNEJ/ncVBVe+c/MLfdkaruEZG/E5EqVR3IaITGmDVvLSa4vB7hSx9+M1987Dhfeeok3znY+YbHK4oLuKKxnOoSHxPTEQ6cHubvnzzJh69p4uK6gENRmzRZ7nz7bwJ/v9ADtihlefrGp8nzCIX5aWt+tSTb1wX4/kvdjIZm09KFwRjjbg/tbedg+zAz0RgekSUvYlmturJCvB6hY8j93+8twWWMMcaYNStZweXWFoUeEerKilKWNBqYnGY2qpSl6e9dV1bEK50jaXltkxYf4jztCUVkHdCrqioiVxNvbT6YyeCMMQZgbGqWukRF81ri9Qi/+e5tfOz6DTx1rJ+BiRlODUzQVFFM0J+PzJmleeOWKv75uTb+7YV2fvXtzSmr+jausOh8exHZqqonEr++FziBWbUzo2ECRW/c15ywvS4+K/jomTGu2VzpaCzGGGe82jlKuT+fxmDmju95Xg91ZYV0ZkEFl7PLEIwxxhhjHDQ8OYtXxNHe+oupKyukZzRMTFe/0LZnJN5qpTxNK+HrE7FqCmI16SUixcDNwL/Pue8TIvKJxK93AK8lZuL+DXCn2n9YY0yGzUTiszLXctVCTWkhH9jVyK++vZkrGoNUFBecc8Hd78vjY9dtpCjfyyP7O4nG7OM6Vyxlvj3wKRE5JCIvAZ9hgfaEZvm6R6ZcUT26I1GVebhnbJEtjTG5KDQd4UTfOJeuL8t4wr0hWETHsFVwGWOMMca41nBohnJ/Ph6HV2ZeyPryIvaeGmJ4cobKEt+qXitZCZauL+vrygqZicQYSkGsJr1UdRKonHffV+bc/hLwpUzHZYwxc/WOhVGgzOEZONmgxJfHbVfU84297exrG+Jaq/TIGUuYb//rGQ9qDegZDVNT6uz57EN721FVSnx57H6pG1+e12bnGbPGHOoZI6Zw2fryjL93Y9DPU8f7M/6+y2UVXMYYY4xZs0ZCM66dv5VUlxgY35WCNoXdo+mt4Eq2ROpJvI8xxhizGsnjyVqu4FqOHXUBNlUV88SxPiLRmNPhGJO1ojHlzFjYFZ89IsL68qKUfBcwxmSflzpGqCwuoL488+2aGyv89I1PE56NZvy9l8MSXMYYY4xZs4ZCs5S74IvrhdSW+vBIapJGPSNTFOZ7KCrwpiCycyVPui3BZYwxJhV6RhOVx1bBtSQiwju21TAejvBiu83ENGal+sbDRGPqihaFAOuDRfSPTzMTscS1MWvJqYFJTg1McuWGoCPzABsr4gtYO13eptASXMYYY4xZk2YiMSanI1QUu7uCK8/roTZQeLa94Gr0jIapLytK28nxurJkgsvdJ8DGGGOyQ7Jiodzl1dZu0lxdzLpAIS+0DTkdijFZK3neXV7kjs+e9eVFKHaO7WYicquIHBORFhG5d4HHfSLyrcTje0VkY+L+ShF5QkQmRORL855zpYi8mnjO34gTGQ7jqIf3d+AReHNT0JH3bwz6AegcDjny/ktlCS5jjDHGrEnDoRkA17coBKgvK6J7ZArV1Q2N7xqZoi6NrQ2qin3ke8UquIwxxqRE98gURfleCvLs0sVSiQi7NgbpGpmyi+HGrFDXiLvao64vz44qirVKRLzAl4H3ADuAD4nIjnmb3Q0Mq+oW4IvA5xP3h4E/AH5zgZf+e+BXgK2Jn1tTH71xq9lojEf2d7KttpSAQ9WkDYkEV4fLP3vsLNEYY4wxa9LwZCLB5fIKLoC68kImZ6KMhSOrep2ukSnqE3Oy0sHjEWoDhfTYjABjjDEp0D0Sdn0rYTe6oqEcr0c4cHrY6VCMyUo9Zyu43PH5EyjKp9SXl5KODiYtrgZaVLVVVWeAbwK3z9vmduDridvfBt4lIqKqk6r6DPFE11kiUgcEVPV5ja9y/Bfg59P6VxhX+cnhXgYmptm1scKxGGpKfRTkeegcsgouY4wxxhjXGTpbweWOL64Xkly1uZovtaGZCP3j02ysKk5VWAuqLyuyCi5jjDEp0T0y5ZoLzNnE78tjR12AF9tHiERtZo8xy9U9MkVpYR6F+emZW7sS64NFdFqCy63WAx1zfu9M3LfgNqoaAUaBykVes3OR1wRARO4Rkf0isr+/v3+ZoRs3UlUe+GkrTRV+tq0rdSwOj0doKC+iw1oUGmOMMca4z/DkDPleocSX53Qoi1pXVogA3atoNXR6MH5SuqHSn6KoFraurNASXMYYY1Kie2SKsixoJexGb24qZ2o2ysn+CadDMSbrdCfm1rrJ+vIiBsanmZheXUcHk3tU9QFV3aWqu6qrq50Ox6TAgdPDvNg+wn+/cRMeh0evNVT4Xd8e1RJc5zE8OcPo1KzTYRhjjDEmTYZDswT9BWTDrF5fnpfKkgJ6RlaeODo9OAnAxsr0VnDVlRdyZjS86nlhxhhj1rbx8Cxj4YhVcK1Qc3UJvjwPh7rHnA7FmKzTPTJFfRrn1q5EY4UfBV7uGHE6FHOuLqBxzu8NifsW3EZE8oAyYHCR12xY5DVNjvrq060E/fl84MrGxTdOs8ZgER3WojD7hGej/OWPj/H5Hx21JJcxxhiTo4ZDMwSzaFV4fXnRqloUtiUquJrSXMFVFyhkJhpjMDHjzBhjjFmJZDVwWRa0EnajPK+Hi2pLOXJmnJgtOjFmWeIJLndVcDUG/QjYbD132gdsFZFNIlIA3AnsnrfNbuCuxO07gMf1AisCVbUHGBORayW+IvNjwPdTH7pxm9e6RnnscC8fu24jRQXOt0ltCPoZDs26unrUElwLeHh/B8lPmKePW+9SY4wxJteoKkOTMwSLsyjBVVbEyNQsoRWeWJ4enKSyuIBAYXovFK4PxhNobm9jYIwxxt26Eos6rIJr5XbUB5icjtA+6O6V18a4ydRMlOHQrOsSXEUFXmoCPktwuVBiptangEeBI8DDqnpIRO4XkdsSm30NqBSRFuAzwL3J54tIG/AF4OMi0ikiOxIP/Q/gH4EW4CTww0z8PcZZn//RUYL+fO6+cZPToQDQWBH/LHRzFZf7h0444PGjfdSU+ggU5nNqYNLpcIwxxhiTYlOzUaYjMSqyaFX4+mD8xLJzZIqLapc/aLZtIJT26i144wnwFY3laX8/Y4wxuSnZlrc8i6qt3WZbbSleEQ73jLGxKr0tio3JFcmZt/XlhUzNxByO5o2aKoo52D5MLKZ4PO5vs76WqOoeYM+8++6bczsMfOA8z914nvv3A5ekLsrc9dDe9hU978PXNKU4ktX5WcsAPz0xwO+/9+K0L0xdqsbEAtaOoRAX1wUcjmZhVsE1j6ryWtcoDUE/G6r89I6FrU2hMcYYk2OGQ/FjezZVcDUEi/DI67O0luv04GTa52/B6yfAVsFljDFmNbpHpvB6hNJCW5e7UoX5XjZXF3P0jM3hMmapki3B68rcVcEFsKHCz3g4wom+CadDMcak2Ewkxh/9x2HWlxfxkWs3OB3OWY0V7v9+bwmuefrGpxmYmKG+vJCNlcUocNDKf40xxpicMpyYD5VNM7h8eV7WlRVyegVthsKzUXrGwmzIQAVXsS+PiuICOobd28LAGGOM+3WPTLEuUIhHrEphNS6qLWVgYobhkM3GNGYpkufaTRXpP29eruS5vLUpNCb3PPD0SY71jvNHt+2kMN/52VtJQX8+xQVeV3+/twTXPK91jQKwvryI9Yl+u0fPjDsZkjHGGGNSLHmRpyKLKrgANlQU0zEcIhpb3rD4zuEQqmSkggvi1WZu7tFtjDHG/bpGpqgvL3Q6jKy3paYEgBar+DBmSdqHQhTkeVgXcN/nT0VxAZXFBZbgMibHtPSN8zePt/DeS+u4aUet0+G8gYjQEPTTMWQVXFnjeG/8pK82UEhhvpcSXx6t/XYiaIwxxuSSockZivK9rloZtRQbKv3MRpWe0eWdXLYNhM4+PxMag35XtzAwxhjjft2jU9SXu69FWLb4n8aJAAAgAElEQVSJzxfPswSXMUt0enCSxmCRK2dciQhv3hDkYLsluIzJFVMzUT75jRcp8eXxh+/f4XQ4C2qsKKLTKriyR/tQiIrigrMXvKpKfLQOrGzWhTHGGGPcaTg0Q7DYHUNblyPZKmW5bQrbEnO7MlbBVVFE1/AUsWVWmhljjDEAsZhyZjRsCa4UEBGaq0s42T9BTO24bMxiTg+G2JChc+aVuHJDkFMDkwxOTDsdijFmlVSV+77/Gsf7xvniB6+gxoWVowANiQWs6tLziLQmuETkVhE5JiItInLvAo+/VUQOikhERO6Y99hdInIi8XNXOuOcq3M4dHZ4GkB1aYFVcBljjDE5ZnhyNqvmbyWV+wsoK8rn9DLb/53sn6Dcn0+5PzNJvYagn5lojL5x++JtjDFm+QYmppmNqiW4UmRrbQmhmSg9I2GnQzHG1VSV9qGQK+dvJe3aEATgYPuIw5EYY1bry0+08MiBTn7tHVt420XVTodzXo0VfiamI4yEZp0OZUFpS3CJiBf4MvAeYAfwIRGZX2fXDnwceGjecyuAPwSuAa4G/lBEgumK9Q0BDYVoDL5+El1V4mM4NHt2GL0xxhhjspuqMhyaoSILE1wAGyv9tA1MLmv11LEz41xUW4pIZlqtJM+l3DyI1hhjjHt1jcTb3K63GVwp0Vwdn8N1os/mixtzIQMTM4Rmohlr670Sl6wvo8DrYV/bkNOhGGNW4Rt7T/OXPz7Of3vTej5900VOh3NBbv9+n84KrquBFlVtVdUZ4JvA7XM3UNU2VX0FiM177ruBx1R1SFWHgceAW9MYKwDRmNI9MvWGCq6qEh8ApwatTaExxhiTC/rHp4nElGBxdia4mqtLmJiO0LvE6ihV5XjvBNvXlaY5stclz6Xc3KfbGGOMe3UnKo3qyqyCKxVKC/NZFyi0OVzGLKJ9KH7tz80JrsJ8L1c0lfPcyUGnQzHGrNBXnjrJ//7ua7xjWzWf/8XLXDnzb66GYPwzsWPInXO205ngWg90zPm9M3Ffup+7YmfGwsxG9Q2lyMmLXzYo3RhjjMkNyVVH2diiEGBLTXwV9sklXqTqGpliYjrCtgwmuNYnWkq59QTYgIi0icirIvKSiOxf4HERkb9JtBp/RUTe7EScxpi1qTtRwWUtClNna00Jp4dCzETmry82breE8R+fEZHDieP1f4nIBifizAXJObdNFe6dwQVwQ3MVr3WPMurSdmHGmIWFZ6P81iMv87kfHuV9l9Xx1Y/uoiAvrROkUqKxYu1WcKWdiNwjIvtFZH9/f/+qX68zMc+iYU6LwmBiVoWtQDbGGGNyQzLpEizOzDyqVCv3F1BZXLDkVdjHzsTbEW2rzVyCqzDfS02pj45lzgozGfcOVb1CVXct8Nh7gK2Jn3uAv89oZMs0OR3hy0+0sPvl7rMXxo0x2atrZIoSXx6BwjynQ8kZW2pKiMaUNutOk1WWOP7jRWCXql4GfBv488xGmTtOD4YQef1irltdv6USVXj+lFVxGbOYmUiMQ92j/PDVHh56oZ1v7D3Ndw528syJfg6cHiIaW3rr/9U42D7MbV96hkcOdPI/37mFv77zTVmR3IJ4JXjQn0+7S7/fp/NssQtonPN7Q+K+pT737fOe++T8jVT1AeABgF27dq36/8YzY8k2CIVnL3758rxUFBfYCmRjjDE5TURuBf4a8AL/qKqfm/e4D/gX4EpgEPigqraJyM3A54ACYAb4LVV9PKPBL1Pb4CRC9lZwQfwi1YvtI0s6GX+1axQR2F4XyEBkr2us8Lv2BNgsye3Av2h82NvzIlIuInWq2uN0YPOFZ6N85Gt7ebF9hHyvsL9tiI9fv5HNiZkzxpjs0zM6RX15YcZmR64FGyqL8Ypwsn+CizK46MWs2tnxHwAikhz/cTi5gao+MWf754GPZDTCHNI+FKIuUIgvz+t0KBd0eUM5Rflenjs5yLt3rnM6HGNcKTQT4acnBnju5CAz0Rh5HqHcn49HhNBMlAOnh9nz2hkqigt45/Ya3nPJOm7cWr2kpNNDe9uXHMfw5AyPH+3jYPswpYV53HXdRtaVFfGtfR2LP9lFNlYVc9qli2TSmeDaB2wVkU3EE1Z3Ah9e4nMfBf5URIKJ328Bfjf1Ib5R31h8lkVN4I2DbBuCRVbBZYwxJmfNWRl6M/G2wPtEZLeqHp6z2d3AsKpuEZE7gc8DHwQGgPerareIXEL8GJ72tsKr0TYwSZk/n3xvdqyWWsiWmhL2nhpaUoXUK52jbKkuocSX2VXwm6qKefr46ivsTdoo8GMRUeCriYVjc52vZfgbElwicg/xCi+amprSF+0F/NVPTvBi+whf+vCb6B4J88DTJ3nkQCefftdWfPnuvkBljFlY5/CUtSdMsYI8D40Vflr73XlxypzXQsfjay6w/d3ADxd6wA3HbLc7PThJk4vnbyUV5Hm4alMFPzs54HQoxrjSkZ4xvnOwk6mZKJc2lLFrQwWbqorxzpl1NTEdob68kMcO9/LooTN8+0AngcI8btm5jvdeVsdbtlSt+JpBNKac6BtnX9swx86M4RHhhi1VvHN7DYVZ+v1kY2UxL5wacjqMBaXtSoeqRkTkU8QvdHmBB1X1kIjcD+xX1d0ichXwXSAIvF9E/khVd6rqkIj8MfEkGcD9qpr2f4O9Y2GK8r2UzrsA1BAs4mjPeLrf3hhjjHHKoitDE79/NnH728CXRERU9cU52xwCikTEp6rT6Q97ZU4Nhqgszt7qLYDNVSV4BI73Xvj8RFV5pXOEt11Uk6HIXre5uphvH+hkPDxLaWF2toPMcW9R1S4RqQEeE5Gjqvr0cl8k1R0Vlqt7ZIoHnz3FL7x5Pe+7rJ6H9rZzx5WNfOWpkzzXOsjbt2X+/31jzOqoKu2DIa7cEFx8Y7MszTXFPH6kj6mZKEUF2XmBzZyfiHwE2AW8baHHnT5mZ4P2oRDv2l7rdBhLcn1zJZ/74VH6xsPUlBYu/gRj1oCYKo++doaftgxQV1bI3W/ZRF3ZwgtmSnx5vO+yet53WT0zkRjPtPTzn6/08Ohr8WRXuT+fmy6u5aqNQd7UFGRTVfF5E15TM1HOjIXpGZ2itX+S1oEJwrMxin153LCliuubqygryu7vxBsq/XzvpS7Cs1HXJenSupRXVfcAe+bdd9+c2/uItx9c6LkPAg+mM775esenqQ34zmmD0BD085MjfcRiisdjLRKMMcbknKWsDD27TWIRyyhQSbyCK+kXgYNuTm5BfGVmtrfmKSrwsqmqmNe6R1HV87Zw6hqZYmBihssbyzIcYTwJB3BqYJLLGsoz/v7mwlS1K/HPPhH5LvFE99wE12rajWfM159rIxKN8Rs3XXT2vqYKP9tqS3mmZYAbtlQ5F5wxZkVGp2YZn47QVOH+Kops01xVwn/Rx6mBCXbUZ/7cwKzIko7HInIT8L+Bt7n9XNythidnGJiYYUtNdrQ4vr65EoDnTg5y+xWubqBhTEZEYjEe3t/Ja12jXLOpgvdeVkeeZ2kVWAV5Ht65vZZ3bq9lOhLlp8cH+M9XunnscC/fPtAJgAhUFvuoKilgbGoWBMKzMSanI0xHYmdfq9yfzyX1ZWxfV8q2dYE3VI1ls01VxahCx1CIrS67nmITW+foGwuf054QoDFYxEwkRv/ENLULPG6MMcasdSKyk3jbwlsusI3jbVFGQjOMhGazvoIL4JL1ZXz/pW6O9Y6zfd3C87X2tsYL4HdtqMhkaAA0VxcD0NpvCS63EZFiwKOq44nbtwD3z9tsN/CpREXnNcCo2+ZvhWejPLyvg5t31NI470L49Vsq+adnxzncM+ZQdMaYlUrOb5y/X5vVa6goIt8rtPRPWoIreyw6/kNE3gR8FbhVVfsyH2JuONE3AcCW2uxIcO2sLyNQmGcJLmOItwT81r4ODnWP8Z5L1vGWLVUrnuPpy/Ny045abtpRi6pyamCSlzpGaB8KcWY0zODkDB1DIVShusRDiS+P0sJ8agOF1AZ8lBXl5+QM0Q2V8e/3bYOW4HK1vvFpdtafe4GoIRg/se4cDlmCy6x5InIr8NfEW4/+o6p+bt7jPuBfgCuBQeCDqtomIhuBI8CxxKbPq+onMhW3MeaClrIyNLlNp4jkAWXE93FEpIF4y+GPqerJ872JG9qitA3GL5pVlficePuU2lEXYPdL3ex59cx5E1zPtQ4S9Oezfd3qT0CXM0gX4BevXI9HoLV/YtXvbVKuFvhu4otXHvCQqv5IRD4BoKpfId6F4eeAFiAE/D8OxXpez5wYYDg0y51Xn5swb64uodyfz4HTww5EZoxZjY6hKYA1V8G13OPsSuR5PGysLLZjcxZZyvgP4C+AEuCRxLG9XVVvcyzoLHWiL976+3D3GD0jYYejWZzXI1y7uZJnbQ6XWeOiMeXbB+LJrfdeWpfSDg4iwubqEjZXvzHxnYljtttsTMwnPD3ovlmeluBKUFV6x8K8c/u5ffobgvFenZ3DU1y5IdORGeMeIuIFvgzcTLyF2T4R2a2qc+f03A0Mq+oWEbmTeEXHBxOPnVTVKzIatDFmKRZdGUq8muMu4DngDuBxVVURKQd+ANyrqs9mMOYVaRuIn4xV5EAFV2lhPhurivnhqz185uaLznlcVXnu5CDXbq50pMWyL89LY4WfkzbM3nUS8/YuX+D+r8y5rcAnMxnXcv3o0BlKC/O4ofncL7EeES5vKOenJ/oZDc1S5s/unvfGrCVWwZVezdUl/OjQGcbDs06HYpZoCeM/bsp4UDnoRO8EBXkeyrNoTs4NW6r48eFe2gYm2VhV7HQ4xmScqvLZ3Yd4uXOUd++otfbkaVTuL6Dcn8+pAfd9v19aI8o1YGI6QmgmSm3g3BXdyQqujsSJtjFr2NVAi6q2quoM8E3g9nnb3A58PXH728C7JBdrc43JIaoaAZIrQ48ADydXhopIcvXn14BKEWkBPgPcm7j/U8AW4D4ReSnxc+5qEZdoG5xEJDcSXACXri/jRN8Er3WNnvPY4Z4xukameOtF1Q5EFre5qpiTtkrcpMFsNMZPjvRy88W1FOQt/JXm4roAMYUnjlm3JmOySftQiMriAkp8th43HZoTq9BbbQGKMW9wom+cmlJfVrUWe/u2+Hn+k3auY9aof/5ZG//6/Glu3FrF27a59jJEzthQWczpQfflRyzBldA7Fp/BuVALwqICL1UlBXQOT2U6LGPcZj3QMef3zsR9C26TuGg+ClQmHtskIi+KyFMicuP53kRE7hGR/SKyv7+/P3XRG2POS1X3qOpFqtqsqn+SuO++RNsTVDWsqh9Q1S2qenWiAgRV/T+qWqyqV8z5ce03rLaBSerLisj35sYp0OUN5RTle/nX506f89gPXunB6xHevXOdA5HFba4uoW1wkljMkY6UJoftbR1iJDTLuy85///fDcEiSn15PHa4N4ORGeNuIvKgiPSJyGtOx3I+HUMhq95Ko7ryQgrzPbYAxZh5TvROUFPq/jbmD+1tP/vzbMsgm6uKefK4XTcxa8/jR3v54/88zC07ah39zruWbKz002YtCt2rbzzeX7f6PAez9UG/JbiMWZ0eoElVB0XkSuB7IrJTVc+Z/u6GOT3GmNx0ajDExqrcuWhWVODl59+0nu8c7OTTN2+lrizeVnk2GuP7L3VzfXOlo9VqzdUlhGdjdI9Ona2INyYVfnSoh6J8L2+7QIWiR4TtdaU8eayP6UgUX543gxEa41r/DHyJ+MxcV2ofCnF5Y7nTYeQsjwibq0oswWXMHKOhWfrGp3lzU9DpUJbtbduqeWhvO+HZKIX5dq5j1oaWvgl+7aEX2VEf4K/uvILvvdi9otdZi7O0VmNDZTH/8XK3675b5cby5RTou0AFF0BjsIjOYfeV4BmTYV1A45zfGxL3LbiNiOQBZcCgqk6r6iCAqh4ATgLnDo0xxpg0Oj04yYbK3OpP/z/e3oyq8oUfHz9733+83E3XyBQfv36jc4EBm6vj/66tDZJJtSeP9XPj1qpFL+RcvC7A5EyU504OZigyY9xNVZ8GhpyO43wi0RhdI1M0VRQ5HUpO21xdzHBo1sYwGJPQ0j8OQM0CY0vc7u3bapiOxHiu1c51zNowOR3hE//3AIX5Xv7hY7vwF1j9TqZsqvITU/eNcbIEV0LvWLyC63wJroagn66RKaLWYsesbfuArSKySUQKgDuB3fO22Q3clbh9B/C4qqqIVIuIF0BENgNbgdYMxW2MMQxNzjASmmVzjg1gbqzwc/dbNvPIgU4e2d9B53CIP91zhB11Ad7hcB/y5JyPE322StykTsdQiM7hKa5vrlx02+aaEnx5Hp4+PpCByIzJHU61DO8ZDRONKU3WojCtksdnS/4bE3e8N36uWlO68DVBN2sbmCTfKzzwVOvZ1oXG5CpV5be/8wqt/RP87YfedLaDicmMzVXx84eWPnctYLUEV0Lv2DTFBd7zDrJtCBYxG9WzrQyNWYsSM7U+BTwKHAEeVtVDInK/iNyW2OxrQKWItACfAe5N3P9W4BUReQn4NvAJVXXt6lFjTO5pSSRZmmtKHI4k9T5z80Vcu7mC3/r2K7z1z59gaibKX995BR6Ps0Oyq0t9VJX4ONx9TjdaY1bs+cQK5euaqxbdNt/rYdfGID87aQkuY5ZDVR9Q1V2ququ6+vytQFMtuSLYZnClV02pjxJfHs/aZ6MxABzvHaco30u5P9/pUJYt3+thc1UJx3vHnQ7FmLR78Nk2fvBKD7/17u1cv2Xx7wImtZLXUtzW5thq+BJ6x8Pnrd6CeIILoHN4yrLDZk1T1T3Annn33Tfndhj4wALP+w7wnbQHmCL/daSX3S93MzQ5w1u2VFFamH0nusaYNzrRF//St7WmhJ6R3FqwUpDn4V/+32v41r52ekbD3HlVE02V7rg4uKM+wOEeS3CZ1Hm+dYiK4gK2LjFZfX1zFX/x6DEGJqapKsm+1kPGrCWnEwkuq+BKLxFhc3UxPzs5iKoi4uyCGGOc9mrnKDvqA3iydF+4aF0px14et3Mdk9P2tQ3xZ3uOcMuOWj7xts1Oh7MmlfjyqCsrPLt42C2sgiuhf2ya6tLzHwSSK8hsDpcxue+Bp09y99f387OTgzzbMsBXnjrJSGjG6bCMMavU0jdBUb6X+hxdqFKQ5+Gj123kt2/d7prkFsDO+gAtfePMRGJOh2JygKryfOsg126uWHKFYrKV4fM2m8IY12vtn8CX58nZY7WbNFeX0D8+7bpV2MZkWiQa47XuUS5rKHM6lBXbVlsKwLEzVsVlclP/+DSf/MZBGoJF/OUvXW4LMxy0pabEElxu1TsepuYCFVzry+Mn2B1DU5kKyRjjgP1tQ/zZD4/y3kvrePZ33skn3tZMaCbKIwc6idkMPmOyWkvfBM01xY637VtrdtQFmI3q2Qo6Y1ajc3iKrpEprt28+PytpEvXl1Hqy+PZFktwGSMi/wY8B2wTkU4RudvpmOZq7Z9kU5UdqzMhOYfrZzaHy6xxLf0ThGdjWZ3gqiguoKbUZ10TTE6KxZTffORlRqdm+fuPXEnAOiw5qrm6hJP9E666RmotComvBO0dC7MucP4KrsJ8L9WlPqvgMiaHxWLKfd8/xLpAIX/xgcsoyPPQEPTz3kvr+PcXu/j+y138tzc1OB2mMWaFWvomlnVR3KTGjvoAAIe6x9hZn70XDow7PJe4EHvdMvblPK+HazZX8JzNmjEGVf2Q0zFcyMn+CTtWZEhFcQENwSKebRngY9dtdDocYxzzSscoAJc1lLO3NXvHhO+sD/DksX4mpiNOh2LMqjy0t/0Nvz9zop+njvdz2+X1vNg+wovtIw5FZiBewRWaidIzFj5bEOQ0q+ACxsIRwrOxC87gAmgMFtE5bBVcxuSqx470crhnjN+5dTv+gtfz/2/eEKSurJAvPnaCSNRabBmTjSamI/SMhtmyxJk9JnU2VhZTlO/lcLetKDWr93zrIJXFBcvel69vrqJtMETXiJ3LG+NW05EoHcNTNFcXOx3KmnF9cyXPtw4RddEqbGMy7ZWuEUp9eWyqzO7Pnh31ZShw1Kq4HCEit4rIMRFpEZF7F3jcJyLfSjy+V0Q2znnsdxP3HxORd8+5v01EXhWRl0Rkf2b+EnfpGpni0UO97KgLcM2mCqfDMXB2DrKb2hRaggvoG4sPmr9Qi0KAhqCfDqvgMiZnffWpkzRV+HnfZXVvuN8jwju21dA+FOLxo30ORWeMWY2TiZMvS3BlntcjXFxXai1TzKqpKs+1DnLt5spl991PVm/utTlcxrhW+2CIaEzZXG3H6ky5vrmK0alZjtgx2qxhr3SOcsn6sqxvjVpfVki5P9/OuR0gIl7gy8B7gB3Ah0Rkx7zN7gaGVXUL8EXg84nn7gDuBHYCtwJ/l3i9pHeo6hWquivNf4brTEeifPOFdop9Xn7hTett7pZLbLEElzv1jk0DUFt6/haFAA3BInpGwlbBYUwOOtw9xsH2ET523QbyvOd+NF5cF2BdoJD/O69U2hiTHVosweWoHfUBjnSPoWorxM3KtQ+F6BkNc23z8luNbl9XSllRPs9bgssY1zrZPwm8PhvKpN/1ic/Tn1kLV7NGTUeiHOkZ47LG7G+NKiLsqAvQ0jdhbQoz72qgRVVbVXUG+CZw+7xtbge+nrj9beBdEs/Y3A58U1WnVfUU0JJ4vTXvP1/uYWhyhl+6qhG/z6YsuUVliY+gP98SXG7Tm6jgWqxFYUPQTySm9I5PZyIsY0wGPby/g4I8D7/45oVnbHk9wp1XN/L08X7aB62S05hsc6JvgnyvsKHC73Qoa9Kl68sYn45wst89J8Em+7w+f2v57Uk8HuGaTRU8n8WzNYzJdcljxCZrUZgxNYFCttSU8GyLJf/N2nS0Z5zZqHJ5Q7nToaTEzvoyIjHlqWP9Toey1qwHOub83pm4b8FtVDUCjAKVizxXgR+LyAERued8by4i94jIfhHZ39+fG//tX+4Y4UD7MG/fVsPmKlv44jZbakpo6Rt3OoyzLP0J9I4nWxReuIKrsSI+OK1zKOSaIWrGmNWLxZQ9r/bwzm01BIsLzrvdnVc18bePt/DQC+3c+57tGYzQGLNaJ3rH2VRVvGCFpkm/qzfFV4i/cGqYLTWlDkdjstXzrYNUlfhWXN1x7eZKfny4l66RKTuXN8aFWvsnWRcopMRWaWfU9c2VfPtAJ9ORKL487+JPOI+Hltjp4sPXNK34PYxJtb2n4sndXRuDDkeSGhsq/RQXeHn00BneO2/0gslKb1HVLhGpAR4TkaOq+vT8jVT1AeABgF27dmV9y4zh0Azfe6mLpgo/79xe43Q4ZgEX1Zay++VuVNUVrSPtKg/QNzZNaWEe/oILn0g3BOOrvjuGbTi1MbnkxY4R+sanec+l6y643bqyQt5+UTXff6mLmA1iNiarHOoeY2d99rceyVYbK/1Ulfh44ZStEDcr8/r8rYoVf4myOVzGuNvJ/gk2W/VWxr11azWhmSj7Tg07HYoxGfd86xCbq4upKb1wR6ds4RHh4roAjx/tIzwbdTqctaQLaJzze0PivgW3EZE8oAwYvNBzVTX5zz7gu6yB1oWxmPLvBztR4Jd2NeLN8tl4uWpnfRnj4QidLsmRWIKLeIvCdYu0JwSoLy9EBDqHrT2ZMbnk0UNnyPcK71jCypD3X15Pz2iYg+32BdCYbDE4Mc2ZsTA76gJOh7JmicTbw+1rs89OszJtgyF6x6a5bgXzt5JsDpcx7qWqtPZP2PwtB1y/pZKCPA+PH+1zOhRjMioSjbHv1NDZBTC54rKGciamI/zkSK/Toawl+4CtIrJJRAqAO4Hd87bZDdyVuH0H8LjGBxTvBu4UEZ+IbAK2Ai+ISLGIlAKISDFwC/BaBv4WR31j72lO9k/yc5fUUXGBDkvGWTvq49dWDveMORxJXFoTXCJyq4gcE5EWEbl3gcd9IvKtxON7RWRj4v6NIjIlIi8lfr6Szjh7x8KLzt8C8OV5qS0tdE120hizeqrKj147w/XNVQQK8xfd/qYdtRTkefjPV3oyEJ0xJhWSJ1076y3B5aSrNgbpGpmyhUJmRZLzt1ZzESo5h2vvKZvDZYzb9I1PMxaOsKXGElyZ5i/I47rNlTxxzBJcZm15sWOE8ekINzRXOR1KSm2uLqY24ON7L3Y7HcqakZip9SngUeAI8LCqHhKR+0XktsRmXwMqRaQF+Axwb+K5h4CHgcPAj4BPqmoUqAWeEZGXgReAH6jqjzL5d2Xa6cFJ/nTPUbbWlHBVjrQNzVXbakvxCBzuzvEEl4h4gS8D7wF2AB8SkR3zNrsbGFbVLcAXgc/Peeykql6R+PlEuuIE6B2bXnT+VlJDsIiOIbswY0yuOHpmnPahELdecuH2hEklvjzesa2aPa/2ELU2hcZkheRJ1w5LcDkqOYdrX5slF8zyPd86SE2pj81Vq2tfds3mSk4PhugesQVrxriJHaud9c7tNZwamOTUwKTToRiTMU8e68PrEd6yNbcSXB4Rbru8nieP9TE8OeN0OGuGqu5R1YtUtVn1/2fvvuPbqs/Fj38eSd57O/GIs3fIJoEACRuaEmigJBQKvfRSbqEFLm1vx6+3LR2U3raMQqGsQqEQdhtmgCRsCNl7J45H7Njxjvf4/v7QcRCOncixpCPZz/v18svS0dHRo/E94zuer/mttex/jTFLrdtNxpgrjDEjjDEzjTH7PJ77W+t5o40xb1rL9hljTrH+xndus7/q6DD88IVNuJzC16ZmB8W8TqpnUeFOhqXFBs0ILn/O3joT2NNZYEVkCbAAd4t0pwXAL63bLwL3S4B/wcYYyuq8G8EF7gYuTa+jVP/x1pZSRODcsRleP2f+pMEs23qI1fn9L52BUv3R1oO1ZCVGkRjdP1MceDupu91GZ8YRF+ni8/2VXDYl2+5wBjQRyQH+gbtnqAEeNsbc22WducC/gf3WohFULFEAACAASURBVJeNMXcEMs5OnfNvzR6W0ueL3VnDkgH3pPL6O1QqeHRWkIzJjLM5koHp7DHp/GLpVlbsKOP6OUPtDkd1ISIXAvcCTuBRY8zvuzx+JnAPMAlYZIx5MfBRhp4VO8qZlptEQtSJM7mEmkunZPHIh/t5bXMJ18waYnc4Sp3Q3z/J5/P8Sv54xSm0tHXYHY7ywrhB8aw9EBxtJP5MUZgFFHrcL7KWdbuONZyzBuisLR4qIutF5H0ROcNfQVY1tNLabsiI83YEVzSltU20tWthU6o/WLa1lBlDkknzch8AcM7YdKLCnLy2SYf8KxUKtpXUMlbn37Kd0yGcNjyFlTvK6dARsHZrA243xowDZgE3dZNpAeBDj4wKtjRuAew7XE95XbNPOpWMzYx3z8O1V0cSKhVMth2sJTc5mjgvUoYr38tJjmZkeizvbtM5e4KNl9mRCoDrgGcCG13o2n+4nu0ltZw/3vuOrqFk3KB4xmTG8eyqAtzTPCkVvPaWH+EPb+3gnDHpLJzatelABatxg+Mprm6kusH+kaJ+nYOrD0qAXGPMFNx5SZ8RkWNqpkTkBhFZIyJrysvLT+qFDtU2AXg9gisnOYr2DkNJTdNJvZ5SKnjkH65nR2kdF3iZnrBTdLiLs8em8+bmUm3sVirINba0s6/8iM6/FSQuGJ9JaW0Tm4pr7A5lQDPGlBhj1lm363DPFRC0V5Od82/NHt73Bi6HQ5g5NJnP9lf0eVtKKd/ZVlLLOO2MYqsLxmeyan8FFUea7Q5FfdnR7EjGmBagMzvSUcaYfGPMJkAvTr30xmb3nNoXTxxkcyT+ISJcM3sI20pqWVcQHCMslOpOe4fhBy9sJDLMyZ1fm6ipCUNI53lbMKQp9GcDVzGQ43E/21rW7Toi4gISgApjTLMxpgLAGLMW2AuM6voCxpiHjTHTjTHT09LSTirIzgaudK9TFEYD6DxcSvUDy7aWAnD+uN732vrqpEFU1Lfw2T7tAa5UMNteWkuH0Tk9gsXZY9JxOoS3rf2vsp+I5AFTgFXdPDxbRDaKyJsiMr6H5/e5w9mJfLqvgoz4CPJSon2yvVk6D5dSQeVIcxv5FfV6rLbZRRMz6TDwjo7iCjbeZEfySiCO2aHAGMNL64qYPiSJwYlRdofjF8+sKqClrYMIl4M7Xt124icoZZNHPtzH+oJq7lgw3uu6eRUcOs/bthb37wau1cBIERkqIuHAImBpl3WWAtdaty8HVhhjjIikWcOwEZFhwEhgH35QVuvunZQR7116suFpsQDsOlTnj3CUUgG0bGspE7LiyUnufYXZ3NHpxIRrmkKlgt2GgmoAJmUn2ByJAkiMDmfWsOSjHQyUvUQkFngJuNUY0/XKZB0wxBhzCvAX4F/dbcMXHc6Op6PD8OneCk4fnuqzHp2e83Appey3s7QWY9ARXDYbNyie3ORo3tyix+j+yt/H7FCx5kAV+8rr+fqMnBOvHMIiXE6mDkliS3EtZXWahUoFn12H6vjz27u4aEIml5wy2O5wVC+lxkaQnRQVFKNEXf7asDGmTURuBpbhngjzcWPMVhG5A1hjjFkKPAY8JSJ7gErcjWAAZwJ3iEgr7iHWNxpj/DJMonMEl7fz72TER5AcE872Em3gUiqUHaptYl1BNbefd8zgUK9Ehjk5f3wmb24p5Y4FEwh3BWvGV6UGtrUHqshKjGJQQv/snRnMnllV0O3y5JgIPt5TwT3v7iI97su99K46NTcQoSlARMJwN2790xjzctfHPRu8jDFviMhfRSTVGHM4kHFuK6mlsr6F00ek+mybnvNwXTYl22fbVUqdnG0H3bsbHcFlLxHhogmZPPbRfmoaWkmI1vnQgoQ32ZFULzzxcT5xkS7mT+qf6Qk9nTYshVX7Kvjryr388pJuB+MrZYvW9g5uf34jsZEufn3pBE1NGKKmD0ni470VGGNs/Q79WiNrjHnDGDPKGDPcGPNba9n/Wo1bGGOajDFXGGNGGGNmGmP2WctfMsaMtya0nmqMedVfMR6saSIlJpwIl9Or9UWEcYPigyK/pFLq5L1tpd64sJfzb3maP2kQNY2tfLwnoHV9SikvGWNYc6CSaUOS7A5FeRg3KB4BNhRW2x3KgCXuq4/HgO3GmD/3sE6mtR4iMhP3dUPAhzx1HmPnjPRdA5fOw6VUcNlWUktidBiDEjQ1kd0umjiItg6jI62DizfZkVQXz6wq+NJfp/2H63lzSwlXzxpCdLjf+vwHjZTYCKYNSeKfqw7oVCsqqDz43l42F9fw20snkBrr3aATFXym5SVTXtdMUZW9qd8H/JCDoqoGspN616t73OB4dh6qo61d5+9UKlQt21LKsLQYRqTHnvQ2zhiZRnyki1c3appCpYJRUVUjh2qbmZ6nDVzBJCEqjFEZcaw9UEV7h7E7nIHqdOAa4GwR2WD9XSwiN4rIjdY6lwNbRGQjcB+wyBgT8C/soz2HGZkeS4aPc/LrPFxKBY/1BdVMzErQ3ttB4JTsBIalxfDiuiK7Q1EWY0wb0JkdaTvwfGd2JBG5BEBEZohIEXAF8DcR2WpfxMHtzje2Exnm5Fun59kdSsCcPSYDEeHud3fZHYpSAGw9WMN9y3dzySmDuWhi/x9J2Z9Ny3XXtaw54JfEe14b8A1cxdWNZPWygWv84Hha2jrYqfNwKRWSqhta+HRfBReMz+zThXS4y8GFEzJ5e9shmlrbfRihUsoX1h5w54LWEVzBZ+bQZOqa2thysMbuUAYkY8xHxhgxxkyyMiZMtjIvPGSMecha534ro8IpxphZxphPAh1nU2s7n++v9OnorU5zrJSHH+wq9/m2lVLeq2loZeehOmbkJdsdisKdsWbh1Gw+31/JgYp6u8NRFi+yI602xmQbY2KMMSnGGM1F142Pdh/m7W2HuGneiGPSZPdnCVFhXD9nKC+vK+aj3Zp9Rtmrpc2dmjApJpw7FuiuKtSNzowjNsJ1tO7FLv1/PO5xGGMormrk3LEZvXredOvke/X+SsYP1knrlQo1y7eX0d5huHD8yacn7PTVUwbz/Joi3ttZ3qd0h0op31tzoJLYCBdjMnVOj2AzOjOO1NgIPthVziTtta96sPZAFc1tHUcbo3xpVEYsgxMiWbGjjEUzde43peyytqASY9DR1kHka1Oz+OPbO3lpXTH/fZLzFSsVbJpa2/nVq1vJTY7m+jlDgZ7ni+2PbjlnJMu2lnLb8xt49eY5ZCZEdvv+vZkPt+vzdA5d1Rv3Lt/FjtI6Hv3mdBKjw+0OR/WR0yFMyU1kTb42cNnm8JEWmts6yErs3QiurMQoshKj+Dy/kutOH+qn6JRS/vLmlhIGJUQyKbvvDdSzh6WQEhPOv9YXawOXUkFmTX4VU3ITcTq08STYOESYOyqNF9cVsbm4hknZicDJVTToRXX/tXx7GeEuB7OGpfh82yLCvDHpvLK+mOa2dq/n41VK+dbn+6sIcwpTcrSBK1gMSohizohUXlpbxC3njNTzKNUv3PnGdnaXHeGJb80gMmzgHfMjw5w8dPU0LnvgY65+bBX/+I+ZdoekBqC1B6p48L29XDk9h3PH9W6wiQpe04Ykce/y3dQ0tpIQFWZLDAM6RWFRlXuCxd7OwQXu1Dqf76+kQ+eOUCqk1Da18sGuw1w8cZBPRgy4nA6+NjWLd7cfoqyuyQcRKqV8obyumR2ldczUlEdBa3JuIpnxkby1pZRmTfOqujDG8M72Uk4fnkJMhH/65M0bnU5DSzur99vb41CpgWxNfiUTshKICh94Fc7B7Bun5lJc3cjbW0vtDkWpPtt2sIYnPz3At+cMZe7odLvDsc2ojDgev24GpTVNXHTvh2wursGG6VXVAPLMqoKjf098nM8N/1hDQlQYozPjvvRY1z8VWk4bnoox8Mke+1KgDvAGLvek0r2dgwtg7ug0Dh9pYX2hXhArFUre3XaIlvYOvjLJdxNZLp6ZS1uH4cW1OhmzUsHifWtenXljBu5FbLBziLBg8mBqGlt5fXOJ3eGoILPr0BEKKxv92rvztBEphLscrNhR5rfXUEr1rKm1nU1FNTr/VhA6b1wmOclRPPrRfp9s71BtEx/sKuf1TQd54uP9lNZox0AVGNUNLby0rpiJWQn86MIxdodju1OHpfDa9+aQmxzNs58X8OD7e9lZWqsNXcrv3txSQmV9CwunZQ/IUZT92dTcROIiXazcad811YBOUZh/2D1pam5ydK+fO29MOmFO4a0tpUwboifkSoWK1zeVkJUYxZScRJ9tc1haLLOGJbPk80JuPHM4Dk3joZTtVu4oIz0ugvGDdf6tYDYkJYazRqXx3q5y0uIiOGNkmt0hqSDx7vZDAL2eK7c3osNdzB6WwrvbD/Hz+WN1LjilAmxjYTUt7R3awGWj4/WUPyU7kdc2lbCuoIqpuSeXQrKxpZ1XNx1kQ2E1AGFO4eO9Ffz2je3811nDueXcUZoCUflNe4fhudWFdBjDXxZPIdw1oPv4H5WXGsPL3z2NH724ifd2lvHkpwfISYrivHE65YLyj12H6li1v5I5I1IZlhprdzjKx1xOB2eOSmPlznKMMbZcUw3ovfu+w/UMTogkOtz7dr7O4ZKvbSzhjJFp/GvDQVraOvwYpVLKV2oaW/lgdzkXT8z0+Q538cxcCiobeG+X9gJXym6t7R18sKuceaPTtcI6BJw7LoMJg+N5c0sp724/RIf2IFXA29sOcUp2AhnxkX59nYsnZlJQ2cCmohq/vo7qmaamGbg+2nMYh6DphIPUtCFJRIY5uH/FnpN6fmlNE39ZsZtNRdXMHZXGTy4aw68umcDKH8xl/qTB3LdiD995ag31zW0+jlwpt+U7DnGgsoEFk7PIS42xO5ygEuZ0MCMvmf8+bzSXTc6itqmNxz/ezzce/Yxdh+rsDk/1Iw0tbby8roj0uAjO03m3+q2zR6dTXtfM1oO1trz+wG7gKj/CsLSTbzm+ZtYQyuuaeUPT6igVEl7fVEJru2H+pME+3/ZFEwaRlRjFfcv3eDW8XytxlPKftQeqqGtuY94YHQ0UChwiXDkjl6m5SazYUcYjH+7jYHWj3WEpGxVWNrCxsJrzx/u/J/GF4wcR5hSWbjzo99dSSn3Zyp1lTBuSREK0PROSq+OLcDmZOyqdFTvK+LiX82oUVzfyyIf76DCG75w5nPPHZxIX6f6eh6bGcPeVk7ljwXhW7iznhqfWaKdh5XO7y+p4f2c504ckMTknUTtT9MDpEGYMTea/zxvFVyYOYktxLRff+yG/eW0bdU2tdoenQpwxhpfWFlHf3M4V03MIcw7oZoh+7azR7rqX92xKUzhgf1nGGPaV1zMs7cS9OHo6AJ41Ko2R6bHc/e4umtt0cnSlgt1zqwsYnRHHpOwEn2873OXg5rNHsKGw+ujcP0ope6zYUUaYUzh9RKrdoSgvOR3CwqlZLJyaTVltM/ev3MOjH+5jfUGVnmMNQC+tK0IELp2S5ffXSogO46xR6by26SDtHTp6UKlAOVTbxJbiWuaO1rkyg9ns4SlkJUbxm9e3e72PLKxs4LGP9hER5uCGM4eT08OUEN+cncddCyfx8Z4KfvjiRjp0H6x8pKy2iefXFJEWF+GXzq39UZjTwekjUln5g7lcMT2bxz7ez7w/vs/L64p0fi510j7eW8H20jounJBJVmKU3eEoP0qNjeCU7ASWbT1ky+sP2Aau8iPN1DW3MawPw5QdDuHn88dxoKKBe97d7cPolFK+tr2klo1FNVw5I8dvKcsWTs0mKzGKu9/ZpRdoStmko8Pw6saDzBmRerSnsAoNIsK0IUncfv4oLhyfSWVDCy+sLeJ3b2xnyeoCdpTUagPEANDRYXhpXRGnWZWqgXDJ5MEcqm1mdX5lQF5PKQVvbSkF4HxNVxTUwpwOfnzRGLaX1PL4R/tPuP7+w/U89vF+osNd/OcZw0iOCT/u+pdPy+aHF4zm3xsO8sDKk0uFqJSn9g7Drc9toKWtncUzc3XerV5Kjgnnzq9N4l/fPZ2spCj++/mNXPHQp2w9qKmcVe9sKKxm2ZZSxg2K57ThKXaHowLgkslZbC6uYWdp4NOcej/5VD/T+WGPSI876W10juiakZfEg+/tJTc5msUzc30Sn1LKt55bXUi408FlfuwNHu5ycNt5o/jBCxt5fk0hi3R/oFTArdpfSUlNEz++aIzdoaiTFB3u4sxRacwZmcqBigY2FlWzuaiGTUU1RIc7mZSdwBkj0kg6QaWZCk2r8ysprGzktnNHBew1zx2bTmyEiyWfFzBrmF6AKxUIr28qYXRGHCMzTv56XAXG/EmDWLrxIP+3bCdTchOZ3sOcaTtKannm8wISo8O5fs5QEqK862j03bnD2VN2hD+9s4sxg+J1jhbVJ/ev2MMneytYODXL7/N4hhJvUzJ6rnfFtGyumpnDXW/t5Kt/+YivTc3mpnkj/BWi6kfK65r57tNriYtysXBqts6LPUBcOnkwd76xnRfWFPL/5o8L6GsP2K4Mm4vdvQ8mZMX3eVvzJw1m3ug0fvLyZh7+YK8O31UqyNQ2tfLSuiIumJDp9wrRhVOzOHVoMr9+bRv5h+v9+lpKqWM9t7qAuAiXVo70Aw4RhqbGcOnkLH5y8RiumTWEEemxrMmv4s/v7OJfG4qp1bkB+p0lqwuJCXdy4QT/z7/VKTrcxRXTs3l9cwmHapsC9rpKDVQFFQ18nl/J/EmD7A5FeUFE+L/LJ5GVFMX1T65h7YEvj3Zta+/gL8t38/SqA2TER3LDmcO8btzq3P6dX5vIxKwEbntuA3vKAt/zW/UPH+85zL3Ld3HZlCym5ibZHU7I65wnd+Xtc/nW6UN5deNBzvnTe7ywppDyuma7w1NBqrmtne88tYaqhlauPnUIUeFOu0NSAZISG8E5Y9P514ZiWtsDO7fmgG3g2lpcS05yFInRfa/sDnM6mDcmnQmD4/ndGzu45P6PvRq+r5QKjKc+PUBdUxs3nDHM768lIvz5ysmEuRz8x5OrqTiiJ35KBUrFkWbe2FzKwmnZRIcP2EHq/ZLL4WDsoHgWzcjl9vNHMz0viTX5ldzz7i7+ueqApoXtJworG1i68SBXzsgNeBm+7rQ82joMT392IKCvq9RA9PyaQhwCl0/PtjsU5aXE6HD+8R8zSYwO4+t/+4yfvrKZVzce5MlP8vnq/R/zp3d2MSErgW/PGUpsRO/335FhTv52zTQiwxz85z/WUtOgHVhU7+wpq+PGp9cyIj2WX186QUeM+FBCdBg/nz+OD/9nHt8+YxhbDtZwz7u7eHldEQ3NbXaHp4KIMYafvryFdQXV/OnrpzBY590acK6YlsPhIy0s3x7YubgGbAPX5uIaJmYl+Gx7LoeDRTNzuWBcBlsP1vDAyj2sPVDls+0rpU5OQ0sbj320n7mj05iY7bsyfzxZiVE8fM10iqsaueKhT9lRWhuQ1w11z6wqOPqn1Ml44pN8Wto7uHrWELtDUX6UEBXGgslZ3HruKAYnRvGzV7aw6OHPdNRsP/C3D/biELjhTP93SOlqSEoM547N4J+rCmhsaQ/46ys1UDS1trNkdSFzR6czKEErvkJJTnI0S2+aw5UzcnhxbRHfe3Y9v1i6ldb2Du6/agqLZuQSEXbyPfUHJ0bx4NXTKKpq4PtL1uu8m8prRVUNfOuJ1US4HDx+3YyTamRVJ5YeF8lPLx7LDy8Yw+kjUllXUMWf393FS2uLjslk5Xltr9f4A4Mxhj8s28lL64q49dyRXDxRR2kPRHNHpzEkJZr7lu8JaCfUAdnAVVbXREFlA5OyE326XYcIZ41O5/o5w2jrMFz+0Cf84t9bOKI9GpSyzVOfHqCyvoWbA5wreubQZJ7+9qnUNrXylfs+4nvPrue9nWXUNGpvRKX8obqhhSc+yefC8ZmMSI+1OxwVAKmxEVx/+lD+cPkkdpTWctG9H/LkJ/k6mitEHapt4vk1RVw+LZvMBHvmzLjxrGFU1rfwtw/22vL6Sg0EL60r4vCRZr59xlC7Q1EnISE6jN9dNpGN/3s+b916Bh//+Gzeue1M5k8a7JPtz8hL5peXjOf9XeX8YdkOn2xT9W/7yo/w9Yc+paahlceunUF2UrTdIfV7sREuLp44iJvnjSQ1NoLbX9jIVY+sYm/5EbtDUza6b/keHnxvL984NZdbzhlpdzjKJi6ng1vOGcm2klre2loauNcN2CsFkY92HwZgzohUv2x/aGoMt54zkgOVDTz5aT6vby7hxrOGc/WsIUT2oUeTUqp3iqoauHf5bs4ek97jZMj+NCMvmWW3nsnDH+zjmVUFvLrxIOA+IQx3OYgOdxId7mTV/goSo8JIj49kRHosO0vqSI4NxyHCVafmBjxupULRvct3U9/cxi3n6sn0QCIifH16DmeOTOPHL2/iF0u38taWUv5w+SRykrWCI5Tc9eYOjDH811n2TV4+bUgy8ycN4qH393LF9ByyNK2KUj7V0NLGfct3MyU3kdnDUuwOR/VBVLiTMZl9n8+8O984dQjbDtbyt/f3MW5QPAsmZ/nldVRoM8bwyvpi/vffW4lwOXj2hlmMHxyYjC3KLTPBPeeeQ4Tfv7mdi+75kBvPGsaNc4fbHZoKIGMM96/Yw93v7uLyadn8eoGmCB3oFkzO4oGVe/jzO7s4b1wGYU7/j68akA1cH+wqJyUmnHGD/HNCBhAR5uSXl4zn0ilZ/N+yHfzm9e08/ME+rj0tjwing2hryLRWXivlH8YYfv6vLQDcsWB8wF/fcwj+kJQYVv3sHNYXVLOpqIayuibWHaiioaWd+uZ2Ptx9mIaWNppav5iEMSrMSW5yNNWNLUwfksyk7ARtIFeqBxsLq3nyk3yunJHLWD8e21XwykyI5O/XzeD5NYX8+rXtXHjPB/zsK+NYPDNHL7BCwPu7ynl5fTHfnTuc3BR7GyZ/cvFY3t1+iF/8eyuPfHOa/n6U8qH7V+zhUG0z9181VcuWOq5ffHU8uw7V8T8vbWJ4WiwTfDi9hApt7R2Gj/Yc5i/Ld7PmQBUz8pK4+8rJOnLLJp2dcs8bl8FvX9/GfSv28PSqAqbmJjFtSBIJUWF2h6j8qLW9g18s3cozqwq4bEoWdy2chMOhx/eBzukQ/ufCMdzw1Foeem8v3wvAiL4B18DV2NLOih1lnDs2w++FrrOC+ysTBzN2UDzv7Sjn/5btxOUQJuckctpw/4wgU0rBk5/ks3JnOT+fPy4oTnajw12cPiKV062Ro93loG5ubaf8SDOlNe40qgcqGvjDWzsBCHc6mJSdwPS8ZGbkuU8WE6PDA/oelApG1Q0tfH/JejLiI/nJxWPsDkfZSES4ckYup49I5UcvbuKnr2zmra2l3LVwos7zEsQOVjdy+/MbGZURy/eDIJ1JVmIUPzh/NL95fTt/fW8vNwU4xfFA1dbeQfmRZmobW2ls7SAu0kVCVBi5ydFkJUUFpOen8q81+ZX87YN9XD4tmxk2ZFZQoSXc5eCv35jGgvs/4urHVvH4dTOYmptkd1gqADyvk40xXDghk6KqRjYX17ChsJoPdpVTVtdMfKSLS04ZzMyhyUFxvT/QpcVFcM+iKVwzO4+/rNjNu9sPsXz7IUZmxDI1N8mvAwyUPYqrG/n+s+tZe6CKm+YN5/bzRmvjljrq/PGZzJ80iLvf3cWErATmjUn36+sNuAauNzaXUNvUxhXTcwL6usNSYxk2J5ZDtU18sreCDYVVrDlQxer8Sq49bQhzR6fr6AylfOS1TQe547VtnDs2g2+dlmd3OF6LCHOSnRRNdlL00ZSKF07IZE1+JWut/cUjH+zjoffd88skRYcxPS+ZsZlxjM6MZ0dpLcnR4Ucnd9YRoqq/q21q5fon13CwupElN8wiPlJ7CCrITorm6etP5elVB7jzjR2cf/cH3HruKL5xaq6eawWZ8rpmrvv75zS1tvPAVacGzfdz/ZyhbCqq4Y9v72RwYiSXTcm2O6R+p6m1nc/2VfDeznKWbS2lrLaZdo8J6p9fU3j0tsshTMhK4NRhyZw+PJWZQ5OD5reivLOv/Ag3Pr2OnKQofj5/nN3hqBCRFhfBkhtmc83jq/jGI6v4/cKJmq6wH+roMJTUNpF/uJ59h+t5a3MJh4+0UFHfTHVDKz+zsrIAJMeEMzMvmaSYcMZkxmnnhyA0bUgST3xrJn9Zvpt1BVWsK6hmyepCosOdFFY1sGhmLsPTdL7kUNbeYXj6swP88e2dGAP3LZ7CJaf4Zh5G1b/ctXAS+RX13PTMOp7+9ql+7aji1wYuEbkQuBdwAo8aY37f5fEI4B/ANKACuNIYk2899hPgeqAd+L4xZllf42lua+eBlXsYkR7LrGH29BrLiI/ksilZXDA+gzX5VWwqqubGp9cRE+5k7ph0Zg9LYWJWAqMz4/TCTQWtYCvbnVraOnjo/b3c/e4upg9J4r7Fk0O+B8lbW9yTMg5JiWFISgwLJmdRVNXIgYp6Smqa2H+4nuXbD9HxRZ0Q0eFOkmPC+WTvYQYlRJIWF0FaXASbi2qJi3QRF+Hi+jOG2pYa5khzG6U1jZTUNFFS00RpTRMlNY2sPVBFbWMbre0dPPrRPlwOIcLlJCkmnJSYcJKiw0mJDSc9LoKM+Egy4iNJj4sgMTpM09z4QLCW655sPVjDTf9cR2FVI/cvnsK0IdobXH3B4RC+OTuPM0em8f/+tYVfv7aNv72/lxvOHMbCqdkkxegI2L6UeV/4fH8l3392PdWNLTx+7QxGZsT5atN9JiLctXASh2qbuO25jewvr+d754zUirQ+OlBRz8odZazcWc5n+ypobusgwuUgJzmaOSPjGJQQSWJ0OJFhDr4ycRBVDa0cqKhnb3k9q/Mrefyj/fzt/X1EhjmYNSyFs0alcdaoNIamxuh5QC+dqPz70mf7Krj5mfUYY3j02umarkr1Sm5KNC/cOJv/enodtyzZwLvby/jx5fZAPwAAIABJREFURWMG3ByJdh+z+8IYQ21jGwdrGimpaeRgdRPF1Y3kH65nv/XX3PZFqv4wp5Aa677eG5MZz9lj0hmcGMW4QfHkJEchIt1mQ1HBJSU2gvPGZXLO2Az2lB1hdX4lf/84n0c+3M/sYSlcPWsI548PzNw8geKP6+lAHq9PpLmtndc3lfDAyj3sLa/njJGp/ObSCQxJibErJBXkYiJcPH7tDBY+9AmL/vYZv7hkHFfNzPXLebvfGrhExAk8AJwHFAGrRWSpMWabx2rXA1XGmBEisgi4C7hSRMYBi4DxwGDgXREZZYxpP9l4OjoMv3p1G/sO1/Pkf8y0/SIoOtzFmaPSuP+qKXy8t4IHVuzh/Z3lvL6p5Og6yTHuity0uAgSo8NJjg4jKSacZOtvUEIUWYlRpMdFhHwlvgodwVa2wd0L+M0tJTz43l52HTrCJacM5q6Fk4gK73+NxGFOB0NTYxia6j6JuOrUXJpa29lTdoR/riqgsr6FqvoWKhta2FJcw7vbD31pbq9Od761g9gIF3GRrqP/4yLDiI1wER3uJDLMSUSYgwiXk0iP/5EuJ6vzKwlzOnA5hTCHg4smZtLeYWjrMLR3GJpa26lqaD0aR8WR5qMNWaU1TdQ1tx0TT2psOOEuB4nRYYRblV0dHYbG1naq6lvYV36EyvoWGlqO/amEOx2kx0ccbfiqamglPtJFfGQYcZEuFs3M1YawEwjGct0dYwxbD9byxCf5vLK+mNTYcJbcMEtTHake5aXG8PS3T+XTvRXc/c4ufvP6du56awfnjMng7LHujkU5yQMvrU1fynxfXrejw7CpuIbHPtrP65sOkpsczYvXnhaUc6tEhTv5x/Uz+enLW7hvxR5e3VTCTfNGcOGETGIjBlwSjF4zxpBf0cCGwirWF1Tz0e7D7DtcD8Cw1BiuOjWXuaPTOXVoMi+vKz7m+Z0NnjOHfrF/b2xp57P9Fby/s5wPdpXzq1fdP9ec5CjOGpXGnBFpTMiKJysxSo/3x+Fl+e8TYwybitxlfenGg+SlRPPotdMZkR48DdkqdKTHRfLcDbN4YOVeHli5h2VbSrlsShZfm5rFtCFJuPpRBXl37Dpm90ZjSzv7D9eTX1FPYWUDRVWNrNpfQVVDKzWNrbS0ffl61CGQHBPB5JwE5oxIZWia+/p2WGosy7cfOmYfXlnfwkd7Dvf4+trg5T99/WwdIozKiGNURhznjcvghbWF/POzAm56Zh3pcREsnJbNeeMymJydGNL1mv64nrae49fj9YlUN7Tw6d4KPth9mDe3lFDd0MqYzDgeunoqF4zP1PMtdULp8ZEsvWkOtzy3gZ+9soVnVhXwnbOGc86YdGJ8eE3lz6uzmcAeY8w+ABFZAiwAPAviAuCX1u0XgfvFXToWAEuMMc3AfhHZY23v05MJpLCygVuWrGddQTU3njWcs0alndQb8geX08FZo9IormpkweTBVDe0UlzdSFldE7WNbdQ2tZJ/uJ6GllrqW9q6ragOcwqZCZEMthq8BidGMSgxktgIl7uS2uVARGhp66C1vYPmtnYaWtpptP7WHKg6+tjwtFh3pbXTgcshuJwOwjrvW5XZLqfgdAgigkPcByyHuHu7CtZ9h/t/d+s4PJbh8Zh7/c77nuu7/7d3GDqMuwL96J8xdFi33Y/xpWWdcTgdcnS7LofgdFr/HYLL4bD+u+87HYIx0GG+eL2OzvtHX/PLr9N5u8NwdFsup8e2rc8szOP+yPTYUD0YBE3ZfmltES+tK2J9QTWNre0MTY3h0W9O59xxGSf1xvzJXyfdntud2KWC8KpTczHGUN/STlltE89+XkhdUyt1TW3kpUZzpKmNuqY26prb2FN2hOLqRppaO2ht66C1o4O2dnej1Yk8/OG+Hh9zihAT4SQ+KoyEqDAWTstmUEIkmQmRbCmuJSEqjPhIl9cXpi1tHdQ1tVLb1Hb0vWQnRVFW18yh2iZ2lx2hqKrhS/vKv3+SD7hz+afHuRvCEqLCiI8KO9qw19kYFhvhIqxzv+dyEO50HL3fuY8S3P/hi/1W5/KkmDDS4yK9ei9BJmjKdVt7B+sLq93fc6P7ey6qbmRv2RG2FNdSWttEhMvBN2cP4eZ5I0iJjTipN6wGltnDU5g9fDbbDtby0roi/r3hIG9tdY+QTYuLYHhaDENTY0mPiyA5JpykmHBiI5xEutyN/ROzEgl39asKtJMu88aYEx8YuvHMqgLueXcXZXXNxIQ7+c8zh3HzvBHEBXFq0QiXkz9eMYn5kwbxuze284MXNvKzVzYzKTuBcYPiGZQYRWa8+5zbs1OIw+P8rvOm+yzZ436X5QAG90fb+Qkf/d9l+fEeM7gbFzpvf/l5X6zb9bHuntO57S7/jj6ntb2DWutYXNfURlVDC4WVDRRWuecRrWtyd2iJDncyPS+Zb852p2XPSz25nr5R4U7mjU5n3mh3Dv+Cigbe31XG+7vKeXldMU9/5j4niotwMTozjsFWR8D0+AgSo8KJCncSHe4kyurEkxobMVB7HXtT/r1ijOHTfRVUN7RS1dDCodpm9pYfYW1+FaW1TcRGuLjxrOF87+wRPq3EUAOPy+nglnNHcvn0bO5fsYd/byjmuTWFxIQ7mZSdyNC0GPJSokmLi2De6PT+NldxwI/ZVfUtHKxppLXdHK0ramnroKGlncr6Zg4faaH8SDMHKurZV+7OLOIpISqMmHAnabERjEyPJcG6FkyMCiMhOpy4SBcOkW5T6odoHYnyQlpcBN+dO4LvnDmc93eV8dSnB3j4g308+N5eUmLCmZidwOiMOHKSo0mKDicxOozIMPf1eGJUOLkpQd0pzR/X03ixzV7ZfaiO+pb2o2W6pfN/WwfVDS1U1LdQcaSFgsqGo3VEADHhTuaNSeeK6TmcMSI1pBsjVeAlxYTz9+tm8Mr6Yu5bvpvvP7uecJeDyTmJ/PTisUzOSezza/jzLDMLKPS4XwSc2tM6xpg2EakBUqzln3V57kknW06MDqOxtYM/XnEKC6cGV85mz8ppESHJqlSB7nuytncYGlraONLcRk1jKyMz4jhY3cjB6kY2FFazs7SO2qZWvKiTPsopQpjL3YhVWNVAW7uhtb2DptaOow03J3VWpI5r7+8uxhmax4SgKduFVQ3UNLZy5Ywczh+XwezhKXpC3IWIEBvhIjYt9uioLzh2fq6eGuA6jOHyadk0t3bQ1NZOU2s7L68rprXd3QDW2tHBmSPTjjbiuhwOwl0OkqPDSYoJY+mGg1/6Tjxft7659wN8wl0OUmIjvtSo0d176a4hrK6plZTYCMrqmig/0sy+w/XUNbVR29jqVUOeN749Zyj/LzTnlgiact3abrjioS+3jYVbIxen5yVx1qg0zhmbQbKmmFMnYdzgeMYNHsfPLh7L7rIjfLr3MFsO1rKv/AhvbSmhqqG12+d9/rNzQrXxuid9KfM9d6E+joSoMGbkJXP2mHTOHZtBQnTwNmx5EhHmjUnnrFFprC+s4o3NpawvqOKldcUc6WZE8kAW7nKQnRRFTlI0k3MSmTA4gcm5iYxMj8Pph4qQ3JRorpmdxzWz82hua2dLcQ3bS+rYWVrHzkN1bCispqyuqdsOggCXTcni7isn+zyuEOBN+feKiPDtJ9ccHWEvAjlJ0UwbksQZI1O5eNIgnR9T+VRWYhR3fm0iP58/lpU7yvl032G2FNfy5uYvjuHLbj2zvzVwBfyY/dqmg/z831t7fFwEkqLDGZISzezhKQxLdXcUykuNJic5mvjIMB1VpXrkdAhnj8ng7DEZ1DS08t6uMt7fWc720jo+2VNBS/uxx+0zR6Xxj/+Y2c3Wgoa/rqd9crzu9K0nVlNU1XjcdRKjw8hKjGJ6XhJXZeQya1gyk7IT+1U6SRV4Todw+bRsLpuSxer8St7Zdog1B6qI9lHmrZDuRiUiNwA3WHePiMjO463/Vu82n8pJngwESDDHF8yxQZDE57qr28WdsQ0JaDBBprdl+w3gV/4L5+jv5Rv+ew2/OF683TzWY7m4xjfh9PS6gd6mX8v/z++Cnx9/FS3bvSjbnnYDb+PO0eADQXEcOIFQiBGCIE4v9wF9ijOj+2O2pwFbtntbrv/q/5CO+a5D7fjtBdvLXU92+2g73wjAe7wHuGfRcVcZsOUaTv6YnQ98iF/LetD+/r0UcvF32Yf2OX5/75PH9HzMHvDX2n05F+8qH1jfx3g8fgshVy6OY8C9F2/LdF/K/lPAU9cfd5UBW67Bt2X7ALCxb+H0pzLQnX79/gJxDu5p6c0nXMWrsu3PBq5iIMfjfra1rLt1ikTEhXvYUoWXz8UY8zDwsA9jPkpE1hhjpvtj274QzPEFc2wQ3PEFc2weQrps91aIfCd9pu9zwPN7uYbgKduh8DsIhRhB4wxhfSnzXxIs5brTQPiu9T2qPgq583FPof7b0PjtE8Kx99tjdqcQ/m6Ooe9lwPLX9XRIXWdD///d6PsLTv4cX7gaGCkiQ0UkHPeEeUu7rLMUuNa6fTmwwsoRvBRYJCIRIjIUGAl87sdYlVLe07KtVP+j5VqpgaUvZV4pFdq8Kf9KqeChx2ylgp8/rqf1eK2Ul/w2gsvKJ3ozsAxwAo8bY7aKyB3AGmPMUuAx4ClrAr1K3IUVa73ncU+c1wbcZIzp/WQtSimf07KtVP+j5VqpgaUvZV4pFdp6Kv82h6WU6oEes5UKfv66ntbjtVLeEe3U0T0RucEa4hmUgjm+YI4Ngju+YI5toBoo34m+TzWQhMLvIBRiBI1TBZ+B8F3re1QDWaj/NjR++4Ry7P1df/pu9L2oga6//270/QUnbeBSSimllFJKKaWUUkoppZRSIcWfc3AppZRSSimllFJKKaWUUkop5XPawNUNEblQRHaKyB4R+bHNseSIyEoR2SYiW0XkFmv5L0WkWEQ2WH8X2xhjvohstuJYYy1LFpF3RGS39T/JhrhGe3w+G0SkVkRutfOzE5HHRaRMRLZ4LOv2sxK3+6zf4SYRmRqoOFVw7Qd8LVjLrC9oGVPHE6zluje/Wzsd55wkqGIVkUgR+VxENlpx/spaPlREVlnf/3PinqxZhagTlWdxT9T9nPX4KhHJC3yUfePFe7xORMo9zmm/bUecJ6u7fV+Xx/U4PYB5e8wWkYUiYkRkeiDjOxFv4heRr3scU58JdIw98WLfk2udD6y3yqZtdRHd0X1LaBERp/Vbes3uWPpCRBJF5EUR2SEi20Vktt0xnSwRuc3aL20RkWdFJNLumFTwC9Zr7b6SflZ/1pu6h1A6XmoDVxci4gQeAC4CxgGLRWScjSG1AbcbY8YBs4CbPOK52xgz2fp7w74QAZhnxdF5YfFjYLkxZiSw3LofUMaYnZ2fDzANaABesR6267N7Ariwy7KePquLgJHW3w3AgwGKccALwv2APwRdmfWRJ9AyproR5OX6Cbz/3dqpp3OSYIu1GTjbGHMKMBm4UERmAXfhPv6PAKqA622MUfWBl+X5eqDK+r7vxv39h4xe7LOe8zinfTSgQfbdExy77/Okx+kBytvfv4jEAbcAqwIb4fF5E7+IjAR+ApxujBkP3BrwQLvh5Wf//4DnjTFTgEXAXwMb5Qk9ge5bQsktwHa7g/CBe4G3jDFjgFMI0fckIlnA94HpxpgJgBN3OVeqR0F+re0L/an+7An6YZ2ZNnAdayawxxizzxjTAiwBFtgVjDGmxBizzrpdh/sgmWVXPL2wAHjSuv0kcKmNsQCcA+w1xhywMwhjzAdAZZfFPX1WC4B/GLfPgEQRGRSYSAe8oNoPBEiwldmTomVMHUfQlute/m5tc5xzkqCK1SrTR6y7YdafAc4GXrSW2x6n6hNvyrPn7/JF4BwRkQDG2FdBu8/ylR72fZ70OD1wefv7/zXuxuumQAbnBW/i/0/gAWNMFYAxpizAMfbEm9gNEG/dTgAOBjC+E9J9S+gQkWzgK0CoddD4EhFJAM4EHgMwxrQYY6rtjapPXECUiLiAaIKsjKug1O/PW7sIquvf3uivdWbawHWsLKDQ434RQdKgJO7UKlP4oofazdYQwcdtHg5pgLdFZK2I3GAtyzDGlFi3S4EMe0I7ahHwrMf9YPnsoOfPKmh/iwNAf//sQ6HM+pKWMQWh930HdZnsck4SdLFa6W42AGXAO8BeoNoY02atEuzfvzo+b8rz0XWs770GSAlIdL7h7T5roXVO+6KI5AQmtIAJtf228p0TfvdWmpwcY8zrgQzMS978dkcBo0TkYxH5TESON+IokLyJ/ZfA1SJSBLwBfC8wofmM7luCxz3Aj4AOuwPpo6FAOfB3K93ioyISY3dQJ8MYUwz8ESgASoAaY8zb9kalQkB/3q8OhPqzkK8z0wauECEiscBLwK3GmFrcwwKH4069UwL8ycbw5hhjpuIeuniTiJzp+aAxxuDeIdhC3HNsXAK8YC0Kps/uS+z+rNSAEdRl1p/683tT/Vew/W67OSc5KlhiNca0WymKs3H3KBxjc0hK+cOrQJ4xZhLuhtwnT7C+Uv2CiDiAPwO32x1LH7hwp/yZCywGHhGRRFsj8t5i4AljTDZwMfCU9Z0o5TURmQ+UGWPW2h2LD7iAqcCDVurOekIrZdlRVgfwBbgb7QYDMSJytb1RKWWrAVV/FqrvR09CjlUMePZ+zLaW2UZEwnBXJP3TGPMygDHmkFV50wE8grvyxhZWD4/OtAqvWLEc6hy2aP23M+XCRcA6Y8whCK7PztLTZxV0v8UBpF9/9iFQZn1Ny5iC0Pu+g7JMdndOQpDGCmClh1kJzMadUsFlPRTs3786Pm/K89F1rO89AagISHS+ccL3aIypMMY0W3cfxT3nbH8Savtt5Tsn+u7jgAnAeyKSj3teyKUiMp3g4M1vtwhYaoxpNcbsB3bhbvCymzexXw88D2CM+RSIBFIDEp1v6L4lOJwOXGKV4SXA2SLytL0hnbQioMgY05lt6UXcDV6h6FxgvzGm3BjTCrwMnGZzTCr49dv96gCpPwv5OjNt4DrWamCkiAy1Rv4sApbaFYw1V8BjwHZjzJ89lnvmvLwM2BLo2Kw4YqzJfbGGYJ9vxbIUuNZa7Vrg33bEZ1mMR3rCYPnsPPT0WS0Fvilus3APDS/pbgPK54JqP+BLIVJmfU3LmILQK9dBVyZ7OichyGIVkbTOXvAiEgWch3u+sJXA5dZqtsep+sSb8uz5u7wcWGH1SAwVJ3yPXc5pLyFEJ7Q/Dj1OD1zH/f0bY2qMManGmDxjTB7wGXCJMWaNPeEew5t91L9wj95CRFJxpyzcF8gge+BN7AW457hGRMbibuAqD2iUfaP7liBgjPmJMSbbKsOLcB+nQ3KkkDGmFCgUkdHWonOAbTaG1BcFwCwRibbO/c+h/51fKN8LtWttrwyg+rOQrzNznXiVgcUY0yYiNwPLACfwuDFmq40hnQ5cA2y25pIA+CmwWEQm4x42mA98x57wyABecR/3cAHPGGPeEpHVwPMicj1wAPi6HcFZO6Dz+PLn8we7PjsReRb3hUyqlbP8F8Dv6f6zegN3yoc9QAPwrUDFOdAF4X7Al4K6zPaVljHVk2Au17383dqpp3OSYIt1EPCkiDhxd+Z63hjzmohsA5aIyG+A9VgTgavQ01N5FpE7gDXGmKW4v9+nRGQP7omUF9kXce95+R6/LyKXAG243+N1tgV8EnrY94UBGGMeQo/TA5aXv/+g5WX8y4DzrWNTO/BDY4zto0y9jP123CkVb8N9TX1dMHUg0H2Lssn3gH9alfv7CNHflTFmlYi8CKzDfX6xHnjY3qhUsAvma+0+6nf1Z/21zkyC6DxEKaWUUkoppZRSSimllFJKqRPSFIVKKaWUUkoppZRSSimllFIqpGgDl1JKKaWUUkoppZRSSimllAop2sCllFJKKaWUUkoppZRSSimlQoo2cCmllFJKKaWUUkoppZRSSqmQog1cSimllFJKKaWUUkoppZRSKqRoA5dCRDJFZImI7BWRtSLyhoiMEpFGEdkgIttE5B8iEmatP1dEXrNuXyciRkTO9djepdayy+16T0qpY4nIZVaZ9vzrEJH/ssrs9zzWvV9ErrMxXKWURUSOWP/zjldWReQJEdkvIhtFZJd17M7uuh2P+9eJyP3W7dEi8p61X9guIg8H5M0ppY53Lr6ly3q/FJEfeNx3iUi5iPy+y3rzRWS9tS/YJiLfCdR7UUodyzp2/8nj/g9E5Jce928QkR3W3+ciMsda/t8i8rjHet8QkdcDGrxS6oREpN06h94iIq+KSKK1vPPc/Tce66aKSGvnObhSKvh41GuP8Vg207pe3i0i60TkdRGZaD32SxEp7lLXlmjfOxh4tIFrgBMRAV4B3jPGDDfGTAN+AmQAe40xk4GJQDbw9R42sxlY5HF/MbDRf1ErpU6GMeYVY8zkzj/gr8CHwDKgDLhFRMJtDVIpdSInKqs/NMacAowG1gMrvCzX9wF3W/uHscBffBOuUup4TnAufiLnAbuAK6ztYHVIexj4qrUvmAK854/YlVJeawa+JiKpXR8QkfnAd4A5xpgxwI3AMyKSifvYPFVETrcqyn4DfK/rNpRStmu0zqEnAJXATR6P7Qe+4nH/CmBrIINTSvXaYuAj6z8ikgE8D/zUGDPSGDMVuBMY7vGcuz3r24wx1QGPegDTBi41D2g1xjzUucAYsxEo9LjfDnwOZPWwjQ+BmSISJiKxwAhgg/9CVkr1lYiMAv4XuAboAMqB5cC1dsallDohr8qqcbsbKAUu8mK7g4Aij+dv7kuQSimvnfBc/DgWA/cCBcBsa1kc4AIqrG01G2N2+jRipVRvteFueL6tm8f+B3fnlMMAxph1wJPATcaYNuC7wAPAH4DHjTH7AhOyUuokfcqX684agO0iMt26fyXuinKlVBCy6rXnANfzxWCOm4EnjTGfdK5njPnIGPMvG0JU3dAGLjUBWHu8FUQkEjgVeKuHVQzwLnABsABY6ssAlVK+ZfXufga43RhT4PHQXcAPRMRpT2RKKS/1pqyuA8accC24G/dorzdF5DZNqaBUwBzvXHy4Z6oT3CM7gKPn5+cCrwLPYvUwNcZU4j4XPyAiz1opzfSaTyn7PQB8Q0QSuiwfz7H7gDXWcqzKtO24y/sf/B2kUurkWefm53BsndgSYJGI5ADtwMFAx6aU8toC4C1jzC6gQkSm4T4mrzvB827zOG9f6fco1ZfoxY46nuHWxfQhoMQYs+k46y7B3bK9CPdFtlIqeP0a2GqMec5zodUjdBVwlS1RKaW80suyKifanLXNvwNjgReAucBnIhLRhzCVUn23t0tq4Yc8HpsPrDTGNAIvAZd2NnobY76Nu4Ltc+AHwOMopWxljKkF/gF8vzfPs3qSTwfCgDQ/hKaU6rsoq+6sFHeK4Xe6PP4W7rTCi4DnUEoFs8W467ix/i/uuoKIrLLmrb7XY7FnisJ5gQhUfUEbuNRWYFoPj3XOwTUcmCYil/S0EWPM57jn6kq1WrmVUkFIROYCC3EPse7O73CnSjlRpbhSyl7eltUpuHt+AzR2mY8rGTjceccYc9AY87gxZgHudEoTfBivUqp7xzsXP57FwLkiko979EcKcHbng8aYzVaa0vNwH/eVUva7B3fKoxiPZds4dh8wjS/m6PkV8DTwW9yjrZVSwafRqjsbgvvc3HMOLowxLbiP1bcDLwY+PKWUN0QkGff59KPWOfYPga/jPiZP7VzPGHMq8HOg66hsZRNt4FIrgAgRuaFzgYhMAnI671v5wH+Me8Lr4/kx8FN/BKmU6jsRSQL+DnzTGFPX3TrGmB24L7S/GsjYlFK9c6KyKm7fxz23VmeK4feBq63Ho3CfrK+07l9opS/Fmtg+BSj253tQSgFenIt3JSLxwBlArjEmzxiTh7sybbGIxFqdWTpNBg74I3ClVO9YKUSfx93I1ekPwF0ikgIgIpOB64C/ishE4Cu4UxM/DOSJyHkBDVop5TVjTAPuUZq3i4iry8N/Av7H2g8opYLT5cBTxpgh1jl2DrAf96jM60TkNI91o22JUHVLG7gGOGOMAS7D3QN0r4hsBe7EPbTa07+AaBE54zjbetMYo3lGlQpeNwLpwINd5vS4sst6vwWyAx6dUqq3uiur/yciG4FdwAxgntVrFOAW4GtWuf8MeMEY84H12PnAFuu5y3BPeN/1XEAp5WO9OBf3dBmwwhjT7LHs37gbvJ3Aj0Rkp1XWf4W7slwpFRz+BKR23jHGLMWdRvQTEdkBPIK7M0op8CBwmzGmyRjTAfwXcG+X0dhKqSBijFkPbKJLWjNjzFZjzJP2RKWU8tJi4JUuy16yll8J3Ckie0TkE9yNYfd7rOc5B9cGEckLRMDKTdzXVEoppZRSSimllFJKKaWUUkqFBh3BpZRSSimllFJKKaWUUkoppUKKNnAppZRSSimllFJKKaWUUkqpkKINXEoppZRSSimllFJKKaWUUiqkaAOXUkoppZRSSimllFJKKaWUCinawKWUUkoppZRSSimllFJKKaVCijZwKaWUUkoppZRSSimllFJKqZCiDVxKKaWUUkoppZRSSimllFIqpGgDl1JKKaWUUkoppZRSSimllAop2sCllFJKKaWUUkoppZRSSimlQoo2cCmllFJKKaWUUkoppZRSSqmQog1cSimllFJKKaWUUkoppZRSKqRoA5dSSimllFJKKaWUUkoppZQKKdrApZRSSvVDInKhiOwUkT0i8uNuHo8Qkeesx1eJSJ7HYz+xlu8UkQu6PM8pIutF5DWPZUOtbeyxthnuz/emlFJKKaWUUkoppZQ2cCmllFL9jIg4gQeAi4BxwGIRGddlteuBKmPMCOBu4C7rueOARcB44ELgr9b2Ot0CbO+yrbuAu61tVVnbVkoppZRSSimllFLKb7SBSymllOp/ZgJ7jDH7jDEtwBJgQZd1FgBPWrdfBM4REbGWLzGpeTPvAAAgAElEQVTGNBtj9gN7rO0hItnAV4BHOzdiPedsaxtY27zUL+9KKaWUUkoppZRSSimLy+4AfCU1NdXk5eXZHYZSPrF27drDxpg0u+MIBlq2VX8SwLKdBRR63C8CTu1pHWNMm4jUACnW8s+6PDfLun0P8CMgzuPxFKDaGNPWzfo90rKt+hM9brtpuVb9iZbrL2jZVv2Jlm03LdeqP9Fy/QUt26o/8bZs95sGrry8PNasWWN3GEr5hIgcsDuGYKFlW/UnoVy2RWQ+UGaMWSsic09yGzcANwDk5uZq2Vb9RiiXbV/SY7bqT7Rcf0HLtupPtGy7ablW/YmW6y9o2Vb9ibdlW1MUKqWUUv1PMfD/2bv3KKnO887336fv9ysNdAPNHSyQZJAx8hkndo58w5MEnDnSGNnOkbN0jsZrSSuepcyJ5UmW4lFGc2Jnjp3MiZxYiZzjRJaxIufCxNiKYsmRk0gILGEQIKSmQdAXoO/3a/Vz/qhdqNTqhuru2nXp/n3W6sWuvd+96ynW2rV37ed9n3dN3OvVwboZ25hZHlAJdF1j3/cDe83sPNGSh7eZ2ePBPlXBMWZ7LwDc/VF33+Xuu+rq1MFORERERERERETmTwkuERGRxecIsNnM1ptZAbAfODitzUHgrmD5duBZd/dg/X4zKzSz9cBm4CV3/6K7r3b3dcHxnnX3zwT7PBccg+CYfxfmhxMREREREREREVGCS0REZJEJ5sO6D3gaOA086e4nzewhM9sbNHsMqDWzJuB+4IFg35PAk8Ap4IfAve4euc5bfgG4PzhWbXBsERERERERERGR0CyaObhERETkLe5+CDg0bd2DccujwB2z7Psw8PA1jv1j4Mdxr5uB3QsKWEREREREREREZA40gktERERERERERERERESyihJcIiIiIiIiIiIiIiIiklWU4BIREREREREREREREZGsogSXiIiIiIiIiIiIiIiIZBUluERERERERERERNLIzPaY2RkzazKzB2bYXmhm3w22HzazdXHbbjazF8zspJmdMLOiVMYuIiKSLnlhHtzM9gB/COQCf+buvzdt++eAe4EIMAjc4+6ngm1fBO4Otv26uz8dZqxheOLwhWtu/9StjSmKRETCdL1zfS70vSCS3ZL1faDvAhHJdPqtIyJzEfvO0HfDzMwsF3gE+AjQAhwxs4OxZ2SBu4Eed99kZvuBLwOfNLM84HHgV939Z2ZWC0yk+CMsmK4rIpIsM32f6Dtk8QptBFfcxfnjwDbgTjPbNq3ZE+5+k7vvAL4CfDXYdxuwH9gO7AG+HhxPRERERERERERkMdkNNLl7s7uPAweAfdPa7AO+FSw/BXzIzAz4KHDc3X8G4O5d7h5JUdwiIiJpFWaJwutenN29P+5lKeDB8j7ggLuPufs5oCk4noiIiIiIiIiIyGKyCrgY97olWDdjG3efBPqAWmAL4Gb2tJm9bGa/OdMbmNk9ZnbUzI52dHQk/QOIiIikQ5glCme6ON86vZGZ3QvcDxQAt8Xt++K0fadf2DGze4B7ABobNcxQRERERERERESWlDzg54D3AsPAj8zsp+7+o/hG7v4o8CjArl27/B1HyXCJlAJXCTIRkaUnzBFcCXH3R9x9I/AF4LfnuO+j7r7L3XfV1dWFE6CIiIiIiIiIiEh4WoE1ca9XB+tmbBPMu1UJdBHtFP68u3e6+zBwCLgl9IjTYHQiwndeusCDf/cq/+NHb3C2YzDdIYmISJqFmeBK5OIc7wDwiXnuKyIiIiIiIiIiko2OAJvNbL2ZFRCdl/7gtDYHgbuC5duBZ93dgaeBm8ysJEh8fRA4laK4U2bKnW+9cJ6TbX3c0ljN5NQUf/4v5zhzqf+6+4qIyOIVZoLruhdnM9sc9/IXgTeC5YPAfjMrNLP1wGbgpRBjFREREREREcl4ZrbHzM6YWZOZPTDD9kIz+26w/bCZrQvWf9rMjsX9TZnZjlTHLyLvFMypdR/RZNVp4El3P2lmD5nZ3qDZY0CtmTURnerjgWDfHuCrRJ/DHQNedvfvp/ozhO3wuW7e7Brm392ymk/sXMW9v7CJlZVFHDhykd7h8XSHJyIiaRLaHFzuPmlmsYtzLvDN2MUZOOruB4H7zOzDwATQQ9ATJWj3JNEeJ5PAve4eCStWERERERERkUxnZrnAI8BHiJYlO2JmB909frTG3UCPu28ys/3Al4FPuvu3gW8Hx7kJ+Ft3P5baTyAis3H3Q0TLC8avezBueRS4Y5Z9HwceDzXANIpMOc+/3sG62hJ2rqkCoDA/lzvf28j/ePYN/v54O59539o0RykiIukQWoILEro4f/4a+z4MPBxedCIiIiIiIiJZZTfQ5O7NAGZ2ANjH28uR7QO+FCw/BfyRmVlQyizmTqLTBIiIZLxT7f30jUyw990NmNnV9bVlhXxwSx3/ePoKLT3DaYxQRETSJdQEl4iIiIiIiIgkzSrgYtzrFuDW2doElVX6gFqgM67NJ4kmwmZkZvcA9wA0NjYuPGoRkQU43tJLeVEeW1eWv2Pbv9m4jH8928Uzpy6zurrkmsf51K36PhMRWWzCnINLRERERERERDKImd0KDLv7q7O1cfdH3X2Xu++qq6tLYXQiIm83PjnF65cH2N5QSU7c6K2YovxcPriljjeuDPJm11AaIhQRkXRSgktEREREZJEysz1mdsbMmszsgWu0+9/MzM1sV9y6Lwb7nTGzj6UmYhG5jlZgTdzr1cG6GduYWR5QCXTFbd8PfCfEGEVEkuZsxyATEWd7Q8WsbW5dX0tRfg7/erZr1jYiIrI4KcElIiIiIrIImVku8AjwcWAbcKeZbZuhXTnweeBw3LptRB+Cbwf2AF8Pjici6XUE2Gxm682sgOh5enBam4PAXcHy7cCzsfm3zCwH+Pdo/i0RyRLNHYPk5Rhra2YvP1iQl8OutTWcbOujf3QihdGJiEi6KcElIiIiIrI47Qaa3L3Z3ceJPtCeac6d3wW+DIzGrdsHHHD3MXc/BzQFxxORNHL3SeA+4GngNPCku580s4fMbG/Q7DGg1syagPuB+NGbHwAuuntzKuMWEZmvsx1DrKstJS/32o8wb11fw5TDkXPdKYpMREQygRJcIiIiIiKL0yrgYtzrlmDdVWZ2C7DG3b8/132D/e8xs6NmdrSjoyM5UYvINbn7IXff4u4b3f3hYN2D7n4wWB519zvcfZO7745PZrn7j939femKXURkLobHJ7nUP8r6utLrtq0tK2TLijKOnO8mMuUpiE7k+q5XLtzMCs3su8H2w2a2Llj/aTM7Fvc3ZWY7Uh2/SDZQgktEREREZAkKSpV9FfiN+R7D3R91913uvquuri55wYmIiMiS19IzAkDjNcoTxtu1tob+0UnOdgyGGZZIQhIsF3430OPum4CvEa2qgLt/2913uPsO4FeBc+5+LHXRi2QPJbhEREQWofn2FAu2fTFYf8bMPhasKzKzl8zsZ2Z20sz+S1z7/8/MzsX1LlPPMpHM0AqsiXu9OlgXUw7cCPzYzM4D7wMOmtmuBPYVERERCVVLzzAGrKoqTqj9u1aWU5yfy7GLveEGJpKYRMqF7wO+FSw/BXzIzGxamzvR3Jkis1KCS0TmJIGH5h8ws5fNbNLMbo9bv8PMXggejB83s0+mNnKRpWMhPcWCdvuB7cAe4OvB8caA29z93cAOYI+ZxZc4+r9iPczUs0wkYxwBNpvZejMrIHpuH4xtdPc+d1/m7uvcfR3wIrDX3Y8G7fYHyfD1wGbgpdR/BBEREVmqWnpGWFZeSFF+bkLt83JzuGl1JSfb+hibiIQcnch1JVLy+2qbYJ7NPqB2WptPAt+Z7U1UMlyWOiW4RCRhCT40vwB8Fnhi2vph4H9399hD8z8ws6pwIxZZshbSU2wfcMDdx9z9HNAE7PaoWK2P/OBPxe1FMljwI/k+4GngNPCku580s4fMbO919j0JPAmcAn4I3OvuelIkIiIiKdPaO8LqBEdvxexcU8VExDnZ1h9SVCKpY2a3AsPu/upsbVQyXJY6JbhEZC6u+9Dc3c+7+3Fgatr61939jWC5DbgC6MorEo6F9BSbdV8zyzWzY0TP32fc/XBcu4eD0ZlfM7PCmYJaSj3LLvWN8vTJSxxv6cVdeUBJH3c/5O5b3H2juz8crHvQ3Q/O0PYXgtFbsdcPB/ttdfcfpDJuERERWdqGxiYZGJ1kZWXRnPZrrCmhprSAly/2hBSZSMISKfl9tY2Z5QGVQFfc9v1cY/SWiCjBJSJzk8hD8+sys91AAXA2SXGJSAq4eySY5HY1sNvMbgw2fRF4F/BeoAb4wiz7L4meZe19I3zj+bM8/3oHB45c5EevXUl3SCIiIiIiWeVS/yjAnBNcZsbONVWc6xiid3g8jNBEEnXNcuGBg8BdwfLtwLMe9JA0sxzg36P5t0SuSQkuEUkpM6sH/hL4NXefmqXNkhnlIRKShfQUu+6+7t4LPEe03Cju3h6UMBwD/pzoaM8lyd35u2NtFOTm8J8+tpUda6p47rUrtPWOpDs0EREREZGscTmW4KqYW4ILYMeaKhz42cXeJEclkrgEy4U/BtSaWRNwPxA/1/0HgIvu3pzKuEWyjRJcIjIXiTw0n5WZVQDfB37L3V+crd1SGeUhEqKF9BQ7COw3s0IzWw9sBl4ys7rYvHlmVgx8BHgteF0f/GvAJ4BZ64Mvdk0dg1zoHua2G5ZTXVLAL9/cQEFeDs+/oWS9iIiIiEiiLvWNUlKQS1lh3pz3rS0rZE11McdalOCS9LpeuXB3H3X3O9x9k7vvjk9mufuP3f196YpdJFsowSUic5HIQ/MZBe3/BvgLd38qxBhFlryF9BRz95PAk8Ap4IfAve4eAeqB58zsONHvgmfc/e+DY33bzE4AJ4BlwH9NxefMRD99s4fi/Fzes7YagOKCXHavq+FESx/9IxNpjk5EREREJDtcGRhjRUUR0T50c7djTRWX+8e41Dea5MhERCSTzL0bhIgsWe4+aWaxh+a5wDdjD82Bo+5+0MzeSzSRVQ38spn9F3ffTrRu8AeIPlD/bHDIz7r7sdR/EpHFz90PAYemrXswbnkUuGOWfR8GHp627jiwc5b2ty003sVgYHSCU239vGdtNXk5b/Uhes+6an7S1MmJ1j7ev2lZGiMUEREREckOnYNjbG+onPf+N62u4vsn2jl2sZc9lSuTGJmIiGQSJbhEZE4SeGh+hGjpwun7PQ48HnqAIiJp8s9vdDI55dy8uupt65eXF1FfWcTxll4luERERERErmN4bJLh8Qh1ZQXzPkZZYR6blpfxs5ZePrp9BTnzHAkmIiKZTSUKRURERJLg+Tc6KMzLobGm5B3btjdUcrFnhIFRlSkUEREREbmWzsExAJaVFy7oODvWVNE3MsGbXcPJCEtERDKQElwiIiIiC+Tu/NOZDjYtLyM35529Q7euKAeg6cpgqkMTEREREckqHYPjACwrW1iC64b6CvJzjWMXe5MRloiIZCAluEREREQWqKVnhLa+UTbWlc24vb6qiNLCPF6/PJDiyEREREREskvn4Bg5BtUl8y9RCFCYl8u2+gpebe1jcmoqSdGJiEgmUYJLREREZIGOvtkNwNrad5YnBMgxY8OyUs6rPIqIiIiIyDX1DI9TVVIwY2WEudqxpoqRiQivX1IlBRGRxUgJLhEREZEFOnq+h/LCPFZUFM3aZm1tCX0jE/QOj6cwMhERERGR7NI7PEFVSX5SjrVpeTklBbkca1GZQhGRxUgJLhEREZEFevlCLzsaq8ix2XuZrq0tBdAk1yIiIiIi19AzPL7g8oQxuTnGzasrea29n4HRiaQcU0REMocSXCIiIiILMD45RdOVAW5cVXnNdisriijIzeHN7qEURSYiIiIikl0mIlMMjE5SnaQRXAA7VlcxOeX88NVLSTumiIhkBiW4RERERBbgjSsDTEScbfUV12yXm2OsqSnWCC4RERERkVn0DUdHWSVrBBfAmpoSakoLOPiztqQdU0REMoMSXCIiIiILcLp9AIAbrpPggmiZwkt9o4xNRMIOS0REREQk6/QE89VWJTHBZWa8e3Ul/9LUyZWB0aQdV0RE0k8JLhEREZEFONXWT1F+DuuXlV63bWNNCQ5c6NEoLhERERGR6XqujuBKXolCgHevqWLK4W9faU3qcUVEJL3y0h2AiIiISDY73d7P1pUV5ObYdduuqS4BoLVnhM3Ly8MOTURERESyhJntAf4QyAX+zN1/b9r2QuAvgPcAXcAn3f28ma0DTgNngqYvuvvnUhV3svUMj5NjUFGc3ATX8vIi3rO2mgNHLvJ//vwGzK5/7y4ii8cThy8k3PZTtzaGGIkkm0ZwiYiIiMyTu3P6Uj/b6hNLVhUX5FJTWkBr70jIkYmIyGJlZnvM7IyZNZnZAzNsLzSz7wbbDwcPv2PbbjazF8zspJmdMLOiVMYuIjMzs1zgEeDjwDbgTjPbNq3Z3UCPu28CvgZ8OW7bWXffEfxlbXILogmuqpICckJIQH3yvWto7hjiyPmepB9bRETSI9QEVwI33veb2SkzO25mPzKztXHbImZ2LPg7GGacIiIiIvPR3jdK7/BEQvNvxTRUFdPep9r/khoJ3I9/LnjIfczM/jn2MM3M1pnZSNz9+J+kPnoRmW4hD8HNLA94HPicu28HfgGYSFHoInJtu4Emd29293HgALBvWpt9wLeC5aeAD9kiHIbUMzROVZLLE8b80s31lBfmceBI4iM5REQks4WW4ErwxvsVYJe730z04vyVuG0jcb1P9oYVZyr0j05worWP4fHJdIciIiIiSXS6vR+AbXNJcFUW0T00zsh4JKywRICE78efcPeb3H0H0Xvxr8ZtWzS9wUUWkYU8BP8ocNzdfwbg7l3urouRSGZYBVyMe90SrJuxjbtPAn1AbbBtvZm9Ymb/ZGY/H3awYeodnqC6pCCUY5cU5LF3RwOHTrTTN6L8vojIYhDmCK7r3ni7+3PuHptl/UVgdYjxpMXw+CT/77NNfOelC/zZT84xOqHfDyIiIovFa5cGANi6MvH5tBqqigFo61OZQgldIvfj/XEvSwFPYXwiMncLeQi+BXAze9rMXjaz30xBvCISvnag0d13AvcDT5jZO3pfmdk9ZnbUzI52dHSkPMhETESmGBibpDqkEVwA+9/byOjEFH93rDW09xARkdQJM8GVyI13vLuBH8S9LgouvC+a2Sdm2iEbLs7PvnaF4bFJbnvXci71j3K4uSvdIYmIiEiSnO0YZEVFIeVFif8Iv5rg0jxcEr6E7sfN7F4zO0t0BNevx226bm/wbLgfF5Gr8oCfAz4d/PsrZvahmRrq3BZJuVZgTdzr1cG6GdsEJUcrgS53H3P3LgB3/ylwlmhC+23c/VF33+Xuu+rq6kL4CAvXNxwdVVUV0ggugJtWV7K9oYLvvHQRd/XrERHJdqHOwZUoM/sMsAv4/bjVa919F/Ap4A/MbOP0/TL94hyZco5d7OWm1ZV8+IYVbFpexgvNXUSmdAEVERFZDM51DrFhWdmc9ikrzKOyOF8JLskY7v6Iu28EvgD8drA6od7gmX4/LrIIzfshONEk9/Pu3hlUUjkE3DLTm+jcFkm5I8BmM1tvZgXAfmD6fPQHgbuC5duBZ93dzawuKEuMmW0ANgPNKYo7qfpGowmuyuLwRnAB3Lm7kdPt/Ry72Bvq+4iISPjyQjx2IjfemNmHgd8CPujuY7H17t4a/NtsZj8GdhLthZI13uwaYng8wo0NlQDcur6Gbx++wLnOITYtn9vDMBEREcks7k5zxxC/dHP9nPdtqCyirXc0hKgy1xOHFz6Z96dubUxCJEtKQvfjcQ4AfwwQ3JePBcs/DUZ4bQGOhhOqiCTo6kNwoufzfqKdQuPFHoK/wNsfgj8N/KaZlQDjwAeBr6UschGZlbtPmtl9wNNALvBNdz9pZg8BR939IPAY8Jdm1gR0Ez3/AT4APGRmE8AU8Dl37079p1i4/mBerIo5VEeYj307Gvhvh07zxOEL7GysDvW9REQkXGGO4Lpu7xMz2wl8A9jr7lfi1lebWWGwvAx4P3AqxFhDcebSAHk5xpYV0Xk5Ni8vJy/HeO1S/3X2FBERkUzXPTRO38gE65eVznnf+qpiOgfHGJ+cCiEykasSuR/fHPfyF4E3gvWLpje4yGISzKkVewh+Gngy9hDczPYGzR4DaoOH4PcDDwT79gBfJfrdcAx42d2/n+rPICIzc/dD7r7F3Te6+8PBugeD5BbuPurud7j7Jnff7e7Nwfrvuft2d9/h7re4+/9M5+dYiIHRSQAqisLsjw/lRfns27GK/3m8jb4gqSYSBjPbY2ZnzKzJzB6YYXuhmX032H7YzNbFbbvZzF4ws5NmdsLMilIZu0i2CO2KkWDvk98HyoC/MjOAC+6+F7gB+IaZTRFNwv2eu2ddgutC9zANVcUU5EXziAV5OWysK+O1SwP80s1pDk5EREQWpLlzCICNdXMfld1QWYQDl/pHaawpSXJkIlEJ3o/fF1RUmAB6eKv00aLpDS6y2Lj7IaLlBePXPRi3PArcMcu+jwOPhxqgiMg89Y9OUJiXQ2F+bujv9elbG/nOSxf4m5db+Oz714f+frL0BJ3FHgE+QrRM8BEzOzjtGffdQI+7bzKz/cCXgU8GJYYfB37V3X9mZrVE79dFZJpQu0QkcOP94Vn2+1fgpjBjC9tEZIrW3hFuXV/ztvWblpdx5vIAvcPjaYpMRESWAjPbA/wh0Yfaf+buvzdteyHwF8B7iM7L8Ul3Px9s+yLRG+0I8Ovu/nTQW+x5oJDo/cNT7v47Qfv1REub1QI/JXoTvugvdOc6ogmuDXVzH8G1srIYgPa+ESW4JFQJ3I9/fpb9vgd8L9zoRERERN7SPzJBeYjlCaeXzF5dXczXf3yW/NwczEzlsCXZdgNNsdGWZnYA2Mfbq5TtA74ULD8F/JFFR4F8FDju7j8DcPeuVAUtkm3CLFG4pL3WPsDklLNm2kOrdUEZo/Ndw+kIS0REloC4nmIfB7YBd5rZtmnNrvYUIzr/xpeDfbcRLWO2HdgDfD043hhwm7u/G9gB7DGz9wXH+jLwteBYPcGxF72znYPk5xqrq+eeoKouyacwL4dLfUtrHi4RERERkdn0j05SURxuecJ4u9fVcGVgjDf1jE7CsQq4GPe6JVg3Y5ugDHEf0Y6jWwA3s6fN7GUz+83Z3sTM7jGzo2Z2tKOjI6kfQCQbKMEVklfb+gDe8dCrvrKIwrwczgdljUREREJwtadYMJIq1lMs3j7gW8HyU8CHgp5i+4AD7j7m7ueAJmC3Rw0G7fODPw/2uS04BsExPxHWB8skzR1DrK0tJTfH5ryvmbGysoh2JbhERERERAAYGJ2gIsQRXNPdvLqKwrwcXjqvKsyScfKAnwM+Hfz7K2b2oZkauvuj7r7L3XfV1dWlMkaRjKAEV0jeuBzt1V1V8vYLc44ZjTUlXOhW7xAREQnNQnqKzbqvmeWa2THgCvCMux8O9ukNjjHbexHsv6h6ljV3DLJh2dzLE8bUVxZxqX+UKfckRiUiIiIikn3cPTqCqyh1I7gK8nLY2VjFq619DI9NXn8HkblpBdbEvV4drJuxTTDvViXRKQRagOfdvdPdh4mWHL8l9IgXKXfnp2/28INX27kyoE6mi40SXCF548oAy8uLyLF39upuqCrmysAoY5ORNEQmIiIyP+4ecfcdRG/Md5vZjXPcf9H0LJuMTHGhe5gNdWXzPkZ9RTHjk1P0DC366cpERERERK5peDxCZMqpKE7dCC6A3etqmZxyXr7Qk9L3lSXhCLDZzNabWQHRqQAOTmtzELgrWL4deNbdHXgauMnMSoLE1wd5+9xdMgf/cOoy33u5hX9p6uRPn2+mW7/BFxUluELSdGWQ5eWFM25rqCpmyuH1S4MzbhcREVmghfQUu+6+7t4LPEd0jq4uoCo4xmzvtei09Y4yEXHWL5v7/FsxKyuLAFSmUERERESWvP7RCQDKU1iiEKL35I01Jbx0vhtXZQVJoqDKyX1Ek1WngSfd/aSZPWRme4NmjwG1ZtYE3A88EOzbA3yVaJLsGPCyu38/1Z9hMWjvG+H51zu4pbGaz39oCxNTzj+cupTusCSJlOAKwcDoBO19o7MnuIIHWrF5ukSyiZntMbMzZtZkZg/MsP0DwQSYk2Z2+7Rtd5nZG8HfXdP3FZGkWUhPsYPAfjMrNLP1wGbgJTOrM7MqADMrBj4CvBbs81xwDIJj/l2Iny0jxEoNN9bMv0ThiooiDLjUrwSXiIiIiCxt/SPREoGpLFEYs3t9DZ2D47zQ3JXy95bFzd0PufsWd9/o7g8H6x5094PB8qi73+Hum9x9t7s3x+37uLtvd/cb3f030/UZst1P3uikIC+HX7ypnrryQt63vpYTLX10DY6lOzRJEiW4QnCucwiAulkSXDWlBRTl5/BqqxJckl3MLBd4BPg4sA2408y2TWt2Afgs8MS0fWuA3wFuBXYDv2Nm1WHHLLIULbCn2EngSaLlD34I3OvuEaAeeM7MjhNNoD3j7n8fHOsLwP3BsWqDYy9qVxNctfMfwVWQl0NtWaFGcImIiIjIkjcQjOBKdYlCgJtWVVKcn8u3D19I+XuLSHgGxyY50drHLY3VFBfkAvC/bKwF4GctvekMTZIo9d0iFpEnZrnwHQ9OkJrSmRNcZkZ9ZTGvtvWHFptISHYDTbEeJWZ2ANhHXB1gdz8fbJuatu/HiD4Q7w62P0O0vNl3wg9bZOlx90NEJ6KNX/dg3PIocMcs+z4MPDxt3XFg5yztm4l+PywZF7qHyc81VlYULeg49ZVFtPQMJykqEREREZHsdLVEYWHqH1Xm5+ZwS2MV/3DyEp2DYywrm/l5nohkl5NtfUSmnPeuq7m6rrI4n7W1JZxo7eO2d61IY3SSLBrBFYLYRHXVpcnatyEAACAASURBVLP3OllVVcxr7f1MRqbnAEQy2irgYtzrlmBdUvc1s3vM7KiZHe3o6JhXoJnuYvcwxy72MDoRSXcoIjIPF7uHWV1dQm6OLeg49ZVF9AxP6LtARERERJa0/tFJSgpyyctNz6PK966vYSLi/NXRlrS8v4gk38nWfpaVFbCi4u1J6xtXVXK5f+zqM3zJbkpwhaB7aJzSwjwK83JnbdNQVcTY5BRnO4ZSGJlIdnD3R919l7vvqqurS3c4SfdCcxd//E9nefJoC48+36wH2yJZ6EL3MGtq5l+eMGZlMC+nyhSKiIiIyFLWPzJBZRrKE8YsLy/i1vU1fOelC0xNedriEJHkGJ2I0Nw5yLb6Ssze3jF1Y10ZAM0dg+kITZJMCa4QdA+PU1Ny7YtyfWUxgObhkmzTCqyJe706WBf2votG1+AYPzjRzpYVZXxqdyNXBkb5h1OX0x2WiMzRhe5hGmuKF3yc2P3Apb6RBR9LRERERCRbDYxOUl6U3plUPv2+tVzoHuafmzrTGoeILNybXUNMOWxeUfaObcvLCykrzKO5UwNPFgMluELQMzROTWnBNdssKyukIC+HM5cHUhSVSFIcATab2XozKwD2AwcT3Pdp4KNmVm1m1cBHg3VLSuxG+d/tXM2Nqyq5pbGaI+e76RuZSHNkIpKovuEJ+kYmaEzCCK6KojxKCnI1gktERERElrSB0QnKCtM3ggvgY9tXUFmcz1+/rDKFItmuuXOI3BxjTfU7f7ebGRvqSjWCa5FQgivJIlNO7/AE1ddJcOXmGJvqyjhzSQkuyR7uPgncRzQxdRp40t1PmtlDZrYXwMzea2YtwB3AN8zsZLBvN/C7RJNkR4CHgnVLxthEhFcu9nLz6koqgtILH9xSR2TKOXahJ83RiUiiLvYMAyQlwWVmrKws4lK/ElwiIiIisjS5O0NjEcoK0zuCqzAvl39700r+4dRlhscn0xqLiCzMuc4h1lQXU5A3c/qjsaaE/tFJdThfBJTgSrLBsUkcEqobvHVluRJcknXc/ZC7b3H3je7+cLDuQXc/GCwfcffV7l7q7rXuvj1u32+6+6bg78/T9RnS5bVLA4xPTrFrbc3VdbVlhTTWlHCspTeNkYnIXFzsjia4kjEHF0B9RRGX+0eZctX6FxEREZGlp29kgog7ZWkuUQiwb8cqhscjPKOpBESy1uhEhNaeEdYve2d5wpjVVdHpAlqDDqySvZTgSrL+IOtbUZRYgutS/yh9w8oUiywFp9r7KS3Mo7H27Q/F3726ksv9Y3QOjKUpMhGZiwvJTnBVFjMRcToH9R0gIiIiIktP5+A4QNpHcAHsXldDQ2URf/vKkpsyXGTReLNrCAc21JXO2qa+qpgcg4s9mg872ynBlWSxYY0JjeBaUQ6gebhEloDIlPP65QFuWFlOjtnbtm1dWQHA61f0XSCSDS50D1Ndkp9QZ5ZErKwsAuCS5uESERERkSUo1tErExJcOTnGL+9o4Pk3OulSBzSRrHSt+bdi8nNzWFlRRFuvElzZTgmuJOsfDUZwJViiEJTgElkK2npHGJucYnOQ2I5XU1pAbWkBTVc0uaVINrjQPZy00VsAy8sLyTFoV4JLRERERJagqwmuDChRCPCJHauITDmHTrSnOxQRmYeL3SM0VBbNOv9WzIpgugDJbkpwJVn/yAS5ZpQU5F63bX1lEeVFeZy51J+CyEQknZo7hwBYv2zm4dGblpfR3DnEZGQqlWGJyDy09o6wuro4acfLy81heXmRRnBJKMxsj5mdMbMmM3tghu2fM7MTZnbMzP7ZzLbFbftisN8ZM/tYaiMXERGRpSJWrj8TRnAB3FBfwdYV5fztsbZ0hyIiczQ15bT3jdBQdf3f7CsqiugfnWRkPJKCyCQsmXHlWET6RycpL857RwmymZgZW1eU8/oljdoQWezOdQ6yvLxw1hv2tbWlHD7XzWuXBrhxVWWKoxORRLk7bb0j3LZ1eVKPu7KyiOYO3Q9IcplZLvAI8BGgBThiZgfd/VRcsyfc/U+C9nuBrwJ7gkTXfmA70AD8o5ltcXf9+hMREZGk6hwcJ8dIqLN4mJ44fOHqcmNtCc+cusyf/Pjs1SpNn7q1MV2hiUiCWnqiFZQaKq+f4FpeUQjAlYFR1tbOPl+XZDaN4EqyvpEJKucwJ8eWleW8dqkfdw8xKhFJJ3fnYvcIa2tnL2kW2/byhZ5UhSUi89AzPMHoxFRCvcHmor4y2nNsaGwyqceVJW830OTuze4+DhwA9sU3cPf4UgKlQOymdB9wwN3H3P0c0BQcT0RERCSpOgfHKC1IrLN4qmyvj86VfapdVZdEssmp9j4A6quKrtt2RUUwH7bKFGY1JbiSrH9kIqH5t2LetbKc/tFJLvdr4kqRxapneIKRicg1H4hXFedTUZTH0fNKcIlkstgEtMlOcK2sjN5Yax4uSbJVwMW41y3Burcxs3vN7CzwFeDX57KviIiIyEJ1Do5lzPxbMcsriqgrK+RkW1+6QxGROTjZ1k+OvZW8upaq4nwK8nK4oufyWU0JriRyd/pHJ6iYw0V5y4pyAF7TPFwii1Zr8EB81TUeiJsZq6tLeFU3zyIZLZHzeT7qg/IJl/pGknpckUS4+yPuvhH4AvDbc9nXzO4xs6NmdrSjoyOcAEVERGRR6xgcz5j5t+Jtb6jgXOcQw6qyIJI1TrX1s6yskPzc66c9zIzl5YVcHlBH02ymBFcSjU5MMRHxOY3g2hokuF6/PBBWWCKSZm29Iwn1HqmvLIrePI/r5lkkU701guv6vcHmoqwwj/KiPI3gkmRrBdbEvV4drJvNAeATc9nX3R91913uvquurm6B4YpIIsxsj5mdMbMmM3tghu2FZvbdYPthM1sXrF9nZiNmdiz4+5NUxy4iMpPOgbEMTXBVMuVw+pKe2Ylki1Pt/XOquLKivEiV1bKcElxJ1Dc6AUDlHBJc1aUFLC8v5DVdLEUWrbbeEVZUFF2390h9ZRHucEbfByIZq613hMK8HGpKC5J+7PrKItX+lmQ7Amw2s/VmVgDsBw7GNzCzzXEvfxF4I1g+COwPHpSvBzYDL6UgZhG5BjPLBR4BPg5sA+40s23Tmt0N9Lj7JuBrwJfjtp119x3B3+dSErSIyDW4e7REYQYmuBqqiqgqzleZQpEs0T00TnvfKPWViXdIXVFRyNCY5sPOZkpwJVH/SDTBVVGUeIILYOvKco3gElmk3J3W3pGEeo/ESpRpEluRzNXWO8qqqmIshAmwV1YUc6V/jMmpqaQfW5Ymd58E7gOeBk4DT7r7STN7yMz2Bs3uM7OTZnYMuB+4K9j3JPAkcAr4IXCvu0dS/iFEZLrdQJO7N7v7ONGRl/umtdkHfCtYfgr4kIVx4RIRSYLBsUnGJqcybg4uiJYv295QQdOVQcYmdBskkulOtUWfp8WeryVieVBtSWUKs1fmXT2y2NUE1xxGcEG0TOFfvvgmkSknN0e/O0QWk76RCYbHIwnN11NVkk95UR6nleASyViJJqzno76yiIg7HQMqjyDJ4+6HgEPT1j0Yt/z5a+z7MPBweNGJyDysAi7GvW4Bbp2tjbtPmlkfUBtsW29mrwD9wG+7+09mehMzuwe4B6CxsTF50YuITNM5OA6QkSO4ALY1VPIvZ7t4/cpgukMRkeuIjbZsmMMIruXlhQB0DIyxYVlZKHFJuDSCK4n6R2MjuOZ2Ud66spyxySne7BoKIywRSaO35uu5/gNxM+OG+oqrPU5EJPO09Y4kff6tmJXBTfglzcMlIiLhaAca3X0n0RGbT5hZxUwNNb+eiKRK52C0c1emJrgaa0oozs/ldU0lIJLxTrX301BZRMkcvk8qivPJyzG6g2S7ZB8luJKof2SSkoJc8q4zz850W1eWA5p3R2QxausbxYCVFYk9EN9WX8FrlwaYmvJwAxORORubjHBlYCy0EVzLygrJyzHaleASEZHZtQJr4l6vDtbN2MbM8oBKoMvdx9y9C8DdfwqcBbaEHrGIJMTM9pjZGTNrMrMHZtheaGbfDbYfNrN107Y3mtmgmf2nVMWcDJ1B9YJMLFEIkJtjbFpexutXBnDX73SZm/me12a2zsxGzOxY8PcnqY49G51q62dbw4x9d2aVY0Z1aQFdQ0pwZatQE1wJnMT3m9kpMztuZj8ys7Vx2+4yszeCv7vCjDNZBscmKZ/HBXnz8nLM4Izm4RJZdDoGxqguLaAgL7Gv2xvqyxkej/Bm93DIkYnIXF3ui/74DivBlZtjrKgo0gguERG5liPAZjNbb2YFwH7g4LQ2Bwnm0wNuB551dzezOjPLBTCzDcBmoDlFcYvINQTn5iPAx4FtwJ1mtm1as7uBHnffBHwN+PK07V8FfhB2rMmW6SO4ALasKGdgdJLT7XpuJ4lLwnl91t13BH+fS0nQWWxkPMLZjkG2NVTOed/a0gK6leDKWqEluBI8iV8Bdrn7zUQnv/1KsG8N8DtEa4nvBn7HzKrDijVZhsYmKS2Y+wW5uCCXtTUlGsElsgh1DIxRV1aYcPtt9dELsebhkoVYSO9PM/tisP6MmX0sWLfGzJ4LOqWcNLPPx7X/kpm1xvUs+7ep+Izp0BqUHE1kTr35WllZRHvfiHqHiojIjNx9ErgPeBo4DTzp7ifN7CEz2xs0ewyoNbMmoqUIY/cCHwCOm9kxor+/P+fu3an9BCIyi91Ak7s3u/s4cADYN63NPuBbwfJTwIfMzADM7BPAOeBkiuJNmtioiZJ5PE9Llc0rovPy/Pj1K2mORLLMgs5rmZszlweY8mhlpLmqLS2ga2hMv8OzVJhXj6snMYCZxU7iU7EG7v5cXPsXgc8Eyx8DnondbJvZM8Ae4DshxrtgQ+OT1FfO76HX1pXlGsElsshMudM5OMam5YlPUrlxeSkATZrAVuYproPJR4hOPH/EzA66+6m4Zld7iZnZfqK9xD4ZdETZD2wHGoB/NLMtwCTwG+7+spmVAz81s2fijvk1d//vqfmE6TOXOfXmq76yiJ++2UPHwBjLEyxtKiIiS4u7HwIOTVv3YNzyKHDHDPt9D/he6AGKyHysAi7GvW4h2ul7xjbuPmlmfUST2aPAF4je/89antDM7gHuAWhsbExe5AvUPTROVUk+uTmZ+0y/oiif+soi/upoC1XFBbO2+9StmfP/Khlh3ud1sG29mb0C9AO/7e4/melNMvXcTrXYfPbbGyr4yRudc9q3pqyQiYgzMDZJRVF+GOFJiMIsUTjTSbzqGu3v5q2h1Anta2b3mNlRMzva0dGxwHAXbmgsQmlh7rz23bqinPOdQ4xORJIclYikS+/wBJNTzvLyxEdwlRTksaqqmOYOJbhk3hbSS2wfcCCYo+Mc0ATsdvd2d38ZwN0HiPYYv9Y1fVGKJbjqK8NLPK0Mjn1KozhFREREJDFfItrh7Jo/It39UXff5e676urqUhNZArqGxqkpmT1plCm2rCjnzS49t5OUaQca3X0n0dHYT5jZjEOTMvXcTrWTbX2UF+WxunruHVJrS6PfQV2DKlOYjUKdgytRZvYZYBfw+3PZL5NO4MiUMzIRmfeQ6q0rK5hyjdoQWUyuDETn0ambQ4ILYENdKWc7hsIISZaGRDqJvK2XGBDrJXbdfYNyhjuBw3Gr7wvm0/zmtUoKZ1rHlLlq6xthWVkBRfnz68ySiPqK6M246vuLiIiILCmtwJq416uDdTO2MbM8oBLoIjoi5Ctmdh74j8B/NrP7wg44WboHx6kpzY4El57byRzN+7wOOp12Abj7T4GzwJbQI85ip9r72VZfwXwqPMYSXN1DY8kOS1IgzARXIicxZvZh4LeAve4+Npd9M8nw+CQApfOcFHPrymgJM83DJbJ4dAxEv9LmmuDaWFdGc8egav9KxjGzMqKljf6ju8eGGP0xsBHYQbSX2f8z2/6Z1DFlPlp7R0MtTwjReTmrS/J5tbUv1PcRERERkYxyBNhsZuvNrIBo2fCD09ocBO4Klm8HnvWon3f3de6+DvgD4L+5+x+lKvCF6hnOjgRXY00JhXk5vHFFz+0kYfM+r82sLph+ADPbAGwGmlMUd9aJTDmvtQ+wrWHu828BVJUUkGNvzQko2SXMBNd1T2Iz2wl8g2hyK36mxqeBj5pZddAT/KPBuow1NB4dolxaML9e3etqSynIzeF1zcMlsmhcGRijtDBvziM7N9SVMjQe4cqAeo7IvCyk9+es+5pZPtHk1rfd/a9jDdz9srtH3H0K+FOiJRIXpbbeERrmOdfmXKyuLuHYxd7Q30dEREREMkNQVeE+os++TgNPuvtJM3vIzPYGzR4jOudWE9GSZQ+kJ9rk6hoap7Ys8xNcuTnGxroy3tAILknQAs/rDwDHzewY0WkFPufu3an9BNnjXOcQIxMRtjdUzmv/3ByjqqRAJQqz1PyGGyUgmBgvdhLnAt+MncTAUXc/SLQkYRnwV8HwwQvuvtfdu83sd4kmyQAeyvSTeHhsYSO48nJz2Li8jNc0gktk0egYGJvT/FsxG5ZFR3SevTLIiorw5vqRRetqBxOiyan9wKemtYn1EnuBt/cSO0i0tvdXgQaivcReCubnegw47e5fjT+QmdW7e3vw8leAV0P6XGnl7rT1jvCBzeGPPGusKeFEax+X+0f1HSAiIiKyRLj7IeDQtHUPxi2PAndc5xhfCiW4kLg7PUPjVGfBHFwAG+tKOdXeH405C0adSfrN97x29+8R7WAqCYjNYb2tfn4juCBaprBbI7iyUmgJLkjoJP7wNfb9JvDN8KJLrrdGcM3/v/RdK8t5sbkrWSGJSBq5Ox0DY9y0au69RzYuLwXgbOcQ/2bTsmSHJotcgh1MHgP+Mugl1k00CUbQ7kngFDAJ3OvuETP7OeBXgRNBDzKA/xxc579iZjsAB84D/yFlHzaF+kYmGB6P0FAVfsKpsaYEgFcu9LDnxvrQ309EREREJB36RyaZnPKsKFEIsL4u2hm1uXOQ95TWpDkaEYk51dZPfq6xaXnZvI9RU1rAxR5VUslGoSa4lpKhqyO45j/x/JYV5fzNK630DU9QWZKfrNBEJA2GxiOMTETmPP8WwMqKIkoKcmnuUOkDmZ+F9P5094eBh6et+2dgxpla3f1XFxpvNmjtHQFgVchzcAHUVxZRkJvDKxd6leASERERkUWrezg6WqK2rICR8ak0R3N9y8sLg9/qQ7xnrRJcIpniZFsfW1aUU5A3/9mYakoLGJ2YYiQYxCLZI8w5uJaUWIJrrnPtxHvXynIAXteElSJZr2swOn/WsnnUEjcz1i8r5WzHULLDEpF5ausdBaAhBQmuvNwctq+q4JUL6j0mIiIiIotX91D0d3O2lCjMMWPDslKaO4dw93SHIyJEKyidautfUHlCeOt7qGdYZQqzjRJcSTI0Pklxfi65OTN2cE/IliDBpXm4JNOZ2R4zO2NmTWb2joltzazQzL4bbD9sZuuC9flm9i0zO2Fmp83si6mOPVW6grq9taVzH8EFsLGuTCO4RDJIWzCCKxUJLoCda6o53trLRCTze7KKiIiIiMxH1+DCfjenw/q6MvpGJugZnkh3KCICXBkYo2tonG0NyUlw9SrBlXWU4EqSobHIgsoTAjRUFlFemMfrSnBJBjOzXOAR4OPANuBOM9s2rdndQI+7bwK+Bnw5WH8HUOjuNwHvAf5DLPm12HQPjWNAVen8yo1uqCultXeE0QkNjRbJBG29IxTk5VCbovkBdjZWMToxxRndE4iIiIjIItUddAytmUflk3TZsCw6Z7Y6pIpkhlNt/QBsb6hc0HGqg+mClLzOPpqDK0mGxicXVJ4QomXJtqws18MsyXS7gSZ3bwYwswPAPuBUXJt9wJeC5aeAPzIzAxwoNbM8oBgYB/pTFHdKdQ2OUVmST17O/PoRrK0twR1aekYWNEmmiCTuicMXZt32r2e7KC/M48CRiymJ5Za11QC8fKGHG1ct7EZdRERERCQTxebgqsmSEoXw1jxcb3YNs2ud5uESSbdT7dHHiu+qL1/QcYoLcinIy1GJwiykEVxJMjwWobRw4fnCrSvLOXN5QLV8JZOtAuKf8LYE62Zs4+6TQB9QSzTZNQS0AxeA/+7u3WEHnA7dQ+MLGunRWFMCwMXu4WSFJCIL0DcyQWXJ/EZkzkdDZRHLyws1D5eIiIiILFrdg+MU5+dSXLCwikipZGY01pTwpn6ri2SEk219NNaUUFG0sN/rZkZVcb5GcGUhJbiSZGhsktIkXJDftbKcvpEJLvWPJiEqkYyzG4gADcB64DfMbMP0RmZ2j5kdNbOjHR0dqY4xKbqGxqlZQB3xxppo2YMLumkWyQi9w+NUFaeuZ6mZsbOxilcu9KTsPUVEREREUql7aJyaFJUAT6a1NSV0Do4xNDaZ7lBElrxTbf1sX+D8WzHVJQWagysLKcGVBO7O0PhkUkZwxeqFHm/pW/CxRELSCqyJe706WDdjm6AcYSXQBXwK+KG7T7j7FeBfgF3T38DdH3X3Xe6+q66uLoSPEK6R8QjD45EFjeBaVlZAcX607IGIpFdkyhkYnaQqhSO4AHY2VnO+a5jOwbGUvq8sHma2x8zOmFmTmT0ww/b7zeyUmR03sx+Z2dq4bREzOxb8HUxt5CIiIrIUdA+PU5tF82/FNNaqQ6pIJhgcm+R81zDb6pOU4CrNV4nCLJRQgsvM/trMftHMlBCbwejEFFNOUkZwbW+oIC/HOHZRJYkkfPM8t48Am81svZkVAPuB6Q++DgJ3Bcu3A896tO7mBeC24L1LgfcBry3kM2SiqxPlLiDBFSt7oBtm0TU4/fpHJnCgqji1Ca7d66M1/V86tygrucoczfW7wMxygUeAjwPbgDvNbNu0Zq8Au9z9ZqJlhL8St23E3XcEf3uT8BFEZBpd40UWJ53bieseGqc6i+bfilldXUyOoQ6pS4zO7cxzOph/a1sSR3CNTkzRN6Iyhdkk0RPy60RHXrxhZr9nZltDjCnrDI1HhyQnYwRXUX4uN9RXcExzbkhqzPncDubUug94GjgNPOnuJ83sITOLPQB7DKg1sybgfiDWa/wRoMzMThJNlP25ux9P7kdKv66h6GiLhfZEW1NTojm4BHQNTrve4OY2lXNwAdy0qpKSglwON3el9H0lY831u2A30OTuze4+DhwA9sU3cPfn3D12oXmR6KhsEUkdXeNFFied2wnqGlzY3NXpkp+bQ0NVMRe6h9IdiqSWzu0Mc6otmuCKVURbqKog4d7aM5KU40lqJJTgcvd/dPdPA7cA54F/NLN/NbNfM7PUPu3JQLGau8lIcAHsWFPFidY+IlOelOOJzGa+57a7H3L3Le6+0d0fDtY96O4Hg+VRd7/D3Te5+253bw7WDwbrt7v7Nnf//fA/ZeolYwQXwNra6Aiu6OA3Wap0DU6/WA3uVM7BBdEfzu9ZW82LzRrBJfP6LlgFXIx73RKsm83dwA/iXhcF82G+aGafmG2nxTBvpki66Bovsjjp3E5cts7BBdF5uFp6Rpicmkp3KJIiOrczz6m2fmpKC1hRUZiU41UHnVpbetTZPJskPKTSzGqBzwL/B9FyJn9I9IR+JpTIssjQWASA0oLkJLjevaaKwbFJznYMJuV4Iteiczv5uobGKSvMozBvYWVLG2tKGJmI0Dmo+r9Lnc7T9IqVJ6hMcYlCgPdtqOXM5YGriXNZ2sL6LjCzzxCdEzO+48lad99FtJfqH5jZxpn2zfZ5M0XSTdd4kcVJ5/b1jYxHGJmIUJOFc3BBtOLK5JRzuU/z5S4lOrczy8n2PrY3VGBmSTlerGRqi0ZwZZWEMjJm9jfAVuAvgV929/Zg03fN7GhYwWWL4aBEYUnhwufggugILoBjF3rZsqI8KccUmYnO7XAkqxdaY00JABe6h6grT05vFMk+Ok/Tr3d4gpKCXAryUl9q/X0bYvNwdbHnxvqUv38qRKac759op6VnmP9163JuSNIEwYvNPL4LWoE1ca9XB+umH/fDwG8BH3T3q09o3L01+LfZzH4M7ATOJuGjiEhA13iRxUnndmK6gyoJNVk4BxfA6uro7/WW3mFWVRenORpJBZ3bmWUiMsXrlwb5tfevS9oxSwpyyc81JbiyTKJPav40KCf2f8dOXjMrBAh6di5pV0sUJmkE14ZlpZQX5XGsRfNwSeh0boegZzg5Ca41VxNcGhq9xOk8TbPekXGqUjz/VsxNq6oozs9d1GUKf3T6Mi82dzE0Nsm3D7/Jpf7RdIeUqeb6XXAE2Gxm682sANgPHIxvYGY7gW8Ae939Stz66tixzWwZ8H7gVBgfSmSJ0zVeZHHSuZ2A7sHklPZPl+qSfEoKcjVXz9KiczuDNF0ZZDwyxbaG5HWQNDOqSwpo7dVzuGySaILrv86w7oVkBpLNhsYj5Oda0np25+QYO9ZU8fKbPUk53v/P3p2HOXaXB77/vlKVSlKtqqW7q6uqqzc3drdtvJQXlkASs5gkY0MCwTAhzDPc68wFPzMZJplxcifEQ8JcyGSAeSbk3ngCE2JgwDjAdIIZNhNgAJvu9tqLl3Z319ZblaTapCqt7/1DR+Vyubprk3SOpPfzPPW06uhI+qnsn87ReX/v+xpzGTa3SyyXV2bmM4t1ezejPxJCBEaidsJc52yeumwqmal4/62iQEOxD1fUldcvt0Qqy/8+Ocl1Ax186Bf30uj38b3jF9wellet67NAVbPAPcC3gRPAg6p6TEQ+KiJ3OLv9J6AF+KqIPCkixQDYVcBhEXkK+AHwcVW1AJcxpWfHeGNqk83tNYgmConjXVVaolBE6OsIWaZHfbG57SFHx6cBOFDCABcUyhTavK4ul005EpFtFJpRh5wVnsWClm1AuMxjqxqJVJbmptJkbxXdvLOT//zd54kn0kSqdDWL8S6b2+UzM58hry/V7d2MYKOfbW1By+CqUzZPvUFVmZrPsGdLi2tjuHV3J3/+H2F44wAAIABJREFUndo8J3h8JE42r7xhXw/hpgZu3d3Fj56fYHYhQ2vQ+jTD5j4LVPVh4OFl2z6y5PabLvG4nwLXbGLYpg6MT83zD0+dRQR+7drtbO+w8kxrZcd4Y2qTze31iRdLFDZXbzn+vkiIHz0/QTqbd6WcuakMm9vedHR8muaAn13dpf2u3hFu5NnzsyV9TlNeq0Vl3kqhcV4/8Mkl22eBPyzTmKpOIp0tWXnCotfs6YLvwmM13HPDuMrmdpkU64h3lKiO+EBnmJFYoiTPZaqOzVMPWMjkSWfzdIQqG2z50mMji7dn5gulkP/zd55j//b2dT3Pe2/ZUdJxldpTY1MMREJsawsChT6kP3x+gmNnZ7h1d5fLo/MM+ywwnhNPpPmbn57BL6DA3/7sDP/yl68gXOJFfzVsU/NaRG6n0NTeD/y1qn582f1NwN8CNwJR4N2qembJ/TsolB29T1X/fDNvxBjzMnbMXofoXHX34ALo7wiTVzg3Pc9gV7PbwzHlY3Pbg54Zn+bA9nb8Pll953WIhANMz2eYWcjQZosuq8Jlv4Go6ueBz4vIb6jq31VoTFUnkcrR3OQv6XNe299BsNHHo6diFuAyJWdzu3ymnABXKUoUAuzoDPPjFyZK8lymutg89Yap+dIGrTeiPxKi0S+cmkysO8DlZbMLGc5OLfCW/VsXt21tC9LT2sTRs9MW4HLYZ4Hxok9/73nm01k+9Et7ySt85gcn+empKG+6auvqDzabmtci4gc+A7wZGAMOicjBZWVEPwDEVXWviNwFfAJ495L7Pwl8a1NvwhjzCnbMXp9YIk2DT2gLVe/iiL5IIXt5fMoCXLXM5rb3ZHN5jp+b4b03D5b8uYtVU8bj87T1WoCrGqxWovC3VPULwE4R+fDy+1X1kys8rO4k0lm2tJY2pTrQ4GNosLNme24Yd9ncLp94MoMA7SUMcF2YSbGQyRFsLG0g3XibzVNvmEpmACqewbVUg9/HYGczpyZqK5vz+QtzAOzb2vqy7fu2tPDY6RiZXJ5Gv5V6sc8C4zXT8xm+emSM6wYi9LYXLuxdta2VR09FeeO+Hpu3a7DJeX0zcFJVTznP9WXgTgoZWUV3Avc5tx8C/kJERFVVRN4OnAZq66BijAfYMXt9Yk75bZHSZl9UUnuokdZgg/XrqXE2t73nxYkEC5k81/SXtv8WvLRgfSw+z1W9pX9+U3qrLZMoLj9wr/FEFdhoD66l5YdWEg74efb8LNG5FF0t1VuT2HiSze0yiSfStIUaafCV5uLOYFehnPNYPMneLa2r7G1qjM1TD5iaLwS4ShW03qjdPc185/gF5lJZWmqkBNjpyTmaA35624Mv2767p4WfvBhlNJ5kd4nrqVcp+ywwnvKNJ8ZJpnO8ds9LWZY37+rixPlZXrw4x5V2IWAtNjOv+4DRJb+PAbdcah9VzYrINNAlIgvAv6OQ/fV7l3sREbkbuBtgxw5vl7s1xkPsmL0OsUS6qssTFvV1hBi3AFets7ntMc+MTwNwTV/pK5wUq7eMxZMlf25THquVKPwr59//UJnhVJ90Nk8mpzQHSp9Zsau78Pn52OkYv3KNlSk0pWNzu3ziyUzJyhNCoQcXwHDUAlz1xuapN0wn0/hFXA8q7e5pAS5wejJRlpN4NwxHk+zoan7Fqt2dXc0IcHoiYQEu7LPAeM83nz7Hq7a2sr0jtLhtz5Zmgo0+jp6dsQDXGrg4r+8DPqWqc6tlTKjq/cD9AENDQ1r+oRlT/eyYvT6xRJrO5uoPcPVHQjx3fpaFTM7toZgy2czctr6Z5XF0fJpwwM+uMnxfbA74CTb6LHBdRdaUYiAifyYibSLSKCLfF5EJEfmtcg+uGiTThcbv5Wio3B8J09rUwI+et/47pjxsbpfeVDJNpISr0HYsCXCZ+rTReSoit4vIcyJyUkTuXeH+JhH5inP/YyKyc8l9f+Bsf05E3upsGxCRH4jIcRE5JiL/asn+nSLyXRF5wfk3Upp3776p+Qzt4UZ8LpdO6esIEWjwcWpiztVxlMpcKks0kV78jFsqFPCztS3IqK2Yexk7ZhsvuDCzwKHhGL967csX3zX4fFy5rY1nz8+QV4uFrNUG5/U4MLDk935n24r7iEgD0E7hotktwJ+JyBngd4E/FJF7SvBWjDFL2DF7bWKJNJ0t1R/g6usIo8DZKbsYXuvWO7eX9M18G7AfeI+I7F+222LfTOBTFPpmLmV9M1fw9NgUB7a34feV/nu6iNAfCVvp0Sqy1hpab1HVGeDXgDPAXuD3yzWoapJIFVZoNAdKH+Dy+4Q37OvhkWcvovZF0ZSHze0SyuWV6fnMYjpzKXQ1BwgH/HZgrW/rnqebOZF29rsLOADcDvyl83xZ4N+o6n7gVuBDS57zXuD7qnoF8H3n95owlczQ7mL/rSK/T9jZFa6ZPlxjsULwaqUAFxQCemPxeTv/eTk7ZhvX/eNzF1GFtxzY+or79vS0kEznuDiTcmFkVWsj8/oQcIWI7BKRAIVj9sFl+xwE3u/cfifwiBb8gqruVNWdwKeB/6iqf1Gat2KMWcKO2WsQTaTpqoEMrr5IIaPZvrPXhfXO7cW+maqaBop9M5e6E/i8c/sh4DZx0qyX9M08VrJ3UAOyuTzHz81wTV8HUGgBtPxns/o6QoxN2YLLarHWAFcxevOrwFdVdbpM46k6CSeDq7mp9CUKAUKNfi7Opvjz7zxflglr6p7N7RKans+gUNIShYWVIyHLZKhvG5mnmzmRvhP4sqqmVPU0cBK4WVXPqerjAKo6C5yg0ONj+XN9Hnj7et+kV03PZ+jwQIALChePJ+ZSzCxk3B7Kpp2dXgBg+7L+W0V9kRDJdG6xB5oB7JhtPOCHz0+wrS3Iq7a+smzy7p5CefVTk7WRaVoh657XqpoF7gG+TeFY/KCqHhORj4rIHc5un6XQc+sk8GFqaOGJMVXCjtmryObyTM9nSlr9xC0tTQ10hBo5O20Brjqw3rm9Ut/Mvkvt4xzji30zWyj0zbSSp8u8OJFgIZPnmv7ylcXuj1hvvWqy1rSjfxCRZ4F54P8SkR5goXzDqh6JVDHAVZ7eHPu2tSLAs+dn6FtS596YErG5XULxZBqASIlXoQ1EwozGLMBVxzYyTzfcgN7Z/uiyx77sJNwpZ3g98JizaauqnnNunwdeubS/CuXyysx8ho4SBq03o9iP6tREgusGOlwezeZcmFkgEm6kqXHlBUL9S1bC1sKFjxKxY7ZxVS6v/PiFSX71mt5X9M4DiIQDRMKNnJpI8No93S6MsCptaF6r6sPAw8u2fWTJ7QXgXas8x30bGbAxZk3smL2KeLKwiKmrBkoUAmzvCFmJwvpQybl9H2vsmykidwN3A+zYsaNMw/GOZ8YLccVy9qbuj4SJJzPMpbKu9+M2q1tTBpeq3gu8FhhS1QyQ4JUrwetSIl2+EoVQWAlSbFhpTKnZ3C6teMIJcJX4YuxAZ5hxK9VVt7w2T52VZH8H/K5TnuFltPA/6or/s4rI3SJyWEQOT0x4v7/kzEIhK7OUZUc3o7cjSLCxNvpwnZ9ZYFvbytlbAFvbggiFQJgp8Npngak/J87NMLuQ5TV7ui65z2BXM2OWdb5mNq+NqU02t1cXc747d9ZAiUKA7R1BonNp5pxF8KY2bWBuV6Rvpqrer6pDqjrU09Oz7vdVbY6OTxMO+NnlLAAth2LpUcviqg7ricpcCex0JlvR35Z4PFUnkcriEwg2rrXa4/pd2dvGd49fYGY+Q5tHyiSZmmJzu0TiyQwCJe/X0x8JMZvKlry/l6kq652n6zmRHlt2In3Jx4pII4Xg1hdV9WtL9rkgIr2qek5EeoGLKw1KVe8H7gcYGhryfMR2yllZ6pUShT4RdnW3cGqyuvtwZXN5onMpDmy/dEmJRr+PzuYAFy3AtZwds41rDp2JAXDzrs5L7tMfCfHk6BTTVl50PWxeG1ObbG5fRjRR6NdYMwGu9hBKYTHITTsvfZw0NWE9c3uxbyaF79R3Ae9dtk+xb+bPWNI3E/iF4g4ich8wZ30zC54am+Lq7e34fZfPbNuMYkWR8akkr9r2ytLcxlvWFOASkQeAPcCTQM7ZrNjBmUQqS3OgYcUyHaVyYHshwHX07LSV+zAlZXO7tKaSadpDjSU/yPZHwgCMxuYtwFWHNjhPN3wiLSIHgS+JyCeB7cAVwM+d/lyfBU6o6icv8Vwfd/79nxt5r14zPV9YWdrukRKFALu7mzlxboapZLpqPw8uzqbIayFL63K2tAW5MJuq0Ki8z47Zxm2Hz8Tp6wjR237psunFcxbL4lobm9fG1Cab26uLJwoLIWomwOW0FDk6Pm0Brhq23rnttAIo9s30A58r9s0EDqvqQQrfsR9w+mbGKHx3N5ewkMlxdHyaD7x+d1lfZ2nJfON9a83gGgL26zrrY4nI7cB/oTCJ/1pVP77s/jcAnwauBe5S1YeW3JcDnnF+HVHVO/CgRDpXtv5bRVtag2xrC/L0mAW4TMltaG6blcXLdMF5oLNwYB2NJ7mmv3w1ho1nrXuebuZE2tnvQeA4kAU+pKo5EXk98D7gGRF50nmpP3T6gHwceFBEPgAMA79ZgvftupcyuLzzxXt3TzNQ6MN1w6B3xrUexbKDlytRCLC1tYnnzs+Qzedp8JUvU76K2DHbuEZV+fmZGK+7THlCgN72ID6xiwHrYPPamNpUrmtoTRQupN9IofLCu1X1jIjcjFMlARDgPlX9+ibfQ1nFaiyDqzXYQHNTA8fOvqKCu6ktG/lubn0zN+BLj42suP2KrS1kcsqNg5Gyvn53cxOBBp+d01aJtUZmjgLbgHOr7VgkIn7gM8CbKTSoPyQiB1X1+JLdRoB/BvzeCk8xr6rXrfX13JJIZQk3rdwgvZSu7W/nO8cvEE+mrdm6KaV1z21zafFkht3dzSV/XlsNXfc2NE83cyKtqh8DPrZs2/+m8IV5pf2jwG3rGV81mEpmCAf8BBq8E1zZ2hYkHPBzanKOG8p8Ul8uF2YW8IvQ3dJ02f22tAXJK0Tn0qtme9UJO2Yb14zEkkzMprjpMuUJoVBetKe1ifPTVl50jWxeG1ObynUN7QNAXFX3ishdwCeAdzuvN+QscusFnhKRv1dVzzaEipapf7VbRIS+jiBHx6fdHoopLztuu+zwmThA2QNcPp/Q3xGyHlxVYq0Brm7guIj8HFisFbNKVtXNwElVPQUgIl+m0Hhv8eCsqmec+/LrG7Z3JNPZy5bpKJVr+zv4zvELPDM2zRv21X7DQFMxG5nbZgXZfJ6Z+QyRMqxAaw810hZsYDRmB9Y6ZfPUJVPzac/03yryibC7u5kXJxKoallLJJfL+ZkFelqbVi3nuqW1EAC7MLNgAa4C+ywwrjnkXExYS9mlrW1BRqK2KGeNbF4bU5vKcg3N+f0+5/ZDwF+IiKjq0g/dIIWSaZ4WS6RpCzbQ6PfOQrLN6m0P8ZOTkyxkcgQby78Q3rjCjtsuOzIcZ3d3c0WyP/siIVtoXiXWGuC6bwPP3QeMLvl9DLhlHY8PishhCuWRPq6q39jAGMoukcrRXIEMrs7mAP2REM+MW4DLlNR9bg+gVkwnMygQKVOvnoHOMKN2YK1X97k9gHo1lcysmmXkht09LRw9O0MskabLg+NbzYWZFLvWkO3a09qEUOjZZQD7LDAuOnwmRnuokb09LavuWyytPrOQoS3orUUCHnSf2wMwxpTFfRt4zFquoS3u42RrTQNdwKSI3AJ8DhgE3rdS9paI3A3cDbBjx44NDLF0qvU89nK2d4TI5pXnL8xybX+H28Mx5XGf2wOoZ6rK4yNxbrtyS0Verz8S4jtWdrQqrGmphKr+EDgDNDq3DwGPl3FcAIOqOgS8F/i0iOxZvoOI3C0ih0Xk8MTERJmH80q5vDKfydEcKG8PrqJr+toZn5onOmcXekxpuDS3a1K82KunTCUW+iMhq/1bp2yeukNVmUpm6ChT0HozFvtwTSZcHsn6pbN5pucz9LSufkGj0e+jsznAxRkrdQb2WWDc9fMzMW7aGcG3SuYlvNRf7/nzs+UeVtWzeW1MbXJjbqvqY6p6ALgJ+AMReUX6u6rer6pDqjrU0+PuwulYIl0z/beK+joK1Z2sD1ftsuO2u6JzaWKJdNnLExb1R8JEE2nm07mKvJ7ZuDUFuETk/6SQ/vxXzqY+YLWMqnFgYMnv/c62NVHVceffU8A/AtevsI+rB+dkurAgprmpcgEugKfGrKavKY0Nzm2zgqlkoYZ4Z5kCXAORMGPxJNaDvP7YPHXHfCZHOpf3XIlCgJ6WJlqbGjg1Mef2UNYtVvysXOMFjS1tQS5YBhdgnwXGPdG5FKcmEgytoTwhwNb2wjXVZy3AtSqb18bUpjJeQ1vcR0QagHYgunQHVT0BzAFXb2Ts5fSlx0YWf05NJEimcy/bVu0i4UZagw3Wh6uG2XHbXcOxQlWjygW4CkHr8SmrpuR1ay12+yHgdcAMgKq+AKyWD3gIuEJEdolIALgLOLiWFxORiIg0Obe7ndc+fvlHVV4iVYjghgOVqa3bEQ6ws6uZJ0en7CK3KZWNzG2zgngyjU+grUwXwwc6wyxk8kxYBmc9snnqgqkyZ2Vuhogw2BVmJFZ9J9rxxPoWA2xpbSI6lyKbr9p2raW0oc8CEbldRJ4TkZMicu8K939YRI6LyNMi8n0RGVxy3/tF5AXn5/0lfC+mihweLvbfWtvFhI5QI00NPp6zANda2DHemNpUrmtoB4Hi8fidwCOqqs5jGgCc4/iVFLJMPCuRytJcoWtplSIiHNjeZhlctc2O2y4aiSVoDzWyZw0ls0uhmJU5atWUPG+tAa6UqqaLvzgHzstGWJx6v/cA3wZOAA+q6jER+aiI3OE8z00iMga8C/grETnmPPwq4LCIPAX8gEIPLu8FuCqcwQVw/Y4OJudSjE/Z5DIlse65bVYWT2ZoCzXiX0Ppno0Y6CwcWK1MYV2yeeqCYlamF0sUAgx2NRNPZpiZz7g9lHWJO3/XyBozuLpbmsjrSwHHOrfuzwIR8QOfAd4G7AfeIyL7l+32BDCkqtdSWJH6Z85jO4E/ptD/42bgj0WkMssljaccOh2jqcHH1U41idWICFvbghbgWhs7xhtTm8pyDQ34LNAlIieBDwPFhSuvB54SkSeBrwMfVNXJkr6jElJVEulsRa+lVcqB7e2cODdDNmeLs2qUHbddNBxNcsOOjjWVzC6F/kgYgHG7Dud5az2a/FBE/hAIicibgQ8Cf7/ag1T1YeDhZds+suT2IQpp18sf91PgmjWOzTWJVOUDXNf0tfP3T53l8ZGpxYlmzCZsaG6bV4on00TKmOlRnO+jsSQ37LDri3XG5qkLpua9m8EFsKOz8JkwHEsuljCuBrFEmka/rHnFbpcTCIvOpemusUbkG7CRz4KbgZNOyW9E5MvAnSypjKCqP1iy/6PAbzm33wp8V1VjzmO/C9wO/I8SvBdTRQ4Nx3n1QAdNDWtfab+tLciz52dQVUQqcxGiStkx3pjaVK5raAsUFogvf9wDwAObHXSlLGTy5LWy19Iq5eq+NlLZPKcmE+zb2ur2cEzp2XHbJfPpHBdnU/x2hcoTQqGaSKNfbKF5FVhrBte9wATwDPA7FA64/75cg6oWCafJXCXTqoONfq7qbePpsSkr12NKYd1zew2ljppE5CvO/Y+JyM4l910rIj8TkWMi8sxKjW+r1VQyQ6SMmR7F2r92YK1Ldgx2wVQys65ATKX1dgRp8Akj0YTbQ1mXuNNQfK0XvLtaCgGuWMLKs7Kxz4I+YHTJ72POtkv5APCt9TxWRO4WkcMicnhiYmKV4Zhqk0xnOTY+vebyhEXb2oPMLGQ5P7NQppHVDDvGG1ObbG5fxuJicY+eZ2/Gge2FhWfWh6tm2dx2yUis8L33xsG19YQtBZ9P2N4RYixefa0B6s2alkuoal5EvgF8Q1Xtm6ujeFAOByq76uT6HR08Mz7NCxeqr7m88Zb1zu0lpY7eTOFC1yERObishOgHgLiq7hWRu4BPAO92Ure/ALxPVZ8SkS6gJmpOZfN5ZuYzZc30CAca6G4J2IG1Dtkx2B1TyTTtobUHYiqtweejPxJebLRbLeLJzLqyXVuaGgj4fUQT6dV3rnHl/iwQkd8ChoA3rnNc9wP3AwwNDVmJlhrz5MgU2bxy0871XUzY0lrIuDx5cY7e9lA5hlYT7BhvTG2yuX15brT7qJTd3c0EG30cOzvDr9/g9mhMqdncds9wLIlP4NUDla1e0h8JWZugKnDZDC4puE9EJoHngOdEZEJEPnK5x9WLZDpLqNFftp47l3LFllaaA36eGIlX9HVN7djE3F4sdeTUHS6WOlrqTuDzzu2HgNukcIX4LcDTqvoUgKpGVTVXqvfkpulkBoWyligE6IuEGY3ZgbVe2DHYXVPzGc/23yoa7ApzdmqeTJXU+FdVYon0mvtvQaGXT2dzgOhc/Qa4NvlZMA4MLPm939m2/DXeBPzfwB2qmlrPY01tO3QmjgjcsM5yMMWSoqcmqivLtFLsGG9MbbK5vTZzLrT7qJQGv4+retssg6vG2Nx230g0SW97qOJJJv0ddh2uGqxWovBfA68DblLVTlXtpNBo+nUi8q/LPjqPS6RyNDdVPqXa7xNePdDBifOzTFvTdbMxG53baylXtLiP0yh3GugC9gEqIt8WkcdF5N9e6kWqrdxR3JmH5SxRCDAQCTFqGVz1xI7BLoonM3SEvB3g2tEZJq/VU7o0kc6RzuXpXOdigK6WALH6zuDazGfBIeAKEdklIgHgLuDg0h1E5HrgrygEty4uuevbwFtEJCIiEQoLVb5dmrdkqsXh4RhXbmujLbi+z8PWYAPNAT+nJqzixCXYMd6Y2mRzew0Sqcq3+6ikA9vbOH52hnzeEttriM1tF+Xyymg8yY6ucMVfe6AzxORciqSTeWq8abUA1/uA96jq6eIGp0n1bwG/Xc6BVYNEKktzhSPHRdcPRMjllW8+c86V1zdVz4253QC8Hvinzr/vEJHbVtpRVe9X1SFVHerp6SnTcEonnixceC13BtdAZyFbI2cnyvXCjsEuyeTyJFLZspYdLYUdnYUT/GrpwxV3glSd68jgAuhqDhBLpslr3X72bfizwFlocg+FwNQJ4EFVPSYiHxWRO5zd/hPQAnxVRJ4UkYPOY2PAn1AIkh0CPupsM3Uim8vz+HB83f23oJB9ubunhVOT1fH55AI7xhtTm2xur0EtlygEuHp7O7OprC1OrS02t110fnqBTE4Z7AzzpcdGXvFTTju6mgEsi8vjVgtwNarq5PKNTp1Rby9rroBEOuvaAXl7R5AtrU187fExV17fVL2Nzu21lCta3Mfpu9UORClke/1IVSdVNUmhGWdNVKWOJ9P4BNrKnO3RHwmRySkXrGF7vbBjsEuK2dFeL1HY3NRAd0tT1fThim1wMUBncxO5vDI9X7dZ65v6LFDVh1V1n6ruUdWPOds+oqrFQNabVHWrql7n/Nyx5LGfU9W9zs9/L+F7MlXg2NkZEuncuvtvFe3uabYShZdmx3hjapPN7TVIpLIEGnw0+le7JFmdDmwv9Ag6Oj7j8khMCdncdtFwrHA+OegEmypp0FlUOlwli0rr1WpHk8vVg6nrWjFQSKsOu5RSLSJcP9DB4eG4TTKzERud26uWOnJ+f79z+53AI6qqFFaPXyMiYSfw9Ubg+IZG7zFTyQztocay9+MbiBQOrKNVcjHbbJodg10Sny/8eb0e4ILCCfdILIlWQXZTMYMr0ry+v2tXSyEgVsdlCu2zwLjipy9GAbh1d9eGHr+7u4Wz0/MsZGqi5Wqp2bw2pjbZ3F6DQjWk2ixPCLBvWwsNPuHYWevDVUNsbrtoOJqkPdRIuwstBAadsogjdh3O01ZLP3q1iKy05ECAYBnGUzVUlaSLGVwArx7o4DsnLvD1J8b53Tftc20cpiptaG6ralZEiqWO/MDniqWOgMPOavDPAg+IyEkgRiEIhqrGReSTFIJkCjysqt8s6btySTyRrkgpswFn5chofJ5byv5qxgPsGOySYgZXJOTtEoUAO7rCHBmJE51L093a5PZwLiueTNMc8NPUsL4LGl1OScPoXN1+d7TPAuOKn52Ksm9rCz0b/GzZ3dOMKpyeTHBVb1uJR1f1bF4bU5tsbq9BIp2r2fKEAE0Nfq7Y2srRs5bBVUNsbrtoJJZcLM9fae2hRlqDDQxHLcDlZZc9oqhq7S6p2KSZ+Sx5dbdmcEc4wGt2d/H1J8b5V7ddgUh5s0dM7djM3FbVhymUF1y67SNLbi8A77rEY78AfGGjr+1V8WSavVtay/462zuCiMCY1fKuC3YMdk88mUEof9nRUuiPhAAYn5r3foArkSGyzv5bUPjv4PcJ0USqDKPyPvssMG5IZ/McOh3j3TcNrL7zJezqLpSROTVhAa7lbF4bU5tsbq9NIpWlLej98+zNuHp7G488exFVtWt1NcDmtnumkmmm5zOLmVSVJiIMdoWrpi1AvardJRNlVrzI4nZa9a/f0M/vffUpHh+Jc+PgxurjG2M2LpXNMbuQJVKBUmZNDX62tgZrqrllKRuCvveWHSV7LlPfppLpxaCK121pDdLgE8an5nn1QIfbw7msWDJNX0do3Y/zidAZDtRzBpcxFffU2BTzmdyGyxNCIYML4NTEXKmGZYwxpgYkUll629d/TlhNru5r56tHxrgwk2JbuyX4GLNRxdKAg52V779VNNjZbCVHPa42OzpWQLEPhNtp1bdfvY1go4+vPT7u6jiMqVfnphZQIFKBEoUAA50hRi2Dy5iymprP0FEF2VsAfp/Q2x5kfMrbge9cXplKpuncQAYXQGdzgHjSAlzGVMpPT0YRgVt3b3wBXTjQQG97kFOT1i+41ETkdhF5TkROisi9K9zfJCJfce5/TER2OttvFpEnnZ+nROSExq1FAAAgAElEQVQdlR67Maa+qapTorC2E2IObC9kLttFcWM2ZziWpNEvrgaKd3SFGYvPk8t7v+91vbIA1wYtBrgC7ga4WpoauP3ANv7h6XOkstbA2ZhKG4sXLip3NFfmYvhAJMx43NsXso2pdlPJNO0VyMoslb5IiPGpefLq3RPuc9Pz5BU6N7gYINLcSCyRRj38Ho2pJT95cZL9vW2b7jG6u6fZMrhKTET8wGeAtwH7gfeIyP5lu30AiKvqXuBTwCec7UeBIVW9Drgd+CsRsaouxpiKSWXz5PLq+rW0cruqtw0RODpufbiM2YyRaJKBSNjV6iqDnWGyeeWsxxeV1jMLcG3QSxlc7q86eccN/UzPZ/jBsxfdHooxdafYD6tSGVz9kRDnpufJ5PIVeT1j6k1elen5TMXmdCn0dYRJZ/NMznm3R1WxtMRGenBBITCWyuaZns+UcljGmBXEE2kOn4nxS6/asunn2tXdzOnJhAWnS+tm4KSqnlLVNPBl4M5l+9wJfN65/RBwm4iIqiZVNetsDwL2H8YYU1GJVOEjyO1qSOXW3NTArm4ra2bMZqSzec5Nz7PDpf5bRcXXH7E+XJ5V20eUMoo6Aa6wB1advG5PF1tam/i7x8e5/epet4djTF0Zi8/jEyrWJLe/M0xe4ezUPINd7tUgNqZWzS5kySt0VFMGl9PXajw+z5ZWb9b4H3N6B260RGExMDYam990Rokx1W61/pWb7Un5/Wcvkld464Ftm3oegJ1dzcwsZIknMxue/+YV+oDRJb+PAbdcah9VzYrINNAFTIrILcDngEHgfUsCXi8jIncDdwPs2GF9To0xpZFIFyoPeWGxeLldvb2dI8Nxt4dhXCYitwP/BfADf62qH192fxPwt8CNQBR4t6qeEZGbgfuLuwH3qerXKzdy943Gk+TV3f5bADs6CwGu4WiS1+11dSjmEiyDa4NiiTSNfiHQ4P6fsMHv487rtvOPz11czCwzxlTGWDxJe6ixYunSA5Gw87qWGm0ubaO9OZz7/sDZ/pyIvHXJ9s+JyEURObrsue4TkfElPT1+pZzvrdymnD5P1dKDC6CntYlGv3i6ZMJoPIkA7Rv8uxYz6mzVnDHl951j5+ltD3J1X9umn2tXd+GCxGnrw+UZqvqYqh4AbgL+QERWXBmhqver6pCqDvX09FR2kMaYmrWYweWBxeLldmB7G+NT88TtOl3dsrLCm1P87lcMMLmltz1EwO9jOGbns17lfnSmSsUSaU+lVP/6Df1kcso/PH3W7aEYU1fG4pXNJuiPFDI1Ru0ir7mEzZxEO/vdBRygcBL9l87zAfyNs20ln1LV65yfh0v5fiptyimBV01ZQn6f0NseYszDAa6RWJL28MYXAxQzP0bj9tlnTDnNp3P86IUJ3rJ/KyKbX7yz0wlwnbEAVymNAwNLfu93tq24j3MxrJ3CqvBFqnoCmAOuLttIjTFmmXopUQhwdV87AMfOWh+uOmZlhTdhJJpkS2sToYC7GZ9+n7CjK8zpCTuf9SoLcG1QLJH21IqTq3rbuHJbK197fPl3G2NMOY3F5yvaq6e3PYjfJ3aR11zOhk+ine1fVtWUqp4GTjrPh6r+CIhV4g24aSrpBLiqKIMLoC8S4uzUPHmP9rkZjSXp3MRnZbDRT6jRbxlcxpTZj16YYCGT5y0lKE8Ihcxzn8CZqF0QKKFDwBUisktEAhQWphxcts9B4P3O7XcCj6iqOo9pABCRQeBK4Exlhm2MMfVVovDA9kImtPXhqmsrlRXuu9Q+TkCrWFYYEblFRI4BzwD/4lJlhWtRXpXhWIJBl/tvFe3paebFiTm3h2EuwQJcG1TI4PLWAfnXb+jjydEpm3DGVEgqm+PC7AKRCvbqafD72N4RtBKF5nI2cxK9lseu5B4RedopYxi51E4icreIHBaRwxMTE2t42sqbSqYJNfppavTWMX41fR0hMjllYjbl9lBWNBqfX+yjtVGdzQHLXjWmzL72+BjdLQFu3tVZkucLNPjoj4Q5ZRlcJeMct+8Bvg2cAB5U1WMi8lERucPZ7bNAl4icBD4MFMsVvx54SkSeBL4OfFBVJyv7Dowx9SyRytLgEwL+2r8c2REO0NcR4qhlcJkNWmtZ4Wr4nr1eE7MpFjJ5drjcf6toT08Lw9EkmVze7aGYFXgnBanKxBJptrQ2uT2MlzWZzucLXQf/9B+O8+b9hVWXm20ybYy5tHNTC6hS0QwugP6OsF3kNV7y/wJ/QqFkwp8A/xn45yvtqKr34zTKHRoa8mSq0VQyQ0cFg9al0tdRKF86PjXP1rYVv/e4Zj6dY2I2xav7Ozb1PJHmgAX3jSmjidkU3z9xkX/++l00lvDC487uZitRWGJOOeCHl237yJLbC8C7VnjcA8ADZR+gMcZcQiKVpaWpoSRlcKvB1X1tlsFV39ZTVnjscmWFRaRYVvjw8hephu/Z61Ws3OGdDK4WsnllJJZkT0+L28Mxy9T+kokyiSZSnqsZ3BZqZO+WFp4YnfJsiSJjaknxQmtHc2Uvhg90hhi1i7zm0jbTm2Mtj30ZVb2gqjlVzQP/DaekYbWKJ9NV1X+rqKe1iYDfx7gHPxvGnJKqnZvN4AoHGIsnyeXtHMeYcvjCo8Nk88pvDg2svvM67HYCXGrfT4wxpu4l0lnPXUsrpwPb2zk9mWAuVTeV5czLWVnhDRqOJgkH/HRt8jtkqezuKWSSnbI+XJ5UP0eVEppP51jI5Gl2ucndSq7f0cGDh8cYjibZ1e2NNE5jalXxom2lM7gGImEnXTtHsMrKqJmKWDyJphCcugt477J9iifRP+PlJ9EHgS+JyCeB7cAVwM8v92Ii0quq55xf3wEcLdk7qTBVJZ5Mc8WW6luR5RNhW3uQc9MLbg/lFYo9Azs3mRkXaW4kk1MuzCyw3clYM6bSllZPWEm1Vk9IpLI88Ogwt125hb0l/gzc2RUmkc4xMZdiS6u3MkyNMcZUViKV81y7j3K6pr8dVXhmbJrX7OlyezimwlQ1KyLFssJ+4HPFssLAYVU9SKGs8ANOWeEYhe/vUCgrfK+IZIA8dVZWeCSWYLAz7Jlsz91O1taLE3O8ma0uj8YsZwGuDYgmCv0tvLjqZH9vO4GGsxwZjlmAy5gyG40n8fuEtmClM7gKKdpj8fmSX4Qy1W8zJ9HOfg8Cx4Es8CFVzQGIyP8AfhHoFpEx4I9V9bPAn4nIdRRKFJ4Bfqdy77a05lJZMjnddKaRW7a1B3l6bApV9cwXAYDRWCGrrBQ9uKBQrsICXMaU1l//+DSxRJoP/tLekj/3Tuc7yZnJpAW4jDGmziXSWU+0+6iUGwYK7YkfH4lbgKtOWVnh9YvOpZicSzM0WJqesKXQHmqkp7WJFy/OuT0UswLvRWiqQCyRBrwZ4Ao0+Lh+oIMjw3Fuv7rX7eEYU9PG4vP0tgfx+yp7Ibk/UriwOxpPWoDLrGijJ9HOfR8DPrbC9vdcYv/3bWqwHhJ3ju+bDcS4ZVtbkJ9n8kzNZyqeWXo5I7EkwUYfLZs8b+p03tNoLMmtu+0CgTGbsTQTbXIuxV/84AUObG/jufOzPHd+tiSZaMXXiM4VFgd++ecjnFxyUaBas92MMcZsXCJVXyUK28ONXLGlhSPDcbeHYkzVeHxkCoAdnd7ov1W0p6eZFycswOVF1oNrA6LOBbCwB0sUAty6u4tsXjl8Jub2UIypaWPx+cVgUyUtZnA5TTeNMaURSxaO750eCg6tR297ITPivMfKFI7GkgxENl9eoj3ciE8Kz2eMKY1MLs9XDo3i9wm/du32srxGRziAT176DmWMMaY+pbN5Mjn1ZLuPcrpxMMLjI3Hy1kfWmDU5MhzHL0KfC9fbLmdPTwsvTlhfWS+yANcGxOa8m8EFsLUtyJ6eZh47HSOby7s9HGNq1lg8SX+k8itKelqaCDT4GIvPV/y1jallsRrI4AI814drND6/GJjfjAafj972EKP22bdmInK7iDwnIidF5N4V7n+DiDwuIlkReeey+3Ii8qTzs7wZtqkB6WyeLz42zNmped55wwDtofKUXPb7hM7mAJNOJpcxxpj6lEhnAe9eSyuXGwYjTCUznJpMuD0UY6rCkeEY2zuCNPq9FbbY3dPC9HzGFm15kLf+T6kSxQtgmy21U06v3dPN9HyGbz5zzu2hGFOTUtkcF2ZSrmRw+XxCf0eI0bhlMRhTSvFEhrZgg+dOpNeqqdFPZ3OA89PeCQCpqpPBVZrPyv5IiBHL4FoTEfEDnwHeBuwH3iMi+5ftNgL8M+BLKzzFvKpe5/zcUdbBmoobiSX5r4+8wAsX5nj7dX3s395W1tfram4iOmcXA4wxpp4lUvUZ4Lpx0OnDZWUKjVlVKpvjqbFpBrua3R7KKxRbhLxwwcoUek11XsFx2WQiRcDvo6nBu3++V21rZUtrE//1kZOWBm1MGZydKmRIDLiQwQXQ3xlmNOadi9jG1IJYMu2p3lUb0dse9FQG11Qyw1wqW5IMLijUYbcShWt2M3BSVU+pahr4MnDn0h1U9YyqPg1Yyn+deOHCLB/84hH+vx++SC6v/PPX7+KmXeVv4N3dEiCaSJG3ki7GGFO35pwAl5cXi5fD7u5mOsKN1ofLmDU4Oj5DOpv3XP8tgKt6WwE4cW7G5ZGY5bwbofGw6FyarpbApntJlJNPhF++cgsnL87xraPn3R6OMTVnzMmeciODC2AgElocgzGmNGKJNJ1VWp6waFt7kFgiTTrrjXhFMdO0VAGugc4wF2dTLGRyJXm+GtcHjC75fczZtlZBETksIo+KyNsvtZOI3O3sd3hiYmKjYzVlNhpL8m8efIq3fvpH/Oj5SX7pVVv4l7ddwZ6eloq8fldLE5mcMruQrcjrGWOM8Z65hfoMcIkIQ4MRfn4m5vZQjPG8I8OFeTLY5b0A15bWIN0tTRy3AJfn1NdRpUSq5QLY1X3t7DkT49Pfe563HthKQ5WWXDLGi4olsvo7w7w4Ufla2v2RMHEnM6LeviAYUw6pbI6Z+UzV9t8q6m0LosD5mQVPrHorflYORMIlKU9WfE9j8SR7t7Ru+vnMZQ2q6riI7AYeEZFnVPXF5Tup6v3A/QBDQ0OWnuMBX3psZPG2qvKzU1H+l7Pg7bV7unnjvp6Kl4fqbmkCYHIuVbZeX8YYU+1E5HbgvwB+4K9V9ePL7m8C/ha4EYgC71bVMyLyZuDjQABIA7+vqo9UdPBrsJjBFay/76+37u7ieycucm56nt52dxbJGlMNDp2Js7MrTGvQm+eL+7e3cfysBbi8pqwRj002tn6/iLzg/Ly/nONcr+hcii7nS5qX+UT4/bdeyQsX53jg0WG3h2NMTRmJJgk0+OhtC7ry+gOdhZNiK9VlTGmMx+dRqIoFLJezzfnCfM4jfbiKpVSLn1mb9dJnnzfen8eNAwNLfu93tq2Jqo47/54C/hG4vpSDM+WXzeX50s9H+Ienz7Gnp4V/85ZX8SvX9LrS+6SrpfDZan24jDFmZWvsnfkBIK6qe4FPAZ9wtk8C/0RVrwHeDzxQmVGvz2wqS1ODr2r73W7Ga/d0A/DTk1GXR2KMd6kqR4bjDO0sf/nsjdrf28YLF2c9UzHFFJTt282Sg/ObKZREOSQiB1X1+JLdio2tf2/ZYzuBPwaGAAWOOI/1RMHaybl0xcp5bFZ0LsXeLS184n89Syanr8j0eO8tO1wamTHV7Uw0wUAkhM/nTqnSYu+vsfg8V/WWtzG8MfVgNF4ImFR7D65IuJGmBh/nPdKHazSeJBJuLNkKvGKpwxEL7q/FIeAKEdlFIbB1F/DetTxQRCJAUlVTItINvA74s7KN1JRcOpvni48N88LFOd529TZev7fb1fLq7aFGGnxCdC7l2hiMMcbjFntnAohIsXfm0mtodwL3ObcfAv5CRERVn1iyzzEgJCJNquqpD925hdqvPrI0i3qpvCrhgJ+fvhjlN27sr/CojKkOpyYTxBJphgYj5D1aF2L/9jYyOeXkxTn2b7drcV5RzmUTm2ls/Vbgu6oac4Ja3wVuL+NY10xViSZSi6sQvU5E+LVre0ln8/z9U2dRa+xsTEkMR5MMdjW79vrFi7yWwWVMaRQDJtWewSUibGsPeifAFUuWrP8WQE9LE8FGn332rYGqZoF7gG8DJ4AHVfWYiHxURO4AEJGbRGQMeBfwVyJyzHn4VcBhEXkK+AHw8WWL1IyHqSoPHRnl5MU5fv36Pn7hih7Xewf7ROhsDjCZsAwuY4y5hLX0zlzcxznOTwNdy/b5DeBxrwW3oFCisB7LE0LhOLi7p4WfvThp1+WMuYQjZwp5LV7P4AKsD5fHlDPAtZnG1mt6rBtNrZPpHAuZfFWUKCza0hrktqu28sz4NE+MTLk9HGOqnqoyEku62t8mEm4kHPAzGreLvMaUwmgsSYNPaK2BL9297UHOzyyQ98CX51IHuESEgUjYMrjWSFUfVtV9qrpHVT/mbPuIqh50bh9S1X5VbVbVLlU94Gz/qapeo6qvdv79rJvvw6zPPz4/wdGzM9x+9TZPXSDoamli0jK4jDGmbETkAIWyhb9zifsrfg1tqXrvH72np5mz0wsMR+081piVHB6OEQk3sqfHvcXkq9nV3Uyw0ccJC3B5SlUXvlXV+1V1SFWHenp6KvKaxbrxXVW2wvuN+3rY2dXMwafPcmHGG6u6jalWk3NpkukcO7vcC3AVL/KOxa0PjTGlMBJNEgkH8Lmc5VAKvW0hUtk8U8mMq+PI5ZXxqfnFkqqlMtAZXiwpaYx5ueFogu8dv8Cr+9t5/d5ut4fzMt3NAWKJtCeC78YY40Fr6Z25uI+INADtQNT5vR/4OvDbqvriSi/gxjW0peqhROHl7OkutDr58clJl0dijDcdPhPnxsGI65UHLsfvE161rY3jZy3A5SXlDHBtprH1pppil9NkorDqsLuKMrigkA79m0P9BPw+/vZnZ5hLZd0ekjFVaziaAHC1RCHAQGfIynQB2Xyen52K8m8feooHD4+S82qxZuNpo/EkkebS9Ily27b2IADnpt0NAl2YWSCTUwY6QyV93h2dYcZiSSvvYswymVyev3t8jPZwI2+/vs9zFwe6WprI5ZVpl4PvxhjjUYu9M0UkQKF35sFl+xwE3u/cfifwiKqqiHQA3wTuVdWfVGzE65DO5pnP5Oq2RCFAV0uAwa4wj5y44PZQjPGc6FyKU5MJz1Qf+NJjI6/4Kbqmr41nxqft2pOHlDPAtZaD86V8G3iLiEScJtdvcba5LuZkcFVjj46OcID33TrI7EKWB352hlQm5/aQTJUSkdtF5DkROSki965wf5OIfMW5/zER2bns/h0iMiciv1epMZdSsaTADhczuAD6nQyuer7Im8nl+e8/OcPfP3WW7xy/wL996Gk+9MXHyeaWt3Y05tJUlZFosiqP7SvZ2hZEgHMu9+EqlhEsdQZXfyTEbCrreoaaMV7z/RMXmZxL8+vX99PU4Hd7OK/Q7fQwLi4YNMYY85K19M4EPgt0ichJ4MNA8bv4PcBe4CMi8qTzs6XCb+Gyos5nf2tTbSwo2wgR4bYrt/KTF6Mk07bo3JilDg87/bcGIy6PZHVDg53MpbI8d37W7aEYR9kCXJtpbK2qMeBPKATJDgEfdba5rnhQ7mqpzotgA51h3n3TAONT83z+Z2fsoGrWTUT8wGeAtwH7gfeIyP5lu30AiKvqXuBTFOqAL/VJ4FvlHmu5DMeS+KRwkdVNA51h5lJZ4nV8kfebT5/jzGSCd97QzxN/9Gb+/a9exf86dp7/+shJt4dmqsj0fIbZVJbOcHUe25cLNPjoaglw3uUAVzHDtNT9Cos9vawHoTEvic6l+MnJSW7cEWHvlha3h7OiYg/jYsl3Y4wxL7eG3pkLqvouVd2rqjer6iln+586/TSvW/Jz0c33stzkbOGzv55LFAK8af8W0tk8P37ByhQas9SR4TiBBh/X9Le7PZRV3egE4Q4PeyJUYShzD66NNrZ27vucc9Deq6r/vZzjXI/JxR5c1VWicKkD29v5zaEBhqNJPvA3h5lPWyaXWZebgZOqekpV08CXgTuX7XMn8Hnn9kPAbeLUyRGRtwOngWMVGm/JjUQT9LaHXF8dvau7cJH39GTC1XG4ZSSa4OdnYrxubzc3OHWa/49f2M07ru/jMz84yZk6/buY9StmGkVqJIMLYFt7yPUShaPxeURge0fpSxTCS//djDHwraPn8fuENx/Y6vZQLqkt2ECjX4jOWQaXMcbUm0nns7+eSxQC3LSzk9ZgA987bmUKjVnq0JkY1/a1u36dbS36IyG2tQU5dCbu9lCMo6wBrloUnUvTHPATCnh/wl3Otf0dvGuon0dPR7n7gcMsWLlCs3Z9wOiS38ecbSvu42RzTlMopdAC/DvgP1zuBUTkbhE5LCKHJyYmSjbwUjkTTTLocnlCgF1Ok9p6DXB978RFmpsaeNNVL7+Y9we/ciWNfh+f/O7zLo3MVJszTtnRWilRCNDbHiSezDC74F6G52gsSW9bkEBDaU83FzO4Yu4G8IzxijOTCY6fm+GNr+qhLejd0k8iQldz0+KCQWOMMfVjYtYJcNV5Blej38cvvWoLjzx70fr3GONYyOQ4Oj7NjTu9X54QCue0QzsjHDljGVxeYQGudYomUovlNarddQMR/tM7X82PX5jkg198nHTWetaYsrsP+JSqzl1uJ1W9X1WHVHWop6enMiNbh5GYNwJc/ZEQDT7h9ORl/5w16ezUPCcn5njDFd2vuHi+pTXI+14zyDefOcf4lF0AN6srZvtVc3b2ctvaggCu1gUfjSUXg1Gl1NLUQGdzwDK4jHH84LnCgo/X7el2eyir6moJLK7iN8YYUz8m5izAVfTWA9uIJtI8dirq9lCM8YSnx6bJ5JSbBjvdHsqaDQ1GODu9YNecPMICXOsUnUtXbf+tlbzzxn4+9o6reeTZi/zuV54gm7Mgl1nVODCw5Pd+Z9uK+4hIA9AORIFbgD8TkTPA7wJ/KCL3lHvApTS7kCGWSDPY1ez2UGj0+9jRGa7LDK7Dw3EafMLQJU6A3v/anQB84dHhCo7KVKvTkwm2t5c+08hNve2FANcJNwNc8fIEuAAGIiHGrAeXMYzH53nh4hyv39NVFZ9h3S1NxJNpW7VujDF1ZnIuRVODryqOVeV221VbaGlq4OtPLL+MYkx9OuRkQhV7W1WDoZ2Fa1GHLYvLE+zIsk6Tc6maWuEN8E9vGeSPfm0/Dz9znt9/6Gny9oXTXN4h4AoR2SUiAeAu4OCyfQ4C73duvxN4RAt+QVV3qupO4NPAf1TVv6jUwEth2CllNlimi7brtbunmVMT9RXgyuTyPDU6xf7tbZcsF9vXEeIX9/XwP58Yt880s6rTkwl2drsftC6l9lAjwUYfJ87NuPL6C5kcF2ZSDETKFODqDFsGlzHAD5+/SLDRxy27u9weypp0NQfIK0wlrUzhZojI7SLynIicFJF7V7i/SUS+4tz/mIjsdLa/WUSOiMgzzr+/XOmxG2Pq0+Rc2rK3HMFGP7dfvY1vHT1v7UKMAY4Mx9m7paWqemJf1dtGa7CBn5ycdHsoBgtwrVs0kaariibcWn3g9bv4vbfs4+tPjPP/fOuE28MxHub01LoH+DZwAnhQVY+JyEdF5A5nt89S6Ll1Evgw8Iov3tWqGODa4YEShQC7ups5PZmoqyDOiXMzzGdyl8zeKrrjuu2cnV7g8LA1/jSXd3oywa4aC3CJCNvaQjzrUoBrLF4o1bCjK1SW5x/oDHN2at6yQExduzi7wLGzM9y6q4tgY3X0By6Werc+XBsnIn7gM8DbgP3Ae0Rk/7LdPgDEVXUv8CngE872SeCfqOo1FBajPVCZURtj6t3kbMoCXEu84/o+5lJZvnfigttDMcZV+bxyZDjOUBVlbwH4fcIb9/XwyLMTdXU9zqsswLUO+bwST9RWicKl7vnlK/jt1wzy3358moeOjLk9HONhqvqwqu5T1T2q+jFn20dU9aBze0FV36Wqe1X1ZlU9tcJz3Keqf17psW/WcKyQLeWFEoUAu7pbSGXznJtZcHsoFfP4SJyOUCO7ey7/3+BNV20l1Ojnfz5ppR/MpcUTaabnMzUX4IJCmcJnz8+6csI96pQPLFcG147OMJmccr6OPvuMWe7Hz0/S4Bdeu9f7vbeKup3vUdGE9eHahJuBk6p6SlXTwJeBO5ftcyfweef2Q8BtIiKq+oSqnnW2HwNCIlJb5UmMMZ40OZeiJWgBrqJbd3exta2Jrz1u31VrnWVdX97zF2eZns8slvyrJm+6aiuTcymeHp92eyh1zwJc6zA1nyGbV7pbavc7wB/92n5es7uLP/zaMxyxrAdjXmEkmqS7JeCZ1WfFi/Kn66RM4UImx4sXE1zT345P5LL7Njc18Ob9W3n4mXNk6rC/4EZPpJ37/sDZ/pyIvHXJ9s+JyEURObrsuTpF5Lsi8oLzb9UsvzodLcydWg1wJdM5V0r5jTmvWb4eXIXnHYlamUJTn6aSaZ4YjTM02OmZc5K1aGlqoKnBZxlcm9MHjC75fczZtuI+TvWFaWB5HcvfAB5X1RWjjSJyt4gcFpHDExMTJRm4MaZ+XZhZoDXY6PYwPMPvE9514wA/eO4io1Z2u2ZZ1vXqHn0xCsCtu6svwPXGfT34BB6xTEzXWYBrHS7OFlYJb2mr3QBXo9/HX/7TG+jtCPLBLx4hnrAvn8YsNRxNssMj/beAxSym05NzLo+kMp6/MEtOlf29bWva/45XbyeezPAz56SpXmzmRNrZ7y7gAHA78JfO8wH8jbNtuXuB76vqFcD3qaKypMXgcK314ALY1h4E4NnzlS9TOBJLEmjw0VOmRUEDnYXSh8VMMWPqzY+dev+/cEX1ZG9BoXxqV0uA6JxlcLlJRA5QOO7/zqX2UdX7VXVIVYd6enoqNzhjTM2ZT+eYWcjSZhlcfBs/KYkAACAASURBVOmxkcWfcMAPCv/3148ubjM1x7KuV/HoqRgDnSH6y1T5o5wizQFuHIzwvRMX3R5K3bMA1zpcmCl8EdvSGnR5JOUVaQ7wmffeQCyR5t9/4yiqVkvUmKLTkwlPXQjf0tpEOODn1GR9ZHA9e36WcMC/5qyQ11/RTVODj0eerbsTjg2fSDvbv6yqKVU9DZx0ng9V/REQW+H1lj7X54G3l/LNlNOZaAK/T8pWSs9NW1qD+ASOn5ut+GuPxAqLAXy+y2dabtT2jhA+wVa8mro0l8py+EyM6wYidISrr3R6V3MTUVtEtxnjwMCS3/udbSvuIyINQDsQdX7vB74O/Laqvlj20Rpj6l6xpHRbyDK4luoIB7iqt43Dw7G6rDhSJyqSdV2t8nnlsdNRtrQEXxb8raZg721XbeX4uRnGbOGlqyzAtQ4XnYPy1hrO4Cq6uq+d333TPr75zDm+Yf1rjAEKF5TOzyywp6fF7aEsEhF2dTdzug4CXLm88uz5Ga7c1rZqecKiYKOf1+7p4h+fq7sA12ZOpNfy2OW2quo55/Z5YOtKO3mx3NGpyQT9kRCBhto7JQo0+NjZ3cyz5yqfwTUcTTJYxmzXRr+P3vaQBbhMXfrpi5Nkc8ob9lVX9lZRd0uAeCJNOmsX8zboEHCFiOwSkQCFrOuDy/Y5SKGcEcA7gUdUVUWkA/gmcK+q/qRiIzbG1LULxQCXlSh8hVt3d5FM53hqdMrtoRiPWkvWtRe/Z6/F8xdniScz7Fqlv7qX/eo1vQD83RG7du6m2ruaU0YXZ+sjg6voX7xxDzftjPCRbxxjfGre7eEY47pTE4UygF4KcAF1E+A6E02wkMmzv7d1XY/75Su3cCaarIu/kRdoIe13xdRfL5Y7OjOZqMn+W0VX9bZxosIlClWVkViybP23inZ0hl3pL2aMm2YXMjx6Ksr+7W1V+52kq6UJBZu/G+QsSrkH+DZwAnhQVY+JyEdF5A5nt88CXSJyEvgwL5UOvgfYC3xERJ50frZU+C0YY+rMSwEuK1G43J6eZnrbg/zw+QnyVj2pFlUk69qL37PXoth/q5q/jw90hnnd3i6+emSUfN7msFsswLUOE7MpWpsaCAX8q+9cA/w+4ZO/eR3ZvHLfwWNuD8cY151yevXs8djqkt3dzYzGkjW/EvqFC3P4RdizZX0Bxl98VeG6TZ2VKdzMifRaHrvcBRHpdZ6rF6iKP3Y+r5yu8QDX/t42RmPzzC5kKvaak3Npkukcg13lDXANdIYYjdsCHFNfvvDoCAuZPL+4r3pjEt3NhbKKZ2zhyYap6sOquk9V96jqx5xtH1HVg87tBVV9l6ruVdWbVfWUs/1PVbVZVa9b8lMVx2xjTPW6YCUKL0lE+MVXbSGaSHN0fNrt4ZjSs6zry3j0VIz+SIhIFZbcLvrSYyP0dYQZi8/zp988UXUlFleyvFxkNbwfC3Ctw8XZBXrqoDzhUj9+YZI37uvhu8cv8EffOFqV/5MbUyovTszh9wk7ynzRdr12/f/s3Xd8m9XVwPHflSx5W957zzjOdjZJCJuQsvfedNNJS3fLW96Ot4VSoOxRKKuEFTaEBMjey07ieO+9p2xJ9/3DdkjTDA9Jj8b9fj7+4GgehUiPnnvuOScqEJv0/J3QZS09JIb74+szvk0GSeEBZEYHsc67ElwT/iI9cvk1QghfIUQakAVsO8XzHf1YNwPv2OE1OFxdZz99g1Yyx5k0dSdTYocrHosanDeHq6pteNHa0Qmu5PAAmrvN9A9aHfo87k4Icb4QokgIUSKEuPc41y8TQuwSQliEEFccc93NQojikZ+bj72v4lwDQ1ae2VBOVnQQCWH+WoczYRFBw+dTFa0qwaUoiuINGjrNBBr1+Bm8Y7P4eOXFhxAV7MvnRc2qAsTDqKrrExudv7Uw/dhxY+4nLz4EP4OOHZXHG1euOINKcI1DU5eZ6GDvSnABLM6MIDrYl3f31Xl8hYiinExpcw/J4QHjTrA4Wlrk8OK8J7fgGxiyUtveP+H2kKdnR7Gtoo2BIe9YCJ/MF2kpZSHwb+AA8BHwbSmlFUAI8QqwGcgRQtQIIW4feaw/AucIIYqBs0f+7PJKmobbjmZFj6/tpTvJjQsB4KAT53CNJtuTwx1bGTfaAlEN9D0xIYQeeBRYAUwFrhVCTD3mZlXALcDLx9w3HPgNsACYD/xGCBHm6JiVE3t9RzUtPWZOz3Gf1jPHE2DU42/Qe/T3FkVRFOUrjV0DxIS4Z1tdZ9AJwfLsKBq6BvjMuzZlegVVdX18xU09tPcNeUSCy6DXkZ8cRkFtJ229g1qH45VUgmscmrrNXnlQ9tHpuHhWAh19Q6wr8pjPUkUZt7LmXtJdsJVZWsRwTKMzwjxReUsvEib8939aZgSDFhu7KtvtG5gLm+gX6ZHr7h+5X46U8sOjLr9WShknpTRIKROllM+MXN4qpTxLSpklpTxbSukWW5dGE1yeXMEVZ/LD5G/gQL3zKrgqW/sQYriFoCONJriqVYLrZOYDJVLKMinlIPAqcPHRN5BSVkgp9wHH7mI6D/hUStkmpWwHPgXOd0bQyn8bstp44ssy5iSHHjnuuyshBBFBRlXBpSiK4iVUguvUZiSGEhZg4OG1xUg1i0vxAptLWwBYkBaucSRjd7KuZkuyohBC8MXhZg0j9F4qwTVGUkqauge8soILhgf+zU4KZUNxCy09Zq3DURSns9okZS29457/5AymAAORQb5HFus9UVlzDz46cWRBe7zmpYaj1wk2l7XaOTLFnZU09RARaCQ80H17fp+KEIIpscEcanBiBVdrH3Ehfg6vdk0KCzjyfMoJJQDVR/25ZuQyu91XCHGXEGKHEGJHc7M6oXOUt3bXUtPez3fOzEQIoXU4kxYZ5EtFi3rvKoqieIOGrgFiTSrBdTJ6neDMKTHsq+nkw4IGrcNRFIfbUNJCcnjAhNd4XI3J30B+Shi7Ktvp6FNVXM6mElxj1G22MDBkIzrYsw7Kx8s+n2i+1nnTYvHRC97fV69RtIqindr2fgYtNjKiXHPXdE5sEIcbnVeh4WxlLb0kRwRg0E/ssBXsZ2B6golNpSrBpXyluKnHJZPW9pYbF0JRQ7fTevpXtvU55UQlMsiIv0FPdXu/w59LOTEp5ZNSyrlSyrlRUe7dOs9VWaw2Hl1XwvQEE2fkeMbohYhAI7Ud/V7TOlhRFMVbSSmHx3142Tz7iZidHEp2TBD/93ERQ1Y1HkTxXIMWG5tLW1mWHal1KHZ1evbwudCag6r7mbOpBNcYNXUNVy1580E5xM/AmVOiKWrsdupOcEVxBaUtw9VR6ROcAeVo2THBHG7s8cihtH1mC/WdA6RHTu7vfnFGBHurO+gxW+wUmeLOpJSUNPWQ5RUJrmD6Bq1HZmM5WmVrHykRjk9wCSFICvd32utyU7VA0lF/Thy5zNH3Vexo9d46Klv7uPusLI+o3gKICBo+p6pUFZiKoigerb1viEGrjVjVovCUdELw0/OnUN7Sy2vbq099B0VxYScrpNhZ2U7voJVlWZ61OS4swMhpmRHsqmpnV5X3jMdwBSrBNUZNXQMARHlpi8JRizIiiAzy5f199VjUjhLFi5SOtP9zxRlcMJzg6h+yUtvheZUMZSND6CdbPbc4IxKLTbK9wi3GQykO1txjprN/yKPnb43KjQsB4GC94zen9A1aaOkxk+KkGUFJYQFUqwTXyWwHsoQQaUIII3ANsHqM9/0YOFcIESaECAPOHblMcSKrTfLI2hJy40I4O9czqrdguAIThmdsKoqiKJ6roXN4LU0luMbmzCnRzE8N529riulVGzMVD/VlcTM+OsGijAitQ7G7M3KiCfHz4TfvFGL1wA3orspH6wDcRVP3SAWXh7UoHC8fnY6vzYjj+U0VbCxp4XQPaZOiKKdS1NBNZJDxyI5jRzm2NehYVY0Man/yy7Iji9kA1y1ItktcWipr6cGo15EYNrmKkPyUMIx6HZtLWz2mxZMycSWNw0nrrOhgjSNxvOyYYHRiOMG1YnqcQ59rtJoq2Um91JPCA9hS1oqU0mMqW+xJSmkRQnyH4cSUHnhWSlkohLgP2CGlXC2EmAe8BYQBFwohfielzJNStgkh/ofhJBnAfVJKtUNgksZynD/62L16by1lLb08fsMcj/o3Hjnyfaq02XPnhyqKoijQ2D2c4IoO8aO9b0jjaFyfEIKfrpjC5Y9t4tkN5Xz3rCytQ1IUu/vycDNzUsII9jNoHYrd+Rr0rJgex2vbq3lqfRnfOD1D65C8gqrgGqOmkYNyjBe3KByVHRNMblwI64qa6exXX1AU71DU2E1OrOsuhEeP7IhrHKk29SRlzb2kRgag101uYc/fqGd2ciibSlvsFJnizkpGFlW9oYLLz6AnLTKQgw2On9NX3jycbE9zUrVrUngAvYNWtWByElLKD6SU2VLKDCnl/SOX/VpKuXrk9+1SykQpZaCUMkJKmXfUfZ+VUmaO/Dyn1WvwVgNDVv7y8WGmJYRw7tRYrcOxKz+DnniTn0fPD1UURVGgcbSCy+Tdm8XHIz8ljHOnxvDEl2W09Ji1DkdR7Kq520xhXdeReVWeaEaCifPzYnngk8NqxI+TqATXGDV1mfE36AnyVUVvACunx2GTko8LG7QORVEczmaTHG7sJicm5NQ31oifQY/J33Ck2tRTdA8M0dRtnvT8rVGLMyIprOuiUy2Ge73Djd0E+/p4zcaV3LgQp7QoHK3GcFaCa7RSTM3hUjzRC5srqO3o5+crctFNcpOHK8qODabICYl3RVEURTsNXQMIAVEO7oTiaX66YsrIRpcirUNRFLvaUNIM4HHzt44mhOD+S6cR4u/DD17by6BFjfhxNJXgGqOmbjPRIb4e1RpkMsIDjSzNimRPdYeaZ6N4vKq2PgaGbOTEunalR0yIr8dVcI3O30qf5PytUYszI5ASNpe12uXxFPd1sL6bKXHBXnNcnxofQk17Px19gw59nrLmXuJMfgQ6aUNQWuRwgqu8RbU5UzxLR98gj6wtYXlOFIszI7UOxyFyYoIpa+5Vc30VRVE8WG17PzHBfhh91PLjeGREBXHL4lRe21HN/ppOrcNRFLv58nALEYFG8uJddwO5PUQE+fKHy2ZwsL6Lv605rHU4Hk8dYcaovrOfOFVS/R9Oz47G5G9Qg/MUj3doZHdxTqxrH4BjQvxo6jZ71PuxrLkHP4OOOJO/XR5vZmIo/gY9W1SCy6vZbJJD9V3/Ma/O081ICAVgf61jT5BLW3rtlpAei5SIQHx0gpImleBSPMvDa0voMVv42YpcrUNxmOyYYAatNipaVQWmoiiKp6pp7ycxzD7nct7g5a1VR37iTP4EGPR85+VdvLSl8sjliuKubFKyvriZJVmRHtmd4FjnTI3hqrmJPPZFKdvKVXGII6kE1xjVtvcTH6oOykcz+uhYMS2WA/VdvLJNHWQVzzU6HyLLxWf1xIT4YbVJWj2oT3dZcy+pEYGTnr81yuijY25qGJtLVYLLm1W399E7aPWqBNf0BBMA+xy4A1RKSVlTDxlRzvusNOh1JEcEqASX4lEONXTxz00VXJmf5NLzPydr9LWpOVyKoiieq6ajTyW4JsjfqOfcvFgq2/rYq6q4FA9Q095PS88gZ+REax2K0/z6wjySwwP4wWt76BpQozIcRSW4xmDIaqOha4BEleD6L9MTTCxMD+cvnxQ5vO2RomilqKGb5PAAp7XcmqjRKtP6Ts9oU9jRN0hr76DdF8sXpkdQ1NjtUYlAZXxGZ1F5U4LLFGAgNSKAfTUdDnuO5h4z3WYL6U6avzUqMyqI0uZepz6nojiKTUp+/uZ+QvwN3LtiitbhOFRGVBBCoOZwKYqieCiL1UZ9xwCJYQFah+K28lPCSAj15/399fSZLVqHoyiTcqCuCx+d4Iwp3pPgCvL14W9Xz6Kha4Bfv12gdTgeSyW4xqCxawCbRFVwHYcQgt9elEdX/xB//UT1FFU806GGLrfYQR0d7IdeJ6jr6Nc6FLuw9/ytUYsyIgDYUqZKxL3VwfpudGJ4/os3mZEY6tAe/mXNo+9Z51a7ZkQHUdHSy5Ca46N4gO0Vbeyq6uAXF+QSFmjUOhyH8jfqSQkPoLhJJbgURVE8UWO3GYtNqgquSdAJwWVzEugftPD+/nqtw1GUSTlY38XC9AhM/gatQ3G4o9uNHqzvZnlOFG/vqeOdPbVah+aRVIJrDGrbhxeLE9RB+bimxIZw48IUXtpaqYZfKh5nYMhKRWufWyyE63WC2BA/ajs9JMHV3EOAUU9MiH3nH85IMBHk68Om0ha7Pq7iPg7Wd5EaGYi/Ua91KE41I9FEXecAzd2OqV4sbR5uE5jh5HaumVFBWGySSjXHR3ESKSUlTT28u7eO1XvrOFTfhU1Ofv5lV/8QHxc2cFpmBJfNSbBDpK4vOyZYVXApiqJ4qJq24e9mqoJrcuJM/pyeHcXu6g7V1ldxW03dAzT3mDk3L0brUDSxPDua/JQwfvlWATXt6rzV3hya4BJCnC+EKBJClAgh7j3O9b5CiNdGrt8qhEgduTxVCNEvhNgz8vO4I+M8ldqRaogEVcF1Qj88N4fIIF/uWbWXQYvaQa14jsON3Vht0m1amcWH+lPX0Y+0w0KblqSUlDX3khYZiE7Yd/ioj17HvNQwNpepOVze6mBDl9u8p+1pRmIogMPaFJY19+Jn0BFn56T0qWSOJNRGE2yK4kgWm41VO2t4dmM5O6va2VXVzgtbKnl6fTndk+irb5OSVbtqsNokv79kOsLOxz5XlRMbTEVrHwNDVq1DURTFAY7ewa6c2iTW0CKEEOuEED1CiEecHfeJ1KjN4nZzRk40UUG+vLW7ls5+NcdHcT8H64eTs2fnemeCS68TPHjVLCTww9f2YrW595qdq3FYgksIoQceBVYAU4FrhRBTj7nZ7UC7lDITeBD401HXlUopZ438fMNRcY7FaLsv1aLwxEz+Bv730ukcaujmkXUlWoejKHazb6QqcUaiSeNIxiY+1I+BIRsdfe79pbetd5CO/iGHtTpblBFBWXMvjV2eMa9MGbvugSGq2/qZ6oUJrrz4EHTiq881eytr7iEtMgidzrkL86NtTEuaVIJLcSwpJW/vrmN3dQdnTonmlxfk8quVU7l0dgK1HX08/kXphBOtm0paKGnq4YLpcaQ5eY6dlnLjQrDapNqRriiK15vkGtoA8Cvgx04Kd0xGE1zxoc7d/OSJfPQ6rshPpHtgiJ+u2uf2G1oV73OgrpOEUH+vXlvfUNLC+dNi2VbRxjf/tfM/NoGoDSGT48gKrvlAiZSyTEo5CLwKXHzMbS4G/jny+yrgLOGC2xVrO/qJDDLiZ/CuVkbjdfbUGC6dncA/1pVQWKdaFSqeoaC2k9AAg9v0DY83DcdZ6+ZzuEbnb2U4aJFvUXokAFtUFZfXOVDXBUBunOu3HbW3QF8fMqODHFbBVdLcY/eZeWMR7GcgNsSPUpXgUhzsrd217Kpq58wp0ZydG4OPXodeJ5iXGs6dS9MZtEqufmIzJeOcKVXe0svHhY3kxoUwPzXcQdG7pmnxwxuICmq7NI5EURRHMQ9Z2VPdwY6KtklVunqBCa+hSSl7pZQbGE50uYya9j5iQnzx9VFrafaQFB7AeXmxfFTYwHMbK7QOR1HGrGtgiOr2fq/sonKs2UmhTE8wseZgo9uv27kSRya4EoDqo/5cM3LZcW8jpbQAnUDEyHVpQojdQogvhBBLj/cEQoi7hBA7hBA7mpub7Rv9UWra+1V7wjH6zYVTCQ0wcs/r+9Swdw81ibYJ5wghdgoh9o/890xnxz4R+2o6mZ5gcptWQbEmP3QC6tx8Dldpcw/Bvj5EBfs65PGnxocQ4ufD5lKV4PI2X1VlhmociTZmJIayv7bT7rs+e8wWqtv6yY3VJnGYER2oWhQqDtXRN8h97x0gJTyAM6dE/9f1iWEB3LU0HSEE1z61dcz/Htv7BnlpayVhgQaumJPoNt837CUp3J9gPx+1OU5RPFR9Zz8PrS3m3zuqeXN3LQ98epi1hxq1DstVTXYN7ZSctYY2qqa9X83fsrPTMiM5OzeG379/gHWHmrQOR1HG5GD98EYmb+yiciwhBJfMSiDQ6MNbu2tUq0I7cegMrkmoB5KllLOBHwIvCyH+610gpXxSSjlXSjk3KirKYcHUdvR7dQnleIQGGPn9JdM4UN/F45+Xah2OYmeTbJvQAlwopZwO3Ay86JyoJ25gyMrhxm6mJ7hHe0IAg15HdLDfkdaq7khKSXlzL2lRgQ5b6NPrBAvSI9ikElxeZ09NB4lh/kQGOSZ56upmJJpo6RmkrtO+G3yLGoZPWqbEanPSkhUdzOHGHnWCoDjMw2tL6Oof4uJZCSecDRkV7MvLdyzAZpNc99QWKkaqkU+ks3+IZzeUY5OSGxem4m/0vh3uQgjy4kMoqFMVXIriaeo7+3l+YwVSwu1L0vjumZlEBBn5xou72O+gdsnKyTlrDW1UTUef23RCcRc6IXjomllMjQ/h2y/vYk+1YzozKIo97avpJDLIl5gQ7zwHP5a/Uc/XZsZT1zGgZsPbiSMTXLVA0lF/Thy57Li3EUL4ACagVUppllK2AkgpdwKlQLYDYz0hKSV1HaqC62SO7RXa1jvI9AQTf1tTzAOfHNY6PMW+JtM2YbeUsm7k8kLAXwjh0ke3Qw3dWGzSbeZvjUoI9aemvd9t+3I3d5vpNlvIiHTM/K1Ri9IjqGrrU2XhXmZvdQczvbR6C76qXNtv5zaFo0ODp2jU+nFqXAj9Q1YqW0+eUFCUiejsG+LlrVVcOjuRWNPJ54hkxQTz0p0LGLTYuObJLew9wcJTc7eZp9eX0WO2cMuiVIdVLLuDafEmDtV3YVHdHxTFY0gp+cmqfQxYrNyyOJWMqCDiTP7ctjiNiCAj33ttN4MW9Z4/xoTX0JwS3ThZrDbqOwZUgssBAn19ePbmeUQG+XLj01vZVdWudUiKckKd/UNUtPQyM9F9OiM5w7T4EKbEBvPpgQbaewe1DsftOTLBtR3IEkKkCSGMwDXA6mNus5rhSg6AK4C1UkophIgaqRRBCJEOZAFlDoz1hNp6BxkYsqkKrnG6cGY8fgYdq3ZWq1aFnsVebRMuB3ZJKc3HexJnt044kf21wzsLp7lRBRdAckQAfYNWWnvc8yBZMtLWKSPawQmujOF/lqpNofdo6TFT097PzCT3ek/b05TYYIx6Hbuq7JvgOtTQRbCfj2YbgqbGD1eOjSbaFMWeXt9ZTf+QlduXpI3p9lNiQ3j5zoXodYIrH9/M41+U0mu2ADBktbGptIVH1hXTP2Tl1sWpJEc4f3adK5mWYMJssVHarBLUYzWJluERQoh1QogeIcQjzo5b8R4fFjSwvriF8/NiiQn5amNAgK8P/3vZdMqae3lhc4Vm8bmoCa+hOTHGMWvoGsBikySEqhaFjhAd4serdy0kIsjIDU9vZc0B1fpTcU37azuReO+IgBMRQnDRzHgEgnf21rrtBnVX4eOoB5ZSWoQQ3wE+BvTAs1LKQiHEfcAOKeVq4BngRSFECdDG8AEcYBlwnxBiCLAB35BStjkq1pMZ3dmfoHadjEuQrw+XzE7gpa1VPLK2hB+co0kBnuKChBB5DLctPPdEt5FSPgk8CTB37lzNPuX313QQHmh0uwrO5PDhk4jKtj6NI5mY4sYeIgKNhAcaHfo8OTHBhAUY2FzayhX5iQ59LsU17BupWvLmCi4/g54ZiSa2V9j3a9Wh+m5yY0M025WXGR2Ej05woL6TlTPiNIlB8UxWm+SFzZXMTw1nanzImFsB5caF8N53l3DPqn388cNDPPDpYWJDhlsIW2ySzOggrpiTSIi/wcGvwPXljSSoC2o7ydFojp87Oapl+DkMbzbbLoRYLaU8cNTNjrQMF0Jcw/B376uBAeBXwLSRH0WxO4vVxp8+OkROTDAL0v97PNQZOdEszYrksc9LuWFhCn4G72vPejyTXENDCFEBhABGIcQlwLnHfC44VflIm960SO/exOEIL2+tOvL7NfOTeXFzJXe+sIPzp8WyJDPylN/Hr1uQ7OgQFeWIfTUdxJv8jnQrOPrfr7cLDTByztQY3t9fz/7aTpUEnASHzuCSUn4gpcyWUmZIKe8fuezXIwdmpJQDUsorpZSZUsr5UsqykcvfkFLmSSlnSSnnSCnfdWScJ1PbPpLgcrMFbleQF29iVlIoj64rUT22Pcek2iYIIRKBt4CbpJQuP6Rtd1UHM9ywjDoq2Bc/g46qNvfbCW2x2Shv6SXTwdVbADqdYGF6BJtLWzxyt8xEd3ePXPezkcuLhBDnneoxhRDPCyHKhRB7Rn5mOfr1TcSe6k50wv2qMu1tbmo4BbWd9A9a7fJ4UkoONXRr1p4QhhN3mdFBHFBzfBQ7+7yoiaq2Pm5anDLu+4YFGnn65rm88c3F3Lo4lVlJoSxMj+COpWncujhVJbdGpEcF4WfQUVCnzhfGaDItw3ullBsYTnQpikN8UNBAZWsfPzw3+4QzC7+5PIPW3kHe3n3sqaR3m+ga2sh1qVLKcCllkJQyUcvkFkBp02hXDpXgcqQQPwN3Lk0nL8HEhwUNvLGrRnVRcjHeXHXdOtJBRSVuTmxRRgTxoX58sL9ete6dBIcmuDxB+cgsh5QIVVY9ERfOiCciyMiPXt/DwJB9FtIUTU2m9Wgo8D5wr5Ryo9MinqD23kGKm3qYlxqudSjjphOC5PAAKlvdr4Kruq2fQavNKQkuGP4yUdc5QJWbVrudyFG7u1cAU4FrhRBTj7nZkd3dwIMM7+5m5HbXAHnA+cA/hBD6MTzmPSMbU2ZJKfc48OVN2N7qDrJjggn0dVgBu1uYnxbGkFXabSh1TXs/PWYLU2JD7PJ4EzU1LoQD9SrBpdjX85sqiA3x47y82Ak/Rn5KGD+7IJe/XzubC6bHzl+s5wAAIABJREFUkR4Z5HabZxxJrxPkxZtOOK9M+S/2ahmuKHYnpeSxz0uJCvKlufu43eiB4Vm4efEhPL2h3CM3minDbedD/HyICvLeGZPOYvTRcc28JM6cEs2uqg6eWl9GV/+Q1mEpTO68nK+qrn/spHDtbt/I2A93m2vvTDohuHBGPF0DFr44rN2IFnenElynUNHSS2SQkWA/tcNyIvyNev54+QwON/bw4JrDWoejTNLICfJo24SDwL9H2yYIIS4audkzQMRI24QfAqM7VL4DZAK/PqrKI9rJL2HMdlQOD2p1xwQXQHJ4IE3dZjrd7IttcVM3OgHpkc5JcC0emcO1yfPmcE14d/fI5a9KKc1SynKgZOTxxvKYLstqk+yqamdWkto9lp8cjhCww05tCg+OJJVyNazgguE5XI1dZlp7TrygpijjUdrcw/riFq5fkIxBr06bHCk/JYyC2i7MFrUhzlW4ykxcxb18WdzCwfoulmVHnrB6C4Znj9yxNI2Sph61oOehSpt6yYhWGzqcRScEZ+fGcP2CZJq6zDz6eQnVHraJ0015bdW1lJK91R2kRAQQGuDY8RPuLiUikBmJJtYXN9PeO6h1OG5JnamdQkVLH6lePvh5suo7BpiXGsaTX5Txhw8O8vLWqv/4UdzLJFqP/l5KGXhUhccsKWWTlq/lZHZUtGHU69x2p8lo1am9KjScpaSph8SwAPyNzunFnxEVRGyIH1963on1ZHZ3n+i+p3rM+4UQ+4QQDwohjrtVU8vFsoP1XXQPWFiQ7p5Ja3syBRjIiQlmm50SXAfquxACsmM0TnDFDVeQHazv1jQOVzOJtiipQoj+ozalPO7s2LX24uZKjHod18xXsyocbU5yGINWGwW1qgpzDCbVMnyspJRPSinnSinnRkVFTSJcxZs8u6Gc6GBfZo5hQ9HK6fFEBhl5bXv1KW+ruJ/S5h4yopyzaVH5Sl68iW+cnoFBr+PpDWUUN6nvxRpzStW1K25Kqe3op6nbrDaYjtH5ebEIAR8WNmgdiltSCa5TKG/tJVUNxZy0C6bFYQowsGpnjeopqriFbRVtzEg0ue3Q48Qwf3QCtpa5T2VS36CF2vZ+p7UnhOHdo8tzothQ3KJ6lU/Oz4ApwDwgHPjp8W6k5WLZ1vLhZM6CNNWhCWBuahi7Ktux2OHf/e6qDrKjtW/9mDuS4CpUc3yOmGRbFIDSozalfMMpQbuIHrOFVTtrWDkj7shQbMVx5qQML37sGqmgV05qwi3DnRij4oWqWvv4sriZa+cn46M79VKT0UfHRTMTWHOwUe1Y9zBdA0M0dZudel6nfCXW5MfXl6UTEejLC5srKWpQSS5P54qbUnZWtuOjE8xIUAmusQgNMLIsO4qC2k62uNE6nqtQCa6T6DFbaO42k6YSXJPma9Bz+ZxEWnsH+VhloxUX1z9opaC2k7lu2p4QwNdHT2JYABtLWrQOZcxKm3uRQJaTT4SW50TRbbZ42qLaZHZ3n+i+J3xMKWW9HGYGnmO4FYNL2VLWSnJ4APGh/lqH4hIWpEXQO2g90hd9omw2ye6q9iOL01oKCzSSGObP3hr3qlx1sMm0K/Vqb+ysocds4ebFqVqH4hWig/1IDg9gp2cdix1iki3DEUJUAA8Atwghao6T9FaUCXl5WxU6Ibh2HFWvl+cnMGSVvLuvzoGRKc5W2tQDoCq4NBTsZ+DOpenEBPvy8rZKqlp7tQ7JWzml6trVDAxZ2VvTQV58iNO683iCpZlRmPwN/O7dA1htal/SeKgE10lUtAwfAFSCyz4yooJYlBHB5rJWSpt7tA5HUU5oT3UHQ1bJ/LQwrUOZlMzoIPbVdtLR5x47Iosbu/H10ZEYFuDU5z0tMxIfneBzz2pTOJnd3auBa0balqUBWcC2kz2mECJu5L8CuAQocOirGyebTbK9oo0Fae6btLa3JZmRCAHrD08uCV7W0kPXgIXZya7xeZmfEsbOynY1sP4rk22LkiaE2C2E+EIIsfR4T+CKLVEmS0rJPzdXMDMpVLVVcaL8lDB2Vqn371hMtGX4yHWpUspwKWWQlDJRSnlAq9eheA6zxcq/d1Rz1pRoYk1+Y75fXryJKbHBvLGzxoHRKc5W2jy8lpYRpdbStORv1HPz4lRC/Az8c3MlbapSUgteWXX9yYFGBoZs5Keo8+/xMProWDEtloP1Xap97zipBNdJHG4cLuPNjlG7TuzlvKmxRAQaeWNXDQNDaoi04po2lbagE7j9wTgrOggpYVOp62/+sUlJUUM3WdFB6HXOLRwI9jMwNzWMdYdcdiTcuE1md7eUshD4N3AA+Aj4tpTSeqLHHHmsl4QQ+4H9QCTwe2e8zrEqauymo2+IBemqPeGosEAjMxJMfFk8uYTErsrhaqk5LpTgauwyU9vRr3UonqAeSJZSzmb4M+JlIUTIsTdyxZYok7WhpIWy5l5uXpSidSheZU5KGM3dZmra1ftXUdzNRwUNtPUOcsPC8X9uXpGfyN6aTkrUrCCPUdrcg0EvSA537sZF5b8F+xm4ZXEqEsnL2ypVW34n89aq69d3VBPqbyBdJbnHbXqCifmp4fzlkyI6+4a0DsdtaDsswcUVNw0flFMi1BvSXow+Oq7IT+TJL8v4sKCeS2cnah2SovyXz4uamZMchsnfoHUok5IYFkCQrw/ri1u4YHqc1uGcVG17P91my5EZOs62PCeaP354iLqOfo9pYSel/AD44JjLfn3U7wPAlSe47/3A/WN5zJHLz5xsvI402sNaVXD9p2XZUfzj81I6+4cm/Hm3s7Kd0AAD6S5S7T6aaNtZ2e70alAXNZ62KDVHt0UZ2TlqBpBS7hRClALZwA6HR62xf26qIDLIyMoZrn3sdKSXt1Y5/Tnnpgy/f7eVt5GkFkUVxa28tKWKlIgAlmRGjvu+F89K4A8fHmLVzlruXTHFAdEpzlbS1ENqRCA+erWn3hVEBPlyVX4SL2ypZPXeOtV+2ckmeV6e6tDgHKCuo58NJS0sz45Gp7qej5sQgt9cNJULH97Ag2sO89uL8rQOyS2oo81JFDd2kxYZiEEdlO0qJSKQpVmRbK9o51BDl9bhKMp/aOkxs7+2k+U57r8LXa8TLEyPYEOJ67eMOlDfhU5ATmywJs9/ztQYAD5RMwI90udFzaRGBKgF02Msy47CapNsmsSsvl1V7cxOCkXn5MrLE5kSG0ygUa/m+Hxlwm1RhBBRQgg9gBAineF2pWW4sJe3Vp30ZyyqWvv47FAT185PxtdHzQxwppyYYCICjWxwo/mhiqJAUUM32yrauG5+8oS+D0QF+3J6dhRv765VM0c8xOHGbrJUJySXMiUuhDNyothZ2c6/VeszxYFe216NlMOdNZSJyYs3cd2CZF7cUqnWzcdIZW5Ooriph6xobRZbPd1ZuTHEhvjx+o4aqtv6tA5HUY74cmQO0+nZ0RpHYh/LsiOpbut3+bl3B+q7SI0IJMCoTWFxRlQQ2TFBfFigElyepm/QwuayVs6cEqN1KC5nVlIowb4+E25T2Nk/RHFTj8u0JwTw0euYlRyqElwjJtkWZRmwTwixB1gFfENK2ebcV+B8L2yuQCcE1y9Q7QmdTacTLMmKZH1xi5rDpShu5IXNFfj66LhqbtIpb3sil81JoKFr4EjVveK+OvuHqGztIy/epHUoyjHOyo0hIyqQ36wuVC1BFYcYtNh4eVsVy3OiCA80ah2OW/vROTkE+/nw29WF6nvxGKgE1wn0mC1UtfWRHaMSXI5g0Ou4bkEyNin51ku71DwuxWV8XtRMZJCRvHhtWuXZ29m5w4v6H7lw4qalx0xzt1mz9oSjzp8Wx/aKNlp6zJrGodjXxpJWBi02zsr1jKS1PRn0OpZmR7LmYNOEdkyPLkLNc7HWj/nJYRys76LXbNE6FJcgpfxASpktpcwYaT+KlPLXUsrVI78PSCmvlFJmSinnSynLRi5/Q0qZJ6WcJaWcI6V8V8vX4QydfUO8sq2KC2fEEWvy0zocr7QkM5KWHjOHGtTCm6K4g66BId7aXcuFM+MJm8Ri5tm5MQT7+vDGrho7Rqdo4UDdcLWBp5xPexKdEFyZn4S/Uc93X9mj1uEUu/u4sIHmbjM3L0rVOhS3FxZo5Efn5rClrI0P9rvuep6rUAmuEzhY34WUMC1BHZQdJTLIlyvzk9hf26ky0opLsNok64ubWZYd5TLttiYrPtSfmUmhLp3g2lfTCWh/EnR+Xiw2CZ8eaNQ0DsW+1h5qJMjXh3mprpWEcRUrp8fT3G1me8X4C3M2FLcQYNS7VAUXDCfcbHJ4jo+ijMeLWyroHbTy9dMzNI3DHq0W3dXSrOEW0esnWFmqKIpzvbmzhr5BKzctmlzVq59BzwXT4/iooIG+QbVBxZ0V1o2e26kKLlcU4m/gL1fO4GB9F3/88JDW4Sge5oXNFSSHB3B6tvuP/NDS6Hd+AcSZ/Pj5W/t5dkO51mG5NJXgOoGC2uGD8rQEdVB2pKnxIXz7jAxe3V7NU+tdeqyD4gU2l7bS3jfEObme1cpsxbRY9td2umQ7UCkle6s7SI0IIDRA2xL23LhgUiIC+GB/vaZxKPYjpWTtoSaWZkVi9FFfeY7njClR+Bl0vL9v/P/uN5S0sCAt3OX+buelhuNn0PF5UZPWoShuZGDIyvObKlieE6V5RbE3izX5kRUdxPpiNYdLUVydlJIXtlQyMymUGYmhk368S+ck0Ddo5ZNCtdnMne2v7SQ2xI+oYF+tQ1FO4MwpMdx6WirPb6rgs4Pq/abYx4G6LrZXtHPjwhSP2TCuNZ0QXDwrga7+IT5y8rx4i9VGdVsfxU3dtPcNOvW5J8K1ViRcSEFtF5FBvkSrg7LD/eicHFZOj+N/PzjEe/vqtA5H8WKr99YS5OvDGVM8q5XZimmxwHC5uKup6xygucfMrCTtK0CEEFw8M54NJS3Ud/ZrHY5iB3uqO2jsMnOWhyWt7SnA6MNZU2L4sKB+XG0KS5p6KG/pdcndeX4GPYszIvn8sKoAUcbumQ3ltPQM8q3lmVqH4vWWZkWxtbyNHtVmVFFc2saSVsqae7lpoX1mFs5PDSch1J83d9fa5fEUbeysbGdOyuQTnopj3btiClPjQrhn1T4auwa0DkfxAC9uqcDPoOPKuYlah+JRksMDOC0zkm3lbWwqcfwGsM7+IdYcbOQPHx7isS9KeW5jBf/3cRFXP7H5SIWuK1IJrhPYV9PBtIQQhFBZZ0fT6QR/vWom81LD+OFre1VLIUUTZouVDwsaODcvBj+DXutw7ColIpCpcSG8u9f1Esh7qzvQC+Ey7WAvz09ESnhzlzqx9gRv7a7F10fHuXkqwXUyK2fE0dIzOK7B7qOtPM/Ni3VUWJOyPCeKytY+ylt6tQ5FcQONXQM8uq6E8/JimO9iM+W80YrpsQxabGpXuaK4uKfWlxERaGTljDi7PJ5OJ7hkdjwbiptpUgvubqmpa4Ca9n6Xa1+t/DdfHz1/v3Y2/YNWfvDangnN41WUUc3dZt7cVcslsxI078zjic6ZGkNkkJHvv7aH5u5Tz4yfaIvzjSUtnPfgl6w91ERqRADXzU/mjqVpnJ8XS2lzD5f+Y9OEOr84g0pwHUdH3yDFTT3kq4Oy0/gZ9Dx101wSw/2584UdHG5Ug6UV5/qiqJnuAQsXzYzXOhSHuCI/kb01nUeG/rqCIauNvTUdZMcGE2D00TocYDgZOD8tnFU7a9RcQDc3ZLXx7t46zp4aQ4ifQetwXNoZOdGY/A28tLVyzPf5qKCe6Qkm4kP9HRjZxC3PHq7EXXdItSlUTu1PHx7CYpX84oKpWoeiAPnJYcSE+PKei55AK4oyPNLhi8PN3LYkza6bAy+dnYhNoqq43NSuqnYA5qSotTR3kBkdxO8uymNTaSsPrTmsdTiKG3t+UzmDVht3LkvXOhSPZNDruHZ+Ml0DQ3z3lV1YrDa7Pv7AkJX73j3A9U9vJdBXz7eWZ3DjolSmJZhIjwxiWXYUn/zgdGYkmPjeq7tZ44Jz61WC6zhGD8r5qeqg7Ayj2eQP9jdw2exEbFJy+WObePizYo8fpK24jrd21xIeaOS0zEitQ3GIy+YkYPTR8co213lPfVzYQPeAhXku9ll7RX4i5S297Kxs1zoUZRK+KGqmvW+IS2claB2Ky/M36rlmXhIfFzZS23Hq9pwlTT3srel06Q0ByREBpEcFsk7N4VJO4cP99by5u5a7lqWTHBGgdTgKw1UcF0yPG9l8NKR1OIqiHMc/Pi8h2NeHGxfZpz3hqMzoIOanhvPKtipsqqLE7Wwpa8PPoCMv3jW6cyinduXcRK6am8jf15bwtkosKxPQPTDEC5srWTEtloyoIK3D8VhxJn/uv2Q6W8ra+MmqfXY7RhbUdnLRIxt4dmM5Ny9K4b3vLiUx7L/PicIDjTx36zymxofwrZd3sa+mwy7Pby8qwXUcOyra0esEs5JU32BnCw80cttpaVhtkmc2ltPhBoPsFPdX29HPJwcauTI/EYPeMz8WQwOMrJwex9u7a+kbdI2ZFv/cVEF4oJHsmGCtQ/kPK6fHEezrw/ObKrQORZmEN3fXEBZg4PQc15sR5YpuXJSClJJ/bTl1FdfrO6vR6wQXz3bdBBfAuVNj2VTaOqY2Dop32lvdwY9e38vMRBN3n5WldTjKUb42I45Bq401qk2horicooZuPixo4KbFKQ6pkr9+YTKVrX1sKh1762TFNWwoaWF+WgS+Pp7V8t+TCSH4/SXTWZgezj2r9qr2wMq4vbilku4BC984PUPrUDze5fmJ/OicbN7cXctP39jHoGXilVw9Zgt/+PAgFz+6kfa+If5523x+d/E0/I0n/vwO9jPw/K3ziQry5dsv76Kz33U2onnmSu4kbSxtZUaiyWVaZnmbmBA/bl2cRv+glWc3ltPSoxamFMd6YXMFUkq770B0NdctSKbbbOEtF9iZVVjXyfaKdhamhaNzsVmHgb4+XLcgmQ/211Pd1qd1OMoE1Hb083FhI5fP8dyktb0lhgVw7tRYXtlWddKKib5BC69uq+bs3Giig/2cGOH4XZGfgNUmeWeP9p95iuv5uLCB65/eSnigkadumovRR31WuJLZSWEkhPrz+o4arUNRFOUY939wkGBfH+5Y4phWVOdPiyU80Diu1smK9ho6Byhp6mFJZoTWoSjjZPTR8cQNc5kSG8I3/rWTTwobtA5JcROdfUM8/nkpZ+REMSNRFYk4w3fPyuLus7J4fWcNVz6xedwzp+s7+3n4s2KW/mktT3xRxhVzEvn0B8s4PXtsG4PDA408fN1s6jsG+OmqfS4z2kNlcI7R3jvIvpoO7j5T7eLUUkKYPzcvSuW5TeXc+Mw2Xr1zIaYANUNFsb/Rxdrz8mKPW4brSeamhDErKZR/rCvlyvwkTRfznl5fjr9BT35KuGYxnMytp6XxzIZynt1Yzm8uzNM6HGWcnttQDsCtS9I0jsS9fHN5Bh8VNvDkl2X86Nyc497m5a1VdPYPcedS1++vnhkdzMykUFbtrOH2JWkIF0umK84hpaR/0EpH/xAfFzZQ097PJ4UNbC1vY1pCCE/dNJfoEPsla1V7bfvQ6QTXL0zmzx8Vcbix2+WqvRXFW31e1MSXh5v55cpcwgKN/3W9PT4DfX30XJmfyNMbyqlp7/P4czRPMdoWemmW6p7gjkwBBv51+wJuenYrX39xJ+flxbI0K/K435+vW5CsQYSKK3riy1K6Bizcc94UrUPxKj88J5vc2GB+smofZz/wBRfPjGfljDjyU8Iw+X+1dj4wZKW9b5DWnkHuf/8A2yra2VfTgZSwPCeKH5ydzcwJdK+bkxzGT87P4X8/OMTrO2q4al6SPV/ehKgE1zE2lLQgJSwbY+ZScZzUyEBuWJDCS1uruPm5bbx4+3yCHdACQfFuo4u1t3nBQrgQgu+fncUtz23njV01XDtfmy+mhxu7eXtPLXctTT9p+bOWYk1+XDQrnte2V/Ot5ZlEBftqHZIyRl0DQ7y6vZqvzYgjIdRf63DcysykUC6aGc8TX5Zx8ax4MqP/c0G5o2+QR9aVsCQzknw3GR5+RX4iv3q7gMK6LqYlmLQOR3GCQYuNosZuDjd0U9/ZT0vPIIPHDGJODPPnlytzuWlRqqrccmHXzkvmoTXFPLexgj9cNl3rcBTF6w0MWbnvvQOkRARw06JUhz7XLael8uzGcp78soz7Lp7m0OdS7OOjggaSwwOYEqs2JLgrU4CBV+5ayNVPbOGjwgbKWnq4dHbifyyYK8qohs4BnttYwUUz49lT3cGeateayeTpVkwfTmg9/kUZr2yr4s2RTk2+PjqkBIvNxtFjuox6HTOTTNx9ZhaXzk4gNTJwUs9/x5J01h1q5rfvFrIgPZyUiMk93mSpM7pjfFTQQGSQUc3fchFZMcE8fN1sCmo7uf7prWoml2JXnX1DPLy2hGXZUcxLdc1KIns7PTuKWUmhPLK2hIEhqyYxPPDJYQKNPi7fo/k7Z2QyaLHx4JrDWoeijMPzGyvoMVvcosLIFf3qa1MJNOr51ku76DqqVaHNJvnJqn30mi38YmWu21RDXTQjHl8fHc9trNA6FMXBugeGWHuokT9+dJBXtlVxoL6LQF8f5qaGsXJ6HNfNT+bd7yxh5y/PZv1PzuCOpekqueXiwgKNXDo7gbd216hzAEVxAX/88BBlzb3cf8l0h39+xpn8uXxOIq9ur6ape8Chz6VMXmffEJtKWzgvL8ZtviMqxxdg9OGaeUlcOCOO8pZeHlxzmM8ONWK2aLN2oLiu+94rxCYlPz5B5w/F8aJD/Pj1hVPZ/etz+NftC/jFBbnctCiF+WnhLMuK4vy8WK6dn8y3lmew77fn8vo3FvODc7InndyC4W4Lf71qJj46wfdf24PFOvF5YPagzuqO0mu28NmhRlZMi0OvUwdlV3FeXiyP35DPofpurnlyi/qCq9jNI+uK6RoY4mcrvKecWgjBT87Pobajn0fWljj9+XdUtPFRYQN3LE07blsTV5IeFcQNC1N4dVsVRQ3dWoejjEF9Zz+PfV7KimmxqlpngqKCfXnkujmUNfdy9RNb2FnZRmVrL3e/uptPDjRy74pccuNCtA5zzEwBBq5fkMLbe2qpGGd/csV9bK9o4/y/rWfNwSbSIoO47bQ0fn5BLreelsbXZsRzWmYk0xJMTE80ERHkqxbf3Mitp6Vhtth4dJ3zv7MoivKVdUVNPL+pglsWp7IkKxIY7oThyLasXz89A4vVxpNflDnsORT7WL23liGr5KKZCVqHotiBEIJFGZF876xssqOD+OxgE3/55DDri5s12ySruJZ1RU18sL+B756ZSXKEaiOrNT+DniVZkdy5LJ1frJzKBdPjODcvlmXZUUxPMJEYFoCfwf7dk+JD/fn9pdPZXdXBPz4vtfvjj4dqUXiUD/bXMzBk48KZ8VqHohzj7KkxPHvLPO58YQeXPrqJJ2/KJy9eLV4qE1dQ28nzmyq4Yk6iWy3W2sPijEgun5PI41+U8rWZcUyJdc7rHxiy8pNV+0gI9ecON6mu+d5ZWby5q4ZfvV3AK3ctVJsfXNwfPzyEVUp+fkGu1qG4tdMyI3n65rl8/7U9XP7YZgB8dMPJ8dvdsJ3rN5an89LWSh5eW8Jfr5qpdTiKHVmsNv6+toRH1haTGBbA15ela94eQ7GvnNhgrsxP5PlNFVy/IMUuO04VRRmfwrpOvvPSLnLjQrjXiRsD0yIDuWxOIv/cXMG1C5LJiApy2nMrYyel5NXt1eTGhTAtwbvOq93VWBPT4YFGrluQQnVbH58caODDggbWFTXR3GPm1sWpdp1hqriPjr5BfvlWAelRgdy5zD3WdZQTv+8nO1PvopnxrD3YyEOfFbNspGOUFlQF1wgpJS9sriQrOoh5qe4xV8LbLMmK5NW7FmK1SS5/bBPPbijHenRDUUUZo75BC3e/spvIIF9+sdI7F8J/uTIXk7+B77+6hx6zxSnP+cCnhylr6eVPl88gyNc99leEBRr51demsq2ijce/0HZHinJyHxU08M6eOr6+LJ2kcLWLbLKW50Sz/idn8LerZ3H/pdP47Een863lmVqHNSHRwX7csDCFt3bXUFDbqXU4ip1Ut/Vx9ZNb+PtnxVwyO4EPvrdUJbc81I/PzcGo1/H79w8ipfruryjOdKihi1ue206Iv4Hnbpl33B3go5Vcjqjm+un5U/Az6Pnt6kL1/ndRm0tbKazr4oaFyapC2kMlhQdw+5J0vrU8g8zoYJ74opQlf1rHvW/so7hRdTrxJlJK7lm1j6buAf529Sx8fVxzprriXL+7eBqxIX5879XdmrUVVwmuEZtKW9lf28lNi1PVQdmFzUwK5d3vLmFBWgT3vXeACx/ewDt7alU/YGXMbDbJT9/YT3lrL3+9aiahAa7dJs9RwgKN/O2aWRQ39fCD1/Zgc3Cy+J09tTz5ZRnXLUg+0tbEXVyRn8jXZsTxwKeH2VLWqnU4ynFUtvZyz6q9zEw08Z0z3TMJ44qC/QxcMjuB6xekuH3i4LtnZhIV7Mv3X9ujWqt4gNV767jgofUcbujmoWtm8cBVs9xm44QyftEhfnz/7GzWHGzkxS2VWoejKF5jXVETVzy2GZ2Af942n1iT86s1ooJ9+dE52awvbuH1HTVOf37l5KSU/O2zYiKDfLl8TqLW4SgOlhgWwHXzk1n7o+VcNS+Rt3bXcs6DX3LjM1tZd6jJ4WsKivYeXVfCpyNt62ckalOpo7gek7+Bv187i7qOfr798i6GNJjHpc4EAatN8qePDhFv8uOqueqg7GqOtxPs3KkxxJr8+OxgI997dQ/+Bj1zU8PIiAoiIdSf8EAjEUFGIgJ9CQ8yEhFodEi/UcW9SCn51TsFvLu3jp+eP4XFGe6VaLG3pVlR/HJlLr979wA/fn0vf7piBga9/fc9bC5t5Z7X9zE/LZzfXDhLD5tGAAAf+klEQVTV7o/vaEII7r90Oocaurnjnzt4+c4F6sucC2nqGuC257cjgEeum6N2kSnHFRpg5C9XzuTGZ7Zx33sHuP+SaWpDkxvqGhjid6sP8MauGuYkh/LQNbNVxaaXuH1JGlvKWvmf9w6QGxfCvNRwrUNSnGT0XHCyLXSUsWvpMfPnjw7x7x01TIkN5rlb5xFn8j9yvSPnbh3PDQtT+ORAI798p4DcuBCmJ6pRBa7ivX31bCtv438umabWW7zIptJWpsaZ+NG5gWyvaGNLWSu3Pr+dyCAjizIi+eNl0wlUG488zktbK/nLJ4e5ZFY8t52WqnU4XsnZx9/xyE8J538vnc49q/Zx7xv7+fMVM5w64kN94gBPrS9jX00nD12jyivdhRCCmYmhTE8wkRQewNqDjWyvaGd3Vc0J263Fm/yYnRzG7ORQFqZHMDUuBJ2ap+M1eswW7nl9Lx8WNPDN5Rl8c3mG1iG5hFsWp9I9YOGBTw/T1jfIX6+cSUSQr90ef/XeOn78770kRwTw5I35bvsZa/I38K/bF3DF45u44emtPHTtbM7IidY6LK9X1drHzc9to6lrgOduna8WupWTWpoVxddPT+eJL8oI8vXhZyumqCSXG/m4sIFfvV1AS4+Zu8/K4u4zM/FxwKYMxTXpdIIHrprFJf/YyM3PbuPJG+e6XUW4ori6A3Vd/HtHNa9ur2LIKvnm8gy+d1aW5okLH72Oh6+dzUWPbOTOF3bwrzvmkxkdrGlMCtR29POrdwqYnmDiuvkqAe2Ngnx9OCMnmqVZkRTUdrGptIV399bxeVET18xL4saFqSRHqPMzdyel5NF1Jfzlk8OcNSWa/7typjqHUo7ryrlJ1HUM8OCaw1htNv58xUyMPs45X/P6BNdHBQ38+aNDrJgWy0Uz47UORxknnRDUtveTExtCTmwIUkrMFhu9Zgu9Zgs9Ziu9gxZ6zBYauwbYWNrC+/vrAfA36EmLDCQ9KpDvnJlJdnSwSnh5ICklaw428fv3D1DT3s8vLsjljqVpWoflMoQQ3H1WFpFBvvx2dSHn/e1Lfn5BLhfNjJ/UwmFb7yD3v3+QN3bVMD8tnCdvzHf7dpCxJj9euXMhX39xJ7c9v507lqTx3bOyCPEzaB2a17HZJKt21vC7dwvR6wQv3D6f/BS1m185tXvPn0Kf2cqTX5ZR19HP7y+Z5vafTZ6usK6TBz45zGeHmsiNC+Hpm+eqKlovZQow8NpdC7np2W3c+vw2vndWFncty3DaibOieBKL1UZFax8H67vYXNbKppIWKlr78NEJLpoVT0p4IFHBvpont0ZFBPnyzC1zueHpbVz+2GYeu34OizNVklsrDZ0D3PzsNixWyUPXzHLqLn3F9fjodMxKCmVWUihVbX3UtPfx7MYKnlpfzsL0cK7IT2LFtFhV1eWG6jr6+fU7haw52MilsxP4w2XTHdL1R/Ec3zs7Cx+94P8+LqK8tY+Hr5ntlES3Qz9dhBDnAw8BeuBpKeUfj7neF3gByAdagaullBUj1/0MuB2wAndLKT+2Z2y9ZgtPrS/j758VMyMxlL+oDLRHEELgZ9DjZ9CfsAqls3+IsuYeylp6KWvu4UB9F+/tqyc80MiMRBOpEYEkhwcQGexLRKBxuN1hoJGwQKP6IB/hyu/tUa09ZtYcbOSlrVXsq+kkIyqQl+9YwIL0CEc8ndu7bkEy+Slh/Pj1vfzw33t5eG0JV85NZMW0OFIjAsb0+WizSfbVdvLWrhpe31nDoMXGt8/I4O6zsty2cutYSeEBvPHNxdz3XiFPrS/njV21XDc/mSvyE0mNdK0ZRY54n57oMYUQacCrQASwE7hRSmn36aKd/UN8XNDA0xvKONzYw/y0cB68ehYJof6nvrOiMPw94b6L84gJ8eVva4qHW6qclsa185MJD/TMRJc7HLOP1dk/xOdFTby2vZpNpa2E+Plw74op3L4kTX0X8yBjabNybFu66BA/XrtrET9/az9/+eQwb+yq5bbTUrloZgKmAO/ZcOKO7+uxstokrb1mGjvNNHYNsLW8le4BC4V1nfQNWukbtFDW3IsQw5sddUKQGhmAUa/D10ePr0GHr8/w70af0d91+Br0Ry739dFh9NFh0Osw+ggMeh2fHmhErxP46HTodQKDXnDjohSMep1brBMMWW30jWzu7Bu00Dv6u9lKR/8QjV0D1Hf2s7Oinfa+IZp7zFhH5uX4+uhIiwzkfy7OY+WMeMIDjUfen67UDmlKbAhvfnMxtzy3jeue3splsxP41hkZHlXN5ervbZtN8mFBA799t5A+s4VnbplHelSQvZ9GcWPJ4QHcu2IK9Z39vLGzhlU7a/jx63v55dv7WZIZxdm50ZyWGUlimL9bfLbag6u/r4+nvKWXf22p5JVtVdik5Jcrc7l9SZrX/D/zJo44zn/7jEzSIwP5yRv7OOfBL7hxYQq3LUkj3oHrNg5LcAkh9MCjwDlADbBdCLFaSnngqJvdDrRLKTOFENcAfwKuFkJMBa4B8oB4YI0QIltKOaGp4DabpKXHTF3nAGXNPWyvaOO9ffV0D1i4eFY8f7hsOgFGtZPAW5j8DSOtCsMAaO8dpKylh7LmXooautlU2sqg5fgD8YJ8fQgNMAz/+BsxBRgIG/l9+HIjof6Gr24TYCTYzwcfnQ6dwCMOBq703u7sH6Khc4CWHjMtPWbqOwc43NDNgfouihq7kRKyooP4/SXTuHpekloUO4Wc2GDe+fZpfHKgkafWl/Hnj4r480dFxIT4MiU2hOTwAOJD/QkwDi8MmC02eswWmroGONzYw6GGLtr7hjDoBRfOjOebp2eQFeM5J5yj/I16/nDZDK6bn8KDaw7zj89LeGRdCSkRAdyxJI0bF6VqHaJD3qcj9znRY/4JeFBK+aoQ4vGRx35sIrHbbJL6rgGaugZo7DLT3D3872tvTQcH6rqw2CSZ0UH8/drZrJwep3aMnoI9vrB62twTIQTfOTOL5TnR/OmjQ/zfx0X89ZMiZieHMT3BRHZMMEnh/oQFGDGNHNODfH3c8hjuSsds4Eil/cCQlYGh4f92D1hoGFl4PVjfTWFdJ4V1XVhtksQwf+45L4cbFqZg8p988sKVFmqViTMFGHj0+jlceqCRv68t5lfvFPKb1YX8f3t3HmVHed55/PvrbrXU2pFACxJGgBWELIRYjHGMCYsdwPaJnBOPQWMnFsMMzgwEJ+MtzJw5xo49CWNixhiPc2yQwSRmCThBdhxsDPhAYhBmESAhm1WAhBaQkNDSUqu7n/mj3hZXre5W39Zdqm7/Puf06VtrP1W3nnrfrrfqrROOmMjc6eM5dto4Zkxs45AxrUxKdfERLU20NmcNGkUvN/KW16Uigu7IGqm6uoOuCLq6gs7ubnZ2ZPm+bdce3kq/X9+W1d83vLWL9W/tYv3WXWzc9nbDy95tBka3Zg1WPQ1TkNUZugM6urrZ09XN7j3d7O7sYndnN+0dXUQfMZbjq/+yiuYm0TaimbbWZtpGNDO6Nfu8tX0Prc1NzJ46jrYRTYxubWFUmj66Nbvhsi09/fTQC5tSLMGpR00iAgLS79g73NnVze7ObDt2dXbtsz09582dHZ3s6Ohi3ZZ2Ojqz+bu6g45BvNB9QtsI2kY0M6FtBLOnjuWjC2Zw7LRxPPHKlr15cfeK9Qe516rrHZNH8y+Xv59v3fcc33vwRX70xFqOnzGB9x4zmeOmj2PW5DEcN318bp48K0eecjsi2NHRxZs7Oti8o4PVm3aw/NUt/HzlBtZuaWfOtHHcfPGpzJk2/mA22RrY9AltXHb2bC4965089vKb/PjJ1/jFqo38YtUGAA4bN5ITZk7kmMPGMOvQMUwbP4qJo0cwaUwrY0e+XW63NjcVuqelPOV1j+7uYFfn23XxLTv3sGHbLta+2c6qdW+x7KXNPL9xOy1N4sPzp/O53z/WrwKwsp1//HTmHzGRb/z8WZb8+0tc/28vcfKRh/DuWZM4bvo4pk9oY/qEUUybMKoi12qr2apzKvB8RLwIIOlWYCFQmsQLgSvT5zuA65RdPVgI3BoRu4GXJD2f1vfQUAJ5ZfNOzrz6l3uHx41s4aw5U1j8vlmclBo5bPg6ZEwrJ4+ZtLd7q57K3I7dnexId7/NPXw8m7d3sKW9g60797ClfQ9bdnbw2pb2vZ+7B/EfVJOguSm703Dll88t6rsjcpPbV939m/0uVvU0xpw/bzpnzTmM42dMKORFyXppahLnzZvGefOm8ermnfzy2dd5dPVmXnh9O4+/8ibbdu3/jruxI1t455SxfHDuVE47ejJnHTuFQxr0SYhSx8+cwJLF72btlnbuWbmeB597I0/HWjXylL7WKWkVcDbwH9M8N6X1DqmBa+eeLt73N/ftM25MazPzZ07kkjOO5tx3TWP+TOe1Hbx5MyZw88XvYdW6t/jXp9fxwHNvcPujr7KzY///GR+64mymTyjkk4K5KbOvvfc5rvnFs8QA9aUJbSN41+Hj+fQZR3POcVNYcMQhhW+MsOr5wNypnHPcFJ5as5V7V23gVy9sYumTr7FtWd/v4+3RpKw7pYULDufr/+GEGkVbUbnJa4CT/+oetu3u3NuoVa6xI1uYOn4k0ye08bvHHMq0CSOZNn4UU8aPYur4UTz0wibGjmwZ0rmgqztrXOvsCjq7g86ubvZ0Bx84bgodnd3cvWL93rg7exrkuoPOrmBPajTr6Ozm6MPG0r6nk50dXbR3dNG+p4tN2zto79jDirVbaU9PlrXv6WJP18D74M7H1w4q9maJtnRT2agRzezu7KKlqWnv02eHjRu593PP02rZU2xNexsCW1uaGNnczKgRTYwbNWK/7jx3d3bz1JqthTvPtrU284Xz5vCfTj+KOx9bw72rNnLjv6/e28j3kz87nXkzJtQ5yiHJTW7f9KvVXPnjZ/YZ19rSxBmzD+UL5x3Lh4+fXtRrGVZjkjhl1iROmTWJK/8geHbDdh5ZvZnHX36Tp9du5YFnXz9gA31zkzjzdw7jhsXvrlHUFZWbvAY48+v3s3rTzn6njxvVwoIjJrLo1HfwkfnTmTp+1FD/lBkzJrbxtx8/gc+cM5u7lq/lF6s2cP2DL9JZUl+88aJ3c2YF3m9fzQauGcCrJcNrgPf0N09EdEraSta90Qzg4V7Lzuj9ByRdAlySBrdL+u1gg1sBfGvfUYcCbwx2+QLy9uXMiK8NOPnIGoUxFLnO7ZeBR8ie766SIh1rh36iRrGuPPhVHPR+/cTBxzBY/ca6BPjjgZetVW5XK0/7WudkYEtEdPYx/z4OJrefIesD8YuDXeDg5T3XaxbfEHOrKvFVMM/Liu/wqw44S17L7VyX2X15CrjlYFZQe3k/V1RT1be9mmX71elnAMM2r6HyuT2QA9Qjh1OODYttTXldtW093mV2z7IVL7OfA24Y3KzD4lguw7DaHwcou4e0L14Ellw04CzDNq+hsmX2Cjj07xvzeG3EPMz7NYEDOqvvMrt0uwaV24Xuly8ivgt8txLrkvRoRJxSiXXlkbfPiqSSuV1pRTrWHGt1FCnWvMlzbveW9+/Z8R2cvMdXJEXK62oYzsfScN724SAvuT2cjjNvq1VbrfLa3+++vD/e5n1RHb4+fmCNuF2NuE0wtO2q5jPFa4EjSoZnpnF9ziOpBZhA9kK9wSxrZvXh3DbLv2rkaX/jNwET0zr6+1tmVh8us80aj/ParDE5t80aj/ParAaq2cD1a2C2pKMktZK9GG9pr3mWAp9Knz8G3BcRkcZfKGmkpKOA2WS9jplZ/Tm3zfKvGnna5zrTMvendZDWeVcVt83MBs9ltlnjcV6bNSbntlnjcV6b1UDVuihM/YZeBvwMaAaWRMRKSV8BHo2IpWTd996cXpS3mSzRSfPdTvbKjU7g0ojY/43flVX37heqzNtnFVHA3K60Ih1rjrU6ch9rtfK0r3WmP/lF4FZJXwWeYNDd8+da3r9nx3dw8h5fRbjMrolhcSz1Yzhve90Mw7weTseZt3UYa7Dc9ve7L++Ptw2rfVHQvG7U76gRt6sRtwmGsF3KGoXNzMzMzMzMzMzMzMzMiqGaXRSamZmZmZmZmZmZmZmZVZwbuMzMzMzMzMzMzMzMzKxQ3MAFSDpP0m8lPS/pL+sdT6VJWi3paUnLJT1a73gOlqQlkjZKWlEybpKkeyQ9l34fUs8YrTHl+VxRpLyQdISk+yU9I2mlpM+k8bmLV9IoSY9IejLF+uU0/ihJy9KxcFt6YawVRLnHoDLXpu/7KUkn1SjOZklPSPpJGu7zuEsvHr4tjV8maVYNYpso6Q5Jv5G0StJ787T/JP1F+m5XSLol5XJu9p8VUz9l7ZWS1qZ69nJJH6pnjNVQpHLbGkee691DUZS6RyXluR5j1dNouVsOl5f7G+x5wOqvUXK3kfOwEfNJZVxXGMiwb+CS1Ax8GzgfmAsskjS3vlFVxVkRsSAiTql3IBVwI3Ber3F/CdwbEbOBe9OwWcUU4FxxI8XJi07gsxExFzgNuDTtyzzGuxs4OyJOABYA50k6DbgKuCYi3gm8CVxcxxitfOUeg+cDs9PPJcB3ahTnZ4BVJcP9HXcXA2+m8dek+artm8DdETEHOCHFmYv9J2kGcDlwSkTMI3uh84Xka/9ZMd3I/mUtZMfVgvTz0xrHVAtFKretARSg3j0URal7VFKe6zFWBQ2au+Vwebm/wZ4HrI4aLHcbOQ8bMZ/Kua7Qr2HfwAWcCjwfES9GRAdwK7CwzjHZACLiAWBzr9ELgZvS55uAj9Y0KBsOcn2uKFJeRMS6iHg8fd5GVoDNIIfxRmZ7GhyRfgI4G7gjjc9FrDZ4QzgGFwI/SMfDw8BESdOrGaOkmcCHgevTsOj/uCuN+w7gnDR/tWKbAJwB3AAQER0RsYUc7T+gBWiT1AKMBtaRk/1nxdVPWdvwilRuW8PIdb17KIpQ96ikPNdjrKoaLnfL4fJyX2WeB6y+GiZ3GzUPGzGfhnBdoV9u4MoO8ldLhtekcY0kgJ9LekzSJfUOpkqmRsS69Hk9MLWewVhDKuK5Ivd5kbogORFYRk7jTY+BLwc2AvcALwBbIqIzzVKEY8H6MchjsB75/3+BLwDdaXgy/R93e+NL07em+avlKOB14Pupi4TrJY0hJ/svItYCVwOvkDVsbQUeIz/7zxrPZakLsSVF7O6kHEUot60hFLHePWg5rntUUp7rMVY9RT9uK8blJVDeecDqqyFzt8HysBHzqdzrCv1yA9fwcHpEnET2qOmlks6od0DVFBFB1qhnZkke80LSWOBO4M8j4q3SaXmKNyK6ImIBMJPszqY5dQ7JKiSvx6CkjwAbI+Kxevz9QWgBTgK+ExEnAjvo1W1AnfffIWR3fR0FHA6Moe9u5cwq4TvAMWTd2K4D/ra+4VRPXs+ZZkUyHPKoAPUYs6oaDnl+ID4PWL01Uh42cD5V7LqCG7hgLXBEyfDMNK5hpDuZiYiNwD+RXaBtNBt6umxIvzfWOR5rPEU8V+Q2LySNIKts/ENE/CiNzm28AOlR6fuB95J1E9OSJhXhWLBeyjwGa53/7wP+QNJqsu4hzibrm7q/425vfGn6BGBTFeNbA6yJiGVp+A6yimle9t8HgJci4vWI2AP8iGyf5mX/WQOJiA3pRohu4Hs0Zj27kOW2FVoR690HlPO6RyXlvR5j1VPk47YiXF7uVe55wOqroXK3AfOwUfOp3OsK/XIDF/wamC3pKEmtZC8hX1rnmCpG0hhJ43o+A78PrKhvVFWxFPhU+vwp4K46xmKNqYjnilzmReor+AZgVUR8o2RS7uKVdJikielzG/BBsj6c7wc+lmbLRaw2eEM4BpcCf6LMacDWkkfmKy4iroiImRExi+xcc19EfIL+j7vSuD+W5q/aHWkRsR54VdKxadQ5wDPkZP+RdU14mqTR6bvuiS8X+88aS6934vwhDVjPLlK5bQ2jiPXuAeW97lFJea/HWFU1XO6Ww+Xl24ZwHrD6apjcbcQ8bNR8GsJ1hX7J9QaQ9CGyviybgSUR8bU6h1Qxko4me2oLskf/flj07ZN0C3AmcCiwAfgS8M/A7cA7gJeBj0fEsHsJuFVXns8VRcoLSacDDwJP83b/wf+DrF/kXMUraT7ZSy2byW4KuT0ivpLOrbcCk4AngE9GxO76RWrlKPcYTJXk68i6udsJXBQRj9Yo1jOBz0XER/o77iSNAm4m6198M3BhRLxY5bgWkL3gthV4EbiIlCPkYP9J+jJwAdBJtq/+M1mf5LnYf1ZM/ZS1Z5J1TxjAauDTRbkIPVhFKretceS53j0URap7VFJe6zFWPY2Wu+Vwedm3wZwH6hmfZRoldxs9Dxstn8q5rjDgetzAZWZmZmZmZmZmZmZmZkXiLgrNzMzMzMzMzMzMzMysUNzAZWZmZmZmZmZmZmZmZoXiBi4zMzMzMzMzMzMzMzMrFDdwmZmZmZmZmZmZmZmZWaG4gcvMzMzMzMzMzMzMzMwKxQ1cth9JXZKWS1op6UlJn5XUlKadKekn6fNUST9J8zwj6af1jdzMeivJ5xWSfixpYq/pyyXd2mvcjZJeSrn9rKQfSJpZ28jNbCCSJqf8XS5pvaS1JcNTJO2R9Kcl84+T9IKk2Wl4hKSnJb2nflthNjz0Kov/UdKMAfK3tZyyW9JFJct2pLxeLulvJC2WdF3JcpdI+k36eUTS6bXeF2a2r5J8f1LS45J+t94xmVnfJG3vY9yxkn6Z8niVpO9KOrekbN4u6bfp8w/SMh+VFJLmpOFlaforkl4vWXZWbbfQbPhJufj3JcMtKQ97rn0v7pWXyyXNlTRLUrukJ1LuPyJpcVrm9yQ91OvvtEjaIOnwmm7gMNFS7wAsl9ojYgGApCnAD4HxwJd6zfcV4J6I+Gaad35NozSzwSjN55uAS4GvpeHjgGbg/ZLGRMSOkuU+HxF3SBLw58B9kuZFREeN4zezPkTEJqAnt68EtkfE1Wn4vwIPA4uAv0vzb5N0BXAdcC7wOeBXEbGs9tGbDTulZfE/ABeUDF9JSf6mceWU3d8Hvp+mrQbOiog30vDiknV+BPg0cHpEvCHpJOCfJZ0aEeurufFmNqDSfD8X+Gvg9+obkpmV4Vrgmoi4C0DS8RHxNPCzNPxL4HMR8WjJMouAf0u/vxQR70nzLgZOiYjLahe+2bC3A5gnqS0i2oEPAmt7zXNb77xMDdAvRMSJafho4EfpGtpNwExJR0bEy2mRDwArI+K16m3K8OUnuGxAEbERuAS4LCVpqenAmpJ5n6plbGZWtoeAGSXDi4CbgZ8DC/taIDLXAOuB86seoZlVwiLgs8AMlTx9GRG3A0j6AvCnwBX1Cc9sWHsQeGcZ85dddvfji2Q3r7wBEBGPk/3zfWkZ6zCz6hoPvFnvIMysLL2viz090MySxgKnAxcDF1Y3NDMbpJ8CH06fFwG3lLuCiHgR+O/A5RHRDdzOvjl+4VDWa4PjBi47oJSkzcCUXpO+Ddwg6X5J/9OPWZrll6Rm4BxgacnoC4BbyQrZRQdYxePAnOpEZ2aVIukIYHpEPEJWqb6g1yyfAa4CvhoRm2sdn9lwJqmF7GaRAS9+lcx/sGV3qXcBj/Ua92gab2b105a6O/oNcD3wV/UOyMzKcg1Zbyf/Kukvencr3IeFwN0R8SywSdLJ1Q/RzA7gVuBCSaOA+UDvXk4u6NVFYVs/6ym9bnYLqYFL0kjgQ8CdlQ/dwA1cdhAi4mfA0cD3yBL4CUmH1TcqM+ulTdJysiewpgL3AEg6BXgjIl4B7gVOlDRpgPX0foLTzPLpArKGLcgq6r0vgJ8HrAPm1TIos2Gupyx+FHgFuGGQ8x9s2W1m+dceEQsiYg5ZGf2DPnpOMbOcSl0FHwf8I3Am8HC6mN2fRWR1dOi7rm5mNZZ6JJtFlo8/7WOW21JZ3fPT3s+q9pbfqVvSsZKOJbvBbZlvMK0eN3DZAaV+RLuAjb2nRcTmiPhhRPwx8GvgjFrHZ2YD6unX/0iywranK6JFwJz0vo4XyLpE+aMB1nMisKqKcZpZZSwCFqfcXgrMlzQbID1pfTlwKvAhvzvTrGbaS/4h/rNBvM+yUmV3qWeA3neJnwysHOTyZlZlEfEQcCjgm0bNCiQiXouIJRGxEOiknxvJ0k0pZwPXp7L888DH3ahtlgtLgas5uG4Ee18363mKy90TVpkbuGxA6YmsvwOui4joNe1sSaPT53HAMWR3pZpZzkTETrIL25+V1Ap8HDg+ImZFxCyyrhL2u3tMmcvJ+ha/u4Yhm1mZJP0OMDYiZpTk9l/zdm5fA/zviFhD1j/4t/0PtVl+DbXs7sf/Aa6SNBlA0gJgMfD/Kh23mQ2NpDlkrwbYVO9YzGxwJJ0naUT6PA2YDKztZ/aPATdHxJGpLD8CeAl4f22iNbMBLAG+fKD36PVH0iyyBrJvlYy+BfgkWcP2XQcZnw2gpd4BWC71dIsyguzuk5uBb/Qx38nAdZI6yRpLr4+IX9cuTDMrR0Q8Iekp4ApgbUS8VjL5AWCupOlp+OuS/hcwGngYOGsQd5ybWX0tAv6p17g7gdskPQS8g9Q1WkT8WNJ/Af4EuKmmUZrZoA227I6IdQdYz1JJM4BfSQpgG/DJAy1nZlXX8783ZE9sfioiuuoZkJn1a7SkNSXD3wBmAt+UtCuN+3xErO9n+UVk78ItdWca/0BFIzWzsqSbQK/tZ/IFkk4vGf5vwGvAMZKeAEaR1a2vjYgbS9a5StIO4LGI2FGdyA1AvR7KMTMzMzMzMzMzMzMzM8s1d1FoZmZmZmZmZmZmZmZmheIGLjMzMzMzMzMzMzMzMysUN3CZmZmZmZmZmZmZmZlZobiBy8zMzMzMzMzMzMzMzArFDVxmZmZmZmZmZmZmZmZWKG7gMjMzMzMzMzMzMzMzs0JxA5eZmZmZmZmZmZmZmZkVyv8HlXmlnrvsvAkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 1728x576 with 14 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "ecbbcf65-6325-4fd9-8028-b6149bec3824",
        "id": "AAM7tQRXyKXJ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 613
        },
        "outputId": "4f5d2093-4939-462f-a887-ddb0a7a203fc"
      },
      "source": [
        "#现在绘制所有特征间的相关系数热力图\n",
        "corrmat = train_data.corr()\n",
        "fig, ax = plt.subplots(figsize = (18, 10))\n",
        "sns.heatmap(corrmat, annot = True, annot_kws={'size': 12})"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f9c6c939ed0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 23
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1296x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "c7bed6a6-4f7e-4541-9a8f-969e71850008",
        "id": "azJei5UOyKXK"
      },
      "source": [
        "其中，LSTAT与price的相关程度最高(r=-0.74)，其次是RM,PTRAIO(|r|>=0.5), TAX,NOX(|r|>=0.4)。\n",
        "因此，房价可能与LSTAT(底层人口的百分比),RM(每个住宅的平均房间数),PTRATIO(各镇的师生比率),\n",
        "TAX(每$ 10,000的全值财产税率),NOX(一氧化氮的浓度)有一定但不算太强的相关性。"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "9bbbfb1f-4541-4c14-8398-3b3a0726ccd2",
        "id": "T18JA-gjyKXK"
      },
      "source": [
        "## 模型搭建思路：\n",
        "考虑到不少特征与房价并无强相关性，若带入所有数据可能会对模型训练引入噪音，故而对所有模型，学生将对全数据集与经特征提取过的数据集分别套用相同模型，用rse、rae、r2_score作为evaluation metrics（以r2_score为主）。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "c2ebab18-4f00-466f-a3a6-dd0bf234461c",
        "id": "VI3VUU60yKXK",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "bc691c0c-3f16-466c-d4d2-7f6e3b0d8a03"
      },
      "source": [
        "#from sklearn.model_section import train_test_split   会报错 ImportError: No module named model_selection\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.linear_model import LinearRegression\n",
        "from sklearn.metrics import mean_absolute_error, mean_squared_error\n",
        "import sklearn\n",
        "print(sklearn.__version__)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.22.2.post1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "3aa44cb2-9d50-4411-a79a-69c274c9c827",
        "id": "_wOp5rKhyKXL"
      },
      "source": [
        "def getCorrelatedFeature(corrdata, threshold):\n",
        "    feature = []\n",
        "    value = []\n",
        "    \n",
        "    for i,index in enumerate(corrdata.index) :\n",
        "        if abs(corrdata[index])> threshold:\n",
        "            feature.append(index)\n",
        "            value.append(corrdata[index])\n",
        "     \n",
        "    df = pd.DataFrame(data = value, index = feature, columns=['Corr Value'])\n",
        "    return df"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "09ce303a-6251-4dcd-a92c-e719f1c97d66",
        "id": "338groY1yKXL",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1030
        },
        "outputId": "5c4e7010-3c8c-451f-cc6b-06c03e57cecd"
      },
      "source": [
        "from sklearn.metrics import r2_score\n",
        "col=['threshold','r2_score','mse','mse']\n",
        "r2=[]\n",
        "mae=[]\n",
        "mse=[]\n",
        "thre=[]\n",
        "threshold=0.1\n",
        "while threshold <= 1:\n",
        "    corr_value = getCorrelatedFeature(corrmat['MEDV'],threshold)\n",
        "    correlated_data = train_data[corr_value.index]\n",
        "    X = correlated_data.drop(labels=['MEDV'],axis = 1) \n",
        "    y = correlated_data['MEDV']\n",
        "    # 分割训练集和测试集\n",
        "    X_train, X_test, y_train, y_test = train_test_split(X,y,test_size = 0.2, random_state = 0)\n",
        "    \n",
        "    ##加载线性回归模型\n",
        "    model=LinearRegression()\n",
        "    ##将训练数据传入开始训练\n",
        "    model.fit(X_train,y_train)\n",
        "\n",
        "    print(model.coef_)     #系数，偏执，有些模型没有系数（如k近邻）\n",
        "    print(model.intercept_) #与y轴交点，即截距\n",
        "    \n",
        "    y_predict = model.predict(X_test)\n",
        "    \n",
        "    df = pd.DataFrame(data = [y_predict, y_test])\n",
        "    \n",
        "    #r2_score：判定系数，其含义是也是解释回归模型的方差得分，其值取值范围是[0,1]，越接近于1说明自变量越能解释因变量的方差变化，值越小则说明效果越差。\n",
        "    r2.append(r2_score(y_test, y_predict))  \n",
        "    mae.append(mean_absolute_error(y_test, y_predict))  #平均绝对误差，用于评估预测结果和真实数据集的接近程度的程度，其其值越小说明拟合效果越好。\n",
        "    mse.append(mean_squared_error(y_test, y_predict))  #均方差,该指标计算的是拟合数据和原始数据对应样本点的误差的平方和的均值，其值越小说明拟合效果越好。\n",
        "    thre.append(threshold)\n",
        "    threshold+=0.1\n",
        "    print(threshold,\"\\n\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[-1.19443447e-01  4.47799511e-02  5.48526168e-03  2.34080361e+00\n",
            " -1.61236043e+01  3.70870901e+00 -3.12108178e-03 -1.38639737e+00\n",
            "  2.44178327e-01 -1.09896366e-02 -1.04592119e+00  8.11010693e-03\n",
            " -4.92792725e-01]\n",
            "38.091694926302296\n",
            "0.2 \n",
            "\n",
            "[-1.25156811e-01  4.48575121e-02  2.01265385e-02 -1.62895522e+01\n",
            "  3.77366409e+00 -2.13185120e-03 -1.40611846e+00  2.66835482e-01\n",
            " -1.21692284e-02 -1.07660161e+00  8.72533635e-03 -4.94425348e-01]\n",
            "38.42706211257525\n",
            "0.30000000000000004 \n",
            "\n",
            "[-9.82292616e-02  7.59436005e-03  1.13203463e-01 -8.57851466e+00\n",
            "  4.28058564e+00  2.47678293e-02  2.60823469e-01 -1.22197011e-02\n",
            " -1.19105581e+00  9.45553301e-03 -4.97973454e-01]\n",
            "25.073916678563204\n",
            "0.4 \n",
            "\n",
            "[ 6.98982840e-02 -3.55728522e+00  4.48966524e+00 -4.38418881e-03\n",
            " -1.07323064e+00 -5.20827038e-01]\n",
            "23.717115033942186\n",
            "0.5 \n",
            "\n",
            "[ 4.26494747 -1.13654724 -0.57552137]\n",
            "24.031093429371218\n",
            "0.6 \n",
            "\n",
            "[ 5.10906846 -0.65494879]\n",
            "-1.2625286844374557\n",
            "0.7 \n",
            "\n",
            "[-0.95648761]\n",
            "34.76519272372788\n",
            "0.7999999999999999 \n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-26-f06ee66db4ba>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     17\u001b[0m     \u001b[0mmodel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mLinearRegression\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     18\u001b[0m     \u001b[0;31m##将训练数据传入开始训练\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 19\u001b[0;31m     \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0my_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     20\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     21\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcoef_\u001b[0m\u001b[0;34m)\u001b[0m     \u001b[0;31m#系数，偏执，有些模型没有系数（如k近邻）\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/sklearn/linear_model/_base.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, sample_weight)\u001b[0m\n\u001b[1;32m    490\u001b[0m         \u001b[0mn_jobs_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_jobs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    491\u001b[0m         X, y = check_X_y(X, y, accept_sparse=['csr', 'csc', 'coo'],\n\u001b[0;32m--> 492\u001b[0;31m                          y_numeric=True, multi_output=True)\n\u001b[0m\u001b[1;32m    493\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    494\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0msample_weight\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/sklearn/utils/validation.py\u001b[0m in \u001b[0;36mcheck_X_y\u001b[0;34m(X, y, accept_sparse, accept_large_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, multi_output, ensure_min_samples, ensure_min_features, y_numeric, warn_on_dtype, estimator)\u001b[0m\n\u001b[1;32m    753\u001b[0m                     \u001b[0mensure_min_features\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mensure_min_features\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    754\u001b[0m                     \u001b[0mwarn_on_dtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mwarn_on_dtype\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 755\u001b[0;31m                     estimator=estimator)\n\u001b[0m\u001b[1;32m    756\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mmulti_output\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    757\u001b[0m         y = check_array(y, 'csr', force_all_finite=True, ensure_2d=False,\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/sklearn/utils/validation.py\u001b[0m in \u001b[0;36mcheck_array\u001b[0;34m(array, accept_sparse, accept_large_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, ensure_min_samples, ensure_min_features, warn_on_dtype, estimator)\u001b[0m\n\u001b[1;32m    473\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    474\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mdtype\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mdtypes_orig\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 475\u001b[0;31m             \u001b[0mdtype_orig\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresult_type\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mdtypes_orig\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    476\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    477\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mdtype_numeric\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<__array_function__ internals>\u001b[0m in \u001b[0;36mresult_type\u001b[0;34m(*args, **kwargs)\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: at least one array or dtype is required"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vkrrs9cPyKXM"
      },
      "source": [
        "def performance_metric(y_true,y_predict):\n",
        "    #calcuates and return performance_score between true and predit based on metric chosen  \n",
        "    score = r2_score(y_true,y_predict)\n",
        "    return score"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lX1J6y-QyKXM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "27f29088-4cce-4b6e-bd2e-98cfec2c5f8c"
      },
      "source": [
        "# Calculate the performance of this model\n",
        "score = performance_metric([3, -0.5, 2, 7, 4.2], [2.5, 0.0, 2.1, 7.8, 5.3])\n",
        "print(\"Model has a coefficient of determination, R^2, of {:.3f}.\".format(score))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model has a coefficient of determination, R^2, of 0.923.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5bjuO3JkyKXM"
      },
      "source": [
        "回答： R ^ 2 = 92.3％ 这意味着92.3％的变化由目标变量解释，并且似乎很高。 潜在的陷阱：这里我们只有五个要点，可能很难得出具有统计意义的结论。"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "8337bfdb-fb56-4c1a-bdce-aebe75a6ebfe",
        "id": "fTVUlWqdyKXN"
      },
      "source": [
        "我们可以得到输出四个特征的关系:\n",
        "## y=−0.01759158x1+4.3701041x2−0.94665828x3−0.60046412x4+20.27487956y=−0.01759158x1+4.3701041x2−0.94665828x3−0.60046412x4+20.27487956\n",
        "官网上还有提到Ridge Regression，我们也来试一试，它其实就是在最小二乘法的基础上对系数的L2范数的平方加了一个λλ惩罚项。\n",
        "对于普通的线性回归，我们的算法思路是：minw||wX−y||22minw||wX−y||22此处为L2范数的平方此处为L2范数的平方\n",
        "## 对于Ridge Regression，它的思路是：minw||wX−y||22+λ||w||22"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "6bb43fc2-a033-4f4a-9192-835fb96ced4c",
        "id": "JfXN1pnVyKXN",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "outputId": "a82937c5-1ab5-4ca6-b455-377089f169c8"
      },
      "source": [
        "rows = 1\n",
        "cols = 3\n",
        "fig, (ax1,ax2,ax3) = plt.subplots(nrows=rows,ncols=cols,figsize=(14,4))\n",
        "ax1.plot(thre,r2)\n",
        "ax2.plot(thre,mae)\n",
        "ax3.plot(thre,mse)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f9c6c02c610>]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x288 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "82407f75-91e2-4c2c-a88f-23d1fb6cee00",
        "id": "Iq9dQO8oyKXN"
      },
      "source": [
        "## 如何进行特征提取？\n",
        "设定一个阈值threshold，与price相关系数绝对值大于该阈值的特征被保留。\n",
        "（阈值太高则保留的特征很少，模型容易欠拟合；太低则容易过拟合）对阈值从0.1到1进行搜索，选取使“正则化且完成了数据标准化的线性回归模型”的rse、rae、r2得分较低的阈值。\n",
        "## 由上图知：threshold取0.5时线性回归模型的rse、rae、r2综合得分较低。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "cf174c55-8efe-454a-929a-3032f71160e8",
        "id": "qQz6UwsSyKXO",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "da5fedf6-f46c-4c18-d9b1-287f57600182"
      },
      "source": [
        "data_X=train_data[['ZN','RM','PTRATIO','LSTAT']]\n",
        "data_y=train_data[['MEDV']]\n",
        "X_train,X_test,y_train,y_test = train_test_split(data_X, data_y, test_size = 0.1, random_state = 0)\n",
        "#查看训练数据集与测试数据集的大小，检测数据集分割是否有问题\n",
        "print(X_train.shape)\n",
        "print(X_test.shape)\n",
        "print(y_train.shape)\n",
        "print(y_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(455, 4)\n",
            "(51, 4)\n",
            "(455, 1)\n",
            "(51, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "9268081b-755e-4bf1-ae99-b34ee62ff825",
        "id": "nuCLfBwXyKXO",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "outputId": "69a537d2-e37e-4397-808b-2808cde6c1c2"
      },
      "source": [
        "model = LinearRegression(normalize=True)\n",
        "model.fit(X_train, y_train)\n",
        "\n",
        "y_predict = model.predict(X_test)\n",
        "print(\"r2:\",r2_score(y_test, y_predict))\n",
        "\n",
        "from sklearn.model_selection import learning_curve, ShuffleSplit\n",
        "\n",
        "def plot_learning_curve (estimator, title, X, y, ylim=None, cv=None, n_jobs=None, train_sizes=np.linspace(.1, 1.0, 10)):\n",
        "    plt.figure()\n",
        "    plt.title(title)\n",
        "    plt.xlabel (\"Training examples\")\n",
        "    plt.ylabel(\"Score\")\n",
        "    \n",
        "    train_sizes,train_scores, test_scores = learning_curve(\n",
        "        estimator, X,y,cv=cv, n_jobs=n_jobs,train_sizes=train_sizes, scoring='r2')\n",
        "    \n",
        "    train_scores_mean = np.mean(train_scores, axis=1)\n",
        "    train_scores_std = np.std(train_scores, axis=1)\n",
        "    test_scores_mean = np.mean (test_scores, axis=1)\n",
        "    test_scores_std = np.std(test_scores, axis=1)\n",
        "    \n",
        "    plt.grid()\n",
        "    \n",
        "    plt.fill_between(train_sizes, train_scores_mean - train_scores_std,\n",
        "                      train_scores_mean + train_scores_std, alpha=0.1,color=\"r\")\n",
        "    plt.fill_between(train_sizes, test_scores_mean - test_scores_std,\n",
        "                      test_scores_mean + test_scores_std, alpha=0.1, color=\"g\")\n",
        "    \n",
        "    plt.plot(train_sizes, train_scores_mean, 'o-', color=\"r\", label= \"Training score\")\n",
        "    plt.plot(train_sizes, test_scores_mean, 'o-', color=\"g\", label= \"Cross-validation score\")\n",
        "    \n",
        "    plt.legend(loc= \"best\")\n",
        "    return plt\n",
        "              \n",
        "title = \"Learning Curves(LinearRegression)\"\n",
        "\n",
        "cv = ShuffleSplit(n_splits=100, test_size=0.2, random_state=0)\n",
        "              \n",
        "estimator = LinearRegression(normalize=True)\n",
        "plot_learning_curve(estimator, title, X_train, y_train, ylim=(0.7,1.01), cv=cv, n_jobs=-1)\n",
        "\n",
        "\n",
        "plt.grid()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "r2: 0.3899959810472349\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": {
        "uuid": "dae255f8-0ca0-4193-aad8-f6afc0002d3a",
        "id": "1Yme29GUyKXO"
      },
      "source": [
        "ValueError: Found array with 0 feature(s) (shape=(404, 0)) while a minimum of 1 is required.\n",
        "\n",
        "数据集分类出错"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "06478962-9803-42d0-bf86-49a94861d1b7",
        "id": "uf33JHT3yKXP"
      },
      "source": [
        "# 模型搭建与训练\n",
        "# 逐步调参前的多模型效果比较"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JTXsvB6FyKXP"
      },
      "source": [
        "#print(X_test.shape)\n",
        "#print(X_test)\n",
        "#print(X_test.iloc[:,0:-3].values)\n",
        "#print(X_test.iloc[:,1:-2].values)\n",
        "#print(X_test.iloc[:,2:-1].values)\n",
        "#print(X_test.iloc[:,3:4].values)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AyLCA_2jyKXP"
      },
      "source": [
        "#ValueError: x and y must be the same size"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "0c72b65a-1030-4d05-8d07-0edf2adc3e77",
        "id": "OOPd5rUMyKXQ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 974
        },
        "outputId": "2acf483c-a457-4338-cdf9-8386754a92d5"
      },
      "source": [
        "print('r2_score: %.2f' % r2_score(y_test, y_predict))\n",
        "# 绘制预测效果图\n",
        "\n",
        "#print(X_test.iloc[:,0:-3].values)\n",
        "#print(X_test.iloc[:,1:-2].values)\n",
        "#print(X_test.iloc[:,2:-1].values)\n",
        "#print(X_test.iloc[:,3:4].values)\n",
        "\n",
        "plt.scatter(X_test.iloc[:,2:-1].values, y_test, color='black')\n",
        "plt.plot(X_test.iloc[:,2:-1].values, y_predict, color='blue', linewidth=3)\n",
        "plt.xticks(())\n",
        "plt.yticks(())\n",
        "plt.show()\n",
        "\n",
        "plt.scatter(X_test.iloc[:,1:-2].values, y_test, color='black')\n",
        "plt.plot(X_test.iloc[:,2:-1].values, y_predict, color='blue', linewidth=3)\n",
        "plt.xticks(())\n",
        "plt.yticks(())\n",
        "plt.show()\n",
        "\n",
        "plt.scatter(X_test.iloc[:,0:-3].values, y_test, color='black')\n",
        "plt.plot(X_test.iloc[:,2:-1].values, y_predict, color='blue', linewidth=3)\n",
        "plt.xticks(())\n",
        "plt.yticks(())\n",
        "plt.show()\n",
        "\n",
        "plt.scatter(X_test.iloc[:,3:4].values, y_test, color='black')\n",
        "plt.plot(X_test.iloc[:,2:-1].values, y_predict, color='blue', linewidth=3)\n",
        "plt.xticks(())\n",
        "plt.yticks(())\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "r2_score: 0.39\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": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "703b7f8d-1e2b-44d4-8c19-c5e06feee3fc",
        "id": "DIW4mTe3yKXQ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d9a09577-f785-43e5-84bf-6dca0692fb47"
      },
      "source": [
        "##导入线性回归模型\n",
        "from sklearn.linear_model import Ridge  #导入岭回归\n",
        "from sklearn.linear_model import Lasso\n",
        "from sklearn.linear_model import ElasticNet\n",
        "from sklearn.neighbors import KNeighborsRegressor\n",
        "from sklearn.tree import DecisionTreeRegressor\n",
        "from sklearn.svm import SVR\n",
        "from sklearn.ensemble import AdaBoostRegressor\n",
        "from sklearn.ensemble import GradientBoostingRegressor\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.ensemble import ExtraTreesRegressor\n",
        "\n",
        "from sklearn.pipeline import make_pipeline\n",
        "from sklearn.externals import joblib  #模型的保存和加载 "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/externals/joblib/__init__.py:15: FutureWarning: sklearn.externals.joblib is deprecated in 0.21 and will be removed in 0.23. Please import this functionality directly from joblib, which can be installed with: pip install joblib. If this warning is raised when loading pickled models, you may need to re-serialize those models with scikit-learn 0.21+.\n",
            "  warnings.warn(msg, category=FutureWarning)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "15a699d4-f42d-461f-902d-4d73df6f8a15",
        "id": "v1EK9ornyKXQ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5862fa2e-d8f5-48e6-ea2a-05f26062f63c"
      },
      "source": [
        "#重新分割数据集\n",
        "data_X=train_data[['CRIM','ZN','INDUS','CHAS','NOX','RM','AGE','DIS','RAD','TAX','PTRATIO','B','LSTAT']]\n",
        "data_y=train_data[['MEDV']]\n",
        "X_train,X_test,y_train,y_test = train_test_split(data_X, data_y, test_size = 0.1, random_state = None)\n",
        "#查看训练数据集与测试数据集的大小，检测数据集分割是否有问题\n",
        "print(X_train.shape)\n",
        "print(X_test.shape)\n",
        "print(y_train.shape)\n",
        "print(y_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(455, 13)\n",
            "(51, 13)\n",
            "(455, 1)\n",
            "(51, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "da98bb4a-ed5c-47a6-9727-4b6271a27d71",
        "id": "93ol-QiZyKXR",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ae6baa92-90b8-444b-9d2e-3e2bfc4af624"
      },
      "source": [
        "#生成模型和训练模型\n",
        "LR=LinearRegression().fit(X_train,y_train) \n",
        "print(model.coef_)     #输出每个特征的权值，系数，有些模型没有系数（如k近邻）\n",
        "print(model.intercept_) #与y轴交点，即截距\n",
        "y_pred = LR.predict(X_test)                        #模型预测\n",
        "print(\"\\n\")\n",
        "\n",
        "#Ridge(alpha=0.01).fit(X_train,y_train)\n",
        "#print(model.coef_)     #系数，有些模型没有系数（如k近邻）\n",
        "#print(model.intercept_) #与y轴交点，即截距\n",
        "#print(\"\\n\")\n",
        "\n",
        "#Lasso().fit(X_train,y_train)\n",
        "#print(model.coef_)     #系数，有些模型没有系数（如k近邻）\n",
        "#print(model.intercept_) #与y轴交点，即截距\n",
        "#print(\"\\n\")\n",
        "\n",
        "#ElasticNet().fit(X_train,y_train)\n",
        "#print(model.coef_)     #系数，有些模型没有系数（如k近邻）\n",
        "#print(model.intercept_) #与y轴交点，即截距\n",
        "#print(\"\\n\")\n",
        "\n",
        "#KNeighborsRegressor().fit(X_train,y_train)\n",
        "#print(model.coef_)     #系数，有些模型没有系数（如k近邻）\n",
        "#print(model.intercept_) #与y轴交点，即截距\n",
        "#print(\"\\n\")\n",
        "\n",
        "#DecisionTreeRegressor().fit(X_train,y_train)\n",
        "#print(model.coef_)     #系数，有些模型没有系数（如k近邻）\n",
        "#print(model.intercept_) #与y轴交点，即截距\n",
        "#print(\"\\n\")\n",
        "\n",
        "#SVR().fit(X_train,y_train)\n",
        "#print(model.coef_)     #系数，有些模型没有系数（如k近邻）\n",
        "#print(model.intercept_) #与y轴交点，即截距\n",
        "#print(\"\\n\")\n",
        "\n",
        "#AdaBoostRegressor().fit(X_train,y_train)\n",
        "#print(model.coef_)     #系数，有些模型没有系数（如k近邻）\n",
        "#print(model.intercept_) #与y轴交点，即截距\n",
        "#print(\"\\n\")\n",
        "\n",
        "#GradientBoostingRegressor().fit(X_train,y_train)\n",
        "#print(model.coef_)     #系数，有些模型没有系数（如k近邻）\n",
        "#print(model.intercept_) #与y轴交点，即截距\n",
        "#print(\"\\n\")\n",
        "\n",
        "#RandomForestRegressor().fit(X_train,y_train)\n",
        "#print(model.coef_)     #系数，有些模型没有系数（如k近邻）\n",
        "#print(model.intercept_) #与y轴交点，即截距\n",
        "#print(\"\\n\")\n",
        "\n",
        "#ExtraTreesRegressor().fit(X_train,y_train)\n",
        "#print(model.coef_)     #系数，有些模型没有系数（如k近邻）\n",
        "#print(model.intercept_) #与y轴交点，即截距\n",
        "#print(\"\\n\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[-0.00857818  4.74311082 -1.09633069 -0.54084703]]\n",
            "[19.93395701]\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "49a5daaf-215f-4e8f-8ae7-ff901bf1e8b6",
        "id": "vIt0YP72yKXR"
      },
      "source": [
        "## 绘制预测结果\n",
        "由于可视化能给我们带来最直观的认知，所以下面将通过可视化的方法，来展示回归模型预测的效果。通过以下代码，可以得到针对波士顿房价数据集，预测房价和实际房价之间的对比图。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "3e316fd6-2b42-4984-a274-57f1334e8a06",
        "id": "UcTlNSIQyKXR",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 305
        },
        "outputId": "909ec1f3-25c0-4e21-c220-9ec47e1f3087"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "\n",
        "plt.scatter(y_test, y_predict)\n",
        "plt.xlabel(\"Price: $Y_i$\")\n",
        "plt.ylabel(\"Predicted prices: $\\hat{Y}_i$\")\n",
        "plt.title(\"Prices vs Predicted prices: $Y_i$ vs $\\hat{Y}_i$\")\n",
        "plt.grid()\n",
        "\n",
        "x = np.arange(0, 50)\n",
        "y = x\n",
        "\n",
        "plt.grid()\n",
        "plt.plot(x, y, color = 'red', lw = 4)\n",
        "plt.text(30,40, \"predict line\")\n",
        "plt.savefig (\"price.eps\")"
      ],
      "execution_count": null,
      "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": {
        "uuid": "72fe1e27-2295-4482-96c3-734f646f0516",
        "id": "I_Qis3CqyKXS"
      },
      "source": [
        "在代码功能实现层面，我们知道，如果预测房价和实际房价一致的话，那么所有的数据点都应该汇集在 y=x 这条线上，但这并不是现实，于是可以看到，除了少数点，大部分点散落在 y=x 附近，大趋势说明预测的结果还不错"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "5da5b4d4-571c-40da-b066-4d3b9a32af07",
        "id": "ElidFk_YyKXS"
      },
      "source": [
        "除了可以利用常规的 Matplotlib 绘制图形，还可以利用前面学习的 Seaborn 绘制更加“炫丽”的线性回归模型趋势图，这时就要用到 lmplot( ) 方法了。该方法用以绘制回归趋势图，描述线性关系，拟合数据集回归模型。hue、col、row 参数可用来控制绘图变量"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GxwJCNFjyKXS",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2c87e675-92cf-4931-d39f-3fbbe039e1d2"
      },
      "source": [
        "print(\"w0 = \", LR.intercept_)    #输出截距\n",
        "print (\"W = \",  LR. coef_)     #输出每个特征的权值"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "w0 =  [38.10606041]\n",
            "W =  [[-1.22077718e-01  4.41268275e-02  3.41761855e-03  2.78687531e+00\n",
            "  -1.65612168e+01  3.62959650e+00 -3.62713611e-03 -1.47823865e+00\n",
            "   3.10442717e-01 -1.35034935e-02 -9.74000918e-01  8.68368658e-03\n",
            "  -5.02132846e-01]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5Bqnm3cfyKXS"
      },
      "source": [
        "#import seaborn as sns\n",
        "#sns.set_style(\"whitegrid\")\n",
        "#tips = sns.load_dataset(\"tips\") #载入自带数据\n",
        "#print(tips)\n",
        "#研究小费tips与总消费金额total_bill在吸烟与不吸烟人之间的关系\n",
        "#g = sns.lmplot(x=\"total_bill\", y=\"tip\", hue=\"smoker\", data=tips,palette=\"Set1\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hick_aPZyKXT",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3362223f-1d40-424a-eedc-8434f8037fca"
      },
      "source": [
        "print(y_test.shape)\n",
        "print(y_pred.shape)\n",
        "#g = sns.lmplot(x=\"MEDV\", y=\"tip\", hue=\"smoker\", data=y_test,palette=\"Set1\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(51, 1)\n",
            "(51, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WZGBHPlKyKXT",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f4d4ec1f-7fb0-4fb1-99b1-7ae31a5ac58a"
      },
      "source": [
        "from sklearn import metrics\n",
        "mse = metrics.mean_squared_error(y_test, y_pred)\n",
        "print(mse)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "38.16545584760873\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "ae4efb8d-9fb8-495a-912f-96f6c3572bf6",
        "id": "t17BMaZ_yKXT"
      },
      "source": [
        "df = pd.DataFrame()\n",
        "#df[{'实际房价': y_test, '预测房价': y_pred},index=[0]]\n",
        "\n",
        "lists = np.array(y_test)\n",
        "lists = list(lists)\n",
        "labels = y_pred\n",
        "\n",
        "df = pd.DataFrame()\n",
        "df[\"list\"] = lists\n",
        "df[\"label\"] = labels\n",
        " \n",
        "#print(df)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0_PtUuTwyKXU"
      },
      "source": [
        "#sns.set_style(\"whitegrid\")\n",
        "#真实房价与预测值的关系\n",
        "#g = sns.lmplot(x=\"list\", y=\"label\", data=df, palette=\"Set1\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "577247bd-422e-4aa8-b0c0-625138e7bec2",
        "id": "RC7ZHWFByKXU",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9c6fd4c0-5c10-4f81-9183-d20f8f311942"
      },
      "source": [
        "#Choosing the best model:GradientBoostingRegressor\n",
        "import random\n",
        "best_model = sklearn.ensemble.GradientBoostingRegressor(random_state=random.seed())\n",
        "best_model.fit(X_train, y_train)\n",
        "\n",
        "# make predictions using the modeI\n",
        "predictions = best_model.predict(X_test)\n",
        "print(\"[INFO] MSE : {}\".format(round(mean_squared_error(y_test, predictions), 3)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[INFO] MSE : 10.733\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/ensemble/_gb.py:1454: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "17740424-66c4-43b1-9bea-2cf09889c303",
        "id": "QqG9S66QyKXW"
      },
      "source": [
        "# plot between predictions and Y test\n",
        "x_axis = np.array(range(0, predictions.shape[0]))\n",
        "plt.plot(x_axis, predictions, linestyle=\"--\", marker= \"o\", alpha=0.7, color='r',label=\"predictions\")\n",
        "plt.plot(x_axis, y_test, linestyle=\"--\", marker=\"o\", alpha=0.7, color='g',label=\"Y_test\")\n",
        "plt.xlabel('Row number')\n",
        "plt.ylabel('PRICE')\n",
        "plt.title('Predictions vs Y_test')\n",
        "plt.legend(loc='lower right')\n",
        "plt.savefig(\"predictions_Vs_ytest.png\")\n",
        "plt.clf()\n",
        "plt.close()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0IqZPzrIyKXX"
      },
      "source": [
        "![predictions_Vs_ytest](/content/Boston_house_price/Demo/predictions_Vs_ytest.png)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BZdazhW305V0"
      },
      "source": [
        "![predictions_Vs_ytest](https://colab.research.google.com/9d5c24d8-e05c-458e-b6c3-077702e20a29)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "69de9132-0dd9-44e2-8d87-0099c10da58b",
        "id": "1BNfpDQ3yKXX",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 598
        },
        "outputId": "f04bb4c1-5088-4b68-82db-b53b3c219d0c"
      },
      "source": [
        "# plot model's feature importance\n",
        "feature_importance = best_model.feature_importances_\n",
        "feature_importance = 100.0 * (feature_importance / feature_importance.max())\n",
        "\n",
        "sorted_idx = np.argsort(feature_importance)\n",
        "pos = np.arange(sorted_idx.shape[0]) + .5\n",
        "corr_value = getCorrelatedFeature(corrmat['MEDV'],threshold)\n",
        "correlated_data = train_data[corr_value.index]\n",
        "                               \n",
        "plt.barh(pos,feature_importance[sorted_idx],align='center')\n",
        "plt.yticks(pos,correlated_data.columns[sorted_idx])\n",
        "plt.xlabel('Relative Importance')\n",
        "plt.title('Variable Importance')\n",
        "plt.savefig(\"feature_importance.png\")\n",
        "plt.clf()\n",
        "plt.close()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "IndexError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-47-6dbc187e81f5>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      9\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     10\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbarh\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpos\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mfeature_importance\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0msorted_idx\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0malign\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'center'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 11\u001b[0;31m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0myticks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpos\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mcorrelated_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0msorted_idx\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     12\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxlabel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'Relative Importance'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     13\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtitle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'Variable Importance'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/pandas/core/indexes/base.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   4112\u001b[0m             \u001b[0mkey\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0masarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mbool\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   4113\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 4114\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgetitem\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   4115\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mis_scalar\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresult\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   4116\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndim\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresult\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mIndexError\u001b[0m: index 1 is out of bounds for axis 0 with size 1"
          ]
        },
        {
          "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": {
        "uuid": "85948368-3dbc-4701-a534-702fb38fc0dc",
        "id": "IiRIGt8QyKXX"
      },
      "source": [
        "# step:\n",
        "\n",
        "1.gridsearchcv遍历每个模型的主要参数，求出每个模型的最佳参数。\n",
        "\n",
        "2.分别求出feature=3(只考虑'RM', 'PTRATIO', 'LSTAT' 3个features)与feature=13(考虑所有features)时各模型在最佳参数下的r2_Score效果，得出feature=3与13时各自的最佳模型。\n",
        "\n",
        "3.比较feature=3与13时各自最佳模型的r2_Score平均值与方差，得出全局最优模型。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "7daf6563-66ab-4f9d-9de9-379506dcd09a",
        "id": "eH8bVGN-yKXX"
      },
      "source": [
        "import pandas as pd \n",
        "import numpy as np\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "\n",
        "class EstimatorSelectionHelper:\n",
        "    \n",
        "    def __init__(self, models, params):\n",
        "        if not set (models.keys()).issubset(set(params.keys())):\n",
        "            missing_params = list(set(models.keys()) - set(params.keys()))\n",
        "            raise ValueError(\"Some estimators are missing parameters: %s\" % missing_params) \n",
        "        self.models = models\n",
        "        self.params = params \n",
        "        self.keys = models.keys() \n",
        "        self.grid_searches = {}\n",
        "\n",
        "\n",
        "    def fit(self, X, y, cv=3, n_jobs=3, verbose=1, scoring=None, refit=False):\n",
        "        for key in self.keys:\n",
        "            print(\"Running GridSearchCV for %s.\" % key)\n",
        "            model = self.models[key] \n",
        "            params = self.params[key]\n",
        "            gs = GridSearchCV(model, params, cv=cv, n_jobs=n_jobs,\n",
        "                              verbose=verbose, scoring=scoring, \n",
        "                              refit=refit, return_train_score=True)\n",
        "\n",
        "            gs.fit(X, y)\n",
        "            self.grid_searches[key] = gs\n",
        "      \n",
        "\n",
        "    def score_summary(self, sort_by='mean_score'):\n",
        "        def row (key, scores, params):\n",
        "          d ={\n",
        "              'estimator' : key,\n",
        "              'min_score' : min (scores),\n",
        "              'max_score' : max(scores),\n",
        "              'mean_score' : np.mean(scores),\n",
        "              'std_score' : np.std(scores),\n",
        "          }     \n",
        "          return pd.Series({**params, **d})\n",
        "        \n",
        "          rows =[]\n",
        "          for k in self.grid_searches: \n",
        "            print(k)\n",
        "            params = self.grid_searches[k].cv_results_['params']\n",
        "            scores =[]\n",
        "            for i in range(self.grid_searches[k].cv):\n",
        "              key = \"split{}_test_score\".format(i) \n",
        "              r = self.grid_searches[k].cv_results_[key]\n",
        "              scores.append(r.reshape(len(params), 1))\n",
        "            \n",
        "          all_scores = np.hstack(scores)\n",
        "          for p,s in zip(params, all_scores):\n",
        "             rows.append((row(k, s, p)))\n",
        "                                                         \n",
        "          df = pd.concat(rows, axis=1).T.sort_values([sort_by], ascending=False)\n",
        "                                                         \n",
        "          columns = ['estimator', 'min_scor','mean_score', 'max_score', 'std_score'] \n",
        "          columns = columns + [c for c in df.columns if c not in columns]\n",
        "            \n",
        "          return df[columns]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1H16NfdryKXY"
      },
      "source": [
        "# 评价模型表现\n",
        "在这个练习中，你将需要将所学到的内容整合，使用决策树算法训练一个模型。为了得出的是一个最优模型，你需要使用网格搜索法训练模型，以找到最佳的 'max_depth' 参数。你可以把'max_depth' 参数理解为决策树算法在做出预测前，允许其对数据提出问题的数量。决策树是监督学习算法中的一种。"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3X3kDhAxyKXY"
      },
      "source": [
        "## GridSearchCV 网格搜索和交叉验证\n",
        "GridSearchCV的名字其实可以拆分为两部分，GridSearch和CV，即网格搜索和交叉验证。这两个名字都非常好理解。网格搜索，搜索的是参数，即在指定的参数范围内，按步长依次调整参数，利用调整的参数训练学习器，从所有的参数中找到在验证集上精度最高的参数，这其实是一个训练和比较的过程。\n",
        "GridSearchCV可以保证在指定的参数范围内找到精度最高的参数，但是这也是网格搜索的缺陷所在，他要求遍历所有可能参数的组合，在面对大数据集和多参数的情况下，非常耗时。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "c1cb5dfa-8639-4f50-8b66-ad4f3b52c589",
        "id": "bMEKoQRAyKXY",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "44cf6f8b-3981-41d7-c500-ad1064a8407f"
      },
      "source": [
        "from sklearn.ensemble import ExtraTreesClassifier\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.ensemble import AdaBoostClassifier\n",
        "from sklearn.ensemble import GradientBoostingClassifier\n",
        "from sklearn.svm import SVC\n",
        "models1 = {\n",
        "    'SVR': SVR(),\n",
        "    'KNeighborsRegressor': KNeighborsRegressor(),\n",
        "    'DecisionTreeRegressor': DecisionTreeRegressor(),\n",
        "    'RandomForestRegressor': RandomForestRegressor(),\n",
        "    \"GradientBoostingRegressor\":GradientBoostingRegressor(),\n",
        "    \"AdaBoostRegressor\":AdaBoostRegressor(),\n",
        "    \"ElasticNet\":ElasticNet(),\n",
        "    \"ExtraTreesRegressor\":ExtraTreesRegressor()\n",
        "}\n",
        "\n",
        "params1 = {\n",
        "    'SVR': {'kernel':('linear', 'rbf'), 'C':[1, 2, 4], 'gamma':[0.125, 0.25, 0.5 ,1, 2, 4]},\n",
        "    'KNeighborsRegressor': {'weights': ['uniform', 'distance'],\n",
        "                             'n_neighbors': range(2,100)\n",
        "                             },\n",
        "    'DecisionTreeRegressor':  {'max_features': ['sqrt', 'log2', None],\n",
        "                             'max_depth': range(2,1000),\n",
        "                             },\n",
        "    'RandomForestRegressor': {\n",
        "    'min_samples_split': list((3,6,9)),'n_estimators':list((10,50,100))},\n",
        "    \"GradientBoostingRegressor\":{'n_estimators':[100], 'learning_rate': [0.1], 'max_depth':[6],'min_samples_leaf':[3], 'max_features':[3]},\n",
        "    \"AdaBoostRegressor\":{'n_estimators': [50, 100], 'learning_rate' : [0.01,0.05,0.1,0.3,1],'loss' : ['linear', 'square', 'exponential']},\n",
        "    \"ElasticNet\": {\"max_iter\": [1, 5, 10],\n",
        "                      \"alpha\": [0.0001, 0.001, 0.01, 0.1, 1, 10, 100],\n",
        "                      \"l1_ratio\": np.arange(0.0, 1.0, 0.1)},\n",
        "    \"ExtraTreesRegressor\":{\n",
        "        'max_features': range(1,4,1),\n",
        "    }\n",
        "    \n",
        "}\n",
        "\n",
        "helper1 = EstimatorSelectionHelper(models1, params1)\n",
        "helper1.fit(X_train, y_train, scoring='r2', n_jobs=2)\n",
        "helper1.score_summary(sort_by='max_score')\n",
        "print(\"\\r\\n\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for SVR.\n",
            "Fitting 3 folds for each of 36 candidates, totalling 108 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  46 tasks      | elapsed:   59.5s\n",
            "[Parallel(n_jobs=2)]: Done 108 out of 108 | elapsed:  3.0min finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for KNeighborsRegressor.\n",
            "Fitting 3 folds for each of 196 candidates, totalling 588 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done 540 tasks      | elapsed:    4.6s\n",
            "[Parallel(n_jobs=2)]: Done 588 out of 588 | elapsed:    5.1s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for DecisionTreeRegressor.\n",
            "Fitting 3 folds for each of 2994 candidates, totalling 8982 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done 956 tasks      | elapsed:    4.4s\n",
            "[Parallel(n_jobs=2)]: Done 5756 tasks      | elapsed:   25.5s\n",
            "[Parallel(n_jobs=2)]: Done 8982 out of 8982 | elapsed:   39.7s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for RandomForestRegressor.\n",
            "Fitting 3 folds for each of 9 candidates, totalling 27 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done  27 out of  27 | elapsed:    2.9s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done   3 out of   3 | elapsed:    0.2s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for GradientBoostingRegressor.\n",
            "Fitting 3 folds for each of 1 candidates, totalling 3 fits\n",
            "Running GridSearchCV for AdaBoostRegressor.\n",
            "Fitting 3 folds for each of 30 candidates, totalling 90 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done  87 out of  90 | elapsed:   10.6s remaining:    0.4s\n",
            "[Parallel(n_jobs=2)]: Done  90 out of  90 | elapsed:   10.8s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for ElasticNet.\n",
            "Fitting 3 folds for each of 210 candidates, totalling 630 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done 540 tasks      | elapsed:    2.6s\n",
            "[Parallel(n_jobs=2)]: Done 630 out of 630 | elapsed:    3.0s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for ExtraTreesRegressor.\n",
            "Fitting 3 folds for each of 3 candidates, totalling 9 fits\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done   9 out of   9 | elapsed:    1.0s finished\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "32e51077-abb3-40d9-8709-4e5d4fa46564",
        "id": "NpxLkOXryKXZ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1c70c23c-ee08-420c-eb97-e8f8c4014cad"
      },
      "source": [
        "from sklearn.ensemble import ExtraTreesClassifier\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.ensemble import AdaBoostClassifier\n",
        "from sklearn.ensemble import GradientBoostingClassifier\n",
        "from sklearn.svm import SVC\n",
        "Xall = train_data.drop(labels=['MEDV'],axis = 1)\n",
        "yall = train_data[['MEDV']]\n",
        "X_train_1,X_test_1,y_train_1,y_test_1 = train_test_split(Xall, yall, test_size = 0.1, random_state = 0)\n",
        "models1 = {\n",
        "    'SVR': SVR(),\n",
        "    'KNeighborsRegressor': KNeighborsRegressor(),\n",
        "    'DecisionTreeRegressor': DecisionTreeRegressor(),\n",
        "    'RandomForestRegressor': RandomForestRegressor(),\n",
        "    \"GradientBoostingRegressor\":GradientBoostingRegressor(),\n",
        "    \"AdaBoostRegressor\":AdaBoostRegressor(),\n",
        "    \"ElasticNet\":ElasticNet(),\n",
        "    \"ExtraTreesRegressor\":ExtraTreesRegressor()\n",
        "}\n",
        "\n",
        "params1 = {\n",
        "    'SVR': {'kernel':('linear', 'rbf'), 'C':[1, 2, 4], 'gamma':[0.125, 0.25, 0.5 ,1, 2, 4]},\n",
        "    'KNeighborsRegressor': {'weights': ['uniform', 'distance'],\n",
        "                             'n_neighbors': range(2,100)\n",
        "                             },\n",
        "    'DecisionTreeRegressor':  {'max_features': ['sqrt', 'log2', None],\n",
        "                             'max_depth': range(2,1000),\n",
        "                             },\n",
        "    'RandomForestRegressor': {\n",
        "    'min_samples_split': list((3,6,9)),'n_estimators':list((10,50,100))},\n",
        "    \"GradientBoostingRegressor\":{'n_estimators':[100], 'learning_rate': [0.1], 'max_depth':[6],'min_samples_leaf':[3], 'max_features':[3]},\n",
        "    \"AdaBoostRegressor\":{'n_estimators': [50, 100], 'learning_rate' : [0.01,0.05,0.1,0.3,1],'loss' : ['linear', 'square', 'exponential']},\n",
        "    \"ElasticNet\": {\"max_iter\": [1, 5, 10],\n",
        "                      \"alpha\": [0.0001, 0.001, 0.01, 0.1, 1, 10, 100],\n",
        "                      \"l1_ratio\": np.arange(0.0, 1.0, 0.1)},\n",
        "    \"ExtraTreesRegressor\":{\n",
        "        'max_features': range(1,4,1),\n",
        "    }\n",
        "    \n",
        "}\n",
        "\n",
        "helper2 = EstimatorSelectionHelper(models1, params1)\n",
        "helper2.fit(X_train_1, y_train_1, scoring='r2', n_jobs=2)\n",
        "helper2.score_summary(sort_by='max_score')\n",
        "print(\"\\r\\n\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for SVR.\n",
            "Fitting 3 folds for each of 36 candidates, totalling 108 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done  46 tasks      | elapsed:   48.0s\n",
            "[Parallel(n_jobs=2)]: Done 108 out of 108 | elapsed:  2.8min finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for KNeighborsRegressor.\n",
            "Fitting 3 folds for each of 196 candidates, totalling 588 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done 300 tasks      | elapsed:    2.9s\n",
            "[Parallel(n_jobs=2)]: Done 588 out of 588 | elapsed:    6.2s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for DecisionTreeRegressor.\n",
            "Fitting 3 folds for each of 2994 candidates, totalling 8982 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done 540 tasks      | elapsed:    3.2s\n",
            "[Parallel(n_jobs=2)]: Done 2940 tasks      | elapsed:   17.2s\n",
            "[Parallel(n_jobs=2)]: Done 6940 tasks      | elapsed:   41.0s\n",
            "[Parallel(n_jobs=2)]: Done 8982 out of 8982 | elapsed:   53.2s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for RandomForestRegressor.\n",
            "Fitting 3 folds for each of 9 candidates, totalling 27 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done  27 out of  27 | elapsed:    3.2s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done   3 out of   3 | elapsed:    0.2s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for GradientBoostingRegressor.\n",
            "Fitting 3 folds for each of 1 candidates, totalling 3 fits\n",
            "Running GridSearchCV for AdaBoostRegressor.\n",
            "Fitting 3 folds for each of 30 candidates, totalling 90 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done  87 out of  90 | elapsed:   11.2s remaining:    0.4s\n",
            "[Parallel(n_jobs=2)]: Done  90 out of  90 | elapsed:   11.5s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for ElasticNet.\n",
            "Fitting 3 folds for each of 210 candidates, totalling 630 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done 540 tasks      | elapsed:    3.3s\n",
            "[Parallel(n_jobs=2)]: Done 630 out of 630 | elapsed:    3.9s finished\n",
            "[Parallel(n_jobs=2)]: Using backend LokyBackend with 2 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Running GridSearchCV for ExtraTreesRegressor.\n",
            "Fitting 3 folds for each of 3 candidates, totalling 9 fits\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=2)]: Done   9 out of   9 | elapsed:    1.1s finished\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e7zH0vL4yKXZ"
      },
      "source": [
        "# 训练模型\n",
        "当我们用数据训练出一个模型，它现在就可用于对新的数据进行预测。在决策树回归函数中，模型已经学会对新输入的数据提问，并返回对目标变量的预测值。你可以用这个预测来获取数据未知目标变量的信息，这些数据必须是不包含在训练数据之内的。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "655b56a0-f91f-41a9-91de-08d48373f0af",
        "id": "1WWO41M2yKXa"
      },
      "source": [
        "from visuals import *\n",
        "import visuals as vs\n",
        "#import sys\n",
        "#sys.path.append('download/Boston_house_price/visuals.py')\n",
        "#python import模块时， 是在sys.path里按顺序查找的。sys.path是一个列表，里面以字符串的形式存储了许多路径。使用A.py文件中的函数需要先将他的文件路径放到sys.path中\n",
        "#from imp import reload\n",
        "#reload(PredictTrials)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "547332b3-a6d8-430a-8d63-780c36c2a5d8",
        "id": "FI2MFrQcyKXa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 475
        },
        "outputId": "fbd6b517-2154-40a4-8b03-5397887ac5cf"
      },
      "source": [
        "def performance_metric(y_true,y_predict):\n",
        "    #calcuates and return performance_score between true and predit based on metric chosen  \n",
        "    score = r2_score(y_true,y_predict)\n",
        "    return score\n",
        "\n",
        "def fit_model(X ,y):\n",
        "    #Performs grid search over the ' max_depth, parameter for a decision tree regressor trained on the input data [X, y].\n",
        "    \n",
        "    from sklearn.model_selection import ShuffleSplit\n",
        "    from sklearn.metrics import make_scorer\n",
        "    from sklearn.model_selection import GridSearchCV\n",
        "    #Create cross-validation sets from the training data\n",
        "    \n",
        "    cv_sets=ShuffleSplit(X.shape[0], test_size = 0.20, random_state = 0)\n",
        "\n",
        "    #Create a decision tree regressor object\n",
        "    regressor = GradientBoostingRegressor()\n",
        "    \n",
        "    #Create a dictionary for the parameter 'max_depth' with a range from 1 to 10\n",
        "    params = {'n_estimators':[100], 'learning_rate':[0.1], 'max.depth':[6], 'min_samples_leaf':[3], 'max_features':[3]}\n",
        "\n",
        "    #Transform 'performancejnetric' into a scoring function using 'make_scorer'\n",
        "    scoring_fnc = make_scorer(performance_metric)\n",
        "\n",
        "    \n",
        "    #Create the grid search object\n",
        "    grid = GridSearchCV(estimator=regressor, param_grid=params, scoring=scoring_fnc, cv=cv_sets)\n",
        "    \n",
        "    #Fit the grid search object to the data to compute the optimal model\n",
        "    grid = grid.fit(X, y)\n",
        "              \n",
        "    #Return the optimal model after fitting the data\n",
        "    return grid.best_estimator_\n",
        "\n",
        "vs.PredictTrials(Xall, yall, fit_model, Xall[:1].values)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-79-3c975e0bc604>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     33\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mgrid\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbest_estimator_\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     34\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 35\u001b[0;31m \u001b[0mvs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mPredictTrials\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mXall\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0myall\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfit_model\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mXall\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;32m/content/Boston_house_price/Demo/visuals.py\u001b[0m in \u001b[0;36mPredictTrials\u001b[0;34m(X, y, fitter, data)\u001b[0m\n\u001b[1;32m    131\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    132\u001b[0m         \u001b[0;31m# Fit the data\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 133\u001b[0;31m         \u001b[0mreg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfitter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    134\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    135\u001b[0m         \u001b[0;31m# Make a prediction\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<ipython-input-79-3c975e0bc604>\u001b[0m in \u001b[0;36mfit_model\u001b[0;34m(X, y)\u001b[0m\n\u001b[1;32m     28\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     29\u001b[0m     \u001b[0;31m#Fit the grid search object to the data to compute the optimal model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 30\u001b[0;31m     \u001b[0mgrid\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgrid\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     31\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     32\u001b[0m     \u001b[0;31m#Return the optimal model after fitting the data\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, X, y, groups, **fit_params)\u001b[0m\n\u001b[1;32m    708\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mresults\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    709\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 710\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_run_search\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevaluate_candidates\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    711\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    712\u001b[0m         \u001b[0;31m# For multi-metric evaluation, store the best_index_, best_params_ and\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36m_run_search\u001b[0;34m(self, evaluate_candidates)\u001b[0m\n\u001b[1;32m   1149\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_run_search\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevaluate_candidates\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1150\u001b[0m         \u001b[0;34m\"\"\"Search all candidates in param_grid\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1151\u001b[0;31m         \u001b[0mevaluate_candidates\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mParameterGrid\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mparam_grid\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1152\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1153\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/sklearn/model_selection/_search.py\u001b[0m in \u001b[0;36mevaluate_candidates\u001b[0;34m(candidate_params)\u001b[0m\n\u001b[1;32m    687\u001b[0m                                \u001b[0;32mfor\u001b[0m \u001b[0mparameters\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    688\u001b[0m                                in product(candidate_params,\n\u001b[0;32m--> 689\u001b[0;31m                                           cv.split(X, y, groups)))\n\u001b[0m\u001b[1;32m    690\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    691\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, iterable)\u001b[0m\n\u001b[1;32m   1039\u001b[0m             \u001b[0;31m# remaining jobs.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1040\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_iterating\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1041\u001b[0;31m             \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdispatch_one_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1042\u001b[0m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_iterating\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_original_iterator\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1043\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/joblib/parallel.py\u001b[0m in \u001b[0;36mdispatch_one_batch\u001b[0;34m(self, iterator)\u001b[0m\n\u001b[1;32m    857\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    858\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 859\u001b[0;31m                 \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dispatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtasks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    860\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    861\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m_dispatch\u001b[0;34m(self, batch)\u001b[0m\n\u001b[1;32m    775\u001b[0m         \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_lock\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    776\u001b[0m             \u001b[0mjob_idx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_jobs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 777\u001b[0;31m             \u001b[0mjob\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply_async\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    778\u001b[0m             \u001b[0;31m# A job can complete so quickly than its callback is\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    779\u001b[0m             \u001b[0;31m# called before we get here, causing self._jobs to\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/joblib/_parallel_backends.py\u001b[0m in \u001b[0;36mapply_async\u001b[0;34m(self, func, callback)\u001b[0m\n\u001b[1;32m    206\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mapply_async\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    207\u001b[0m         \u001b[0;34m\"\"\"Schedule a func to be run\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 208\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mImmediateResult\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    209\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mcallback\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    210\u001b[0m             \u001b[0mcallback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresult\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/joblib/_parallel_backends.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, batch)\u001b[0m\n\u001b[1;32m    570\u001b[0m         \u001b[0;31m# Don't delay the application, to avoid keeping the input\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    571\u001b[0m         \u001b[0;31m# arguments in memory\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 572\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    573\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    574\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    261\u001b[0m         \u001b[0;32mwith\u001b[0m \u001b[0mparallel_backend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_n_jobs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    262\u001b[0m             return [func(*args, **kwargs)\n\u001b[0;32m--> 263\u001b[0;31m                     for func, args, kwargs in self.items]\n\u001b[0m\u001b[1;32m    264\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    265\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__reduce__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/joblib/parallel.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    261\u001b[0m         \u001b[0;32mwith\u001b[0m \u001b[0mparallel_backend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backend\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_jobs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_n_jobs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    262\u001b[0m             return [func(*args, **kwargs)\n\u001b[0;32m--> 263\u001b[0;31m                     for func, args, kwargs in self.items]\n\u001b[0m\u001b[1;32m    264\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    265\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__reduce__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/sklearn/model_selection/_validation.py\u001b[0m in \u001b[0;36m_fit_and_score\u001b[0;34m(estimator, X, y, scorer, train, test, verbose, parameters, fit_params, return_train_score, return_parameters, return_n_test_samples, return_times, return_estimator, error_score)\u001b[0m\n\u001b[1;32m    502\u001b[0m             \u001b[0mcloned_parameters\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mclone\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msafe\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    503\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 504\u001b[0;31m         \u001b[0mestimator\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mestimator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_params\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mcloned_parameters\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    505\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    506\u001b[0m     \u001b[0mstart_time\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/sklearn/base.py\u001b[0m in \u001b[0;36mset_params\u001b[0;34m(self, **params)\u001b[0m\n\u001b[1;32m    234\u001b[0m                                  \u001b[0;34m'Check the list of available parameters '\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    235\u001b[0m                                  \u001b[0;34m'with `estimator.get_params().keys()`.'\u001b[0m \u001b[0;34m%\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 236\u001b[0;31m                                  (key, self))\n\u001b[0m\u001b[1;32m    237\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    238\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mdelim\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: Invalid parameter max.depth for estimator GradientBoostingRegressor(alpha=0.9, ccp_alpha=0.0, criterion='friedman_mse',\n                          init=None, learning_rate=0.1, loss='ls', max_depth=3,\n                          max_features=None, max_leaf_nodes=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=100,\n                          n_iter_no_change=None, presort='deprecated',\n                          random_state=None, subsample=1.0, tol=0.0001,\n                          validation_fraction=0.1, verbose=0, warm_start=False). Check the list of available parameters with `estimator.get_params().keys()`."
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zB5RfdoKyKXa"
      },
      "source": [
        "## 最优模型\n",
        "将决策树回归函数代入训练数据的集合，以得到最优化的模型。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oQlTR3aSyKXa",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "029f0825-7b11-46cd-f07c-3bc8443e48bd"
      },
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "# 3.特征工程-标准化\n",
        "transfer = StandardScaler()\n",
        "x_train = transfer.fit_transform(X_train)\n",
        "x_test = transfer.fit_transform(X_test)\n",
        "\n",
        "# 4.机器学习-线性回归(岭回归)【重要】\n",
        "# 4.1 创建模型 实例化估计器\n",
        "estimator = Ridge(alpha=1)#(alpha表示正则的程度）\n",
        "# 4.2 训练模型 fit 正规方程计算得到最优可训练参数\n",
        "estimator.fit(X_train, y_train)\n",
        "\n",
        "#保存模型\n",
        "joblib.dump(estimator,'./model.pkl')#注意用文本打开是错误的，只能加载！！\n",
        "\n",
        "# 5.模型评估\n",
        "# 5.1 获取系数等值\n",
        "y_predict = estimator.predict(X_test)\n",
        "print(\"预测值为:\\n\", y_predict)\n",
        "print(\"模型中的系数为:\\n\", estimator.coef_)\n",
        "print(\"模型中的偏置为:\\n\", estimator.intercept_)\n",
        "# 5.2 评价\n",
        "# 均方误差\n",
        "error = mean_squared_error(y_test, y_predict)\n",
        "print(\"误差为:\\n\", error)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "预测值为:\n",
            " [15.41474181 27.67385814 38.91096341 17.93850884 30.56432352 38.11959508\n",
            " 25.46426373 10.06599298 14.05346064 31.91178993 28.66252764 19.6186771\n",
            " 13.79776964 31.10106212 16.98103507 21.32689071 20.75609768 37.90251373\n",
            " 18.17275623 15.17723806 13.83741404 24.25234819 32.97306654 39.02298372\n",
            " 31.06539273 22.27774084 18.02434424 21.07951621 23.44156002 20.20969863\n",
            " 25.8758938  33.89326731 15.24125427 21.86549101 23.55451403 34.90492937\n",
            " 29.06373689 18.70135244 14.0201236  38.03364524 30.93213893 18.62584975\n",
            " 19.52976164 39.85791297  7.58525952 27.02562925 19.76925618 25.58930666\n",
            " 15.66646037 25.90725614 34.57589276 24.15257841 20.0943251  12.49389348\n",
            " 23.97640411 18.39910689 10.85719625  2.28168721 32.12428378 13.71626515\n",
            " 20.20688268 22.89727354 21.18323798 18.12016194 24.74959609 26.87247563\n",
            " 25.93490528 21.60301644 24.4940454  30.25172932 20.2314519  28.42329824\n",
            " 15.54882444 24.7681247  20.01515427 17.9731661  17.75043632 16.51074495\n",
            " 28.99782675 -1.19500063 32.33943694 13.05268241 25.15397935 20.58169784\n",
            " 14.170428   23.20931397 18.76986687 24.70365612 19.5171153  31.8981448\n",
            " 15.92175608 34.2338877  22.7820955  25.51279291 31.24324375 28.72677262\n",
            " 18.61781459 31.39275534 28.00034314 36.14219225 16.66155801 16.00413736]\n",
            "模型中的系数为:\n",
            " [-9.61349688e-02  4.78865533e-02  3.89249382e-02  1.08839162e+00\n",
            " -8.54108900e+00  4.38347501e+00 -7.52222105e-03 -1.26565701e+00\n",
            "  2.67244306e-01 -1.32867762e-02 -8.35013825e-01  7.53577625e-03\n",
            " -5.51499341e-01]\n",
            "模型中的偏置为:\n",
            " 26.888573080527635\n",
            "误差为:\n",
            " 28.93216989681372\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dfPsCh1LyKXb"
      },
      "source": [
        "## 预测销售价格\n",
        "使用你优化的模型来为每位客户的房屋价值做出预测。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eshuw_RAyKXd"
      },
      "source": [
        "def fit_model(X, y):\n",
        "    # Performs grid search over the 'max_depth' parameter for a \n",
        "    #  decision tree regressor trained on the input data [X, y].\n",
        "\n",
        "    # Create cross-validation sets from the training data\n",
        "    # ShuffleSplit works iteratively compared to KFOLD\n",
        "    # It saves computation time when your dataset grows\n",
        "    # X.shape[0] is the total number of elements\n",
        "    # n_iter is the number of re-shuffling & splitting iterations.\n",
        "    cv_sets = ShuffleSplit(X.shape[0], test_size = 0.20, random_state = 0)\n",
        "\n",
        "    # TODO: Create a decision tree regressor object\n",
        "    # Instantiate\n",
        "    regressor = DecisionTreeRegressor(random_state=0)\n",
        "\n",
        "    # TODO: Create a dictionary for the parameter 'max_depth' with a range from 1 to 10\n",
        "    dt_range = range(1, 11)\n",
        "    params = dict(max_depth=dt_range)\n",
        "\n",
        "    # TODO: Transform 'performance_metric' into a scoring function using 'make_scorer' \n",
        "    # We initially created performance_metric using R2_score\n",
        "    scoring_fnc = make_scorer(performance_metric)\n",
        "\n",
        "    # TODO: Create the grid search object\n",
        "    # You would realize we manually created each, including scoring_func using R^2\n",
        "    grid = GridSearchCV(regressor, params, cv=cv_sets, scoring=scoring_fnc)\n",
        "\n",
        "    # Fit the grid search object to the data to compute the optimal model\n",
        "    grid = grid.fit(X, y)\n",
        "\n",
        "    # Return the optimal model after fitting the data\n",
        "    return grid.best_estimator_"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nD_eyAPVyKXb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "57075d05-9626-4982-de5c-1bf504a6a1d3"
      },
      "source": [
        "# TODO: Import 'train_test_split'\n",
        "train_data=pd.read_csv('bostonh.csv',sep='\\s+',encoding='utf-8')\n",
        "from sklearn.model_selection import train_test_split\n",
        "prices = train_data['MEDV']#TODO：提取SalePrice作为labels\n",
        "features = train_data.drop(['MEDV'], axis=1)#TODO：提取除了SalePrice以外的特征赋值为features\n",
        "\n",
        "# TODO: Shuffle and split the data into training and testing subsets\n",
        "X_train, X_test, y_train, y_test = train_test_split(features, prices, \n",
        "                                                    train_size=0.8, random_state=123)\n",
        "# Success\n",
        "print(\"Training and testing split was successful.\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training and testing split was successful.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h5CaiqXLyKXb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8b2a44b6-3848-4390-96a1-07fc9a076a21"
      },
      "source": [
        "from sklearn.metrics import make_scorer\n",
        "# Fit the training data to the model using grid search\n",
        "reg = fit_model(X_train, y_train)\n",
        "\n",
        "# Produce the value for 'max_depth'\n",
        "print(\"Parameter 'max_depth' is {} for the optimal model.\".format(reg.get_params()['max_depth']))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Parameter 'max_depth' is 5 for the optimal model.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "E4LKtu-SyKXc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 132
        },
        "outputId": "0dbf32dd-11b9-4518-a9b9-d89bb3ad2302"
      },
      "source": [
        "# Produce a matrix for client data\n",
        "client_data = [[5, 17, 15], # Client 1\n",
        "               [4, 32, 22], # Client 2\n",
        "               [8, 3, 12]]  # Client 3\n",
        "\n",
        "# Show predictions\n",
        "for i, price in enumerate(reg.predict(client_data)):\n",
        "    print(\"Predicted selling price for Client {}'s home: ${:,.2f}\".format(i+1, price))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "SyntaxError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;36m  File \u001b[0;32m\"<ipython-input-90-101e3df801a4>\"\u001b[0;36m, line \u001b[0;32m8\u001b[0m\n\u001b[0;31m    print(\"Predicted selling price for Client {}'s home: ${:,.2f}\".format(i+1, 'MEDV')\u001b[0m\n\u001b[0m                                                                                      ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m unexpected EOF while parsing\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rSkfcT5nyKXc"
      },
      "source": [
        " 我会建议客户按照上面的建议去销售，从房间数量，贫困指数，师生比例来看，建议的价格是合理的。"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "idRsqUz4yKXc"
      },
      "source": [
        "# 模型的鲁棒性\n",
        "一个最优的模型不一定是一个健壮模型。有的时候模型会过于复杂或者过于简单，以致于难以泛化新增添的数据；有的时候模型采用的学习算法并不适用于特定的数据结构；有的时候样本本身可能有太多噪点或样本过少，使得模型无法准确地预测目标变量。这些情况下我们会说模型是欠拟合的。执行下方区域中的代码，采用不同的训练和测试集执行 fit_model 函数10次。注意观察对一个特定的客户来说，预测是如何随训练数据的变化而变化的。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "F6yx7qZryKXd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 451
        },
        "outputId": "b4a1f726-e288-4d06-e792-faf4bb5dba1e"
      },
      "source": [
        "# 请先注释掉 fit_model 函数里的所有 print 语句\n",
        "from sklearn.metrics import make_scorer\n",
        "vs.PredictTrials(Xall, yall, fit_model, Xall[:1].values)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Trial 1: $23.85\n",
            "Trial 2: $25.22\n",
            "Trial 3: $24.85\n",
            "Trial 4: $27.02\n",
            "Trial 5: $26.63\n",
            "Trial 6: $30.55\n",
            "Trial 7: $29.97\n",
            "Trial 8: $27.16\n",
            "Trial 9: $30.38\n",
            "Trial 10: $22.69\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-86-e34dfc8c0926>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# 请先注释掉 fit_model 函数里的所有 print 语句\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0msklearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmetrics\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mmake_scorer\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mvs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mPredictTrials\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mXall\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0myall\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfit_model\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mXall\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;32m/content/Boston_house_price/Demo/visuals.py\u001b[0m in \u001b[0;36mPredictTrials\u001b[0;34m(X, y, fitter, data)\u001b[0m\n\u001b[1;32m    141\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    142\u001b[0m     \u001b[0;31m# Display price range\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 143\u001b[0;31m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"\\nRange in prices: ${:,.2f}\"\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mMEDV\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mmin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mMEDV\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m: name 'MEDV' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bMaX4nt-yKXd"
      },
      "source": [
        "## 决策树基准测试\n",
        "R2 大概在0.7左右"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SCr0uJI_yKXd",
        "outputId": "e3ff6765-523a-4d7f-d48c-154c39a5d6e1"
      },
      "source": [
        "vs.ModelComplexity(X_train, y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 504x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TnBOc51kyKXe",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f8aabd0f-0ff4-4dd5-9191-6bb86f48cca8"
      },
      "source": [
        "from sklearn.ensemble import ExtraTreesRegressor\n",
        "\n",
        "cv_sets = ShuffleSplit(X_train.shape[0], test_size=0.20, random_state=0)\n",
        "scoring_fnc = make_scorer(performance_metric)\n",
        "\n",
        "params = {'n_estimators': range(20, 40, 3)}\n",
        "grid = GridSearchCV(ExtraTreesRegressor(random_state=0), params, scoring_fnc, cv=cv_sets)\n",
        "grid.fit(X_train, y_train)\n",
        "print(grid.best_score_)\n",
        "reg = grid.best_estimator_\n",
        "reg.get_params()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.8809001016534717\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'bootstrap': False,\n",
              " 'ccp_alpha': 0.0,\n",
              " 'criterion': 'mse',\n",
              " 'max_depth': None,\n",
              " 'max_features': 'auto',\n",
              " 'max_leaf_nodes': None,\n",
              " 'max_samples': None,\n",
              " 'min_impurity_decrease': 0.0,\n",
              " 'min_impurity_split': None,\n",
              " 'min_samples_leaf': 1,\n",
              " 'min_samples_split': 2,\n",
              " 'min_weight_fraction_leaf': 0.0,\n",
              " 'n_estimators': 38,\n",
              " 'n_jobs': None,\n",
              " 'oob_score': False,\n",
              " 'random_state': 0,\n",
              " 'verbose': 0,\n",
              " 'warm_start': False}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 91
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9iCrqlJGyKXe"
      },
      "source": [
        "# 实用性探讨\n",
        "简单地讨论一下你建构的模型能否在现实世界中使用？\n",
        "\n",
        "尽管此模型最终预测效果良好，但在尝试使用给定参数预测给定新数据点的房价时，在实际环境中使用此模型之前，我们需要解决一些缺陷：\n",
        "首先，以上预测使用的是1978年房价数据。由于通胀、动乱等政治经济社会因素的影响，若将该模型的特征用于如今的房价预测可能效果不会很显著，\n",
        "也许有一些与如今房价更紧密的特征（如：与学校的距离等）未被纳入此模型。\n",
        "其次，一个社区内的房价也可能有很大差别（美国有不少贫民窟紧挨富人区），而我们的预测模型有将同一社区房价同质化的缺点。\n",
        "再者，在城市中收集的数据不适用于农村城市，因为城市和农村都具有不同的特征，每个特征的值也不同。（总之，统计数据的采集过程可能有诸多细节有待优化。）\n",
        "\n",
        "可以从这几个角度分析：\n",
        "\n",
        "1978年所采集的数据，在今天是否仍然适用？\n",
        "数据中呈现的特征是否足够描述一个房屋？\n",
        "模型是否足够健壮来保证预测的一致性？\n",
        "在波士顿这样的大都市采集的数据，能否应用在其它乡镇地区？\n",
        "\n",
        "答案:\n",
        "1987年的数据在今天肯定是不适用了，但是在学习机器学习的目的来看，还是十分有意义的。\n",
        "数据中呈现的特征已经尽可能去描述房屋的各项指标了，但是肯定是不够的，最终客户还是有必要去看房。\n",
        "模型的健壮性还是不够的，经过十次训练，价格的极差竟然高达 $69,044.61，是房屋平均价格的17%。因此模型还有待改善。"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "uuid": "2d0f0a96-8582-4428-8942-1c5a0402c9a5",
        "id": "943fVu_tyKXe"
      },
      "source": [
        "## 相对平方误差（Relative Squared Error，RSE）\n",
        "## 相对绝对误差（Relative Absolute Error，RAE)\n",
        "## 决定系数 (Coefficient of Determination)\n",
        "\n",
        "事实上，当threshold取0.5时线性回归模型的rse、rae、r2综合得分较低。\n",
        "r2_score——拟合优度，一种回归评价指标，其含义是也是解释回归模型的方差得分，其值取值范围是[0,1]，越接近于1说明自变量越能解释因变量的方差变化，值越小则说明效果越差。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "8d02068a-c487-464c-90d4-83ae39a3bc21",
        "id": "_bKM619EyKXf"
      },
      "source": [
        "#可视化训练过程中的损失值\n",
        "# 迭代轮次\n",
        "train_epochs = 50 \n",
        "\n",
        "# 学习率\n",
        "learning_rate = 0.01 \n",
        "\n",
        "loss_list = [] # 用于保存loss值的列表\n",
        "\n",
        "#data_X=train_data[['CRIM','ZN','INDUS','CHAS','NOX','RM','AGE','DIS','RAD','TAX','PTRATIO','B','LSTAT']]\n",
        "#data_y=train_data[['MEDV']]\n",
        "#X_train,X_test,y_train,y_test = train_test_split(data_X, data_y, test_size = 0.1, random_state = None)\n",
        "\n",
        "for epoch in range (train_epochs):\n",
        "    loss_sum = 0.0\n",
        "    for xs, ys in zip(data_X, data_y):   \n",
        "\n",
        "        xs = xs.reshape(1,12)\n",
        "        ys = ys.reshape(1,1)\n",
        "\n",
        "        loss = sess.run([optimizer,loss_function], feed_dict={x: xs, y: ys}) \n",
        "\n",
        "        loss_sum = loss_sum + loss\n",
        "        \n",
        "        #loss_list.append(loss) # 每步添加一次\n",
        "    \n",
        "    # 打乱数据顺序\n",
        "    data_X, data_y = shuffle(data_X, data_y)\n",
        "    \n",
        "    b0temp=b.eval(session=sess)\n",
        "    w0temp=w.eval(session=sess)\n",
        "    loss_average = loss_sum/len(y_data)\n",
        "    \n",
        "    loss_list.append(loss) # 每轮添加一次\n",
        "    \n",
        "    print(\"epoch=\", epoch+1,\"loss=\", loss_average,\"b=\", b0temp,\"w=\", w0temp )"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5wSQhnljyKXf"
      },
      "source": [
        "# 可以用于训练外部模型"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "d20c3c23-84fa-453d-9110-53636e0870ce",
        "id": "ZOGiG2xhyKXf"
      },
      "source": [
        "!python3 Boston_house_price.py"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xzoEYujjyKXf"
      },
      "source": [
        "# TODO: Import 'make_scorer', 'DecisionTreeRegressor', and 'GridSearchCV'\n",
        "from sklearn.tree import DecisionTreeRegressor\n",
        "from sklearn.metrics import make_scorer\n",
        "from sklearn.grid_search import GridSearchCV\n",
        "       \n",
        "def fit_model(X, y):\n",
        "    \"\"\" Performs grid search over the 'max_depth' parameter for a \n",
        "        decision tree regressor trained on the input data [X, y]. \"\"\"\n",
        "    \n",
        "    from sklearn.model_selection import ShuffleSplit\n",
        "    from sklearn.metrics import make_scorer\n",
        "    # Create cross-validation sets from the training data\n",
        "          \n",
        "    cv_sets = ShuffleSplit(X.shape[0], test_size = 0.20, random_state = 0)\n",
        "    print (cv_sets)\n",
        "    \n",
        "    # TODO: Create a decision tree regressor object\n",
        "    regressor=GradientBoostingRegressor()\n",
        "\n",
        "    # TODO: Create a dictionary for the parameter 'max_depth' with a range from 1 to 10\n",
        "    params = {' n_estimators, : [100], * learning_rate’： [0.1], ' max.depth' : [6],' min_samples_leaf, : [3],}\n",
        "\n",
        "    # TODO: Transform 'performance_metric' into a scoring function using 'make_scorer' \n",
        "    scoring_fnc = make_scorer(performance_metric)\n",
        "\n",
        "    # TODO: Create the grid search object\n",
        "    grid = GridSearchCV(estimator=regressor, param_grid=params, scoring=scoring_fnc, cv=cv_sets)\n",
        "\n",
        "    # Fit the grid search object to the data to compute the optimal model\n",
        "    grid = grid.fit(X, y)\n",
        "\n",
        "    # Return the optimal model after fitting the data\n",
        "    return grid.best_estimator_"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "uuid": "ea78ab4f-09b7-4745-95b0-aa167c816eae",
        "id": "a3TI3768yKXf",
        "outputId": "83d96cd5-c3e6-482d-ba13-5caa93b18dfc"
      },
      "source": [
        "!python3 train.py"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Traceback (most recent call last):\n",
            "  File \"train.py\", line 121, in <module>\n",
            "    traing_data, test_data = load_data()\n",
            "  File \"train.py\", line 42, in load_data\n",
            "    maximums = training_data.max(axis=0)\n",
            "  File \"/opt/conda/lib/python3.6/site-packages/numpy/core/_methods.py\", line 39, in _amax\n",
            "    return umr_maximum(a, axis, None, out, keepdims, initial, where)\n",
            "ValueError: zero-size array to reduction operation maximum which has no identity\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Vlec8_YUyKXg"
      },
      "source": [
        "#Import RandomizedSearchCV\n",
        "from sklearn.model_selection import RandomizedSearchCV\n",
        "\n",
        "# Create new similar function\n",
        "def fit_model_2(X, y):\n",
        "    \"\"\" Performs grid search over the 'max_depth' parameter for a \n",
        "        decision tree regressor trained on the input data [X, y]. \"\"\"\n",
        "\n",
        "    # Create cross-validation sets from the training data\n",
        "    # ShuffleSplit works iteratively compared to KFOLD\n",
        "    # It saves computation time when your dataset grows\n",
        "    # X.shape[0] is the total number of elements\n",
        "    # n_iter is the number of re-shuffling & splitting iterations.\n",
        "    cv_sets = ShuffleSplit(X.shape[0], test_size = 0.20, random_state = 0)\n",
        "\n",
        "    # TODO: Create a decision tree regressor object\n",
        "    # Instantiate\n",
        "    regressor = DecisionTreeRegressor(random_state=0)\n",
        "\n",
        "    # TODO: Create a dictionary for the parameter 'max_depth' with a range from 1 to 10\n",
        "    dt_range = range(1, 11)\n",
        "    params = dict(max_depth=dt_range)\n",
        "\n",
        "    # TODO: Transform 'performance_metric' into a scoring function using 'make_scorer' \n",
        "    # We initially created performance_metric using R2_score\n",
        "    scoring_fnc = make_scorer(performance_metric)\n",
        "\n",
        "    # TODO: Create the grid search object\n",
        "    # You would realize we manually created each, including scoring_func using R^2\n",
        "    rand = RandomizedSearchCV(regressor, params, cv=cv_sets, scoring=scoring_fnc)\n",
        "\n",
        "    # Fit the grid search object to the data to compute the optimal model\n",
        "    rand = rand.fit(X, y)\n",
        "\n",
        "    # Return the optimal model after fitting the data\n",
        "    return rand.best_estimator_"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}