week6_end_to_end_regression.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "week6_end_to_end_regression.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oEVKKQfOVYBC"
      },
      "source": [
        "#**Big Data Colab Week 6: End to End Regression**\n",
        "\n",
        "Back into the code! This week, we'll use a combination of the programming skills from the earlier weeks along with the past lectures to create an end to end machine learning example. We'll go from downloading the data to making and validating predictions"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5pLZomT1WHGl"
      },
      "source": [
        "## Workflow\n",
        "\n",
        "\n",
        "1.   Understand the problem\n",
        "2.   Get the data\n",
        "3.   Explore the data\n",
        "4.   Visualize the data\n",
        "5.   Prepare the data\n",
        "6.   Pick a model\n",
        "7.   Evaluate\n",
        "8.   Present\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f7zoXXE--4X0"
      },
      "source": [
        "import pandas as pd # Dataframes and series\n",
        "from matplotlib import pyplot as plt # Plotting\n",
        "from sklearn import datasets # ML Library full of models\n",
        "import numpy as np # Fast list operations"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0VHXfgGBXDX1"
      },
      "source": [
        "## 1. Understand the problem\n",
        "\n",
        "Welcome to the MAGIC Housing Corp. Your first task is to use California Census data to build a model predicting housing prices in the state. The data includes various metrics (population, income, median housing price, ...) for each block group in CA. Block groups are the smallest geographical unit for which the census publishes sample data.\n",
        "\n",
        "Your model must predict a district's median housing price."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yZAWsb6lA84_"
      },
      "source": [
        "## 2. Get the Data / Setup Environment"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CyOV42aeAPIU"
      },
      "source": [
        "import pandas as pd # Dataframes and Series\n",
        "import numpy as np # List operations\n",
        "import matplotlib.pyplot as plt # Plotting\n",
        "\n",
        "\n",
        "housing_df = pd.read_csv(\"https://storage.googleapis.com/mledu-datasets/california_housing_train.csv\", sep=\",\")\n",
        "\n",
        "from sklearn.datasets import load_iris\n",
        "iris = load_iris()\n",
        "iris_df = pd.DataFrame(data= np.c_[iris['data'], iris['target']],\n",
        "                     columns= iris['feature_names'] + ['target'])\n"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "as9ebu94GXzP"
      },
      "source": [
        "## 3. Explore Data\n",
        "In order to understand how our model is working, we have to understand the dataset. Do you remember some of the operations we can try to get a feel for our data?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6P1SocWwAVwz"
      },
      "source": [
        "#### Take a look at the first rows #######"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KXffeI1fZDGW"
      },
      "source": [
        "**Value?** What are the features? Numeric or categorical?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HsFS1BivGuwf"
      },
      "source": [
        "#### See general info of dataset ########"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WoruGQKRZNBb"
      },
      "source": [
        "**Value?** How many values do we have? How many columns?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XuEn8N2lbwNm"
      },
      "source": [
        "#### Check to see if we're missing any data ######"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MJzyu0b7b0oZ"
      },
      "source": [
        "\n",
        "**Value?** ML algos can't handle missing data."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KqBANfwKG2DY"
      },
      "source": [
        "#### Calculate point statistics (describe???) for the dataset ######"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b9vRLrZBG7ui"
      },
      "source": [
        "#### Visualize distribution for each features as histograms #####"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RFz2cE1NZexV"
      },
      "source": [
        "**Value?** What is the distribution of the features?"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7LEFZZU1HmTO"
      },
      "source": [
        "**Things to notice ...**\n",
        "1.   Scaling for median_income\n",
        "2.   housing_median_age capped?\n",
        "3.   Attributes have very different scales.\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "slPYFkJdI28x"
      },
      "source": [
        "## 4. Visualize data and correlations"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5yK9ksVYI5ij",
        "outputId": "7255d103-c314-4d83-d69a-58f75e7e5ef3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 296
        }
      },
      "source": [
        "housing_df.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\", alpha=0.1)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f3eb18a0780>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEGCAYAAABhMDI9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy9WYxkaXbf9/u+u8eNNSP3ysqq3ma6p6d7FjZnhou4mDBkkzRhwKQgW5YNQwL9YkCAIUjgkwHDfvCT5BdbJmzDMkyAgGUJtGVDpkxzuFgcznT3THN6uqe3qs6q3LfY736/zw83IiozK3KpbYZddX9Ad3ZlZkTcuFF9zvnO8j9Ca01JSUlJSckE+eO+gJKSkpKSv1yUjqGkpKSk5BSlYygpKSkpOUXpGEpKSkpKTlE6hpKSkpKSU5g/7gu4CvPz8/rmzZs/7ssoKSkp+Uzx1ltvHWqtFx70cZ8Jx3Dz5k3efPPNH/dllJSUlHymEEJsPMzjylRSSUlJSckpSsdQUlJSUnKK0jGUlJSUlJyidAwlJSUlJad44o5BCGEIIb4rhPjn4z//jhDiAyHEu0KI/1EIYT3paygpKSkpuTo/ihPD3wHeP/Hn3wFeBl4DPOBv/wiu4RRKadJcodRnS0Dws3rdJSUlny2eaLuqEGIN+BXgvwT+UwCt9f914uffBtae5DWcJUpzdnsRSmukECw3XFzL+FFewkPxWb3ukpKSzx5P+sTwD4G/B6izPxinkP4m8C9mPVAI8ZtCiDeFEG8eHBw8lotRSrPbi7AMge+YWIYojO05EfjjiNAf13Nc9bof9PXKU0hJSclZntiJQQjxq8C+1votIcQvzPiV/wb4Y631n8x6vNb6t4HfBnjjjTceyWplmSJRCqEpIm4pyXKFlAKVKXKtkYhTj3nYCF0pTa41hhAkuXosUX6uNUprTKN4rGlIoiQlynJc00BK8VDX/Fk8hZy8v5P3XVJS8nh5kqmknwF+TQjxy4AL1IUQ/4vW+t8XQvxnwALwHz/B1wegGyS8u9UjVxqVKZRWKEPj2zaGEDS94utJTkbopmGQjQ38+lzlQmN00tAKIMkUVdd8oOc4i1J6Gs1nucI0JMMoZacfoQWYUrLccLEN+UDX/KDv8WEN8uM05A/jyEpHUlLy4Dwxx6C1/i3gtwDGJ4a/O3YKfxv4q8Avaa3vSzE9TrJM8e5WD9eUDOOMP/5knx/uDGg4kpeW67y2NkezYt/3uFkRepxlM08WE84a2jDJ2B/E1CvWlZ5jlgE7aQjTXDFKUrI45yhMuTnvU3OtqUFfabgPdM0P8h4f9mTxOE8kD+OsP4snopKSvwz8OLSS/hGwAfyZKCL1f6q1/s+fxAslSpGrwth+59Yxe70EQ0qEYXL7KKTmDlhpuqeM4awIPcsVUoj7ThYnOWtoHbP4mqQ5rm1e+ByzDNjZE8B2J+BPPzogzRXDOCPNNS+v1qnYJnGWASCFmF5zkubFyeWcJJwhxKnfP+/6Hvb09LCPu+r9vYqjfZyvX1LyLPEjcQxa628C3xz/94/MGdlSYkhBJ4w5CiIECgH4liTINEGi2BtEvLLcAO6P0NNMYZlyaqwvMihnDa3SmsW6Q65hFGfnPsd5BuzkCSBJcr57p0uSaxarLsN4xJ9+fECYZ7y+2sQ0DCyjSCnt9iL6UczRMKHt22x2w5mRspRi+vtxdv71Pczp6VEedx5CF/cqyXJs07jUWT/u1y8peZb4TKirPiymKfnitQZv3j4iyzRZrlloOOQKkjTFtQRLdRctZhvoJFOsNj0sQ14aZU4M7XY3ZBRnmIbkRtvHNuSFOe7zDBjcOwFEKidIMmxDEmY5S3WXrU5AEOZsdkO+drONlAJXGqw1PT49GnGjXZka0PMiZdcyWJ+rXHh9Vz1ZPK7HzWLisFOl2D2KaPs2nm1e6Kwf5+uXlDxrPNWOAaBZsfm5lxZxbcGbtzp8fDBCoHlltcHPvLhAzSmKz7MNdNG59CCpB3Hmq5Tiwgj1PAN28gSgMo1lCBCQ5xohoVGxWF+osFLzsMyi61gpTaIUCLDNq0XKl13fVU8Wj/K4iwrEJx22ZztUbZM4V6w1PUzz/G7rh73ukpKSZ8AxANi2wRs3F1iqVTgcRvTClGtzFWqOfc9YKB4pwpwYMNuUVBzzyjntWSeN1aY3PQFMIvpaxeKbP9zn1sEQ25B8+UaLtUYFKSWGENOoOs8Vu/0ICfjj4vTDRMonjfVVThazuMrjLisQn3XYtmWQKo2+wiU87HWXlDzrPBOOAQoj8cJijZsLVYQGLThlLB41wnzUnLagmLHIzgybTSL61WaFX//qdY6DmOMgwTYNhBC0q/bUKRkSLMNgqe6w3YtY5l4764MYxfOM9cPk5i86kVylQPyoKaHLTkQlJSX388w4BrjcSDxKhPmwBmxiHJXWDOOcJMvZH8R8db1FxTn98di2wbJdYbHuMUoyDgYxB4MYpTX9MCFOFUqDbUqarslCzcG3zAtTLuddz4+im+cqzvRRHHY5w1BS8nA8U47hLLMMx8R5TKQirmpUHtaA5VqT54pOmGJKgevZDMKU7W7I8wvVcx9/NExwTIlpSHqjmH/1yRGOIXEsA8sApcC0JK5pnkrPXGYsf5TdPFd1pg/jsMsZhpKSh+eZdQwXGY6LfnaRYX0YA2YIgRaQZDmuZ5PlCtuSRaH5CsNpSmmORglojTeel3hvb8B81eZoGLNSl9OI/yoSHY+zm+cyJ/QgzvRBUkLlDENJyaPxTDqGiwwHcO7PrmJYr2rAThrN1YbHbi9iEKbYlqTlWdOi8ixOGm8NJHmOZ5vMV222ewEAqRIcDROEECzWXNLxtV9mLB9XN89VI/YnUSAuZxhKSh6NZ9IxXGQ4gJk/u6phvQqzjOZX11tsd0MQIC8pGJ803nmuMITBjbkKQZJz2E8QQtL2LTzbYL8XM+87576vWcbyUY31g0bsj7tAXM4wlJQ8Gs+kYzhpOKQQxFmOGH8fZretwtUN60VcZDSfX6he2M9/8mcnjfdK0+NuJ+DWfp84UyzXLEAwDDMyrVmqu8WQ3gMYy0cx1j/uiL2cYSgpeTSeSccwMRwbRyP2+zEAizWHJFe4ljHTqBhCTCUZTCnvcyZX5SKjaRlypuE8t310bLwtQ/LCfBW0xrIkUaJBQ6xyvrBcp+Zaj91YXlQ/uErE/qQ7hsoZhpKSh+eZdAxKFbLYQsBqw8F1TNBMI/ezRiXJFZvdkFQp7u4HKFW0hE6dibx6t8uDpjmumpbRAizT4IWFOrvdEKU1ubJYb/vT33tcxvLkMJ0WsNrwTrXWXuaEflQdQ+UMQ0nJw/HMOYYozdnuhvSChA/3BizXPSquwWKtEK2bpDtOtq1ud0MMAXXXojNK0BpuLvinnMlVjeyDRu5XTcsYQpBmim5YFJy1hsWajW+f/ogf1VhOHFWuihbbJCuM/Nm5i/OcUNkxVFLyl59nyjEopdk4HHE4jNg4Ctg4GnE0jFhtVQijjLW2f1/kPkoytjohjlUopk4GyNAPnzt/kMj9gU4YArQujL8JiAfQJbpqaudB5i5mOaHHWX8oB9hKSp4Mz5RjSHPFZiekE0b8xWaHbpDQD3NeXgppVT1eWKoC9wyO0HA4iLEMgW3IItrtBiw1HKQUj9TtclnkftLoXVUe2xoruk4eF6b5KYN7XgrnstTOZNgPHm7u4iSPq2OoHGArKXlyPFOOQSnN8Shmtxdy62DEbmdEmmvCYMTN1QbvbFbpBimmlFhmcUJIM8VK05tKT1Q9k7pjESY5UggWa07R5qp4bDINs4zeRSeMk8uF1Ph3zhbHz0vhrDW9C1M7QZzxycGA/UFc6C7VXRZqzgPNXZx9349aBC/TUSUlT5ZnyjEAZCrnrU8P+Hi7z3FafG9nlPPxwQFVKeldX2Ct5fHico0sKwxO1TW51vSIs5yFqsP1VgUtIM0U+2OHcZWo9SpR7kVGzzLu1zw6u1xoGKf0gmKfw8ni+HkpnESpc1M7UZzzrU8O+Xh/iGUK5nwHKcAyJV9ea7Lbj9Bao4DlmvPYHN1l/LjbYUtKnnaurq72GSdKc757p8PvvXWbP/r4nlOYMFTwrU8OOBoMOQrjcYrEoF21iVNFmOaAYGW8B8AQgv1xmsmzDECz0w1PKaOe5KTB9x0TyxCFsRxH+3GaE8YZQZKRqSK9ko7TLJOi+MnnSnM1dVyT56xYBr0gY63l8uJSlaprTl/jZAoHmKZwbHlvvkEpTZhkRceWhs1OwHGQ4LsWdc9mlGQcjRKSLMexDVabHuP1rOwPYqI0v/L7Bq60AGkW572XcoCtpOTx8EycGJTSfLLb57/7ww/51p1w5u/I8b8Ohoql5r2dz55lstb07pPpnkStmYKDQRENx6livuZQc637nv+8KHeUZGx3Q+52ArqjlJZrkaHxLAPXLrSQWr49NXono+9JqsurFpPNQorxaxT1ECkFKlPT+sOsFI5pypkzHWFW7Iy2DEmSFQZfa0E+NsJCF87As41prWBWOudJRPflAFtJyZPlmXAMcZrzvZ1jvr3RPfd3JKCVxLc0Nc8iztR0l8Es2WpDCASw0w3H6SCB1pqDQYxvm1ca+hLAfj/icBgRxhlV1yTKc4ZxSmAarFgeQgDjw8LZNFMydhIV28AwJMNRwkF/RJQnNFwXrTR1z0KMH39eN5RtSCxDstZ0MUx5ouguafkWCE1nlJFmOcv1KmvjVNpV22ifhDxFOcBWUvLkeOodQ5Tm3O0EbO6NmJHpmDJfg6/cbLParvHTz7fxHPNCgyOlYL7msNkJkVJhSMFywyNXemY0PCvKna85fHpYROr9KMOzFPbY6C/UHVbqLrZlECb5KR0nKQsDa5qSmmty+2jExuGAb75/QBCn2LbBV260WG36zFUsXMvgWqtyalr6JLnWxFnOKM6ndQDfMVhrFcKBWa7xbZOlussLCzUqjlmcSK5g8B80un+QFtTJ5r0HKf6XlJRczlPtGCYRtmcZvLzawDHuks1wDq+1JT//xTW+dH2epYY73mlwefnFt02utTwMUaycVEqj9fkyGbYhWWm4QJFfV0rTC1IMIbCkLIbG4pyaa2EIMX3OkwY3zRQHgwghirROmCoMnfNP/vw2+92ESIFnwvZxwE+90OLz19q4tkGmNC8t1mYaT6GL/Q6uKanYJlGScTRMuN6s4JpFLUEKwVqrMh1iexCDf9Xo/kFbUMuW1ZKSJ8NT7RjSXJFkOZYpafkur63V+NbG4NTvzHvwjVfWWJurYRiCXpix1Q1ZbXqXGhkpBavjds9J++os46iUZhCl7PWjorArC6lty5S0qzZCQpqHdMMM1zJ4fsHHt61Tzzl5P4rC+QgBOZpBmPB/fO8u399Ppq/Xy4Aoof/9PQaJxrMWSTO42faxkPcZ6FxrGq5JlCuCJMOQgrmKxW4/wrON6Q7r/UHM+vjUAQ+WzrlsAdKDtqCWLaslJU+Op9YxRGnOTjdktxdxPEqoOgYvLze4vT8gisAoBEhxLZNRGJOpGoMo4/XrVWxD3mdkzktxXGYcozTn9sGQd7d7KFUUhtu+zW4v4strTTzbZN01WW9ViLMcpeH5+WKC+KRW053jgCTLORwkrLUKpxIlGf/fBwe8eacz8x4cxPDDzQ7LdZfr7cLo98LsVIQNRZ2kG2aYBizVXCxTEmdFx485PjmdV0N4EImNiyL8WUXqKEmJshzXNO67r2XLaknJk+OpbFedRJO2KbnW8tBK8+nhEC2g3fBotxxaVRvLMsgyzd1ehC1hse5iG8W6zJMtolGac+c44O5xwJ3j4L62TCnFfa2XkxbUrU7AURDjmRKlIUwyoqwoPO/2o2JATkGqCiO33vYxTTl9Tri3OKjmWliGYH9QnDx2BxGmAWHCuRwHKfuDiCyHvf7pttGdbsh2N8Q2JevtCgLBVjcizTWrDQ9zXMuAR28JvahdF+5vQR1GKTu9iDtHIzaORgRxsRPjvN8vW1ZLSh4fT+WJ4WQ0WRGC5ZZLohU/1azwzt0eMQmjWGEIhbQsyOHPbx8jjHtR9MTIPEzKYhIZJ1k+FuATSEOiVIZpSJI0x6xYMB4WO+/EoZQmynIypfDsogV2pelx5yigFyToDG7O+/g2xNHse3EQwU5nxE8+vwCcPgGM4mIQzndMTANutH0GUcq1podjGSxL8dhaQi+L8E/WLKIk5U4nQADHg5hRlvPRXp+bC1Vso5BFt40iDXcwiIkzVbaslpQ8Rp5Kx3C2RXLBd9juRtQqNr/xE9f4p9/dYhAFVB2HpYbHSquC0oqKbbJ5HHKt5RUFV1kMmV2WsjiZZoJ7Eb5jWjhmzF4/ZrFmc9AvhsAqjkHbtzHHMhKz0jEnpa13+xES8F0LUwrWWh7LdRfXjmj7Nl9an+OPPjxGzboXQJ7njKJ0qu806SKyDImG6fcmcwuT9/Q4W0Inn0mS5dOhvbMR/uT1hlHKB/sDkjTncBizcRTgmBLblqw3fTaORtPrFMB8zZnZIlxSUvJwPJWO4WzHjGka/PxLCxwOYxbqC9xcqPJ7b27S9C2QBr5jEGeKl5dr5IppxHxSg+i8tsyzefN21T7lSFZbFUZpTpxrVhouSmmutT3MceR7WWHVsy1WBGz3IpYBU0pWxoXxSeH73/nqGnudERsHMaMTz2MCtgG+byOAuYrNMM6nJ4CVpgcwvU9prkDDVjc8vRDoIVRPz4ruCQ2+Y/D+Tn/aufX5ldp9j01yxXY3ZGN/SKw0Ao1rGmRKcTxIMIUkzTXPz/u4dlEUPxom+HNP5V/lkpIfC0/t/00no92JppFlSLSAL63NYRqS7210OApS0lzz5fUWtmmQq6KV9KwGUZopTCmmi2mknJ1mOhwU08MTR2JKwQvzVZoVi4NBPO1KWqw553Y9nU27VF2LFV2kkU4WYl3LYK3pEcQ1fuPrN3l/s88fvLdDmEGioGZD07d5ab5OqjSeaTDnO/edANbnKqS5Yqsb4pjywknmy4jSnNuHw2kqTSBo+iZRookzhWdKmlWLfpjxw50B11rZtAPsZG1oruaw2RkxjHJsQ+Jakm6QEWcBUkhuzvvA5Se48hRRUvLgPLWOAe4NQG1NNI1sqzDeo4TXrrVo+Q69UcwgyVmuucSZYrVxL4o+afD7YUqui9rD/iBmWQqMqQRFYcAn+kZLdZdOkE4j86W6y/4gHufy5czWz5PMmhY2DDmzO0cLcEyDFxdqJKkmVop3t44ZhjmWIXhpsUqjYrNcc6f3ZNbwndTF9y7rQroIpTSfHgz5YHeAVoqPDkZkWY5pCpYbDoMo51qzwocHA15ZqeNYEkPcW3Y0cYieXYgWBklOnkc0KxbdMEMphWfbND2L/X7Eetu/b86jnG0oKXl0nmrHAOcXPS1T8uJCjXy+Spzk7PaL6u3+IJ6mg6SQpLkqJCKGCetz3jR9MZGsnhjwTBUiemletKQu14u2T2PsLJIsx7GsU9dwntF9kOExQwgMQ7JUd/lgf8jLq3UW6w57/Yjd4wF5lvP2xjGjRFHzbL603qTt36+E+jikK9JxPcSUgl6sSDIFWqMU3NoLSVVOxbZIM81eN2auZhfO9GSELwSZKqTOkzzHkoAUeI7JzbbP2lwFKQR3jgIGYYplSOZrhVZUOdtQUvJ4eOodw0UGb3KiOBwlp8TgDgcxcZaz2wsRQJLnxKnCNE+nL7SA5YbLTjdksxNim4L1dgVTFqeK9blCUmKnG7LXj+kECcsND1OKS43uVQu/Eyey2QlYn/MI4pyGZ9EPMqJc0h/vUVisx9w+6HM8jHn9RpP1lv9Ae5qvihQCLTSZzotidwZRWuyGnqs6aA37/ZDjQcTNxRrHw4QXFv3pe2xWLN7d6pErjW2Y/OuvrlC1TbZ6IZ5tYJuFwV9recz5NkejhINBzJFI7qvvlLMNJSUPx1PvGC4yeLPaQYvBqkIC+2AQE6c5x0GMa5nUvBGrzcopw25ZktWmR5orap4F+t76y3QcsU7mBHa6IXeOglNdT5dd+1UMmmsZ3Jzzx2J7mk/2B+wNQlpVB5UrRmnO9+52qViSIFXMdW0OB8mV9zSf5KL8vWVIlhouSiv2ejGgsS2Bb1kcjGJqjsFizaYTRNQci7prkuaKg36CWi6K/N0g5cZcBSEFWmnCVLFQM1lv++z2Ikbx6fTcyZrI2fpOOdtQUvJwPHHHIIQwgDeBLa31rwohngN+F2gDbwF/U2t9wYjWozPL4J3XDprlhUz1IM5ZqNp8tD/CNS2iNGMUZzMNu2VIbNMgiDM6QUoynmCerznTCHbWnMDjxDQla3OF84kyhdJF9L3dDclyTZTmfLjfZ7WlWJ+vYLjiynuaJ1yWv5dScKPtYxuSpmez3SskzvtRxudW6niWSaIyPNvi5WsNBmEhv7E/jOjHKTXXmtYYADBgFBcR/9nPcHaKULFQczgaJqUcd0nJI/CjODH8HeB9oD7+838F/AOt9e8KIf4R8LeA//ZJX8RJg3dZO+hSvThhHAUJphQ4toRI49nF/MFZwy7HXUZv3+lgCIFjGbQ8i+NR4e9OzgnY5tUE+h4G1zK40fapWgZvbxxz2E9IM8UoSsgUBHHOMM75eG9I3bOYrztXTrOcvGdSSOKskBy50fbvkwhZqrvkSvOSY6KBhmcxSnL0pI1VSQ77EWGqiNOcOFdsdwJeXm5cWOc45bQUM3/Xt038ObPsSiopeQSeqCSGEGIN+BXgvx//WQD/GvBPxr/yj4F/+0lewyzuRZvF26+6Fit1l9Wmx/pchZprMV+1yVJNjiZKigU8aFHsg55h2K3xwpv1doXVpofvWmhgoeaQ5ppRnJHm+rFEsJM5gVnb4qQULDQ9/vrXbtCu2ViGZK5i8/n5Gu5YOdU1izTNwSCe7mq4jHuLiTRb3ZD9QcxmJ2SUZPdd22SBT71i4zsmYapYb1VYn/f53HKdV6/V2TgMuLU/5PZR0bn0zmaPUZIVS4KSnEGUXni/JinCWff2pETJRfeqpKRkNk/6xPAPgb8HTCaZ2kBXaz2xJpvAtVkPFEL8JvCbAOvr64/1oq7SDnpzvsrRKKWWZQSJwrcMcqVZrp/fHWTKwmFMnvtJRLBX3Rvt2Ra/8cYa/+Tbd9noDDiMYpI0Y843iFKNZQnaFRt9zuWcrSUYolhEdPc4oGKbmFLOXEx0NsVTTDsX6SDHNIjSnIN+kToK4hzfkmRKsHkc8MFej/mqx2Q30UWzHnA1AcOydbWk5MF5YicGIcSvAvta67ce5vFa69/WWr+htX5jYWHhsV7bRdHmhIpj8pXrTdaaFW60PQxRaPMcjpKZu42v8pyPymVCdBNyXexu7o4yDsIU33GpWhYVxyZXirpr0PZs/PHeh8lzTyLrk6KBG0cjBlHKMEo56Ad8uNvnk4MB/ShlcTwbcXIf9UmnG6U5G0cj9voxW92QIM7Y6Yb0opSlalF/CVJNlGQ0PZNP9gO0Vri2gWNK9gfxzEg/ywql2GS8wGiWU8gyxeZxgCG58F6VlJTcz5M8MfwM8GtCiF8GXIoaw38NNIUQ5vjUsAZsPcFrOJerRJuHowRDCHqjjNWWS82zL+yNP6/IXcw3FNpEK1fY83AeV5WaFhqORglhnuFbJuicwBR4tkmYQpxB1TVZqBY1higu5jj6w5jjQUgmNM8t1bENk51uyA+2e3zn4y1+eDAkCTPW5mu88fwSphS0q86prp+Jg5zVwrvdC4tTmpQs1F12+jGGIXAsyVzNJUpy9gcxhkwxpKBiGdP3NnFcx0HM+zt9slwxiHJeWPBpVpxTp4EozdnsBGx1Q6quyXy1OHmUraslJVfjiTkGrfVvAb8FIIT4BeDvaq3/hhDifwV+naIz6T8Efu9JXcNlnNeBc3pQysQcCXpRVojYXWE47WSRe+NoRGeUFPsVckWYZHxuqT5zj/QsTqZ0rjqEpgVYJtzZDugEMUfDGIVmrmLy4lKd19ZqJJlmqxtgSclmJ+Sjgx7/+5t3+PQoRgr4ifUm/+aXr3O95fK/ffs2f/RJbyrS997eMb0wZ33+BbSyC6fHPdnxiY7TpIVXjq9PJDlZptjrx0ig4hikWU7NNZmrWOznxfs8uUXuZruQI7nbCbhzOOTNjQ5Nz8S1LVoVm81uSGu832J9rgIUk9SOIfEdE600h8OYxZpTtq6WlFyRH8ccw98HflcI8V8A3wX+hx/DNVzIychcqaKTKEmLNIuiyFcLzcxNZCdJc8V+P6bmFjWGw0HK3U6IaUjW2/6lJ4dZOfKLhtAmTiRPFZvHEYs1l9Vmhf1+SDdMUUoxV8vY6kYcjRKGsaJdtfn+3Q7/+M8+4Si899p/eKvLXj/gl764yJ9v9E4pt8bAd2/1+Fu/oDgYJkg5wjaNU1H7pIVXKY00xjUXKTBMyVzVohtk3JirkOeaL6zV8W2Lds0hyyFIsiIVleXcORpxMEzIlWKYFhPrmYZuUAwl2kjQTPdnZJkiiFMavs1CzeFgEDOMMhquVUxNl11KJSWX8iNxDFrrbwLfHP/3LeBrP4rXfVjORuatisVOLyJMc0wpaVYsNrvhlYuaSmmORgmmIXAtY5rvvkiq4SJ5h1kpsJNOJM5yPEuSZorOKGG+5uE7DksNiyBMOR7G1FyLesXkeBTxrU/3TzmFCe8fJrgfHZDP2JOtgM1OzIsLDr5TDKpN2lehcK6LNYf9QTx1YvNjQ/3SYn2qvhonOdfaFWwp2eyGGKIoPG92Amquieea6GFEJ0jQGuI0J4gz0ALXSlhteSBAIhhEKe9t9dnuFWKAr6zUWao5xBWLm3P+lU9pJSXPOuX/KTM4W0g2pOSr6y1utH3Wmh7dIL20AAxF1LxYcwjSnCDJyJVmvmbj2uapDXGzONtSe3Kr3NmNcWeL0hXTIEwVvmOAAMeS+I5BmioOg4QkVbSrRUFd60KbaBYVoGqb2GOfd9KFNdwiFdSq2Oz0omn76tEonhau9wdFCuf62Jn5tjndxeBYBoYUWJaBaxqY43bfXBdOLlew3PCwpMS1TDKl6AYZS41J27CmG8Rca7poLZj3bd7b7uOYghvtCgL4i80eca5Ya1VKp1BS8gA89ZIYD8t5xemrLO6ZIKXgxryPaRRCeq4lWcpzZSQAACAASURBVG1W7lMEncWDiNqdLUq7jsmLi1V2uiGeCbePikVBvm2xZhtUHIP5mkUvzInSnLbrIBndt+hnuQ7X2zXmqhV+//0DgnGTccOEf+PLq6zUPPYHEVWnWKCd54r3d/o8N+/jmSZJmrPdC7k55yON4rovSoVN7nl6QrJcSsGcb9MLbMI0IskEL6/UmK85zPsOLy7XcE2DKMuJ0owgued4bLNoLy5bVEtKHozSMVzARGQv17qYtJVXLwBP8v22IXluvspi3eVgEJMrjdZc2sp6majdZUVp1zKougaZgn6QECQ5yw3NGzdarLUrJJnGNyV2zeNv/PQLRLnmTz/qkI5ff7UO3/jcCl9Zn2e7F/BrruDPPthns190Nf2ff7HNnW7M55ZqfPVGm5Zvs1h32ewUS37COGO3X6Tf0LDa9LBMiW2cv8p08r4deW8JUZwVJ7ZvPD/Pdq9INVmWgdCQa6azJyaCXpjhWQa+YzKKM8I0x5blSaGk5EEpHcM5KKUZRCl3DofESuGbJmvzPr5tslhz2O6GIIqaw1mD3QsStrsBQgo8y2Sl6VFzLXz74kG3s0Nl551aLitKCyBNc97bHOBYFqtzPlGcM4oyhklKzbH50vUmliHZ7oX4tsnf/+VX+fqHu2zsD4iihPXFBr5rU/NMqonJYBcORpBT/BMFMPzoCKly6q7JL726gjkuyg/ChA/2hyilMQxJlGa8fafDcsOd3q/LovhZ733dlOz2IrJc3+cohSF4YcFnsxvSCYp6zgsLPsIoi80lJQ9K6RhmEKU5tw+G/NGH+7y71cM2BFIYfO25Fq9dbyIRxTY3Tk/nRmnO+9s9/uCHu/SDlLmazReWmyS54qXF2kMJ1J19zFWK0kppPthLSVSORrPe8tnqh+hcEyaaed+iF2bM+zb7/RhDJNiW5GeeX6DluTDuvIqynO9vdRiEGb1RRHKmCD3KYbc7ZBQ3+OF2n5pnUfdMvvPpMcejhGbFpu5afLA3ZLnu4o23tG12giK9JMV9Tu+sczz53i+aPTGEoFlxaPl2Ub0WoLVA55ogz7ClLOsMJSVXpHQMZ1DjhTu73YD3t3p4lqQ7SjFkzh9/sIfnSOarLjfnq1NdoPWxY7i1P+BfvrfL2592yVA4B5LuMEHpBa41PSpj6Yizxu9BFsxcNORmGbIYBhMazyh2NSeZxpYaV0ocW7JYd/DdQi12u5ez0hhvm0tzRrnm5dUaUZTzf7+3w3Y3ZK8fYQvBURST3Xe3YL+b0Rsk7PsRTd/CNiWjKCfOKYrLpmSrE7DadEnG+5lHcUY87vCyTDl1hMClEhbnOVcpiwL07cMBwzAlU5paxeDD/R6GKPStXl9r0qzYj+FvSUnJ003pGM6Qj3c8j9KcfpxhJZIgzWn5Fv0w5ePdIfkSLNZcqq41NcpKae4cDrlzNMIwoWZZRGnOrcMRddditeni2xb1isUgKkzsxPidXRF6UUH7ZD1BCkGc5Yjx9ydIKbg+7/PGc3PsDyL2+zFIuD5f4fn5Kju9qNiuRrHveaXhooHuKOGgH/HeTq+QrghSRlHG1igmzmYI9gGOI6n6NsejFKXghzt9LEuy6BiA4KBftPU2HJODQYxWRRG+FxbTzWut8b7p4yL1VuxXePDta90g4U8+OuCPPtjn/e0ewyjFMAQ321V+8eUlmhWbtzc6/NxLC+XJoaTkEkrHcIZJMbc/SpFAL0pJ85wozrnWcqi6JijBcZBgj6NdQwgUmlwU6SXPlGRKkylBkqVkueZwFPPR3pB+nPLiYpX1uSqmFKdWhCZZPm3nPK8DaVKU3jgaFQafIp2V5ApX3ouuXcvgjZttXliocudwSIImihWZKiL5lbrL+7sDukGPa3MeLdeiE6Q0fatoDR1lHI8ntqsVByOKkYYmiIq2VQOYqxu8utZkpekzDFM+3B2glEIIQa7Ar0iiVPOFa3WiPOe9rT6KQpDQsw2qns3d4wApixmEuYrFcrOYXj7pHCcNAOfVZrJM8c5Gh3fvdNntBgyiBIRgFCXsdgL+/NY+P/u5RY6GIZ1rddrVckdDSclFlI7hDFIWg1hIySurdd6+00UYJnma89r1Odo1h3bNJoiLdZ+TaVoLyY1WhVbF5tbhkDRTCC2oOxY3F332+zG3DkYc9wu9oFzB55frqKxYe9msWPzFZneqqfT6WvNc42UbRWpkfc7DHuftZ0XXUgraNZeW70yngu92AjzL4NOjITVH0o0ywiSjN0rQ49qJKSW+K9kfafI0RwoDz7ZYrpj4tgmCYhbCd/j8SrM4vRiF+qqQxczE7YMR3/rkgDBR/GCzC1Jy56jPbidCofAchxcWq/z6T6zTcC06o5iDQUyUqVNb8tJMsTWIL0wvJUoRZBmJVoySHEMamIYgTPJi8jlO2Dwa4toOqYKvP9fmc8v1so21pOQcSscwA9cyuNbyeHHR5yefm2P7OGR3EPHiUpWb81VMIYhzNS2gTqQxnl+s8ZPPtzGFYBCnVMZRsVKaf/bWpwxjhURQc03SDK41XCzTRGjY60WF3LTSJCpjuxNQd62ZzmEyGOeON51JQ5ybejrV1joWpXtz45hv3zoqttY5BmuvLqGUxDQEnm3w6mqDD3d65LkiSHMsI0crxfFhwvMtgy9dX8CyTI7DHEMUO7PjNOOlpTrrcx5vf3rM97c6REmOa5u8dafLXjchPnFd9jCmP4qRKue16y2Wm1Wk1IRJPt2Stzxe33lZ7cWWkoppgtYkWUYUZyQqJ4g0oQY3hlwpnlv0GAQJH+0PqDgmz89Xy5NDSckMSscwA8soWio7owTPtri5YPD51QZ1z0Lrol1zrVUh05rd43vSGO2qzaurDV5faRBmOYYUfOf2Ef/Tn97ik4MIRXHDB8k+r11rcHPB5xvPtcm15tbBkNtHQ7Y6EWhF03eouSbXWv59xuuiWYqTjiAZG9LJ9dVdk7uHI75967jo8TcNglTz/76/z5fX5/hcu8ZBP0YaglfXmnTChE/2R+z1smnh+e3dnLd3d3ltyeFrzy/gWCbX52xuHwYkWc6nRwG7w5hBlOIYxbT1MDrtFAASYBjBB3sDqp6N7xYid2ZdslCzWam7aMF9+7hnOUDTlLx2vcnHB30+2nPZ7ccMQ008vt81F6Qp6QTRdKo6SfNSabWk5BxKxzCDk7uLT8pl24acGl2AO8fBqWj2cBAXNQI0QaoIooR37hxz5yia5uUzYKuT8kI7xjQ07+30+fxilR9ud/n4cIApYJgoPj0OEXnOz7+6xIvzdSqOOZXBOG/47aQjEECSKaquiWkYJGnOu1s9tIQ53ybJTIIko2IbZJkiTnL2+xFowcEg5sacz0azwq39/sxupO/vxSTpDn/9Z14kyzWmJRlECXleDKjlGQR5xijO6USz73MEdIY5Wa7ohRkvLVU5HiaFNlU/Qit93z7u82ovNc/iZ19apFmxcN5RDKKUW/sBQoCQBkILRmHO8TBCiDq2ZZRKqyUl51A6hnOY7E8+W/ScRJizpTEUbd/mnc0uguJkYUkBEmwDhIAkgRSoOiZzFZckVby9ccy3Pz3kzlFElOUYUmEZFkJlbA9jluoOL7VrXJuv8LnFOo2KjWsZrDU9EqWwZeEwTjqqMMnYH8TUK0W0rYEky3ENiRACpKbqWliGpFG1MM3iPVlSIoUsagZAmJ6v5/TJcc7/8/0tvvHiIrZp0vJdjoOEZcfmWqvCR/t9+nHKec9gAJ4rmPMsqo4kiFNyVajoFd1J8r593OdNjBtCIIQgyWCpWUUOYly3aI2VeU4qJRXLwJSS59o+a61SabWk5DxKx3ABFw2knZfOAUAXjkIimKu72AakWeEYFOBKuDFXYZSkdAYpb20c0hmlZEohNPRGIEWKjcIxDO4cBnzvTocsU6w2Xf6tL6/xxnMLdIP0VBrrpKNyzOJrkuYgBNudgG6YsVC1ea5d4Zsf7pHlcK3pcHO+SZBoFmoGcxUbw5AMw4Sa5+BaAqLZpj0DekHEViei6TsIComKKFG8sOgjRMZ2N2a0HzFDoBUHqDo2cabZ6ydEqaLp2cRjOVelC+e1oikWHJ1YvTrrs1qqF7MQSa4IUkXb94iSIa5lcHOuwhfXmtxo+/iWoBNGtLWLbZcF6JKSs5SO4QE4O5g22VQ2ijMsQ7JUd4uBMFNOO15eWqzzhdUGP9jukWVQdWC14ZNLi4/3hizVHRRFl5FpSIIkJQNsCSmaTw6HCARxlrIz1Ii7Q/7gB4f8ta9f4z/6Ky/hSIM4y9nvRQh5z1EprVmsO2RKs90tNqm9vtbgaBiTA7/y+goa6Ed5sYmtVmxi68cZFVty9yijXbW51qqxN5idTqrbYFvFiWTON+mOUoQEKaBeMVmu+2S5YLsT0UtPP1YC7VohkDdIc15puSw2KlgS3tro8IUVRcUxaXnWffu4z/tsXMvgldUam92ALNe0fQvT0DRci2tzHhtHQ/7FD+6SRJqlVoVf/MIKf+0nb7Dc8B7HX4+SkqeG0jFckVmSFcA0TaIpuoU0RXR7MG6xbPkO/8kvvcSbG8d0BgmCnFa1UhhiKUkScKTA80ycNMeIwBZF6klrOBqlWAKOk3sa6aGC3/nWFl96rsG1eqPYwZAqXl6pMYrzad3hRtsvUlpKTzepuZZBkOZ8bqmG1nD3eIQQgoWaQy/MGIQpAwFfWmsgZJO5mo1pbPDuxoD+CfnVlg1N32Wh7pJkxUKiqmuhlMJC8P2NHn7FohfnRGecAoArYL5ewbNNlmoezWpRbO9HGTXHIteKMM0Ik5yvrrfOlcw4+9mYUrLS8Gh4FrlSLNUdOoOEv7jT4d2dcHpyOQ4DBukmFVvyH3zjhfLkUFJygtIxXIFZkhVbxwG51ni2ge/Y0+IzgCkF15oecZajdVHszXKBplgzud72yTJNpnLiLOfVa03u9kIGwxQhBfWKwLYNGo7JMA4ZDymjKJyDBSgN//J7O/y7P13Bsyy0KRhEGeutClowNZxZpqYpL9s0EBRpJqU1aTpesykK6YqWa1GxJAjwvWLSGw1fvTnPL35umUGa8sO7R2z1MkwJS00f25TTuYZP+8PiMQIGUcrhMKI3nJ1G8sziPq23KlRdg83jkL1uSJTmmKbBDVnBkJKaY2KNJ5VnOWfbkKc+G0sKVpoeuVJsHAYYUiIEfLQXnrqOCNg5jPngbpfBV1PapWMoKZlSOoYrcFafKFOaO50ApTXNin1i2bxioeZwNExQujDIK83CeK3NVRBa49kmKF0ss08zPtkdkuaKzy/V+MZzbd7eOGQUKyp2MXPgmgZhmvCDvQQoCraFBAYc9kLe/rTLtbkKX1xtAON9z8ZpQ5oqxe5RRNu38exiV8PHe0O2eyESaNXsYq9yN2Sp5nAwSNg+Dvj4cFQM00mDum+Tj+CL1xdYnk9peAZV22avF1LzLLTKeW+vj1SKw2HE4TAjzQoHdjINZVM4t+WmyZfXm7y2Nsc7dzt8sDtAa7Btkxfm/bGGlCo6pJoZQzR7gwjPLKamJ856peGe+mzs8b5pPb7HriX5eLfHrF1EuS5UWY2yZbWk5BSlY7gCZ/WJdrohrikxDHnfsnnfNvHn7pfXnuwXcEzJx3tD+mHCTi+k7lk4lsHhMGKh7vD1FxcJwoz9QUSc5Nxo11BKMYi3udPNpvsSXljwuLHU5Hq7gmVIjkYxS3Vv2oJ58pTj2Q5V2yTOFat1l+1+xGrTxZDgWAZpVux+vtsJMaTAEPD9vSF3jkd8YbXB+lyFO8cBtgGrc1W+4BkM43xcSxF8aa3J2xsddo4GbA3P72KyKWosc77D8ws+dcfimx8c0B3F0+U8hszoBgm//+4OCzUHrQVv3+lQdQyOhglfWZ9joe4wX3VQJzbgRUk2nQL3LJPlqsP+MMa3jaLV2Orcl9KqO/Dl59tUXetx/nUpKfnMUzqGK3BybiDJMtJcs94uNH0OBjGDMMW3Ddbn/PvaWqEw0oYUrNZdslzx+rU6393s4oYmSo/TQ1LQGaa8uFTjuZbPV+0Wb210MQX044yffnGRyt0jsizH0HC9XeHlaw0ExckgSovTyslNc0mW44yLw7ZlkCpNhibJcywhEOOp7d1BRJCmHPVTlFYcBBG9UcRwGLDTtbDaVQypaVc9Xlmt0hml7PQiemHCneOAP/7BXd6+HdA53ycAYEpY8A1uLFbJpGQU59w5GnI4iulHCaYulGFvtHwWmhXIc/aDlHbFZhQZuJbBezs9frY6z24vZKHmkqtC9HDrhG7UjXkfQwqWai7dUYJrWby0VOOD3QFBVqTk2h781VdX+LkXlzBNObN2UVLyrFI6hisya+2kaUiarkkQZUgp2B/ELI8X7Ew4mRdXY+VW3zGn2+E2ewEN1yHLwa8ZJFnGYZphWpLrcx4CyA4DlNK8tNjgxrzPfj/EMgyOBjFpqtFC8/x4idDkNXe6IXv9mE6QsNzwptpDQZzx/c0eQhcF8yAttp65lkOqIv7B73/Ch3sBiSrSVs8fh/z8i4v0opyaa/HR7ohOEPHOrUP+8L0jeg9wDyVQ9xw6QUKQKPY7EQdBSBJnCAVJrkmB7c4IQwrCKEUjWGl4KCXwHJN+mDKMUyxp0vZt9gcxVcekvmhNazr2OJXm2SaOJfnJF9o0PIumbxGmOUtVm597eYnFWoUk1wRxMfNxkR5TScmzROkYHoCzayejJGV3ELM2500nc09q+ZwtWidjJ1GxDZaqDh9u90lzTa5yEqX4cLdPZ5igDYklBJ9fqfHiQo2KZ9KuOni2idbwXLvKe9s9XCmL71sGO72IV5ZybNtgtxdhSsFqs2ifnWgPLVYdvrfZZbFuM4gUvSDi7lHA156fYxCk/OkH+3y4EwDFjEEK3DqIWXL2+StfXGPOd9npjfif/+QjNocPfv+uN4op5CyHmmsyjBOCOGMUgGODluBISLUgTjIqjkWaK+4eBazNVUjSnPmqw1qrgiWLE4TSybS+4NnFSs/JbopJO7Fnmnzlxhxfe36OD/eH+LbJ2lyFxZpLlim2uyGebTyU3HdJydNI6RgegsnpIcpytCjkGuB+LZ9cazKlsIyiC8i2DNpVmzTXNH2HtTmPqmvSCxJcU/L+9gCNpOVZSEvy7dvH9IOI12/MsdcTpErx4e6AzihiGCnqrkmQpOz2cqJ0gNaCN262CNOMQZSR5QpDFqJ915oeuS5E6jSglSJMMgxD0xkl7PVjtnqFU5ASUlWkXAAOhjGdQNGqJPz2H37E4TkSF5exulhDKYs41/SDlFxLPMsis1IsEywhiJVGoBmmCjPKWG/7JHlOnOSYhuCn1trYhslywy0kQi7Yvz2ZXp+vOcW+7Uyx1lQs1x0avoNSmnRs+83xKeOiXRgwu122pORpo3QMD4mUAtcsJBbOM0xpVkSfhhDYVmHwPctkrekxSjIOhhGuFfHx3oAsz+gGKVJCkKaYCHb6MdtHfXYHMUs1h/e3enSDmGutKotVl4/3A77zaYf5moNlSJqeyShKqbjFycI2JUmm8O2cFxeqSA39KMUbq6x+shtwHERopfAtG9+ziprFmQ6e2124vn3Ed25nD+0UahJema/x1uYILaDlG4DEMwUrDYfdQUwS59iGpO6aLDc8luoeq3MeK3WXqmvy1etzzNXcqWYUMFMz6qz0eM21MIRguxuyUHc4HBUpKs82WW147A/icz/Dk5y3frWk5GmjdAyPwHlidpM00v4gPrU6c/vEsNZ2N6Q3Svmzjw/4YK9PriAIE7ohGBKUKmoAnRA2fnB06nWVHDHvVdjpBwyClH6UUnNtbHNA1bMI0pyF8VY2pQshuo3jEaZR1C12OxHvbHUZpik118QybIZJxtdutrmz1+PuiTRRFajVDPpJPk0zPSiegJ97ZZ6hMlhpVcgzxf4oxjMFN+d9rrcrbHdDdo5HJDks1GxSXYgRWkLgOxamFOyNYoZxzvV5H5tC0NA25H17oGetTt0fxHi2Qc2wqDsWca5Ya3qYpmR5vDDpPOcCF+/aLk8OJU8bpWN4RM5bUD+Zfai6FhW7aF+NkxzLlAyilO/cOuS9nT4Hw5QoUYRpyjApBq+Y0XN/ko8OUjbpFdviXGhVDBSCj/b6zLsGN5Zr1J0KFduaymF4drEDOk41Sw2b9NOUum2w1Q0BTSdI6UcJwnKwxyLZFRPadZtMwWLd4p07D35/fv6GzS//xAss1j2E1nTCnJZj8MnRkCjVCAlaQbviUncsgkSRakWvF4NKiBse7213CRPFUsNDCMGr/Tqrc0VX2NnIfVZUf3Z16qRDS4uLP8OTXLRru5TuLnnaKB3DY2CW2N5ZkT2VawxDIjR8ejhk4zgkiHO6YUKuckaJJpw1InwOIUWXT0PCwSBB6ASN5P29Hh8dB7x9u8NL8w6G7XBtzme7G9Kq2MRpxrc+6fAnHx7QDVKCBHwBhge+ZXJzrorQOYeDjExDnKUs1qu8ON/g7Vqf7cElXusEP3HN5/Wbi6RKcDxKqdgmNdfCtCW+6/DyiofvmBwPY7632cUVBqM0ojOI6Y5irLrL8TBmqxdRcU3qFRuJ5p9/f4t/72vrzNWKQbZJ5A7MjOonq1MvShddJJg46/O8KOVUUvJZp3QMT4jz0ky51hyNEpoVm42jIVGaEWQ56SyVukvQwGFQfLWAuqfY6UXsBUUh4Pc/KDqBfuX162it+d7dDtfqLv/qg11ud+9Ne8UaCAAvJ8oVS40qYTpEorje9vnZl5b4yrUmP9gdsf3Dw0uvywF+6eUaS60GYabYOBiwWK8wMHK0UHz9xhyqDm3f5qP9Efr/Z+/NgyRN8/q+z/O895tnZd1HV/U5PT07186xF7ALK0C7XCuQACFBgG1MWDgsG/sPWRGE8SVZhCMUDiIcCnCE0CoCh40QKxPAImAPlr1YZnd2dmd7rr6rq7quzMrzvd/n8R9vVXV1d3VXdU/VzPSQn4iO7s7MN/PJrMrf73l+x/enNL5lsN6N6UcpZdei2U/ohCmeBWmWs97O8U2DRtkhTHPOr/Q5nhXaSCXH2BldGsQptZIN3NzVa7F/LmI/7hU2HDLk3cbQMRwhe4Uo4jTfEbObqHps9lPiRBHHKcnBN+NA8cPbNu8x0AmLyWi7WezA7311kWdPj2IZkpd0xosr4Z7Ptx5qppKM8YrHybESj0+X+YWPnGLU97AMSTfOWNoccH517+sF4EtwTGgFipKvGCuZLHdTKm6KUzKwDJO1QcLjszVeXuoQJhlV3+bsdJU467C4GeI6mtGKQy9M2Rxk9KIUISVBkpH3CsFAUJhCkKYZi70QU2oWWxEb/QTHlJybrhb9Ilu7esu6Mxdxv+z18xxWKQ15NzJ0DEfM7SEKy5BMVBxutENGyw5PzNe5uGpQdkKag5i1YJ/24V2kFCql2+MSbncK22yksN7uE2aCV9bvXlbkA1mmSPKcZ483+PDZCSarJSyj6Ayebfj83eeO8f7NAS9fb9HqhHRDSFKQ1tbfBlQ8k41BTrUUM16xmKo7+I7J8TEPUxqMl4sObQEYUpCkOWXbxjRgvGwjAKEzslxTcgttpEFcONSKY1IvOfSijFdWu1xc7dPqxcR5zrERn/edGiNOFd+63uHp+frOQJ7DMuC7f57DKqUh71aGjuEtRkrB8bEyzX6CMGAQGpydKvPaSo/1zYBrmwNeW43uauRvJ9WFBpGCPWcmbNMbJKwM7u10Pnjc4uPvPcmZuSolYZIKyNPiGBPEGa1eQq4EC6NVnpht8K3rXYI4ZhClXFjrs9aLKTkAOcEgY6UNJ0c9ogze6PXwLYupEZc0V2wOEoQoHMNaN+L8jQ5zdY+ROZsrrQFhljNdc0nzIrn7ntkKM7Uyjil5ZaXL3IjHciuiEyTYlsCxLHpxxrcWN/nwIxP04oypamGoj8KAD6uUhrybGTqGtwHfMXn2eIPaqsn5lR6uafD47Agjp8ZZ7gRc2uhyYWXARqfLxZa+p5PIKX6IngUo6N0lgX2xq+9ZO3O8DD/2vnOcnqiQKc1La108S3J+ucvJMZ9BrMiFZrxq0wtyLm8E2FLTaPjM1ktEySKXN2I2Y9ieUrF8I0KoDaZHSri2wWo/xLElKoepmkOmNa1BihaaPIeJWhHCGivZfOVSk8mKw0TVJ0hS+mnOZNUhyRXT9aIEeLE9QCNolFx6UUI3yhAkXNkIGK86uKZxZAZ8WKU05N3M0DG8TfiOyWMzdVzbxDIECMFia4BhGRyfqPHeY6OsdCO+da3FVy802biLd3AA1ygmpp2brPLClRatPR57r/SFAH76A/M8MllhbsTnr6+1mKu7dKIcD8HVVki9ZGIiQAsypag4BscaPp5p0ByEbAzCPR3YS6sREzWHsuMSJ4pjDZ8s01xrBSy2QkqOgdImE1Vo9QvdI8c2ODFWYq5RwndMuqHJ+nKHyxsBnSCmXnbwTRPXMImSmDwv5kOEcYZrQpZnVNyiSumoDPiwSmnIu5kjcwxCCBf4AoXtMoHf01r/mhDibwH/O0W1ZR/4Ba31haNaxzsZ05TMj5aKmQlpjlJwbqLMxY2AjUGK0oKnFhr4rsV/+MbKnoa3ZMHjs2Vmx6scH/HIVcZnLnTvax3PzpV5Yz3ixubrlB1Y6mScma1Q9hwmSyX6UUKc5QhgxLcJ0xzPspFSMOJbfH2xybduxHd9/leudxDCQAvJIEkp2zag0VojhMSUGte02RjETNc9fNvi+FgFyxBMlh2ubgwoWQa5Uqz3EuJM4VkSJaETJrQGIVGq8B2Lcm6y2OqTKnAMg1MTlSMx4PdbpTRMUg95mDjKE0MMfFRr3RdCWMAXhRCfBv4V8Amt9StCiF8GfhX4hSNcxzsa1zKYqDhcaw1QWrPYjtFoXMsEkeMY8MhUlU88rfjTb67doWbaSuHb1/uYjkk/DPnsfToFgBeu93nhdlW872xiAuemXUYrPu+ZqlD2HaQAgeDcVAXHNnl1ucMf4UKNQQAAIABJREFUvLjIvbIXywEYKx1KjsV6L2FqxiOKbR6bsehuKdMO4oyTYxVOjpcxjULK48rGgGY/ZqMbUfZtHENS8VM2+yHXWn2ubvQJYo0gJ1E5rb7BYmvA+Rsmj88m2GYxR3thtMRaL36gMtN7GfSDNMbBMEk95OHjyByD1lpTnAigKLO3KILPGqhu3V4Dlo9qDQ8D23INJdvEd0w6YbLV5euQ5SZhkgGCsZJPyZd0gjuDQpsZfPa19qGvLQO+fSOiuhIRxjkfe3KauRGfXpQhZdG0d7m1SWvv6tWb7xEo2yYzdY8kLZIgEzUX0xC4g4QoTnFNm8mqi9wVojGEKGYliKJaquKaWBIubQxYbgcEcU6uoL+Vda8aOb4n6UcZ1zcDTnVDFjdtjo+WHqhU9SAGfb/GuGGSesjDiNz/ISCEeEQI8RkhxMtb/39SCPGrB7jOEEJ8E1gD/kxr/VfALwJ/LIS4Dvwc8C/ucu0vCSFeEEK8sL6+ftD389CxHQO3LYOpiospDcIkI800da+YJd0axFxq9rGctyclFGu42uzz5TdWubIxQEjBsRGf6RGP/uBg7dqaHNcS5FsPn6y5GFJSsgsxvcmqi2kURr0fFoOAZuouk1WPs1NVmv2YzV5MnCp8yyBJFVs+Zme30cshThWOLH6tr7WKmRSLmwFJrnbE99TWcB+l7n7O2W3QS06RB1rpRPe8Zi9u5jhuqrcqXZxChgx5p3IgxwD8X8A/ZaufSmv9LeDv73eR1jrXWj8NzAHvE0I8DvwK8ENa6zngt4F/eZdrf0tr/ZzW+rnx8fEDLvPhY3cS03NMzk0V4ZSSY5JlCpBUPRvLlBgH/WkdMgrohjmrnYg3VnqkaY5lSFzTYGHMO9BzpEnOZpDiOgYbvYTL6z1GPBNDCo6NejTKDr5lIASMVxymai4l10JKwdmpKk/P15ka8TjWcJkaKcaZGgZkuw5Q2/vvVOegFGVHMtfw8LZmVCilidJ8K/EdcK0VEKV7O7bDMOhK6R1HkuVq5+9hknrIO52DbkF9rfXXxK2/zAcWcdBat4UQnwM+Djy1dXIA+H+BPzno87wbuT2JaZoG33V6HGNr7ObF9T4XV3tUXZvJis+l5kE7HA4Hl0KoTyKZrDnYVlGZFCQZhhCcm6pTkbCfhFKYa2qehVKal65vstGPmR3xMKRkrGrR6sa4tmTEd9Fao/TNOc6mFJyZrDJRLnIxaa559UaH/voApQr1VtMEQ4DvGNQ8m7GKh21bNHwb2zQYxFkxxvSAYZ3b53xvJ98PatCjNGe5HZLlilxr0kxhmXIopTHkoeCgjmFDCHGKrQJ1IcTfA27c6wIhxDiQbjkFD/gB4NeBmhDiEa3161u3vfLAq3+XcLckpmVIqq7FeM3h8fk6VcfgWrPL0n0I2d0vZyrg2JJupMgBLSUoGK36nB6robVguRvyF6+v8qevLPOlVzfID7Cc5T60u32W2hHrvRhbQKsbcnVjwPV2SJwVA4KeOtbgx56eoewVqrRSiJ05zq5lMFZ18W2Tx2droDSdMEYrwdx4mTPjPlEGszWXkmcx6jssbobU/aKbOs0VmVJ49t6DlXaz7bCvNges7ZonneQKV947cayU5urGgHaYIIRAa03NtZite7fMkhgy5J3KQR3Dfwn8FvCoEGIJuAz87D7XTAOfFEIUQWT4Xa31Hwoh/nPg3wshFLAJ/KcPtvR3F3slMaUUTNe9rSRozJnpKr9YPcU/+/QbBz+u3Sfn5qq4ns16J2ajG1B1HYRpcWaijOtYhElOEKe8sN7j0y9tHHgdCvijb6+jNPQSuL24tWoVjuHlpRaOkfHsiXE+em4ay5BkSmOKIjfQjzKOj5c51vBYGC3TC2K0hLGKiyklSsOjU1UEsNFPaIcp7UGCa5msdiJWuhESdkax3iusYxsSy5DMNzxs62az3H6J4zRXrPViKq65UyK73k84Mc4DO4VhueuQt5IDOQat9SXg+4UQJUBqrXsHuOZbwHv3uP1TwKfud6F/U3Etg7NTVWbqHqvdiEwp5r7wBlcGR/N6f/DKreWu437AsyfGOTVVRUgQ2ERxzh9eObhT2Gbt7q0OBClUXEhyWB+kbAxSrrYC6r5NmBQ6SZNbw4em6x6rnYizUxVWOzZnp8q4lkndt3h1tUemFINYkWY5hiiMcXnLSE8LWO5ETAGmlHcN6yilibIcpTUlpzhhSEM8cHNcGMZc3egyVfGplp37unZY7jrkreaejkEI8d/e5XYAtNZ7Jo6HHC5SCmq+TcW16AYJ9boFg3T/Cw+BZgDnL6/z3EKNM5N1moOYbpAQ75KCFbDTxzAqoPkABTcZsBkVR8tWJ6Q9iPjO9TbPn2wUISUJa52IJFd0o+K9m4bBE3M1Zkd8moOEIFWMeBavrPSwpcS2JefGS7SDjEbZIcsVvmMyXXWZrnu4ZmFc063S2G0HsW2I81zd1wljG8uQTFQdNgcJMlecX27z2VfWsA2Baxv8/IeO8/yJmwUV9zoNDMtdh7wd7HdiqGz9fRZ4HviDrf//KPC1o1rUkL2RsjAcpycafHNp9chfz6Qw+ALI8yKsMlZ2EBpGyw5X2+lOqShAScInnpvk2zdCXl7sst3eYHN35dfbcYHpkRLNQcbmoM1E3WWm5qOBKMkpuwZaF5+FSbFJaQ4KqW3TkEgBU1WXhVEf2yqm1q31OlzZ6GMZBloXZcCuaZBsGdndO3HbkDuG2LOtA58wdiOlYGG0hG1INnsRn391jfGyy1jVod1P+OSXr3B2vEq17Ox7GhhqMg15O7inY9Ba/08AQogvAM9sh5CEEP8j8EdHvrohd2AIwajvINl3AuibJgMMoF62eGSmzFzDx5KS6UrKf/aRMxhfusjLi110DmUPfvL5ed4z2+DUdMJEZRVLSrRO+fq1Htfvpu53GzFwpTXg1LhF2TVIU0WU5tS8onS1F+WcHC+RK41lSAZxRpYrSls9Ho5pYMii+seUkiTNMaRAINBKk+Q5aZ6TJDnXOyGWKbCEIFWKG+2Qmbp3iyEuuxbTmp0TxkF36a5lsDBa2tJxEoxVi/BRvWzTWotphTFl3973NDDUZBrydnDQ5PMkt276kq3bhrzFmKZkesSl6kD7HjH7Q3kt4OyEy999/0kWRso4ZmG8LMvgex+Z4ETD53qrT5CmnJ2psdbJMA2BbQoem2sU2kbdiIlaykZvwN0nQdwkBwYBrHUGjFdqLLcjPMek7llMVhyuNgdcaw12dv8118K2jJsjVLVmouqQaxjEGUprpqounmVwrTWgFaRca4Vc34xQShHlilxpHNPAtwwqbvGV2G2Ija2ejfsN3UgpmCh7uLZBu59QL9u0+wmOJWl4zoFOA8PJcUPeDg7qGP4t8DUhxHbS+O8AnzyaJQ25F4YQ1HyHEc9iEKccRabhRBV+5oPHePL4GGfHGziWyVovZhBnCGCs4hRJ8ekap6eqGEIQpTlr3RZZrhgkikbJ4upawGJzgEZybMzhjY2DebIBsNFLaXgOZVfS6oWMlEyyrmJjkDBluFgGaA1CCqaq7i1aSNthnFxrhIbFzYAbnZB+nONZBu0gJUxSWoMEKSWuZeCYmssbPRCayapLmAhsQyKkYKbmPbAhrpYdfv5Dx/nkl6/QWotxLMnPf+g41bKDUnrP04DQt+Y9XMtgru4RZTlCF6E9pfTQOQw5Mg5alfTPtgTwvmfrpv9Ea/3i0S1ryN3QAuZGfM5MVYmzNr04J0zvo9vwAFzuwj//j4v8448KHpsYw3dM5i2DQZKx3otZ78U0RbITD9/We5qte6z3Y3KVYAiJaRcJ4um6zUpbM1dXtAcp/QN4sziHP3tthRHPIlEwVrJ53+lRzkxU0VrhmlDzXXIFlimZq3skSmFLiWkW3crbu+5Gyebieo8oKxxDo2RjCHAsk2yrGzqIMzSiKHENUiwp8GyD6ZGiGmys4lCyzQcyxs+fGOfseJVWGNPwnJ2qpL1OA3Xf4no7vCXnAHC1OeDqeo/VbsR4zePEaHmntwOG5axDDpcDOQYhxDywwa4yUyHEvNb62lEtbMjeGELQ8B0+dGqcOM3pJSkCiWMKrqx1WT7EMtbf+Ow1Xl/e5H/4O88xVfVo9m8meXfHw7dDImXXKip9NIRpEftfbEYYUmCaBraQVF2DMR9yLYGUxbuIwVY92Axy1rpFbuLaRsSF9R4zVYskV5QRPLowykcfm6Hmmaz3itOIbRrM1L0dgxmlOa1BgillEXryLNphRpbllF2zOHZoTTNIcQ2DYKvi6bW1PrMjDiudkKm6x7Vm4STm6j7+A2hWVcvOnmWqu5sbhYbr7fCWnMONLSdxYbXHX11ugoILG33yvOikPj5a2jOJPixnHfJmOOhv+B9xs/jEA04ArwHvOYpFDbk7UgpmGz7PnmgQ5TnfuNKibBvUfZvvf88UF1f7/M5fH55g7Z+82mPmM6/w3/3Qk3eNh9+SIDUl4xWHG23NWMVhftTn8lqfLNNEeU7JMclzjW+aSMNEq5DbFb+fmHDop4pWmJIBniya39YGmrXBzVTX19ZW+PzrG3zfIxNMjfiMlj3GKg5JmvPIVCHgu9KJsE3JifEyUsJyO6LmmziGRc036IQZSaaIuzG1skU3VKz3Y9Y6IWGcstZLeGTcZ6pRIlWKtW7MM/MjezqHB921bzc3prm64zMexBlBnPHyUgfXMqh4Fr0g5eXlDgtjpfuS+Rgy5KAcNJT0xO7/CyGeAX75SFY0ZF9cy+A9M3VOj1f4sadmWW4HbIYZrmFQsgpF1s+dX2HtINneA/Bn31nhFz5yBpA7+kXb8fFtI7g7JGJIyTMLI+RaM1Vz+WbF4ep6nxev5ZRdm6XNEMPUBLHi5GSNR2YknTCl0x0QaTBtG5knhTYRhQZSfpeipmudjK9fXef7/FlSFdKPEla6IZM1l7Jr7Rha04AzE1UmKgljZYdOmJIrzYjn0CjZzNRdNvoRS5sDXr/RIVEwiIuQ0kovoeo7JLnGNTXXWgNOj1d2QlZwOE1oe1UgWYYEqUlzjWMZpJnCtgySJEPlxV5tWM465LB5IB1nrfU3hBDvP+zFDDk4Ugo8x8RzTCYqHq+tdOnGKaNVB9+x+fhTs5xfanNpdUDzTWao1wK4vNGlUS7RGqS36Bdt70rvpvf02HQd05A0Sjb9OOWN9QFjFQdHGvhOUco5Xy+hG5BO1WiUDC6uBpRMwWaQMAg1qYL91K7b/Zx6VdIKFF6W8+LiJu9bGL3F0CpdVB91owzXMnaM71o/xjYNhJYESU6UaXzHpB9nCAF5nlP3HcK4OF2kSmNKyVzD38mxHGTXvt+JYq+cw3TdYzSxqZeaBFFKkBZlulXfYrbhD8tZhxwJB80x7O6AlsAz/A0fsPNOQspioI2IBZ4pmax5jJdtPvzoJC8vdbiw0uPCUpNXNg/WS3A7BvD7X1/ksblRTk9UmKv72KbEvk0HfC+9J9OUzI+UeGW5x3jV59pmRM21QGtOjVQRQvD8iVFcs5AXLzkmcZzRizO+fGGFr7zR5EorJFOQ3iXD7lsGiUq5vBHiGRbHJ0ug4Ho7YK7uszFIdgztWMVhvRcXjkJp8i2jfnzU59R4maXNkHaYkmQaQxZx/6pno1RGq684O13F2+rEXmoFzDV8lNIkWY5j3V2c76Anir0crGsZ/MQzc3zh9XWSTCHRnJmq0AkzelFO3bdoB+mwnHXIoXHQE0Nl178zipzDvz/85Qx5EHJd7CIXRouYswbWuzG2lEzXPE6NlTg96aO+ucRrD3B8mK4ZnF8ZcKOb8dK1Nt//2DRPzNZ2DN9+O2HHNnh0poJvG1xe7yG2drlogedYNDyTLC9mQC80SpimJIgzwlxxbKxKK0gQKue1G33+5OX1HQE+ATw9U+Y9czU2gowrGwN8y2atX8h2vE+MYwjBRLUwwtaWI1snpj2I6UQZcZqx3kuYH/UxpKBWsnlits61ZkA7hG6cc262ihQCDWwMEkY0xFnOjU5EcxARJook11Q9k6mahynFLbv2+5W12MvBztR9fuLpOQZpxnq/mPi3fUJoBylzdQ8tGFYlDTkUDuoYzmut/93uG4QQPwn8u7s8fshbyHY4QekiDj1d84hTRZjlKA2mMLi8EYEw8WXKHtNB78qxikSaElMYKJ2z2Orx/72Y4toLTNc8clPvuxM2hMC3TGq+xQ89OcOXL2yQK02iFBWh+Oeffo00yzANg5/9wHF+5KlZWoOEimNRtU2m6xkiF3z/43P88g+c4i9ea2JaKadGakzWy3QGKf/mKxep+za+bdENE84v9Xn6+AirvYjrrYDphsf8SBH6itOcl5c7WFIyVnGYqDisdSPmRnyqrsFaJ+c9s1WUBteSGFJwYryYG73aiRjEGQiBRPPt6x1Gyy4KjWUIrjUDZkc8Zuo3ex8OS9bCtg2EIdgM0lsGCMVZhhbsOL4hQ94sB3UM/5Q7ncBetw15G9grNj1T93BNyZoXc3WjTy9IWOuFpAd0CpMSxhouli1Y6ySYVsLVjSKWc9mIeHymTtVxGC3bVD3rnjthKYsQzvXNkHrJ4amFOpNll06Q8K+/eBmtNVN1H5Vrfv8b1zk55VN3XaSEz726Tp4rQPATdZfTEyNwxtqZddAJMuJUkWeCk2MVgkShNbQGCe1eSlumvLbUgVwxM+nz/PFxRjyHuREPtEYBsw2P5c3C4I+VHaJUUXGLoUKZUtzoxjh2RJYqmoMUrRKqroXrGAiRUfFMoiTHMSUjJZvZuoezyzkeZh5g+7mSLN/ZDAxzCkMOm/3UVT8O/BAwK4T4jV13VTncnqohb5LdsWmlNEvtEN8xGQMurfVJVSFpvV8gqSLBt8CwoDWIkGEhiT0YgCnAlEUI54uvr/LIbJkkz6mXbKDYvUZJSpTld0hIlGyTsbLNRj/CMgy6UU6U5WgUoxWPqmshpeTqRp9OL8FUks+9uo5jSIRpEKU5n3ttDSEFcV4YfyGKNoSSY1BxTOJU41qSG51iN/7VC2u8cK3NcjfbqbV+amqZ/+Zvn6M5SOlFKUmmCOIyExWHesmiE4Bnm8RJRjNIcQxJmGSsdWJKjsFUzSXPc3INWils0yDNFZZpFKczKe/YuR+mrIWURRPcy0sdcqUxpODx2dowfDTkUNnvxLAMvAD8GPD1Xbf3KGY3D3kHsR2bVuKm1IJrGjiWwaPTFb692GK//rdEQS/mlkk6FVk4FBOwLaiWHJpBxuJaSMXPsA3NeNUDBOu9BC1uKpHuDitJKbAMg/Gyw0YvwTVNHNMsTgRC0BnEWIakUXVIc02WK2qeQ2uQUPVMBBBGKRthyrGah2UbeKZJEGf8wBPTfP6VVda7MZYpee94ia9eWmepW+xftkUHX1oJ+c0/e5kPnZshzgRxqljeDGiULRzTYLrmMln1ubgZ0Q1TToxXOOmXuLQ+wDAEo2UbU9oMopTNIKNRskgyTdUtchC7Q0i7uVvV1v2ilKYdpCw0fIQshAHbQbrlWIfOYcjhsJ+66kvAS0KI39FaD08IDwm7d6hJViiTPrswyl++vsbmWnjPa/dSM9qeJCoEWMAgTADB66ubrPVzekFCvWRxdqLCjzwzR2VrdsHusNLuBHmuNSdGYZBk/BcfPc2/+eIVrm70sQzJTz03z+nxGtebRVmrbQgaJYtBnDNQiiutkK9e2mCq6lJxLd67UKfk2CyMlvjIuXGuNUN8x6DqmHzp0sbOezDYkg+nmOx2YT0gzTOuNTtcW8kJKY7ANjDiwvyIybmFCc7NVKm5NlUvZrRkc3qigso1vShlekSz0UuQGhplm/lG6Z69C3slle+X7XyFZ5s7b2wQD/sWhhwu+4WSfldr/VPAi0KIOyrJtdZPHtnKhrwptneo6VZc2xDw489N88ofX3rg50wVdGPwbc2ZyRpvrAas9mO00tzohixuhjiW5Cfed5yqa9+SYN2dILe24uy2afC9j0zxzOwIa72QkbLDaLkYZdoMUmbrLq/e6DOIEpJc812nxnh1rUfdt+hHOf0w43JzwMeenOREo8qTMw3Kdo/1XkyWKcbLLq+vF65udwhNCnjp8hpX95hDmACrEazeyPjrG8ustSM+9MgEEkHFNYlTVTTeSUnNNRkruXTDhLVegmfHbAbpPZvb3qym0bBvYchbwX6hpP966+8fOeqFDDl8pBQ4stAOWulEfPDkNB+c3+Ar1+4UKBpxoOabXNnc+2DoAKNlgWVITo+VGKt6LF0akGmNZxbNY61BypcurDMzWuaDJ8cwDWPHYElZNMUtt0MQ+S1Db+oVl3qlEIvbFuSbrrnYpsS3DLpRimUauLZBkip82+L6IGC5HTKIU3phwj/84HEWRivMj5aJkpyNQcoHTo+z0gm52LopozFdBb9c4vWlg4lK/cmrLVypePTYKMubISfGyvi2yVI73OmF6MU5phR4W87gbqWoh9EdPZThHvJWsF8o6cbWP39Za/1Pdt8nhPh14J/cedWQdxq7Tw//y997mpeW1/jDv14kCQZIu8RTC1Wma3W6ccwXXl3lK9fuNJoxsN7XjLg5UzUX05SkqmgAA1BaYEmQUrLeibi2GfCBE2O3jMtc68U7Y0AntqS7b+dmaWcRGCmaySSjJRvXkFwoWVxdG7C4GVKyDTxLopXmM99Z5R98yMOUkpPjFeYaOUttC/N9c/zhN6+z2k4o2VD3PVbawX19fp97tc2rqwEzjTJBkvODj03tNLWJrQohxyqS7VIUc6HTXCH1TcmQLFNcbwU4lsQzzTelaXRY+YohQ+7GQctVf4A7ncDH97htyDuU7dPD3IiPKac4OzbKWjfCtSWvrrXphDFJMmCkZFMVA1INt2cjUmAtghevbfLhs1PMVF2WNgOiVCEE1H2HR6YqzI2WmK64WFtaQrsbvDy7yD+s9WLmrTuH3xhCIIAb7XCnKU3KHNOUGKbBYzM1vrPSIU0zTM9kbsRHCEGcKTY2Q2oVl6mqS5TlfO1Sk+9cH+BZPienPPpRTJgpovT+hlJ3FCSbCc2whRA5nSDi3NwIWgkqtqQXJlQ9DygG/KS5YqldfHrbUtob/ZildkjZNRkrF07xzWga3Z6vuD1ENZThHvJm2C/H8I8oxPJOCiG+teuuCvClo1zYkKNh927TtSW/9qlv8qXLdwbbqwaofO9k9MWNlP/qY3UWJqt849I6Ly93yXPN2dk6T8+PMV3zcGxzJ4x0Pw1eu3sepFQYUjBV88iVZrbuMVv3sA343ReuU3ZMXLuQ16g4Jo5rEmc5L17bxDQE6/0Q15TM1B2agxilJLahqfsm7eT+ailCBWEAX3y9w/XmgLNXmzxzapK1XowhNC9cbfP4TJ2zU2WEEDvy5Ema8/JSh2MjHiXHRCvNRj9mouIcWm7g9hDVtkTGUIZ7yIOy34nh/wY+DfxvwH+/6/ae1rp1ZKsacqRIKUiinN/8/Ot7OgWA7j1klXLAlpL3LYzwtx+doh3GLHUCglTjmiYTFfeWss37TZiWbJPZEQ9DsKPkqnXxPFrA43MNohy+8NoagyhhtOzw9PwINdfmwnqfy+sDkjxjEGUEaYZtFFPaar6FKQW+LYkSzUpw/9pRMfBaM+Nqc5MXFjc5MVbh8bkRbMPgarPHqUkf1ywG+mS5QgvIVREaG9/SaepHGTXXYu4QpLFvl9vYdkQLo/6bDlkN+ZvLfjmGDtABfgZACDEBuEBZCFEeDup5eFkPQ752qX3Px1js3RBXMsCzi8YuaQim6yXmRyvkugjRFOGfWzufb0+YTlSKmcco7jBYUoqdhHmY5HtONntuocF7Z+uEKsczDNYHCUppLqz1KTkmjpaYo5JUDzAotI0qtsmj02VudBLGqiWWNnu8tHzv8t27EQHLfdjo92iGGeO+g+taPDVfY6zk0ezHCCHIt2ZKZ0phSsl42abuWxzf0oR6s9x+GhNSkG9JosNQhnvIg3FQddUfBf4lMAOsAQvAKwwH9Ty0qFSDvLc+hk3RBd3Z5R3qBnzi+VkurAbkhIRJzohv856ZKscbJbQEoQtdn93sDmGlWZFjuFeoY7/JZmu9mPmGT0XaKFVMYOuGCa1BQpwqpITpmkuSKp5bGCFTmmZQSFm8fL1NmGQoINOK9U7MZlD0ORRz5YpT0d0c424SoNkJsYWgG2e8caNL44xbyGnkOZlS+I7J1WZQnHq2OpUPwynAnacxvdUNrbac9LCcdciDcNDk8/8KfAD4c631e4UQ3wf87NEta8hRM172eGyswtXW3SOCj826zI7W6cQxvW7IyYbFYyenmamUuNQa4FlFIjXJcl642uRLF9ao+Q6OafBdp8eYqnm3PJ+UAhQs9eIDKY3ea7LZ7l3wdins4saAilOcZMquRTtIeGSywhNzI1iG5NpmgNga5fmVi01a/Zg8E4xVfEbLYBkalGS5M2AjAkdyIG2pIINenDJR8Vhqx6z3I9JM8cbqAIFGafjw6TFGqx5aaTYHCd5OYv3NGey9TmOPz9ZoBymD+O0pZx0mvh9+DuoYUq11UwghhRBSa/05IcT/caQrG3Kk+L7FL33vaV5Y/Brrt0VTJPDhMyNUHBvbEshI8NjcGPPjJU6OVVjqBGg0limwTUmaKl5f6XNirMxk1SWMMj57foUffnKGimffYhweRGn0IDkKy5RMNzxGqzavr/bJlEIpeGKujmnKHWHBlU5E3bXZ7CfUnEIoT6DpRDkkOSNli9NuBbXYo7XlFFyKGdSDGAa7HIVDccpQGfT7KZ4lOL+sWe4ECC0Yr7lIKZEoLm8GjNU8cuD6Zki61eB3GInhvcpXq671thjnw+jVGPL2c1DH0BZClIEvAL8jhFiDfWV3hrzDeXxhlF/9xBP8P399hU4vxrZMpssuQmga1RIjvk1zEOPZJvPjHo2yy+YgRShB3XUK6exMEWQpSmkqnkmQZFxs9rm6PiBV8N75Oo9MVXeMw4N07h4Xi4QzAAAgAElEQVSkqWv7eUu2ybPzI8RZTpIpBknOoBXsXDPf8NlwJCcmSsgmDLKcKM0xDY1QOa4p+dZSe8cpANR9ODFRY30QE4QxvUDTU7dWbIUptNYSHpmUdIMI0zRQKqfquSRa0RmkRFHGjV6EaQgqnrWTOD6MxPDt5auHIb9xv9zv3Ikh71wO6hg+QZFv+xXgHwI14H8+qkUNeWswhOCpuVFOjZVZbIeAwpEW03WHF662afZjXMvgkckqZ6crlGyLTphQ90ooNN9Z7rHRj6nYBtMjLq6UXFobcL0VIIUk1zmvrfTwHZOTY+Ud4zBatlnvxcSZOnCoY7+mrmS7f6BbmOvxso0h5U7Z6LaRmqg4bHYTNvopQgomSjZfu7zGRr+IyV9strl9VPZKAOd0Qs21GPVtkrri8kqf9m0VrznwympxtUHKRi/lsTlJzSv6GC40+3SCjLmGR5KpN93L8E7jsOZODHn7OZBj0FrvPh188ojWMuQtZmcnDsyPSJq9hNGKjWdbfOLJWVKlWO1FWIakE2b04xQhBKcmKoXDmKgWktOGZDNM+Pwrq1zZ6CGF5IljNSquxeYgpjeIafkWtpS0turrBTBWcSjZ5oF3k7t3wbvj2FDIUJQdk+qEtXNakFKAgCjJME1JniuW2yGOY/D0sRqf/vYy31lqE6SaqgOZhnay92v7rsVEuUSuNDdawR1O4XYU0AtzXl3e5NHJCt91ZpTjdZ+mk2IJ8cC9DO/k+P1Qx+ndw34Nbj1grzZRAWitdfVIVjXkLcO1DObqHleaihMTJWyzCAG0wpT5hk/JtVjpRNQcyByDY3Uf3yl+bTzHxKMwVuOmy8efmCZTRaVPxbVIM8VmP+EvL24wutqnH6e8/3iD6ZESWa5o9hNKjYMeWm+yHcfOc0UYpTiOJFaa8XKR7PZskzxP6QQJ374eoHQhnLcw6lP1bNCQa5iq+7y81EEChm1hKUU72bu3Yaxk43g2Nc/kwsqdWlO3oynyEWlfsexEtAYpnSRHac1KN8Gz5H33MrzT4/dDHad3D/v1MVTudf+QdwdaFF9q27wzBOBaxo74nSUEG4ME05Q7Bmm3sRLAY7MVLm8ErHYjWv2YZj/i3EyN0ZJDmilevN5mtOpiGw8WRtmOY+dK8dL1Tf78/Apxqqj5Jj/+3mM8Ml0jy4sEwWovZnOQYBjFacExJJ5lsBakTJRtLCEoewZhIrDRDPIcgyIktJsTVcmxqRrzY1WWmyHrwV2OFbchAUPCaj/msy8vcWayzFjZL4b/pIq5mleUliq9r/E8yvj9YZ5ChjpO7w6GQ2KH3BICgFtr37fVTj3boOJaWIYoHIHStxirkmNim5KKa/PkTI0z4yWeOV7n1ESF0bJLO0xwbYM4USRJvmeYQSldlKaqu2sZ5VqT54obrYA/f2WVKFMgJZ0g5fe/sch6PyTNNTXfohOmTI8UMhpTNY8rrYBumLLcjkhyODVV5ocfn2Wi4hNmGi0Ex0ct5usG435RdWQDiTD442+ucm1jwEi5mEN9N5xd/5aAUhDEcP7GgL94ZYnl9oAcKLsG1zshi62Aa62AKL13F/ZucUEonLfSeqep8EGJ0pxrreDA6zgIxUCmN1+KO+Tt4/7P8QdECOFSVDE5W6/ze1rrXxNCCIq+iJ+k2Jz9K631b9z9mYYcNfcKAdyrhwC44z7LkEzUHYQUOIZkpXOz4cw2BEJCqjVGrm8JMxw0TLIti9EKE1r9mPGKi5ACSxh0wpSyXexYO1szrjtBiu9IshxMKYq+i1yB0hxrlDjeKHF8vMTF9T5a5TQqLlrDpdUBX359CdOy8B2XTOX80bdu8DPPz+I7Fg4RKUUuYZuTFXhiYYyNTp+vLRb3b+v1RSn89ldXeWkx4KffN49pGjw1Z1FyLJI05/pmcM9u6KOI3z/MVUTv5FzLu4EjcwwU1Xwf1Vr3hRAW8EUhxKeBc8Ax4FGttdqS2RjyNnO3EMB+Bmn3fUlaxNBtKbFNA9MQPH+8wVcvNYmznMmqy48+OctoudhXb89Gvh8DJaVgpubx+koXhSaKFSMVmzDOsUxJ2bYAeGOtS7ufsNHro4Wm5tl895kxmkGRQL/Rjzgx5tOLcmzDZLLqM152WBj1CdOcTpBimi5TdQ/fsciUYqk1wLAsfvL5eYLkMhfWgh3HMN8w+ciZSU5NVGgFVb554xLpHgnq7yz1+MobN/jY0/N0ogzDkDT7CYM4Aw1zDX9Ph3gU8fuHtYronZ5reTdwZI5Ba62B/tZ/ra0/GvhHwD/QWqutx60d1RqG3B971b7vZ5C27+tGMc1+wmjJZrkb7Sh8lhyT7z0zTrVkMeLaKAE3bvtSb0s4HNRA+Y7Jd58e5421Pt+8tkmzF+E6Jh95ZJK67xDGGV+/1mZhrMxkVdGNEtZ7MYoivOPbJnN1j7JjUfVsHEOigNVOxFI7Yqbu8uxCg8+/vo7QAkMKBkGObRg4luDs5Ai/8jGbP3rpOlmqWJgoU7IMgkTTjjISpRkrmwz2KF2SgECCEIRxxmqmMKWg7Jo4lrznjv2w4/cPYxXRw3zKeZg4yhMDQggD+DpwGvg/tdZ/JYQ4Bfy0EOLHgXXgH2ut39jj2l8Cfglgfn7+KJc5ZB/uZZBuVjUNWBj1d6qa2kHKXN1DC26ZEXCtFdzxpZ6re/dtoKq+zc998DiPTlYYJCk11+bZE6OYpqQXFw13vmNQsk1qJQulNevdhCTTmIbgkYmtugoN7lbOYGG0RC9KmRvxma55fPw9M/zp+Ru0w5SSY/D3n5+n6jj0wpSybfGBkxMEccZY2WK5HZOplNGSQ7MbMVpyWG1nd/RECAEIBShudCKqnsVYxWGs7GCbxr7zmw+zce1hrCJ6WE85DxtH6hi01jnwtBCiDnxKCPE4Rc4h0lo/J4T4CeBfA9+zx7W/BfwWwHPPPffmMmxD3jT3Mkh3q2rS4ma4CO7+pdaCBzJQjZLD3zo3RaIUtpQ78fmSZVJxLBabAbZlECc5jmWyMOZRcSyEEGwGCWNlB7HLISmtsc1Cw8ixBD/z/gWeOVajnSSMeS5npmsopVluh2it6Ucpm0HKn55fpxvGTNU8vvfsOE8vjNCOMwwDzi8NGGzlc13g0dkq3/3IHONln5MjBivdmLGyjW1IwiQr5km/hTv2h62K6GE85TyMHKlj2EZr3RZCfA74GHAd+P2tuz4F/PZbsYYhR8dBv6z3epxlyVsMFECaq32NlWlKzNuK60xT8thshW9czVAaTNdkuuayMFqm2U+2pDw0E1UXy5B3dUi+Y/LUwugdRvPkeJk0V4RJxovX2pwcL7HWLzSj/urKBj/8+CzPLoxwbNTjBx/LWGqHSA0np0ucna4zUy3tjAHNBfSjjNagEKyaqBTJcVfeGjM/ymTr2yGf8aA8jKech5GjrEoapxDfawshPIrxoL8O/Afg+4DLwEeA149qDUPeGg76Zd3vcdsGar/k4n5GMteaRsnl44/PEGc5lpRc2wyRwGzdI85ytGan6/peO+a75V2kFiCLORGLzQGf+846ca4QKNrdhPedmcCSEunYfNepKjMNj4VGiY1B0QOx7SBd0yDXmvmGuTOUaDu8th2Gi9Kc5XYIAkwpmaq52Ia8b0ex1+f2MFb3PGynnIeRozwxTAOf3MozSOB3tdZ/KIT4IoUQ369QJKd/8QjXMOQt4qBf1tsfB7eeDPZLLh6kImX7ZCKloOLZZLliouqQawjTYvDPdP1Oh3Q/GELgGQZBnPKn37mBaxmUXJNelPAfz9/gQ+fGeHSqxiDOGK86nB6vFKcb89YTytjWVLftPIc0BN0o5kpzULzfJON6O6Rkm7i2wYhncbU52AnR3e0zuN3g7/W5AQ9tdc/DdMp5GDnKqqRvAe/d4/Y28MNH9bpD3j4O+mW918nAkIJMKXQOXZViS0G61fhmIQ9UkbLXyWRhtHTPXfa2IRWaWxLm93oPx8ZKTJZtwkyhhAFZhm2aRKmi2cuYqRpkW+9Nbz3VdrJ+Oy8ipaApkltKfpv9hIVRnzzXvLrSY7UXMjfiU1cWYZKT5orjW4l+pe9UaL39c52oOKztmoGRpDnXmgMUGksItNIstwMurLV57tgo9Yr7AD/9Ie8m3pIcw5C/2Wx3NMPNsZ93OxmMlWxeudHlpWttOlECCk5NlIlP5xwfLd+92U5xi9G/2wlmL8e1bUjDNNspufVsc98dtEoVj02XsQFLpPi2TS9KSDPoRyFJVkVvdStvn472coa7nZjSmtGSjSklS+0BoBhEOS8vdgiynIZvUPPcYhaGYWBIgW8ZO1U5e32uy50if+HZFlGas96LWWoHXG8FKJXzmVfXWNkMUcDpiTK/8oNnef7E+OH9Agx56Bg6hiFHShBnXFzrsdaPsaRksuqyMFbas3chSlKubmzNcsgy8lzRCxO+vZyDVqx3Y2ZH/FuktKUQpJliaY9RoQc5wWwbUkPAIC7mMQRpTtk171kf/8pym9//xnWiJGNmxONqMyTOMkzT4PkTNYJYshkkzNV9ZurevmGy28eYxlmOzjW9SBXlv2FCkilUpkAXSrgnxsvFCWOQcGarbm+vyi+R5GggSXNWuxFpmrHRjTDRfOaNDa6s9TFMiWcKFlsBv/nZ1zj701WqZeeO9739mb3T4/sPwxrfyQwdw5AjI4gzvnJxg4trfSxT0ig7CIppa/Mj/h0VSnGWc60dsNqLWOtFdIKUTClGDYlhmHTClMmaIs7UziyH28Mk99vwtFuDSGmNbxfDhqQQpFrtWR/f6kT8269cwTFNyo5NvewQpoonZ0qcmqrh2jazIx7PLzQou9bOOnKtyZTCMoyd19w+8ViG3HmdiYrD0mZAkOWYRpFH8B2TEd+gUbZJ8pxUKfpRim0ajPgWiVJIJfas/DIMyUTF4cpGnysbA6Qo5ldUPIvNIEUBlpDFND4laAYpG0G0p2N4GLqOH4Y1vtMZOoYhR4JSmuubAa0goeQUidMgztAoxjOHXOtbBvYICgPoGZLmIGKlHbLSiUnyjBvtiJmqTc0fxTIks1s7cEOIN93wtG1ItSqMSJRkO6eZvUpuldJc2uyT5YrpusXFtT6r3ZhWkPLKjR6xlrz/1Di2YRKnOd1BTDeIsX2TqmNzfTPAknInkSylvOU1ojRnrRcjhWCyUijSdsMUJzMYLdtIBL1IM162mWv4JEnOSi/GaYc7FUt7VX7ZhsSxDGbqLo4hGcQ53SDBNgVagaTYYSepwjZMaq6958/0KLuOD2OXP+yMPhyGjmHIkbBtsB0pSeVNwb08K6qQltoheiv3ULYNMCDMNONVDxODIFFoNI5pYJsGN7oxZ6bzHaG+nS+54o4d8v00PO1OVJccYyfHkCv2LLndToI7pqTZjbi41scUkmMNn5pr0Aoy6mWTtX7Ar37qEi9dbdENi/GHJvDdU/DMuWMsTI0QxC7PLozcUjq6bdQ828K1DaQohAdbg4Qgzik7FsfHS0xUPNJUsdKLmam5lFxrz9DUtpFNc4UQgmONEuu9mGOjPueXEp6cqRMnilY/IVU5YyWbn3rfKUbKdyagj7Lr+LB2+cPO6MNh6BiGHAmGEDimQb1kQQDNICHLciYrLqaUaK1Z7UV8e6nDN69uUvNMbEPyodNjNComM5FDvWSCVgSJwDYkZcdmsuruW4F0vw1PuxPVZ8bvXpUUpTk32iHdJOepY3X+8o0NOmFKxTE5M1mmE2QMBhFfPr/Oi8ttXl+LbplylQCfXYHPrixyZmyVn/vQSZ6er+/cv5dRc22TD5wYY60bkSqFKSVzI4XQXpTlaAEl19p5/F6hqe2fhxQCUwpm6x5jmc1sxSVIM77n7BjXmyGByjjdqPDRx6b3/PyOquv4MHf5w87ow2HoGIYcCVIKpuseSa7QCCqOyVjNYX6kxEYvpjVIWO+EnL/eph1ErPcgVymLrQHzNZeNbkSmFQZFfHyy7HJmskRpj1kIh9HwtF+iett42aZk/v9v782DK8vu+77Puft9G957eNi3Xmfp6emZ4TTJ4TaiKJParF2OLCmSGMVWRalUEquSWC7HFVfkVMklu5xIchQz5T8sl0WVJEsRJUqkRZEUTXJmyNk4090zPb0v2PGAt99315M/7gMG6Aa60Wg0Gj19PlWobtx333vnAHjnd85v+f76M5i64FOPG8RhTN7W8WJBkCTkXYvTsytcWvA3bX24yrmlgM99+yJPjPXxzET/mmtss0Ut75jkHfOm+TmGjqFp21oE1xvQRKb3TQ3mAOjL2hyuFBC6YHxdh75bvcZuVh3v5i5fVUbvDsowKO4ZjqlzdDDPgf4s8K5u0mLLpxtGdELJteUOi02fVhBhaILlVkwlZ7LS6TLT0+Y9t9illLep5Lb+gN/rgqf1i5ehp4J75azFVCXDf3x1msXlNn0Zk6mSy58ttdlOj7dAJlxa6HBsuIhrG7evDN+k3/WdLIJbGdAD/dltG9V7UXW827t8VRl99yjDoLinaJrAvkH3Z7TPZa7WxYtC5psefhDhmAaChKYf8jdvLbDgv3t/CPz1mQV++NkqH3NHtmxms112EuRcXbyCKK2cXhXcOzKYZ6qU5asXFim6BkjBX3xneluvObsc8dqVJSb7XU5MlHFMHUvXGOlVJW/WBW0zX/ydLIKrBjRJJH6vW5vZi9tsl902wvdil68qo+8OZRgUe07GNnjfVIkoiSlnbabDGD+UWIaGpSU0vdQJI3pfCdAI4YW3qowV8kzepor5Vuw0yKlpgoJr8Pq1FeIkDYqfGC+m13M2zx8d5NR0nbYfcngox4WVldu+pmFrLHd8XrhQJWubHKzkWLihHsNCWysO1IXY0he/1cK+mRHshjFXqm0WGqn1HczbTFWy9zWlU+3y9xfKMCjuCxnb4AOT/XzkUIWvnZtFCJ0wjoikRRB3afdODOtbZ3aDDlImm2oFbcdQ3E2Qs+NHvHJ5mYVmF10TDOUd5utdCr06hWLG4gNTZS4utfh7HzvK3MobvDF/YzeGdzlY1MlnbIbyaZbQ1d6csraxNrYr1TaJlCw1A4IgwnF0MrbOaF/qmrudL77jR1yrtvHjBEvXmOxPF//ZmsdKOyDvpB//mhdg1jQO9Ge3tSDfq+IxtcvfPyjDoLhvZDImP/mBSRpByJVlD11KxoouoZR84Y05Wjf0pf+jV6oE8jzvO1DhmckSOdciSeS2ROVg50HOJJG8NVvnpYvLREmMTGAx7xNFkgOV7JqrTOgCy9SZyOT477/vGH99epovn5pnvX0Yy0Mha2NoBhnLwLU1mn5CDO8W2iWSOJHM1jwMXdDshrx4sUo3iNE0+JGnxzg0WLjJFx9FyZoGUxAnfP3cAu8sNGl6MXnH4OhgjucOVwjjBE0TGL2fWRCn1dXbCfaq4rGHA2UYFPeVQwN5fv4jh/C6AZapo2ka52dbnJ9p8Pp8Z8O9XeAvvzPPO3MNuvEBjgzmGSm4LDR8JssujmXc8hSw0yCnH8a8PdvEC2IQCS0vZrreJYhjTh4qY/cWxtXXT6Tk6GAB19T54KFBOn6AYxrMtbrMVbvMND0WmyF5x6QdSIbyNlPlVBSv1Q2ptgPa3YC5hs9AxuJbl5fJWiZ9GYsoivny2wv0ZS0yprnmi691Ak5N14mTdCdvGXBuvkXdSwv26l7AuYUmk6vie4kk6rmowjAhtBNkLOEWa7wqHnt4UIZBcV/RNMGB/tyGXejJw2XOzFW4unyVWviuOykBujF0w5CvvzPLdy4u8PhUgUP9ZSzz9qeAnQY5YylBJuhI6n5qVIwoRkMw3+jS51qpG2RDSqhktC/DifESrqETJAkzNQ8TwULL4+x8k8VmQCVncWK8yNGhAkkieelilboXAKnBulr36PoRxYxFFEvKWYtOEFPOWAzm0wrwKEo4NV3HMdK6h1Y34Oxsk1rHR9dSOfB2N6LZjQjjhPFyliiRLDR82n5IzQvphDYr7ZDjY30UMzdXPa/+HG514rqVi0lpFz1YKMOguO/cGHiMooThQoZi1qRdC1mXoIQPnFmMOLNYSy+8WeW5ySr/5EeOU8m7tz0F7CTI6Rg6AwWXK0sdukkqYjdYcBgpuQAbjNCWqq6JwNA0DF0wWckzmHdpBxFT5exaqqofxmiaYLyUwTJ1vCDi7EyNOElYafk9WXIdx4KinS7eYU9jKk7kWk+HjJW2L3VNCy8KaXkxSZJQzJhYRtoH++hgnvE+l5euVDmUzZKxTbpBxKnpOs/1emffyFYnLiGh2Q1ZbPpr1ezFjIlj6tiGjoSbgur7zf2kDNdGlGFQ7AvWBx4tS+d7Hh/m5SvLrHgL+P6tn/vi1QZ//I23+eEPPUqfa9/2FHCnQU7D0Dh5oMxCs0uy2ME2YaqSZ6TgYBn6TUZoq65vN55WDg3mb1ogNZH6/rVe5fiBgQLvO1Dmb95Z7MUGYj71+DAhkplqGwCZSLwwYqkZU8jYJLFkpOhSzplcXGrT8mIKGYsjgzkmy+8GmIUu0IVGxk4rpx3LoB34BEmy1i71xgXzxjkUMybXVjpcXe6QJAleFHPqeo1L801cS+fR0QJT/QUeGcptkO1Y36Hufp8u9jpu8iAYIWUYFPuSoaLLf/XRQ4wUbc5N1+n4MQutDpfqm9cTv7PUZL7hc7hy82K7G5RzNj/69Djn5hssdQJsXWew4KxJaq9nyw9+LMlYGq6hk12nurqKqWsMFmxW2gFanJAkkqE+h6ODeY5UCrSiCEMIVjohr1+tYRmC4T6X5U7A2dkGDS9E0zSeHCvy/KMDWLrG4crGrKT1Vc2WpqH3usQ5lrEmIGhpqVHYasG8USK8G0ZUm13mG11eu7rMtWqbuZaPqQnemmvy3MEupZyJaxsYurahQ92NC/H9WKT3Mm6yOr8wjAnimIG8Qzlr33Vtzm6jDINiX6ILQcG1eP/BARzDotbuUqg5XKpXN3+CH2IbGnONLgdMfVtd2G7kdju5nGvy1GT5pqZD69lqYbu81OKLp2cJogTL0PjeJ0Y4UMlteK6mvdtpLowTTF1jpGd4HMfASnSuLncQAmxTw9AE1xbbvL3QIOeYHB0q0A1DgjAmY+hYls7R4cKWczIMjeNjfWn9ReCja4LjY30YhnbbBVMjFeZr+yEXF9sstX2uVNtcX0kl0x3TQBNQ70a8dHmRpw+UGS9miJJkrUOdZWx8XWDPg9t7Kbq3+jPtBCGnrtc5v9hCSnhmssSHj1S2jO3cD5RhUOxLNE2kFdL1LsfHCpxbSPsl9LtQ9W6+f2Ik3SG3g4iLS601d8zqwny7RX+7O9XNKrlX2WoxHcxYfPH0LBnTZKhg0PQivnh6ll/44EEcZ+NH0DF1praQqFhdxFxTJ4oSqt2Q5U7AtWqHct5mse7RjWOiOGG+0aGSd7FNfU2enIQNr5ckkqxt8IGpMhESS9PWdq7bWTCFhGorwI8ibEOn3g1p+SGRBEcIhCYwAEOQBrp7fS76sxaWsUkXPthzZdS9FN2LpSQMY87PtZitdxnI2fhxzPRKh9evrfDRwwP75uSgDINi35KxDd43WWJ6pcNo0aUTlPB8n//4xsZTgwFMDGXohjHTtS5jJYfRYgZDSyuFV5v5bLXo75Y7YavFtBGGBFHCUCH9uOVdg5rn04oinE0+glvFQIRMxxpEMQgIowRDghcEXJjzuDC/wnw9JozhlQvzfPjxMZ6aKFKwLUxjo6HczBCuX5TWWpEGEVbPsK4umKtGNpGScsbk/EITQxMcGcxTbfl0goAgjDEMgWUYDPZleXKiwGifi2PoXK95Wy7Ee62Mupeie7oQREjaYYSmC3RdwxGrfyfJhtjO/UYZBsW+RtPSD5AFdLoRDS/BIa1pWOX5I30U3ByS9ENt6hrXltoMlxwarYCZ5SYDRZeCY2+66O+WO+HG3WcQxiRSkjMMLEOj6UXk3fTEYBkaOWP7H78133SSMFPzaXQDIpmQiNSo/eeL9Q1qrl++2GKucYnl5ggfOjrAkcE0HXa25jFccJhtdHstUjc3hEGcpH0zbpDNCHr3JlKmkiWa4MBAhm4IOcdA1zQGsjbnFxvYhs5kOcdHHxkkjAQLTZ/RonvLhfh+KKPulRyHpgkmihlOX6vT9WN0oJAxCcIE29DWYjv7AWUYFPuW1Z28bWjYusZXZ+Z58dLKBqNgAe87UGG0aJOxdFp+xNnZBnN1j4WWx/SSh2PC2ECBH35qlCODhZsW/d1yJ6zffTa6/lrTnyUv5OOPDPLVdxaoef5ajOFGN9L6ea9fpDY28LERUvLa1RVsXfLy5RpvXqtvKvF9ZinAeGeO0XKGoayD1AQzNQ8viFhqBYyX3LUK6BvrEebqXXK2QWHQpOUFtMMIzw9ZbAXYZvr7iOIEKSWa0Cg6GrgGjw3l+dTjg9imYK7m0YokRddiuM9dO8Ft1kholfulmbRXchw51+QTTwzx4oUlLi21aHgRhyo5np4o7Rs3EijDoNjHrN/Jt7shlxfqNKON9wTAWxevcnziBGGYsNT0qXshiw2Ply5WQYBrGiBa/Pnr0/zi8zamZmxY9HfTneCYOuNFl8vV9oYAKxj83PsP0ElicoZxk1FYNQZhlGxwew3mbRAQJQmuZZIkkmo7ACRSCuaaHt1406EAcHXB580ryyw3fUzDwNIFJddguR2w3A4YK7v0Z6wNLUbDOCGIYmzTZKnp88KFJa6vdNCAvKuTz5oULBOZpD2yh4suQoAXSKqdiMG8zVg5y3g5z5Vqm6xjgEx/zkmUbNpIaD0PsmbSdlJRy1mbTx0boROkf8wZy9hXRgGUYVDsY9bv5HUhqHnRpvfNRJKModEMEoI4IevoLLYDTEPDdUxsXbLUjjD0DisNnxNTubWd+OqHeFXuOkkkmibuSIb6Rt41aOlrrO7ITVunops33b/qJorjhNlGl9E+B9cyaHgBr1zpMFywmWv4JPkEoQmmax7dSGI7AksTGAZs1SxCPKsAACAASURBVACilsCF+RVqXsRTkyUG8jlevVbj0eE8S60AL4iYCRPeN5m2GO2GaTB0uuahr3i8Nd+kE0SYumC56fNXpxcJwgQJlHIWedviqZE8Tm8nPF7KrZ04xosuUsL15Q5CCKRMTw/v1W5qd5JqaxgaBWP/ZCHdiDIMin3L+p18nEiePVDiL07fnK56vJSjE0menSilMhIrHUyjFygNJbZlEkcRedfk6HABS9eodwIuVZvM1zzmGi28MMbr+rxzrYYZJjx6pI9PnjjMQCFL1kh3dJvtBIMgZrnZpR2GOI6BY6S78emah9P214Lg611T6w0SbEzRREimVzwMQzBd67DS6lLv2Kx0Q07PSUwhsAyNAxWXejvENnXMWxiGggmm4xBKqHsR42VJFKfxgcGczUjJJY4kpqERRQkX5ps0/BBT17i63OLacotyzqbRCXnl6jKzzWRNomSpGyAIODfXolK0mG34/Mxzk0yU82snAwRICaL373YOAg9CAdiNvNd0pJRhUOxr1vucf7p4mLPTdf7g9cW1xz9+uMRPP38M29DJuiZPT5YIgghL6BSzNi0/ou2HmLrODzw1im0anJ1r8MXTs3zxzWnOL3QJN3vjqwv8xpcX+IFjAzw9VWQga9GftxjqyzLVn8MxdebqHp998RJfO7fAcsun6NocGMjzvqkS48UMM22PlhcxNZBltOASxgnLjS7XltpkMiY5y6I/Z625yzw/YqUVMr3SRgBfPjPDhYU2yz5kdChkdcYKGR4dLXCgkmWpFXCgP0fBMTg1U6PaSDbEX0xA18BFUnQNKnmbuZqX9q6udxkquMzXuxRdizBKuLrc5o3pOgXXYDDvMFXKcHa2SbMV8NZMg+lauEEGHUACnRiW6gFvs8IfviT4vqdGGS+m0uC6lvaYlgJMTcML41sG9R9U9dbdSGDYTwZRGQbFvmfV52zqGr/+X7yf//IjC3z9nRWmhi2eHhsGCWGcfqCKGYtPHBthvD/DF07NUm37SAnf8/gwHzo8wHyjy+WlJl86M8fZhe5NC92N/MWZRV48vwi6oORaPDVZ4sdPTnByvMwffusyn3ttmnYQAxJNRJydreMakjhJMDSNszMrRPSnxV/VNl85u4itQTnv8KknhonjHFJApxsy3/LXXDbfuLDAW/Pvjq8Vg9+ICYMmrgmWrlHOmxwbrTCQs7i82OSvTs/yzkKbmhcjAMeCJ4aLHB3rQ8p0y15wTB4fK6ALbU3iO5GpdPlS06flR3TDiIWGT3/WJGsZLDU9mkHEVqGMGIgi6EYxdT/gerXLaDFDGKW7Zl0ILFOj5JobYhk38iDvuu82gWG/GURlGBQPFJomODExxCPDlTRYHCY3BYsNQ+PYWIlD/XkaQbgmQRFLiR/FzKx0WGh00eC2hgGgFcBAQacvY3BuvsWX3phmIGvwrUvLhEkEQhJGkuW2T0aDi1VBuxtxodqm1U34+rkqh4byzC2nLiJfCGrzTX637vGp40P0Zy3eWahx7lqbwbwOhsn15ZuNVggsdiG71MS2DUiyDOYS2qFkuJTjpz50iBhJ1/eJE8GKF+D5MFZyGSmk2km6SIvYspaxtjutewGz9S59jslwwebCYpsoTsjZBiNFG8fUaPsh0ys+m0d5wNBgIO8wmHNxbY0kSri+0mGoYFP3IvwwZiaIed9kKZ1LL260WQHfXha47RZ3k8CwHw2iMgyKB5LtpDU6zg3ZP0laQJUIgWvoyC33wBvRSF0iGdtkuRUy1w64sNRhoeGz2Ezw1+WK1oHZKx2gQxYo9gka3ZCXznvIGFxbx4sSBAKJ5Mr8CrNNNijIjuSgvql/K+VyEy6/vcKJwQZX5qtUsjZDw31892Mj5E2TpXZa4zCWCA4PZml2ozQtVdOoZC3mGl2iJFnLmNJ6u11NEwwWHKZXPJb9iFgm2IYBImS0mCNvV6n7NxtTAbgOFPMOx8aKRLFkutHF1DVyjkFW1zA1nSBOqHk+nfq777m++56Qe1/gtpvsNNV2PxpEZRgUDyx3mta4Kmk9VcwwVcngRwlz7a32wCkOkLEhY5u0/ZggShjJWQxmLTphtMEo3EgbaN8o+hesGqP0+tImz5ttbW8+byzEgJd+na7xp69c4cmBPANZg8mpfr7/0QkGCy7lbLqUx4lkoekTJglz1S79WQvXMhgvZZDASjvAD2NWvJA+16DPtRBS8vZsSCcIyTs6gR/TXjcGA8iZYBkG7XbA6WvLlPIuI30OQsLbMw2+c22Z5XaAHyZMlDOcmChx8kCZrG3c1H2vmDGpdcINu264+YSx6o8XEqTgpn83u/d2i/Vu+Ph3kmp7KznzzU5We4EyDIqHioxt8NFHhzAtna+emWW55dMJI/Qo4sVLbRq9+0oCPna8zJHBPl66uEy1HdINY54Y6+MHn5lAE2mfBjYPXd8XrtTgSq2ZfvP6Cr9bPs8njw5TyloMVzJkHQtdClxbp+DYCGC04KSLDzErrQ5Lns9AwaTPsfGDiKV2wETF5dXLHgLBjTJVOpBzoOgYzDR9DEPDMTUSCcutLqena8zWuoSJxDF0qk2fmVqHN6d7UuY3dN+rdcI1SW4hoR1EXK22kYkkETBRzGAYGnP1bk9qPMA2oOXH5BydbigZyNm4lrFmVLbju7+fPv6t5Myv17z7FnNQhkHx0JGxDb7rkSFOTpTxkpicbhBr8M5cjbMzDUJiHhsq8shQHxnb4NMfilhseAhLEESQt00anYCcvb+zZS4tw2demtv0scN98ORkhYGcwauX53hj9l0T92gZDg6VGC/nqDZ9ppeWuTIXMb9JQMYHrjeh0e6SzxsEiaQRJHz7UpX+rEmYxOQdEynA0AQr7YBqw8cxDc4tNBFSu6n7nhTp6eb8fINXrq4QhhGdMCZnW7yiCx4ZyTOUd1lpBSw20m54GqlL5vBAWqMy5RjM1jwk3FL6A/aHj38zOfP7OR5lGBQPJZomyGct8uuuPTNZ4fhYGdgoqW2bOoWcDby7szQ0jQ8fHuDsfJvO9kIV+4oLdbjw5maOLDi7DGeXV7BY2ao84iYaCTTrEf22j1bOMlfrsNzSaLVDXNuiFUTEUYJEgADbED1ZDYjiNN6xqi0lY8mZ2RpffnuR8/MNZmseCZKya2JbJmdmGnzi2BC1VogXxfixpOgYXF32cEyPZhAylLfpCbaS7fWg2Mp3v198/OvlzO/3eO6ZYRBCOMDXALv3Pn8kpfzf1j3+m8AvSilzW7yEQrGn3EpSe5X1O7uf/9gh2kHIZ1+a5jZN5h5ItmsUVpHAGws+V5dnKWY1Dg0UyGcsam0fP0kIgoRKwebYWB/PTJXozzrUvDT20PJTt1AxY3BmtsZX31rg2lKbbpDQDWNkAisyYswxaPgR8w2PVjfC0AW6AC+IMTWBroFAY67hMdqXQWwjmL2X0tvbYT+M516eGHzgE1LKlhDCBL4uhPhLKeWLQoiTQOkevrdCcc9Y3dkN5Bx+9sOHmChn+PW/PLdptKEPKOQ05lrJLaMRAjYVwnsQqUVQqydUWzWmKg4T/Vk+NFFBCokEJvuz+JHECyJc02C04HBuoYmhQ7MT8fZcjZm6B0KStXXiRJKKcAiyloFh6Aih4RgG3TCm6BostSP68zZRIii4JkmSZliZunbbFNK9lN7eDvthPPfMMEgpJbCaX2H2vqQQQgd+A/gZ4Mfu1fsrFPcaTRNMlLN8+OgQv9QO+exLl1led3R4ctjlUydGONKf58xcna+8Pcd8vUu9s3E3PpQVjBYyLDa7XG89gH6pLWjGcHq+S8PrMpSzEIZB3rGIpaTjh3S6EScmisRSstwJSYh57UqN1y9XmW342DqUsg5ZO+0f4VoGMoHJUoaJkkspYyGB6eUOYqlNJWsxVsowmLexDJ2sZaBpYlsppHul6rrdzKf7pTK7ipDy3u1TekbgFeAI8K+llP9QCPE/AJqU8l8JIVpbuZKEEL8E/BLA5OTks1euXLln41Qo7oYkkTS7IW/P1nh9ZoVmJ+SRkRzPjFe4tNTB1DQMXSCThItLDQxLo9WKub7SZqkdULRN2mHMcjuk2uqy0upwpfFeOT+kDDhQztuM9WV4/vFhJssZlloBjw/nsU2dd+aavD3b4MxMnZVOl4aftijVhM5In81wKUslY5J1LR4b6uPQQJapyrttUFvdkPlGF8PQsHT9jrJ49kqK4n5kPgkhXpFSnrzT593T4LOUMgaeFkIUgT8RQjwP/B3g49t47meAzwCcPHnyvfUpUbyn0DRBXybtT/30ZD/wbvC6P+swU/Ogl1vfl7ExdMFyO2C8P8tC06eUsfDDNC2zG+WZr3cRFxe43NhOXfaDQSeAjB9xbcXj6nydjKXjGDrdXoX1crvLXL3NufkajU6MlFDpMyg6Fo8PF3j2YD+T/VnG+jJkbGNNAVfTBDKGph9jG2mv78G8ve0FdzcX61sZmP2Q+XQn7ElWkpSyJoT4CvDdpKeH8yINpGSEEOellEf2YhwKxb1ks+B1xjY4NJBbWzBWO6CVMxalrMVzByuYPS1+P4y5tNRiseXz5GiB3/3mBS7W3huuJU0DXRc0uyHnllpohsET4yX8MJUStzWDN643uN6I12ItjeUIW7RwDMnkQJ6DlRwLLR+zGxJHCUIXjBZcltpBr4mRSRSn/SwmTf22C+5uLta3MzD7JfNpu9zLrKQBIOwZBRf4JPDPpZTD6+5pKaOgeK+zvhrW0bb2HdumzpPjJcI4gaODPH9siD/41hX+n69dvV9DvyPGXEAXzLTkTYH0ZgRaM8K2NfpzNmP9GepegBDQnzG4vNyk2vRuCsL7El6+2saPrxBHIZqm0+eaLLZ9kkQy0Z/lUCVPJZ8Ws93Jgrtbi/V2DMx+yDS6E+7liWEE+He9OIMG/IGU8s/v4fspFA8Et5JNWH/qOFQp8IsfPcyp6RW+fqG5l0PcEe0ATGtVBepm6hEcLBiUMjZ9jsVis0vDC7m84kGSEAO2Dt4Nh6QAeGXaY2b5AqWCzWLdJ5Hpz28gZ/Pj7x/ne58cwzGNO1pwd2ux3o6B2SrTCO6f7MWtuJdZSW8Az9zmHlXDoFDcAscwODpY5JsXmttSgt1tBGmzH0dCNWJLdVWAegzlJNky7dYGMkbqVurPGkSJyXDBZao/y0LdI2dq+MHWs5z1YNZbTftK36Xqd/m/v3QeEvjIo0NkbXPbqZ27lRa6XQNzY6ZRwws4s9DEMDVytsVo0d03vSdU5bNCsY/JWAZPjZd4dHCRtxa6t3/CLtJnwPOPV+jPOiw2urw9V6Prx7RCSX2T6jcJJDLVT9rMgOiArqcOhMvLqeqSpmsgBE+Nlzg6WsK/uky3u/mJw9jidWsRfOH0NYbLGT7x6PAdLa67kRZ6JwZG0wQkcGG+we+9dIWGH2LpGh84WCGMEo4O5ffFyUEZBoViH2MYGh9/fIiVTpfPvnSVC0vdbYqF3z2lrM7fff8E882Qr769QMGx8aIQS4aYgdxQsLfaLU4X4BppTOFGXBsqrsXhwSwjRZe6F1NrB6kcugY/eXKCYyM5zi80efPqMtV2WiVrkhod14BWdHMhYLqICZZbITN1jyO9rKXtcieKqFtlHm3XwHTDmCsLLX7v21e4Vu2Qc0x8P+LLZ+exTI0Dlextq+/3AmUYFIp9TjFj8XMfOswnjw3z6vQK/+9/eos3lzauvCN56DZhZRffN0ESJlCwTRxdo5J3CSXM1wME6Qlg1UjpAvIO2IbOuGuw6EVUGzEhaYBxoCB4fLjEswfKxIlGn2tTcGG+3uXaSoeBvMWjg3kKrsHBwTwnJktcXGhx+nqNetunHYKpCSbzBldWNtaQZ0xA6nTDgI4f0o1iHOPdrKTdqlO4XebR7QxMkkhmah6tMKTthwhdEMuEQsZkodFlvp4G1PcDyjAoFPucJJFIAaOlHKOlHB88MMAXTl/j5fPLaMR84vgo33t8EkMIZmptPvfaFf761DVe20wO9Q4YyDl0ooRnxstcWW5zerrBfENiGQbFTIShCWodSZJA1oaJYpZy3uIDvRTcmVqHZttHCMFYOYtlGhyp5Jhr+QRhTMYxGS446AIQsNgKEFIjiQVFx+LIUIGhvEu9G1DJmFxe7ODaBqPFDleX6qw0094L5YLN8fEihYzNucU2GcdcK3KD7club+d3cLeprc1uyJVqG5kkeGGChiBOoOVFSE3QXzD3hRsJlGFQKPY1m+1ShwouP/v+I/zEMzG6ENjrcvYnK3k+/dFH+eDhIT7/xnV+78WZOxbDW+W7Hh3kaKVAOWvz3OEBLENntORyfbnD9HKLph+StyOCWJCzTSwjPVVIBJrQmSznWDQtTA2yjkk5a5NIQMJM3cNoBfQ5BqahM1F2qXsRMoopZ00iaTDVn7qEik6qj1TOmlyvdghkwpmZBgLJ9HKHgmtiGQaG0ClnTDRAJglXq200TeCa+l3XKdxtamvHj3jzep1aO8A2NZ6ZKPH1c4uESCzX5EMTZR4ZLKw1LbrfKMOgUOxTbrVLjaKEbhSTM272p+dck2cPVDg+VuLTHznEC5er/LsvvcXb9Tt7/1PXFvnUiXF0IRgqOJiGzpGBPHGccHahzpVqQJSAo4Fe1jBjk44fstTwODpcxDI1RkouptDQ9TRL53rN4/hogU6U4PkhMZKhgkPBtcjZaV/uVjd1FeWddAetCUHLC8nYJsfHS8RSUs466MB8y0cmCVECdT/g0lKbhVYXz08oZkwylsEjw3kMXcPQNTrdgKYfkjUNDGP7i/DdpLYmiWSm7qFpMFnJMt/wKeUsvvuxATKmQSFrMlXOcaCSUycGhUJxa7bapV5cbPLXb88TRAmWofG9T4xwoLIx81vTBK5tcGCgj8n+Aj/wxDiX5hvUvIA35xr8/jcvMt26tT/7crWNkAnXax7dMM22Kdg6FxbbLNV9wiTNEgoTiKshmUED19DQTJ1ixmC85FLz0ljIRCmDH8WEsSSfsfCbabMeP0pAsrbgJrFMpS1W5yEErW7IbL1LmCTYhs5I0WW8lGGu3qXPNam2AhxD8I3zdcoZi6VmSBBFvDPfZLzoML3S4SNHB0ik5K3ZJuW6h6lrHB/ro5ixtvW7uJvU1tXmO5ahowvBeNGl7ukcH+ljsj+7Qd5jv6AMg0KxT9lslxoGMV85N0/WshgqGDS9iC+enuUXPngQx9n847yq5XRiqp9YSk4eHGAwb/P7L1zgtZmtO0mUXIdmNyLn2NiGxWDB4rVry1yaq7LkSXTSjKHVAHTGSTultdo+3zy3yFDRIe+aGL1q5YxlMJi3ma93cUy9txjGaLrAjxL8KN2FjxRdII0NdIOQqyudNO+/G5EkIUGccHQwv5YFdLhfcnahwVQ5QydMeGd6mdlal7xtMFHOEMSSVy4tgw5TpSzFrE03iDg1Xee5g/1rJ4fbBal3mtqqC4Gua5QyJiud1GgJIZjoz+La+3MJ3p+jUigUm+5Ssxkj3XW76Uc37xrUPJ9WFOHc5uO8mjVjuho/dGKCRwYLnJ2p85evX+Rvrmw0EIdKFh97fIyca2HoGvVOwBdfvchnv/Nu3tP6vKiOhIsLbRrtmGLeouBYWKaJY5mM5i0QMNbn4icJi1draFqCrgmG+1ziRDJWdNE0sWHBnSxn6AQRV1c6ZCwdy9RJEslCw+dAfzaNrSAISXB0naxlMr3SRNd0XEsjnzWptrqM9rn4cYQmdTphghvFOJZBO/DpRjG2gDBKNZYSKRFAJW+vyXZv9jPc6e+x5JpIYTLa55LZp0YBlGFQKPY1N+5SgyDGMjSaXkTeTU8MlqGRM+7so5yxDZ6aKHNstMj3PzXO69NVvvjaDOfnGgwXbZ46OMj3Hx8liCUXFxr82p+d4sy8d8vXXPAgCLo0/S5DhQwDeQvLyFDzI2S9i63rjBZdxkouugDL1ImihLA3N+OGwKumCeJEUm0FdPwI09Dpc2+epy4EpqkzULD45sUuURTTjSR5Swcp6AQJI30uhpGevqqtgKJrkEjJXKOLkDDb6DLa56DrGrM1j+srHmMld9eqke93f4U7RRkGhWKfs0GEzzH43idG+OLpWWqevxZj2MqNdLvXtTUd29T52OFhPjA5gBdESA3ypoll6dSaPr/18pXbGoVVajHUmnCt2cEwDWzD4NhYHwXXxDY1Fpo+wwWHhaZPrRNQbQX0Zy2u17ybUkmTRLLcDhjMW7SDmLAXfH9kKL8he0fTBIN5m2tLbab6s7T9hCfGC8zUuyAkhg5PTvShaYK3ZhusdAIMzaGSs3HNNJ6hidQ4+FFEzjaxNQ1dsKvS2Ds5bdwvlGFQKB4wDlRy/MIHD9KKInKGsSOjcCOrweobfd4BCUutnSW8vni5gWvAwYEcti6QEuI4wTQ0xosul6ttpvozWIZO14+4uNBkqpz63TVNEMtUGmOqkmO+0SWO0zjEUK8+YT2moTFSdukvWJyZbbDSDhnOOTw1VaSUsck7JlEiqWQtMpbOcMFBCAECwiBmpePznWt1kgQyts7xsT4myhm8IN630tj3EmUYFIoHEMcxbhtT2A3cXuB4p8xUG1ytNriw2GKi5DJWzjCQt9OAr0gzdepewKnpOrM1j0rO4fGRPI8MF7B0DU0IDE0wUcrQ8AIWmgHL7YCGF204YehCYGgarqvzoYMV2n5IIuHIYJ5IyjX3kGVoPFLOogFnZup05mKSWPKNC4tkTZ1SzkIIuLTY4kAlg6np+1Ya+16iDINCodiSrGvyY09P8ML5JRa9O5drqNfh25dr6Jrg9Gyd4yM5lloBJ8b6qHkRURjz1nyTeieknLXpyxhcWmqnDY4qubWgbRzHLLYCxooOWce8qVhtfYA3kRLLNBjuczAMDQMYLbqEcULeNdNMryih2gnJmDo13+fSQp0IwYCjc2CwQA2NC4ttDvXnCOIEZx/oF+0l+6PMTqFQ7Es0TfChRwb5R3/78R09fy6BRqtL1jKpuBaXqx61pkfTjyhmDF69XuPyUouaF1JwTGxDJ44lvp9WFa8GbUdKLsN9qVGAtKYjkWl66Sqr906UM0yWMzi9LKbVfgeWoa9pEXlhKukx1ufwxuVlLlY9zs93ePlak//01jxhGPLYSJ6cY6TGZp9oGO0V6sSgUChuiWPqfOyRYX7w2CyfP3PnMn1zjYiYFSpZF6npLHciGl6ElDDe52JqGg0vZKHlobU1gjBisWNxMErWCr8cQ8fQtNtWHq8P8N4oJ1LMmNQ6IX4UIRBUcjbzrQ6vXFtBJGAbqTDgciui0w0IowTHNkhk8tDFGdSJQaFQ3JY+2+LkwUGOlO7cpdJK4FJN8u3pDjO1FnP1Nm9eW+bKcpsE6M9ZBFHCmek6zU7A0eECk6VMWlfQ26mvuorCWNL2I8JY3rLyeL2cSNY2MHVBrRMyWnAY7nOY6s/y2EieN67WWGpEtEOQEZg6GAbMdwLOXGvw8uUqK830xPAwnRrUiUGhUNwWy9I5eaDM//e6AyvtHb/OTFOiXalS7SQ0OwHdoZjjY308PdlHX8ZIezgP5DA0jba/UaTuTmoBNpMTaXR9rq500DRBN4i4stRmtuYRyrR6ux5DvQN5Lc1U+vcvXMQ0NUxd5+9+YJInxko7Vmddz6p7C9h3UhirKMOgUChuSxQlzNe6BOFOtVrfxRKQtTSkJjgzXSdjG7imwUgxgyG0Ne2k9a6i9XIVt1IgXb1PSDbIiQRhTLUVMNWfIYwSTs3UubzY4rXrtZsaH3US6IYJGcek4Jp0A8nnT81wZDjHXJ27qmvohjGXllrM1bup3lLBYaqS3TctPVdRhkGhUNyWIEloBhGd8O69z60uoEGtE1J0LEb6bHK2RcsL8cOYdhCt9VPQNHHbBjmr3CqmkEhJf9YijiVvTtdpdkNq3YCF+s2t5mLg+qLP8UMZNCHoyxjUOgGdToiZM3Ycb0gSyeXFFmfnmph6euJZaHYxDY0DPTG9/YKKMSgUittiaRqWoeH5W4vubRfDgIyW9noe6LMxdJ1OEDHf8iln03TSwbyNpWv4YcxMzdsQK5ird1MpjThZ8/tvFVMYL7pMlDMcKKe78uv1DrqAjGXSZ5sYW2zUaxKqjQ5eEBOGEtPUsGx921LbmxHGCXON1BDknLR/RSqqF2/IrtoPKMOgUChui2FoHBvtY6yYu/3Nt8AVUMpbaKbGgUqOx4cLjBVSNdXxkstAn4tr6lxb6XC52uZKtc30ikfUMwCGruGFEZerba4td7i63KEbxutiCtrafYlMO9+ZuoZhaFTyNjJKTwQZS+fQQI7B3ObFew5Q6/h0g5Ca7/OxIxVc09q21PZWaEIggDiRCARJsv2+DnuJciUpFIptMdaX4fuOD/GdmRY7aRr6WNnk8Egfj48UOD5RhESg6RpeFCOlYLSYum4QsNDwmSy7uI7Jcjtgru4x2Z9NC9N6sQLLeLd50XjRvW0jnaxlMNafYTCyqXcjvCDkbz89xp995zpXau+6lAazUMlmONDv8F2PjVByDE4eGaDoWndlFExdS+U8Gh61TkScSIquxXhpd7SYdhNlGBQKxbawLJ0feXaKqytd/sO3p+/ouSeGLU6M9TPSn+WxkT4ylokfJpwY78MxdSzDw+gtjn6UhoMtU1/rz3C12qHphWiaoD9rYRkbmxdJwW0b6WiaYLToMlfv0q9pxK7JDz7l8vFHB/itL53irWse2YxGIZvBNnUm+nM8NVXE1o27Ngqr7z/Vn8XSNQbzMZoQjJcy+1J+W8h95tvajJMnT8qXX375fg9DoVAAQRBzvdrg337zFL//7cZNWT03cmJA40fff5jhYpYjgznyjoUfxUgJU72g6/rAsQCCKCHnGGu7/yBKGC26a+1B03anveZFsVzLFLpdsx3YmOEU9E4cV6otPv/6NJerbXRNMFXJ8oMnRpko53YlRXWr97/XJwUhxCtSypN3/DxlGBQKxU5IEskL5+f4p3/8oBmxKgAAB+tJREFUKudqW9/3f/zI4zwzWWaynGWpHWyZXbTZgr3ZvdvNUrqTecRSEvqpyqo0oGzZmLb+QPROuBU7NQz77wyjUCgeCDRN8OzBQf7B9z/FP/3TN1nobIw85DT49PMH+MChChO9rKCMbWy5W97Qd0Lbuphtt5verHW2y2hkMjtXkn0voQyDQqHYMY6p831PjvG+qRJ/ceYaXz1zlepCyFNHinz06BTvP1KhP2uvLd530qzmVvc+SE1vHkSUYVAoFHeFpgmGi1k+/dyj/Oz7jxJFCVKQCt8ZKiP+QUQZBoVCsSusbxWqeLBR5lyhUCgUG1CGQaFQKBQbUIZBoVAoFBtQhkGhUCgUG1CGQaFQKBQbeCAqn4UQi8CVPXirCrC0B++zX3jY5gtqzg8DD9t8Yes5T0kpB+70xR4Iw7BXCCFe3kn5+IPKwzZfUHN+GHjY5gu7P2flSlIoFArFBpRhUCgUCsUGlGHYyGfu9wD2mIdtvqDm/DDwsM0XdnnOKsagUCgUig2oE4NCoVAoNqAMg0KhUCg28FAaBiHE3xFCnBZCJEKIk+uuf1II8YoQ4s3ev5/Y5LmfE0Kc2tsR3z13OmchREYI8XkhxNu95/36/Rv9nbOT37EQ4tne9fNCiN8UQjxQgv+3mHO/EOIrQoiWEOK3b3jOT/fm/IYQ4gtCiMrej3zn7HDOlhDiM0KId3p/3z+x9yPfGTuZ77p7tr12PZSGATgF/DjwtRuuLwE/JKV8EvgF4N+vf1AI8eNAa09GuPvsZM7/Qkr5GPAM8BEhxPfvyUh3h53M93eAvw8c7X193x6MczfZas5d4J8A/9P6i0IIA/i/gO+WUp4A3gD+uz0Y525yR3Pu8Y+BBSnlI8Ax4G/u6Qh3l53M947XroeyH4OU8i2AGzeEUsrX1n17GnCFELaU0hdC5IBfAX4J+IO9GutusYM5d4Cv9O4JhBCvAuN7NNy75k7nC5SBgpTyxd7zfhf4UeAv92TAu8At5twGvi6EOHLDU0TvKyuEqAIF4PweDHXX2MGcAX4ReKx3X8IDVCW9k/nuZO16WE8M2+EngFellH7v+18D/iXQuX9DuufcOGcAhBBF4IeAv74vo7p3rJ/vGHB93WPXe9fes0gpQ+CXgTeBGdLd87+9r4O6x/T+lgF+TQjxqhDiD4UQQ/d1UPeeO1673rMnBiHEl4DhTR76x1LKP73Nc58A/jnwqd73TwOHpZT/QAhxYJeHumvs5pzXXTeAzwK/KaW8uFtj3Q3uxXz3O3cz501eyyQ1DM8AF4HfAv4R8M/udpy7yW7OmXTNGwe+KaX8FSHErwD/Avi5uxzmrrHLv+MdrV3vWcMgpfxbO3meEGIc+BPg56WUF3qXPwScFEJcJv2ZDQohviql/PhujHW32OU5r/IZ4JyU8v+82/HtNrs832k2usrGe9f2FTud8xY83XvNCwBCiD8AfnUXX39X2OU5V0l3zn/c+/4Pgf96F1//rtnl+e5o7VKupHX0jpmfB35VSvmN1etSyt+RUo5KKQ8AHwXe2W9GYadsNefeY/8M6AP+x/sxtnvBLX7Hs0BDCPFcLxvp54E73Y0+aEwDx4QQq+qbnwTeuo/juefItKL3z4CP9y59D3Dmvg3oHrPjtUtK+dB9AT9G6kP2gXngi73r/yvQBl5f9zV4w3MPAKfu9xzu9ZxJd8ySdKFYvf737vc87uXvGDhJmvVxAfhtesoAD8rXVnPuPXYZWCbNTLkOHOtd/296v+M3SBfM/vs9jz2Y8xRpVs8bpHGzyfs9j3s533WPb3vtUpIYCoVCodiAciUpFAqFYgPKMCgUCoViA8owKBQKhWIDyjAoFAqFYgPKMCgUCoViA8owKB4KhBC7Ln4ohPhhIcSv9v7/o0KIYzt4ja+uV8lUKPYDyjAoFDtESvk5KeWqHPmPkmoNKRQPPMowKB4qRMpvCCFO9foQ/FTv+sd7u/c/6mn0/4fVfgxCiB/oXXul16fhz3vXPy2E+G0hxIeBHwZ+QwjxuhDi8PqTgBCi0pMkQAjhCiF+XwjxlhDiTwB33dg+JYR4YZ24W25vfzoKRcp7VitJodiCHyfVCHoKqADfFkKsats/AzxBqjT6DdIeFC8D/wZ4Xkp5SQjx2RtfUEr5TSHE54A/l1L+Edwsi7yOXwY6UsrHhRAngFd791dIq7L/lpSyLYT4h6RSyf/7bkxaobgTlGFQPGx8FPislDIG5oUQfwO8H2gA35JSXgcQQrxOKiHQAi5KKS/1nv9ZUl37nfI88JsAUso3hBBv9K4/R+qK+kbPqFjAC3fxPgrFjlGGQaF4l/V9KGLu7vMR8a6r1tnG/QL4KynlT9/FeyoUu4KKMSgeNv4z8FNCCL2nKvo88K1b3H8WOLROy/6ntrivCeTXfX8ZeLb3/59cd/1rwM8ACCGOAyd6118kdV0d6T2WFUI8so35KBS7jjIMioeNPyFV1fwO8GXgf5FSzm11s5TSA/5b4AtCiFdIDUB9k1t/H/ifhRCvCSEOkzZ/+WUhxGuksYxVfgfICSHeIo0fvNJ7n0Xg08Bne+6lF+i1n1Qo9hqlrqpQ3AYhRE5K2eplKf1r0sZF/+p+j0uhuFeoE4NCcXv+fi8YfZq0cdG/uc/jUSjuKerEoFAoFIoNqBODQqFQKDagDINCoVAoNqAMg0KhUCg2oAyDQqFQKDagDINCoVAoNvD/AyNwE0lbefwdAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "awgrbQqXJTu0",
        "outputId": "ca188b35-9de8-40a5-9830-81b656af7095",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 436
        }
      },
      "source": [
        "housing_df.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\", alpha=0.4,\n",
        "                s=housing_df[\"population\"]/1000, label=\"population\", figsize=(10,7),\n",
        "                c=\"median_house_value\", cmap=plt.get_cmap(\"jet\"), colorbar=True)\n"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f3ea4a48ef0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 48
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlgAAAGSCAYAAAA//b+TAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3iVVbrw/+/aPclOryQhJAFC76F3RQRFEXsv2I6OM+M4MzrzXr8pv+N7zniazvGMOuOMOuoRddQZ7CgKKCpFUBDpRUoCAdKzk+zstt4/no2BkISUZ6fA/bmu58rez37KSt131r3WvZTWGiGEEEIIYR5LdzdACCGEEOJsIwGWEEIIIYTJJMASQgghhDCZBFhCCCGEECaTAEsIIYQQwmQSYAkhhBBCmEwCLCGEEEJ0KaXUfqXUFqXUJqXUhvC+JKXUcqXU7vDHxPB+pZR6XCm1Ryn1jVJq7EnXuSV8/G6l1C0n7R8Xvv6e8LmqtXtEggRYQgghhOgOs7XWo7XWheHnvwA+1loPBD4OPweYDwwMb3cBT4ERLAG/ASYCE4DfnBQwPQXcedJ5885wD9NJgCWEEEKInmAh8Hz48fPAZSftf0Eb1gIJSqk+wIXAcq11uda6AlgOzAu/Fqe1XquNauovNLlWc/cwnQRYQgghhOhqGvhQKbVRKXVXeF+61vpI+HEJkB5+nAUcOuncovC+1vYXNbO/tXuYzhapC5spJSVF5+bmdnczhBBCiC6xcePGUq11alfdb8qwYbrS4zHlWtsPHtwKeE/a9bTW+ukmh03TWhcrpdKA5UqpHSe/qLXWSqmIruUX6Xv0igArNzeXDRs2dHczhBBCiC6hlDrQlfc7VlLCf8yfb8q1Ln/pJe9J46qapbUuDn88ppT6B8YYqqNKqT5a6yPhNN+x8OHFQN+TTs8O7ysGZjXZvyq8P7uZ42nlHqaTFKEQQgghsJq0nYlSKkYpFXviMTAX+BZ4CzgxE/AW4M3w47eAm8OzCScBVeE03wfAXKVUYnhw+1zgg/Br1UqpSeHZgzc3uVZz9zBdr+jBEkIIIUTkONxucmbMMOdiL710piPSgX+EKyfYgCVa62VKqS+BvymlbgcOAFeHj38PuAjYA9QBtwForcuVUg8DX4aP+2etdXn48b3AX4Eo4P3wBvBIC/cwnQRYQgghxDnO7/FQ/OmnXXIvrfU+YFQz+8uA85vZr4EftHCtZ4Fnm9m/ARje1ntEggRYQgghWuT3+ykqKsLr9Z75YNFuLpeL7Oxs7HZ7t7ZD0bb0nmg7CbCEEEK0qKioiNjYWHJzcwmndIRJtNaUlZVRVFREXl5et7bF7naT3XUpwnOCBFhCCCFa5PV6JbiKEKUUycnJHD9+vLubgt/j4UgXpQjPFRJgCSGEaJUEV5HTU762kiI0nwRYQgghzmn79+9nwYIFfPvtt60e88UXX3D99dcDsGHDBl544QUef/zxrmpmRNndbjIlRWgqCbCEEEKIM9i/fz9Lliz5PsAqLCyksLDVWpq9it/joURShKaSQqNCCCFMpbXG6w9izK7vvP379zN48GBuuOEGhgwZwpVXXkldXR0ff/wxY8aMYcSIESxevJiGhgbAWP3jwQcfZMSIEUyYMIE9e/YAcOutt/L6669/f123293svaZPn87YsWMZO3YsX3zxBQC/+MUvWL16NaNHj+axxx5j1apVLFiwAIDy8nIuu+wyRo4cyaRJk/jmm28A+O1vf8vixYuZNWsW+fn5Pbq360SKsCsKjZ4rIt6DpZSyAhuAYq31AqXUS0Ah4AfWA3drrf2RbocQQojIK/U0sPSrYirrfSREObhsbBYpbmenr7tz506eeeYZpk6dyuLFi3n00Uf505/+xMcff0xBQQE333wzTz31FPfffz8A8fHxbNmyhRdeeIH777+fd955p033SUtLY/ny5bhcLnbv3s11113Hhg0beOSRR/jP//zP76+zatWq78/5zW9+w5gxY1i6dCkrVqzg5ptvZtOmTQDs2LGDlStXUlNTw6BBg7jnnnu6vSRDc+xuNxmSIjRVV6QIfwxsB+LCz18Cbgw/XgLcATzVBe0QQggRQVprln5VTEMgRGZCNBW1PpZ+Vczt0/M6PZi7b9++TJ06FYAbb7yRhx9+mLy8PAoKCgC45ZZbeOKJJ74PsK677rrvP/7kJz9p8338fj/33XcfmzZtwmq1smvXrjOe89lnn/HGG28AcN5551FWVkZ1dTUAF198MU6nE6fTSVpaGkePHiU7O7u1y3ULv8fDcUkRmiqiKUKlVDZwMfCXE/u01u/pMIwerJ73kyaEEKLdGgIhKut9JMY4AEiMcVBZ76MhEOr0tZsGaAkJCW0+/sRjm81GKGS0JRQK4fP5TjvvscceIz09nc2bN7Nhw4Zmj2kPp7Ox985qtRIIBDp1vUiRFKH5It2D9XvgQSC26QtKKTtwE0YP12mUUncBdwHk5OREsIlCCCHM4LRZSIhyUFFrBFkVtUaa0Gnr/P/yBw8eZM2aNUyePJklS5ZQWFjIn/70J/bs2cOAAQN48cUXmTlz5vfHv/rqq/ziF7/g1VdfZfLkyYAxNmvjxo1cffXVvPXWW/j9p49OqaqqIjs7G4vFwvPPP08wGAQgNjaWmpqaZts2ffp0XnrpJX71q1+xatUqUlJSiIuLa/bYnsrmdpMmKUJTRSzAUkotAI5prTcqpWY1c8iTwKda69XNna+1fhp4GqCwsNCckZJtUFIFz38OF42EEedI39qmTV42b27guuvicDh6Rk0WIUTvo5TisrFZLP2qmMOVdd+PwTKj1tOgQYN44oknWLx4MUOHDuXxxx9n0qRJXHXVVQQCAcaPH88//dM/fX98RUUFI0eOxOl08vLLLwNw5513snDhQkaNGsW8efOIiYk57T733nsvV1xxBS+88MIpx4wcORKr1cqoUaO49dZbGTNmzPfnnBjMPnLkSKKjo3n++ec7/fl2taDHQ5mkCE2lzJrlcdqFlfodRg9VAHBhjMH6u9b6RqXUb4AxwOVa6zP2HRcWFuoNGzZEpJ1NHSiFP30Cl46CSQO65JbdbvXqOtau9fLDHybgcsnEUiFEo+3btzNkyJB2naO1piEQwmmzmBJctaVO1clyc3PZsGEDKSkpnb53V2jua6yU2qi17rI6EEOTk/VL8+ebcq2xL73UpW3vqSLWg6W1/iXwS4BwD9bPwsHVHcCFwPltCa66Wr8U+PWl4GrHJA+vDxr8EH/6P0O9wvTp0UybFtVjKgoLIXo3pRQuu4zG6U1sbjfJkiI0VXcUGv0jcABYE35D/7vW+p+7oR0tak9wBfDuRth1BB66DCwmdwDt3RvkyBGYMsWCxRK5AKg9wZXXq/nggwb697cyfHj7pxsHApq33y4jJsbKBRckSGAnhGhVbm5um3uvwOjxEu0T9HiolBShqbokwNJarwJWhR/3iOrx738BxaVw83xwnBQjeBuMHqmE04blt2xSAQzONj+4Ali1SrNrV4jhwy2cYdJMlzlyJMTy5T6KimwdCrDKy/18/nkNdrtixox4XC4JsIQQortJn6O5ekSw0x12fgf7isB/QWOA5W2AP7wB5dVww1wYlt+2a2UlG1tn1daGOHIkQP/+9u97dS6/3EpVlYWEhK4LQmprQ7z5Zh2FhQ4KChynvd6vn4W7744iLa1jEWVqqp0bb0wlKspy1o/5qqgI8MorFSxalEBGRs8rLihEW2itpac5QiI1Drq9bG43iZIiNNU5G2AFKsFSA74GiHLC6tXw2RdwyA/uRCgpb3uAZZbPPqtn2bI6fvrTBLKzjTfjxERFYmLX/mGrqgqxebOPpCRLswGWxaIYMaLjwYJSitGjT1+iojXFxQ188kkVc+YkkJZ2epvaY/36GurqQsyaFd+p67RFVJSF/v0dxMR0LJAMBDR/+1stkyc7ycuTAE10PZfLRVlZGcnJyRJkmUxrTVlZGS6Xq7ubQtDjoVpShKY6ZwOsiy+A0jKIj4MlSzT//u8hPLWK3IkWbroWJg/v+jaNHeskLs5CRkZkvy2l5bDkbbhoFgzoB34/lJWH8NQE6dfPRmamjZ//PJ6EhJ7Tu7RjRz2rVlUzYEBUpwOslSurqK0NMmVKLA5HZD9Hl8vC3LkdD+SUgqgohc0mb2yie2RnZ1NUVMTx48e7uylnJZfL1SMqu58oNCrMc84GWEOM1RXweuHVV+HQIUXuQE1NAxzeDdGXdn2bUlNtpKZG/lsSCEJ5lZESDQQ0Dz5Uz6erg4we5efWW6KZPt1FSkrP+lWbOjWOrCwH+fmd/0/vjjvSCQR0xIMrM1itikWLeun0VHFWsNvt5OXldXczRIRZ3W7iJUVoqnM2wDrB5YKLLoKghrvuVBQVw5BB3d2qzvH7Na+95iM+XnHxxaf39mSkwq/vMwblHzgQZMuWeqorNX362EhNtXDgAPTtG5lB+x3lclkYPDjalGslJ0uqTQghThb0ePBIitBU53yABXDHHYpFixTp6UZK5ky0hv/dB32iYE5m5NvXXrW1mi1bAkRFWZg3T2O1nv5JnQieMjOtPPjzaNLSLIwe7WTTJvjjH+Huu6GttQVLS0MkJamIlpEQQggROYoIL058DpIAC7DbISOju1thnoQEC/fe68LpVM0GVyez2xVz50Z9/zw/HxYtgrYu/7h/f5Ann/RyxRUOJk6UniEhhOiNLG43sZIiNJUEWB2gFNzUv7tb0bq+fTs2hiouDk5aL/WMUlIsTJ1qJze3Z43Z6k579nh5/fVKCgqcXHZZgvTsCSF6vJDHQ52kCE0lPYI9xPLlQZ57LtDdzWg3t1uxcKGD9HT5UTph5UoPwSCsX19LaWnv+54KIc49J2YRmrEJg7wr9hBuN8RHvixTp/WQmng92tixUdTXh+jf30VionQSCyF6BwmwzCV//XuIyZN7/o/lsWPw5+dg1AhYcFF3t6ZrhEKa994rYdAgNwMHtm39pHHjYhg+PAq7XQb+CyF6B6vbTbSMwTKVBFiizTy1UFkFR0oie59QSPeYwCQQ0OzYUUNcnK3NARaA0ymdw0KI3iPk8dAgY7BMJQHWWaioCD78GOacBzl9zbtufh785D4ivuj0mjXl9OnjIj+/+wtsOhwWHnhgYI8J+IQQIhKkkrv5JMA6C23bAau/0Oze7efOxRby8837Nmd2Qd2voUNjiYnpOT+aElwJIc52yu3GKSlCU/WcdzFhmimTjHILn30aoqam46mqtWth/Zdw4w2QlAR1dUFee62MkSOjGTOmfYs1t0diYufWGhRCCNE+2uMhsFpShGaSAOss5HbDpAmKSRM6t27fd98ZW2WlEWDt3hPiyT8FGdi/jGeficHhUFTXgcsODqkxarrNm71UVQWZMaP7U6VCiLObUmCVHKGpJMBqRSik0ZozVkM/Wy1aZBQdPZEW7JNhZ/78eDLTfVgscOg4PP0+ZCbDPRd3b1vPRmvX1lNSEmDKlGhstnPzZ1AI0TWU2419ukkpwuclRQgSYLXqjTcCNDRobrzx3ExZuVynjrlKS4Pf/ioaMBZdDoaMLRDsnvad7W64IR6/X0twJYSIPI+H0OeSIjSTBFitGDnSQjDYemXNyko/O3d6GD/+3FsS5cPl0Ae4c17zr+/YEeTddwNccYWd3NyeUbagri5ERYUmK6vn94W73T3jayaEOAcosMqfHFNJgNWKQYPO/CZcUuJlzZpyRo2Kw+U6/filS/1oDYsWnV2DlCqq4MV3wWaBO65t/pj9+0Ps3Rvi0KFQjwmwVqxo4IsvfDz4YCwJCT2jTUII0d2U2411mkkpwmckRQgSYHXa4MGxDBzobnGcltMZueVlqqs1W7cGGT3aSlRU1/aexcfCT+8AmxVczuaPmT3bRv/+lh4TXAFMmOAgM9NCXNy51dsohBCt8nhgjaQIzSQBlglaGwQ/f37keq62bQvy4ot+4uIUw4Z1bcrLYoHLzm/9GKdTMXBgz0rFpaVZSUvrWW0SQohup4y/68I8EmD1YqNHW4mLUwwc2LW/FcGgprw8RGqqBCpCCHFWiHHDZJNShH+UFCFIgNWruVyKoUO7PsipqdGUlkqAJYQQZ41aD6yTFKGZJMA6x5VXQkJc+7qGExIsMkBcCCHOJrIYoekkwGpFgx9eWw/ZSTBrSOeuFQrpHlfGweeDtz+C2ZMhJ6u7W9MxPh/s2Q/5OUbdLjP4/RqfD2Jietb3SwghIibGDZNMShH+j6QIQQKsVtU2wNZi8HjbF2CFQppDh/z06WPD4bCwbFkDX33l58c/julRb9oOB1w2F+Jiu7slHbdtFzz1Aiy+FiaONeeaq1f7OHAgyG23RZtzQSGE6OlqPbBeUoRmkgCrGVob6zIlueEn8yC6nYXcd+1q4Mkny7jyynhmzHATFaWIjVU9coZGfFzHzvP5NA5H9weL/XPh2oUwqL951ywstDNkiPxqCCHOMT3wPao3k3eRJmp88NxWmJUG0UEYkNO+8wMBSEuzM3u2m4ICo0DUzJkOZs48e5bb2bIlyD/+EeJHP7KRkNC9QVasG8aPgqIioyfOjCA2Ls5CXAcDTyGE6JVi3DDBpBQhkiIECbBOEQjAx8sgIQY2H4Wt2+GXdxhFNdsiFIL/eQpioq3cdXt8ZBvbjTIyLEycqInuIRm0tevg3Xfgpz+F7Ozubo0QQvRCdR7YKClCM0mAdZLaWtiwHoYPhzkXQkDDul1wwVgjZXgmSkFmBriiIt/W7pSaqrjwwp7zo1M4DtJSoU+f7m5J2/TECQ9CiHOczCI0Xc95l+wB4uPh/vvB7YZAEPaUwM4SmDoMYtowQ00puOaqyLezN/H7oaYGkpJOf622NsQHH3iYPTuGxMSO/2YnJBhbb+DxBHjppRIWLEghK8ukaY9CCNFZ0W4olBShmSTAOkkwCK9+CEPz4YXnAQVjpsJX9TDNeXovVp0XfAFIcHdLc9slEIDjZdAnvWvv++ab8OWX8OMfQ2bmqa81NGgOHw5QXx9qU4CltUa1pSuxB4uOtjJ5cjzJyWfX4t9CiF6uzgNfSYrQTBJgncRigcH5QBA++gIqvLC7HoLTwHEE1n8Ddy5qXNz4ww1w6Bj88PJubXabfLsT/r4MfnQbpDTTmxQp2dlw9KjRK9hUUpKV++5rW2OefiaAvwHuvcfaq4Msi0UxcmQvroshhDh7ySxCU0mAdRKlID/dz8GDIbKzHJRsURw/AhNCsPJ9KCqDXSNh5FDj+Bkjoa6h/fc5dgwOHoSxY7tucc2CfLh+ISSZlErz+2H1ZzC+EGJbiRcmTTK2jtq508ejj9bx9Q47c+c4enVwJYQQPVaMG8ZKitBMEmA1sXKlj23b/FiDNgLVVnYdgp++ADcmg8ULS9+BEUPCdbLioCOdQevWwWefQW4upKS0/3yfDw4VQ16/tgdo0VEweED77wVwoARcDkg/6ZMNhYx2BIMdu2Zb+Xxw4ECQ+CgLCy+KTDRaXGx8H5zOiFxeCCF6vjoPbJIUoZkkwGpi4UIXM2Y4ubrCyiW3QvUoaKiDa34ONp/xht/ZTpTZs2HEiI4FVwC79sBLr8M9iyEnwmUJgkF4dhlkJME9lzbudzph3oWRvTfAiBEOnnkmgdJSzfDhrQdYx8vAam1fL53HA8+9CPPmQqFJleCFEKJXklmEppIAC3j//WqcTgt2ezR33BUgMdnKjAvhySfgo+9g6nBYux9m5EO6CSk2t7txTFIo1P404cD+cNv1kNVKWYJg0Ag2OstqhVvmGj1Y3SUry0rWGdZK1BqeetFYuPpHi9t+bbcbbr7BKPMghBDnrGg3jJYUoZkkwAKsVkVlJVx9tSYYtII9xMb98MZs6JsMsS54fzf0S4SRJtZa2rkT1qyFm29qX5DldEJBC+m+6hpYuRbWb4KrF8CoTi5SDZCfeeZjOioU0mzeHCIz00J6ese7BpWCy+eDswOBYE7fDt9WCCHODvUe+EZShGaSOQPAnDludu+xEbQEwKIgxkHf/nDJVPg/l8L8IfCTqTA0zdz7ZmXB5EnmDnTf+C288QFUVMPGrbBsTeNrK1fCX/8KXm/L59fUwPLlUF9vXptaU14OL70U5JNPOj+Ya/ggGJhnQqOEEOJcZDFpE4D0YAFGjahvd/gYNtHBcauFeTPh9kXw9nq4dCIUlcPKrRAzGjITzbuv2w2DBpl3PYCJo6FfFsS5oSEA1bWNr+3ZA999ZwRPrhZqXFZVwddfw6hRENVKRfrSSqiu63zvVnIy3HWXjdRUmR0ohBDdJtoNoyRFaKZzOsAqKYFvt8K0qYobbool43NFwQC481ooKoU9R4zjdhyBDd/BsCxzA6xIiI6C/JMWqD556NKNNxrBVWIrn0N2NvzsZ833qoVCmmXL/Kxf78Mb6yDgcvDQDZDaiXFpSikKChSHS8B3HNJlLJQQQnS9eg98KylCM53TAdbGr+D9ZZDbD+bOVIwaCgnx8Ok+GJ0Js0YYx80eAnkp0L+Lq6B3RCAA738CY4ZCdpPxYlFRrfdKndA0uAoGYfN2IBjigQca2L8/wOhCH/f8NonYaNiyzxgDNbwT6bnlnxrjp669rOPXEEII0QmS3jPVOR1gzZwB/fMhP98IKvqkg9cPRZVQkAJx4TRalAOGnGEWW0/hD8Deg0aasGmA1VGHj8KTL4IKWMDmJMptI2MofA2kHIDPP4X6APzzzRDbwYWur1xgrDXaG7z3NeSmwtBOlMjQaFSv+YyFEGe9KDeMkBShmc7pAMvthsGDG5+/8C6sXgOj+sOyfXDrld3Xto6KcsH9t3X8/O17jDUWxw1v3JeZDtEWeGe1ojrKTvQoO/v6KNzVkBEL54+HVzbBmoMwt4NjymKim99fXQ2bt8D4cS2PG+tqGQmQGNO5a+zARyJWMs7tX0EhRE/h9cA2SRGa6ZzuENQaNu+D41XGc6sFVn+h+dWvQ6z9XKM1fLkFnv4beMNL4ngbjDRcd6irg61bI1s9/ZP18N4q42tzgtUKhVOhGig7ojlarjm8UVOyDiweeGEbaCeMikA5hwMH4e334EiJ+dfuqLF50KeTY/FysJMqVf2EED2JzCI01Tn577PWxpihCg8sWQUTBsLCKXDDfHjvdXhtI6z9TLNypYJoKK0wgqqAFR59BvKy4bpLz3QXc4VCmt27FS+/DPfdZwxGj4QbLoVAsLFavdbw0g74391Qbg2BguDuBsoO21ix3caDtQo1HmblQFozCzr7/PD1TsjLhLQOrCs0bCg88ENI7wXj39ojRv4KCSF6kig3DJMUoZnOuQDrzRVw4DDcdz0kxcKd8+BvK+DpMrj3Mnj0XxWhIGRkKfLyIC8PZk0wxmhpDaOHQFoHl7jpKK83xOOPlzJyZBT33BNLZqbmr39tYNAgK5Mn2zt17bogHPRBfyfYLRDbJEjyBWFDGUSlQU6msS4gbitEwagCSEmCAQlwz8TmlxDaVwx/WQozx8KNFzXuL62AQ0dhWH9wtPIpWCyQkdGpT1EIIcSZeD2wQ1KEZjrn/o1Oioc+KY3BQP8+0DcNUsKlBtLT4c9PKC6c6WfpUi8NDfr7WXVKwUWzoXBE17bZalVERysqKwP06wcWiyI11UJ8fOcHSX/hgceOwLYWCovurARvAG4bDrHHLLijbOTm2Rg+28awqYqiY7B1Nzz2DhyrOv38fhlwzVyYNe7U/W99Ak/+DbZ/1+lPwRTBIPxjKRw82L7zqmvg6RfgH+8ayx4JIUSvpUzaBHAO9mBNb/JGrxTcNPfUfe4YWPK/PjZu1Fx4oY2hQ9v3Zdq0Cfbtg8suM6dKu92uuPLKBI4e9X+/7+KLzVkccFgU1CRAP2fzr/eLhcvzYUI63HQbFBcr6itgbVmIjZuDOP0WRqdZsVtUswFGlAvmTDh9/4Thxtiuvj0o9RcMnjr2rC0Ol8D2XfCdEy6Y2bjGpBBC9CpRbhgqKUIznXMBVltNmBCitLSBuDgXe/dCQwMMHdq2c48dg/37jTdss5bByciwk5HRuXRgc/o4YFErY6PinTA3XLj0wcXwj7fhz3+Bmt0Big75ccUEGeiKYfJQG3EtBGnNGT7A2HoKqxWuvKL95/XPhQVzITGhMbgKhsewmbkEkhBCRJTXAzslRWgmeQtowZQp0SQnx7N7t43t22HLlrafO3cu3H8/2DsRD23e7OP48QhOF+wgrx9GFcI9t2ocIU2wOki918eKNcYyO+caux1GDNasXO5n/foAfj/8/vewZEl3t0wIIdpJZhGaSnqwWjBsmJW777YydGjrS8u0pLO9Fw6HMfbKbLuPwgdb4YaJHavllJkDuBSXnefEYgmyYoWFEZMd7D4KOw5CH5OKm/Ymr72m+f3vYdWqBl57zUpMDERHn/698/mMnq3OBN5CCBERLjcMlhShmSTAaoHTCdOmdd/9hwwxZ4xVUw1+KK+FQAcGZAeDsHwd1NTB5NGKKVNiOHRIc/V8xZufgqfO/PZ2Fa83RHl5kD59bKjmpkM2Y8te2F8CM2crjh+HCRNs7NoVpLjYx5QpDpr+ev35RWOtyNuuj8AnIIQQndHggd2SIjSTBFjdqLZWc+SIZsCArutTHZ5tLFrdxhjiFFarUSustBLcLhg7FtLTjXIWI4d1vtdu3XrYfwAWLTR68LrSK69UsmWLlzvuSGLIkLaVjH/jEzhWAXcvVPz2t0a31Ecf+aiu1hw7dnoEO3p46yUphBCiW3Vxek8pZQU2AMVa6wVKqTzgFSAZ2AjcpLX2KaWcwAvAOKAMuEZrvT98jV8CtwNB4Eda6w/C++cB/w1Ygb9orR8J72/2HpH4/CTA6kZHj2o2bw61K8DyegM4HFYslo6nDzsSXL2zDA4cgrtvM2bNvf0h3HI5DCsArxee+jMMHwoXnN/hZrF5C+z9Ds6fDSkm1xrbttuo2ZWWfOp+f8Co4J+YaCU6WhEd3fbvxSVTYW8x5J9UwX7mTDsDBljJyjr9OlMndrT1QggRYS43FHR5ivDHwHYgLvz834DHtNavKKX+iBE4PRX+WKG1HqCUujZ83DVKqaHAtcAwIBP4SClVEL7WE8AFQBHwpVLqLa31tlbuYbpzPsDyeIyto8Usg0GjLEPfvpCW1r5z8/Mt5EpYinoAACAASURBVOe3718Gj6cBh8NGXHum7JlEnagHFqHrX3c11NSYH1wFAvDIH8AagMf+BRIS4PNtkJ4AH22Gvqlw6aVxzJ8fi8Nhob5eU18PSUmtf6bjBhnbyex2RW6ulSVvwrFSuPsGo1SFEEL0aA0e2Nt1KUKlVDZwMfAvwAPKGJtxHnBiEMXzwG8xgp+F4ccArwN/CB+/EHhFa90AfKeU2gOcKAy0R2u9L3yvV4CFSqntrdzDdOd8gLVmLezcCff9oGPnHz0KL74IU6fCVVed/noopDvV29RUSkonVxnuoAXzGh9PGAcjhsKAcPkGlwt+8sPO3yM21tjMFgxqhmRpKissNDRAMAQ7S8FuhWlDISEGlFI4HMb36csvgxQVhbj++o7nKcsqoboW/P62B1jBoOatt6rIzLQxcaIU1BJCdDHzUoQpSqkNJz1/Wmv9dJNjfg88CJz4q58MVGqtT6z2WwRkhR9nAYcAtNYBpVRV+PgsYO1J1zz5nENN9k88wz1Md84HWNOmwrixZz7O54Pq6tN7VzIy4M47m5899+GHtWza1MCPfpSAy3X2zF197j2Ii4Ef5XR3S9rGbofbblZER0N1bJBNlYp9NgtzMiEnFiq9cMQDfcIxzaRJVhoaOrcQ853XGsFV06WHWuP3a775po6aGqcEWEKIruV0wwDTUoSlWuvCll5VSi0AjmmtNyqlZpl00x7nnA+woqKM7UxWrIDPPoOf/cxIMZ1gscCwYc2f43YrkpLOPF6qpERTVKQpLOwdQdi154MtHH9orTl6VFPRoIiyKdauhuy+MHVKx8Z6RYLFor5PAa8O+kl3W7h+kIPMcGfg7kqobGgMsBwO1elB9i6nsbXrHJeFn/40/fueNCGE6DI+D+zrshThVOBSpdRFgAtjDNZ/AwlKKVu4hykbKA4fXwz0BYqUUjYgHmOw+4n9J5x8TnP7y1q5h+nO+QCrrUaNMip1t5TCqqk5/bUpU6KZMuXM105JwdQ0YqQNyG58vHt3iNtvr2N/iWbWJVEc+tpG/wGKEcMhPr772tiS+RYnFqsyOoqBsjpj7erCJr1xx8uMIDIx4fRrdFZ1tfExLu7019zuzvWcCSFEh3XRnx+t9S+BXwKEe7B+prW+QSn1GnAlxiy/W4A3w6e8FX6+Jvz6Cq21Vkq9BSxRSj2KMch9ILAeY6jwwPCMwWKMgfDXh89Z2cI9TCcBVhv16dNyEc2dO+HZZ+Haa2HMmPZf22ZTrQ6Q11rz8sslDB/uZuTICAxS6gSnU7F9e4DqGj+WMs0DP4ll2DCF1WqUXMjt190tPJWlSbdarNOoCda0t+2J5wIkxCruv9v8vzjvv2/0fDY3Zk8IIbqF0w353V5o9CHgFaXU/wW+Bp4J738GeDE8iL0cI2BCa71VKfU3YBsQAH6gtQ4CKKXuAz7ACBuf1VpvPcM9TCcBlglCGsqqjMHTwSCUV0NqB6q/t0QpRWamk6SknldEqV8/Cw89ZOPTT4M8/HAMaWmK6uoQyz6ysG4DPHg/pJo8K9BMDiukNxnuVF+vKS+qpd4FWse1WHj00GHYsRcumN6+e154Yc9JnwohBGCkCPd3faFRrfUqYFX48T4aZwGefIwXaPZfUq31v2DMRGy6/z3gvWb2N3uPSJAAywQ+P9hcRpX0dd/Cm5/CD6+G7HTz7jFrVisrMnezn/88lp//3OhZe/11L1995ef6G2JIT7PwXbFR+T2/lwyIB2NW5FVXuHC5IBQyeuOaY7GAvQO/QQkRSDsKIUSn9Y5hwL1GxL+cSimrUuprpdQ74ed5Sql1Sqk9SqlXlVJdXLO77ep84Gk483HDhsBP74PxYyEvE2aNg5R2vomWV0DxkY61s6c4fjxISYmPsWOtDCpQjBoBry+Dd1aaf69QSJt/0TClFNOnOykvd/L44xBqYVmhrAyYNTlizRBCiK6jMJJpZmwC6JoerLZWau1xlu0BXwCuHdn6cTYbDBwAb38Kuw7CD69p/5Io5ZVQWQVZvXix5LKyIMeOBbniimjsdsWqVUFy4uCK+eb/xm3bpsnIgJSUyOXacnON721nlwASQogez+GG3G4fg3VWiWiA1c5KrT3O7DxjfFVbJcQaY686MiFwQF77z+lp8vMdpKTY2LbNQp8+8N57tVRUaKzEMWeOYuBA8+7Vv7/C1Y4K6V4vVFVBejvStlUBGN6BSQtCCNHr+DxwUBZ7NlOke7DaU6m1x0lsQ32sk00fY2znKq0hELDgCy+bee21TsrLNbt3G8vVmCkqSqHbEfzu2QM7dsCVV7b9nECo5fSgEEKcdaS33lQRC7A6W6lVKXUXcBdATk4vGiF9jio6DPFx8JOfaPx+AMXEiUalzfnzI3PPZ581KulffPGZjx02DAoKznzcCUcqoMwHmX4jyJI0oRDirOZwQ46kCM0UyR6s9lZqPUV43aKnAQoLCyM3oll0mscDTz0Ho4bBob3VHDgQ4je/iSc62tKmKvkdlZ8PiW0sh6EUbarOfvQ4bNwMO4thcyn8cTs8eAfMnNS5tnaUz6cpL9dkZEiEJ4SIIL8HiiVFaKaI/dXWWv9Sa52ttc7FKAq2Qmt9A3CiiipEuIqqaL81a7ysWFHfrll6MTFwxSUQG1XPk0/W8NFHPl55VfOv/wpFRZFr6+zZMHq0udfcuhP+/j5s/xqmZUCUHzxVpx9XUQmlZebeuzkffeTn0UfrOXJEcpVCiAiSWYSm6446WF1WRbU7rV1rpJUmdEk5M3P4fJq3367H5wsxerSDpKS2/aYoBSOHwpoqOO88BzNnRlPjseD3Q0Mbylw09cXXsGs/XHtR+9fz66xpE6AgH9CQkgzTxkO/vqcf98laY+D8tZdFtj3Dh1vRGpKSpDKpECKC7G7IlhShmbokwGpLpdazzY4dxhT/3hRgORyKm2+OwevVbQ6uAOrq4Pd/gOHDXDz6qJPYWAter7HmXmtLALXkwGHYexDqvacGWDX18N/vw7yRUDig/dc9QWv480tG8dM5Tf6eOByQndn4fHAL47bmzoC29vFpbQSa7Zn1eEJOjpWcHPmXUAgRYX4PHJYUoZlkYEcnBULw8i7YU3nq/ptvhuuvb/6c9vJ6NQ0NXTMMbfBgB6NHt6/byGaDjHRISVbExho/Uk4nlFZBbV3r564th9/vhaL6xn1XXAA/vx0S4uDD7+DdPUa5jLc/hueegz+8ANU1sGkrPPa08bg9lIKMVEjqxHJGXg31bczaffQRPPwwHDhgLKW0dTd4O9CzJ4QQEWUxaROALJVjiqA+vV6WmbPOnn22FpsN7rrLfeaDu8HBgzBjKgwI9yppDTUe+PsHcOn5MHJI47FV1XDoCAwdaHyNVh2HtSUQ64Pbw8c5HMZW74eVB401Hidlwqbv4EgtbNgOdQ3G7L5gsGNtvvTCzn3OGw4aRWgXjDj9Na2Nbf9+yMmBhoYQmzf72bIFotxO/vEhXDUfBuW3fP11X0JyMgxo5RghhDCNww2ZkiI0kwRYnWSzwI2DInuP3Fwr33zjp6Ym9H0PUU+hNbz4ovH4t7+FQ4c0L74IV12l+MFNEB976vHrN8GyT+Bnd0FyChzbBatXwcfb4S8D4f/cCBePM4KvKDssHgHVdfBfT8NfXtbUHw2xq1yz76iNKSNgbDMBTlc4r6DlFOGzb4ElCAe3wFVXQ60Nhg6vw+FwkpYM994AifGtX/94aXhBaAmwhBBdwe+BEkkRmkkCrF4gMdFKZWWAigqN2w27d0O/fkYarrspBddcA88/D488EuTjjwO4oq2MmWgjOrYxkCgtDbF0qZeBBTbuvM5Beir8/RAsL4LyWMAaYu0XmusOW3jhQcXl443zBibBW9/A028ZM/cIKgIYC2ubTWs4UgLpabS4wPMJtlZeH5ANCW44v9CoHF+vLVz8qwRioo3Xk9qwTuWCCNUOE0KIFslwT1NJgNULTJxoZ/BgG4mJxuDxTz+FuXON9FNHBAIhvN4Abrc562zn5xsB386dmtJSTXxSiBXroCEE2X1g61Z4/HFNZWWQQABmzjDuu7MSbHbAHoIkoCiA97CFj7bavw+wAAbmw6CRUJWoiLUqHAXGQstNLV0G7ujTB663h9Ua7jlqh60l8OVBuH4sOGwwc9ypr48bCsYcaCGE6KHsbsiQFKGZJMDqBSwWRWKi8QbtcsGttxoDyzuqqKia0tJaCgvbv0pRSYkmIQFcrsaAweWCe++FI0es1Pph+nQLl18CznD85vVCbKyFOXNczJ7d+C9S/yjwAokBqDgSgirFkDGKscNOveeQHHjzYbAqY+B8Yjwkx5zetpoa2rV8TlNKGb1X7eX1Q1V9+9atFEKIHiXggWOSIjSTBFhdaF8R7DoAcyZ2LkDqzLkAubkJ5OScYRBQM3w+zXPPhZgxQzF16uk9MunpissX2ejXF6Jcje0cNw4GDVK43fZTjr+8P+TGwJFiCysGKy7obyUtW/FlDRyphz4nVYFPiYX3voRPtsCDVzbfy3TTVS23PRDQ4d4p83uSxvU1tuYEg0YP3uDBrVeSX7MmxOrVQR580N7yQUIIEUk9a4hvrycBVhfatg/WfQtTRkFcN08ItFjaH2g4HIobb7SQnNzSNeGiC4wK56+9AwvnQnR43JG7mc/XYYVJWUAWTBmhWP8ZPPUR2GfC5JRTAyyAKUMgNx2SYk+/1pm8+26Afv0Uo0d37Y+8xwPr1hlrJrZWEywpCbKzJY0ohOgmdjekS4rQTBJgdaELJ8O0Md0fXHVG375nDgKSEmH6hMbgqiWhkLHZbJAeBVYLDHLDggEwsJlzE9zGBrBvP2SknXoPbwP879swahCMbzK7cNIkK7GxXR/AxMfDHXeceVzXoEEWBg2Sfx+FEN0k4IFSSRGaSf6idyG7HRI60PvS21gsp1ZDb8k7K+D3f22sZXXR+fDLH0CmDf7tz0ZBzuZUV8MLr8A3207d7w/A4WNQUX36OenpFqKju6eHKAJZSSGEMJ8UGjWV9GAJ0wSDsG6DJidbkd2G8fO52WC3nVqUVSmIdsGIAkhuoZyB06m57XrISD81comNgV/e2foYte07jR62jgxmj6SqGnj6Fbh4llGEVQghupTNDamSIjSTBFii02ob4A8rNcte8bN7RYBf/9rJXXecuaDKyMHG1lR0FFwxt/H5tm1BNm7UDBkSYseOAM88W0dMnJOLLnczd6ZCKcjNMkos2M8wRvyjVTBkUM8LsKwWoz5WVAfWKxRCiE4LeqBcUoRmkgCrh9Ea3lhu1HOaN6172uDzwTvvGAtVZ2ef+fiQhp17NFu2aSzeIGNHBzGzYl1DAxw8GGLJEi9ff13L0aN+oIbvijxs/y6dkkoLowacefkZjweuWcQpg/SDQaPXzMyljTrCHQO3tzILUgghIk7Se6aSAMsE5TVwzA8by+GSHIhrMh2/ugFcNmPWXEsqamDzHhiWa5RzaLrETFdqaICdO6F//7YFWLEu+PWlislOCzU1muHDjR+rvXuhvh6GD2/bfT0eeOUNuGguZPZp3D9mjBWnE44etaG1jY8/biAQ8LBrs4e7b09kYH8Xh4rg6PHWA6yVK42aXFeFA5lgEP77caNg65VXtK2NQghxVrK5IVlShGaSAKuT6hvgiXegzg6vboTS2XDLbKgIQF4UlHjgya+hTwzcM7bl63yyCd75AsYNhDuvMHqwuktsLDz0UPt6dXJzFHfeaQcac3SbvzEGpDcNsCoq4I03YOFCSE1t3O9wQE62kSJsauhQK488EoXTGcUvfqF54okAGX3sXDDLypAhEAicPvaqohIS4hsHmQ8ZDolxja9bLJCba5RQEEKIc1rQA5WSIjSTdAh20v56COWCrRpK1sDbH8MWDywvN163KOMN3nqGmWRjBkJBFny4Em66B373P/D5hs61ra7OCHA6woyU2WUL4cYbTt+vlBFMNZ1d53DA3PMhLg48Hs3vflfNO+/Uf/96dLTCalX06WMjNVURn+jGH7A3O3Ow+DD8x//At9uN5weK4Y8vw7cnzUxUCi5fBFMmd/5zFUKIXk+ZtAlAerA6rcYPygU/uBLy0mHWWOgXD+PDPSVpMfDQxNbTgwD9MuDuy+CVF+HbNVBeCuNGd65toRD4I7AocltZLM0HagkJcPPNLZ/3zvvw1vua7Zv9lJaGOO+8KKKjobISXn0VZs5MYPbsGLbusOOKhidfNmYjPnQHHDsaJClJkZxk4YJZjeUiYqJg2jgYmNf8PbU2xp6dvID28eNGL1hBM7P6jh03Uox9Itz7dajYmPk497zI3kcIcY6zuSFJUoRmkgCrkyakQmGK0VO1+GLYUg7/uhnuHQKucFAV3cbVT97dCn1ngM8L0ybD3Omda5vb3XwF9Z7OaofioxZSMxz8famH41VR3LHYSW62Ma4rO9vCxRc7GTMG6uohJdFIqdptUFISwmazkJkJs8Nfv8oqeOJ5yM+BzPTm77niU/jkC7jvDkgLpy2Plxo9Yc0FWMfLjLSkWQGW1ppvqyEvBlCKHUeNsXxvrAFbCOae8QpCCNEJQQ9USYrQTBJgmcCiYB9rcRBFtG0UGdFgs2jqCRHVjtl0sS4YMRR+czXkpzWWHPAG4EANFCQ0ptVqfVDphay4lq/XU6xf76e8XDNvXiuL8WGMzaqohDkzjVpVD///Xo4fC7H6cy9Wl5N/+w08+KDRA7Z/f9AoHhqluD/cG1ZSDtVWO4lNlvKx241ZenGtTByIjTUWkT65zMPQIcbWnGHNlJfoKK3hiQPwRl2AFKuP6OPRRB1RzBoMdRqyksy7lxBCtEgGDZlKAiyTfEUdMYSYHwf942A39XxIGdeSTjKtBxYnLBgJc4eCs0mP1/4aeG0P3DMc4p2w8ijsOgz7y+GBSZAZgRmHwSAcPAj9+nV+PJbfb/T2nFBeA3HRYGsSe278GjZ9Aw/8CCaPhxefT2T5Rz5WrXFBAOLjjBSe36956y0f551n/37GIsDGXfD+OsjNgAFZUOkx7hMTDT+9q+X2hTSMHQMTmpmE8J3HGD+XE9O5r0Frth2G1avhmokW3lnn4JOdin4K/u88+GY31IUid28hhADA6oYESRGaSQIsE2g0FvKwnjSDLgU7Y4jF3Y4vsVKnB1cAFg2VtfDyDri8Pxyohb7x4LYaAVckHDgAL74It9xizLTrjKlTGz+pUAj+sQYmDYZhOaceN2MajB/XGNAlJ9soOeLHbfNyzQ0unE7jhZIqxcKrXfQ9qVhoKAQzR0H/TMjLgD2H4ekPYOFEmDoUfCGoCUByM7Huq2VwyAsznZAfCylO43uhNTz/HURb4Z4C+K4WhsSCvZWAMxSC4grISmx7YJoRD7eMVMzsZ2Wsy8pT8TA1EVJi4e6ZENXGFLMQQnRYyAM1kiI0kwRYJlAoFtIPddL0iUTsTKGFtV5aUVIL1T4oSGzcdzw8kW5dMew7DP80AQalNn++Wfr1M4KrnJwzH9seFgtcPgVim5Ri8PmMoCY+/tT9ZWUBduzw09BgAVx4ffCn5dAnUfGDecYx63bBB5vh3gthcLi9sVGQmQTJ4d69z8rh0wpNVmKAA9XwYLaNOIfx/aoKwafH4OkdRlmHuybAXalGe27NM3qwPi2FpSXwo3wY0Upads9ReGYl3D4bCvq0fNzJkt1w0Sjj8cA0yEmEUQOM530SIAisrYQ+1iArvwuxqMBOvFR8F0KYTVKEppIAyyRWk34yNxyDvVWnBljj0iGoYU8pfH7szCUfwEjJvbkKxg6BvDasC9iU1dr5nqsTfD7YtR9iHJDTFxKbDLyvq4MnnjDu+YMfnDqTb/p0G15vkMOlitp6YzbgogmQcFLKzm6DKIcxFu6E9ES4f2Hj8z42mJikeTQYYL/PStlGzR15ineXw+OfQ8MgUCpEfbKfQ1g4WGvHboHccFtPpDPzzlCfLCcZrp1ifOyIRBc8NAmibHC0Hv53LxwJwDYN/Y5Ws2NXkAUDkpC/hEIIU1ndECcpQjNJgNXDXJgD/iZjblw2qArC0l3QzxUiN/nMxUZC2qgNVd8Quba21f5iePQv8MHf4OpF8Nhjp74eDBoV1i0WI8V2sjlzXPTNc/LShxby9sH4YTCuPxw5DnsPQf++MDbf2Fqz8iDkxFiYcNyJwwY2v4UPyuGzg1BRBKEKSMwE25gg+2oVb5RBvB0W94dACDKccHEbZgy6HDAmt11fnu/t5jA+Agyz51BHKfU4+DZgpyA6xP0JMcSkOjiYHSAlWoIrIYTJQh6olRShmSTA6mKhECxfBxnJMKrg9NftVmNr6tXtsLU8ROLIcj5Tilm03kVSXgbV+yF5qjnt7oz+OfDjW6B0J+TlGcvnRJ2UIoyNhR/+0AiwTuwPhYzn3+xU/OXviiLg22Xwn7GwdGmI4ipFVLzi13eDy2kUZd1fDDcsPP3+dQEYlgppFnjjSwsuBxQMgspYmHcVVPaHfSuhX44Ff5WLhoOK80bBpho45oWnN0FhBszLjezX6RhV1GofofIc4uOq2aOj0bZyxseUMydtGhDD1DYsXSSEEB0i/7uZSgKsLub1weebIT+r+QCrOX4/fPSmsdZeYJebaQMDTE6Af6yDGUONsUZNHT8On66G2FS45rLWSxSY5YutEAjCjJHG8/Jyza5dIYYOtTBqmOLqq2DTJqNY6K23nnpuwknD1Q4fhqefhksugVo7vLERSg6GoEjz5u98UF5PVoGbP/zegstp/AgnxkNxKbz5BaTFw8QhjYPMj3thQzncPhD+61JYXQKpbhicAL/6FlSiZsoiqLVCZcjCnD5wLADf1ENhA+wugmwHkBvZr98kBtEQ0jxTDFGl+eT1gTGueGLJ55taGBLV+gB7IYToMIsb3JIiNJMEWF0s2gX3X2f0ujRHa9h+zHgjHRgeyL6zAqrigWoL+rCLgeHjQho2fgsNeZDX99TrjBgB//EY3PvvUO6Bh+6J5GdlCOlTU3xvvhnkq69CTJ+uufpq22npv5Zo3bi9HoSKicABIF5BlQ9bgp9jB2t5/N8dXHKBFatVMWQAvPs11B6DOi84nHAkFiYnQj833DcYDtRB3yhYmG+0tSZgzNAs82r2KM34PJhnseL1wtEj8MAASLXBlEwY2kKH4ab9UOeDKW0MlltjxUKx1cfMfpq/bbQxIQ1+PCCKgIZtdaA7fwshhGheyAN1kiI0kwRY3SApvuXX9pbB81+C1QI/ng7psUZJhscWwvtvG3Wyju0D10S4frqx9mFF1ekBVn0AtgbgigshIQjvrIRQfRB/A1x+mQXVdCHAdvr8c2OmYfZJKatpTRZ1HjhQsXevIi/PuNdVV8HEidC3SVsB3lxhFPm8aDpkZcGvf20Men9rHSTWGItm44KYggYCX9XQUKHxeFLw+6OxWo1grM5r1L3yNoDfBwUxEBf+Ca/0w1+/gyuyYWwiPHUUHD4odMFXVoU/JsScVE1SvZVvisBjgTS7ka69e0rLX4ejVVBr4ji3KkIkxcP/mQbR4ZISOgjeo6D7Il34QojIUNCOutiiDSTA6mHcDuON1W5trH8U74QB/WHGYliQBEcrGo+/5Pzmr/PHT+GtLTA3Gz7/CGJTNF99FgQNly6wnFKxvCPS042xU62ZNs3KlCkWLOHpfVFRMGhQ88fm9DFmA55gDf+i3zUQyuJg61RFuVczIc/C0jUQsvuIjm4gFNKAwmKBa2bBe+tgZH8YPcAYcH5CmhPuzIdEpyYIFLgUq4th/T7ITQmRnFVHTbKPLRtTCAWNXqsSL/Q9Q4HRC0e1/np7FRKuv+CA7buhbyYEFHxTDPmpzddJE0KITrO4IVpShGaSAKuHyYiDn840Sg5EnxQgRFsh3gaZCdCvhaVT/CGjnIO3Bv77f4w0140Pw66p8OCF8O10K0vXKb49oNh/DEblgrcuRE2Vn8LRDqxtqf8QNmBA246zWFq+5o4dITIyFAkJijEtLEmTnQS/uxdsVkWlR5GRnMKU/i6WLPFx220O9pdYKK+EaWNhxzYYnqrJTtMQOnWmpVKQ54bHKmGIHeYlQE4BjM/084jzEPHBGiqtWaSmBqmosvJ5KXxTDb8aCc5wsBcIgK0LfmNqvPDRNtizAS6eBUML4I5OrksphBCtCnnAKylCM0mA1cN8uRuKy+HS8afun5JsbK15uQRKGuD+bJg9BDaFYPpQSEyEBJdi7mwrSVnGoPitB2DZKnj/7RAlOsh/PRBkxgRzfhy8Xs2yZQ2kplqYOrUxStRao7URdNXWapYsCTJrloXzzmu9XzouXIsqOsoY4xWdGcP9/18MC2YpjpXClq9hb7ymvALKgrD64yDXX29lyJDTg7sLo8Bi9bM6WM1n5XG4EuvxOmrx1dixlUeT6PD/P/bOO0yO6srb763qnCfnHDRKM8pZsoQkBAIhgjE4EgzY6xzXn9f2t8n2rvdzWBt7MaxxwgTbYDBBxjYgMiiAEsqjNDmHns7dVff7o1poJM2MBhghAfU+Tz+arq5bdavV4dfnnvM7zMxRKdbBqYAtvSR38Bj85B6onwQ3XzkhT9OoJHVJjyXF9ddYyPEL46KPNUJp1YnQnomJiclEIjBTECYYU2CdZxzrgcMdRhL2mV7rHd2QnXEiqjLFDYV2sFnh374CwSEjInb7L+CR++BfvwSLphr7fmQl/OMTYJuh4tpm5cVnBcvmnTj25s0wOAirV7/xazhyROPppxM4HIKZM624XIbQefBBneZm+OxnFdxuwac+ZTmpevBMSGlUKR5qERTnGdsUoPmoZNc2nUsvFcyfL2hvt5CXZ8xfCPANc16fYhc8Qz8v0E11gc5O6WC1yGSTRaE1qFLsOEznYBmXFbuoHGYqmp0By+dCeeEbfz7eKB6XZEW9Rubxt2cqBf09UFACqnPswSYmJiZvBsUDTnOJcCIxBdZ5xpXzQdNPb4Q8nI5uI6jx03vgilUwd7qxfU46eV7T4ZfbIM8Nw7vq2gAAIABJREFUnyyD982H794KBxuhdJhAWFMODx6APY1x/vmhFJtftPKnPxlJRxkZoL7JV0dZmcrChVZyc9XXxRVAbq5ACMnx/Pr8/PEvSR5oh3tegBuXw5c/euL5cbhh9gWCAq9CWREIIShMX+Nv7wWhwGfSjZ51KXlBxnHgYaki6O734xiysE2JIwv6uNLmZW6sjM0WF65Tnv8MH1y16uRt8aSRK/dWm2Gfig2Faob1wrHZYJaRaa/r0DsEtz6WZF9PnNtvdpHhNH92mpiYvEX0ECTMJcKJxBRY54CmGOwYgkuyT27vAsaX9Vhf2Edb4ad3w6Xvg/evgeoRegWqCqyfDO1d8MRWeN9iaO6GaVMgGDUiwV4nWPugxCKYWaewb5uFRx5PsGOHnYYGCzU1b/76XC7BNdecHmlZsmR8QuChhzSqqwXTpp3YX6SvC062uGjqhAefh4+vFSeZlwJcvPrEcxmXkue1BH8XMSYJC0t0H3d1Cv56TKFbZjG/18NNc6zkuS1cOGwpNpKAjqCR96YOm76UcNsGmFwCa2aN67LeMs29klvuaCRi6ebZ5xZgKXdwU3uKCytH6GBtYmJi8kYwlwgnHFNgnSUSCbj/71BbBnNOsS8IpqAzYSSkj5EDPiLZAVg0E2rKoShv9P1mFEKwC1r6we+Fj10JXjc88qpRrbeoXCcykOSCBTZWLnLRH4sw2KPS3CJpGFYZd7DNyNlyT2BzYU2DpjYoKRg5abyhQSHzlET+mgL41rDcp/4gbNwMS2fBl66G4lOaXx9uh0d3wNbd0FAK6z6s8bCWYJ5iZ4VF4W5xlNZ8D8FgJvE9dlpsDvr7IO+UPLeHdsKmY/DROTAl3xCo+QFj6fF90yD3jffzPonWTiPHzDtKtWK4uxvVasURCPD12yV/e6iCvHIrLqlQUSF5Om5hicZpETcTExOTN4TwgN1cIpxITIF1lojEYE+j8YPgVIE1zWPc3gweN1w5zryoZTOMf4Mh+P3f4NKlsHKqIQ5+97Cgr0/laxcKurvhP77pwuaCwCkeXU3d4LZPrMA63AS33ws3vB+mjmDQedw3S0rYdxAqSsFxyvn7g/DqHqivPTmKF9cMk9Ynt8Mr++HBZyVbS3Uu/4Dg43YnmQmFkOjHJruYFfYTtQhecUisPo0ei8rwysPuIcPxPeCAbA9s3AtbDsPX1xl2CTOrznytff2Gi/5o1YddvcZ1jiawLA4HisVCTxC6kwq2DPBll/Jf10FRuUq3Bra3ZmlmYmJiAjKETJpLhBOJKbDOEgEffOUGw7n9XOPzwNolRnRGUYzkd0UIJlUZ//3798OLL8IXv8hp/lgr34DPk6afvIw2GmVF8NEroKrs5O1RglhxYMFY8orHjUpHZzZ8bO3Jz2VlMfzTzUZl4XE643DbMZjigfmT4Fgn3HStTuzCCNutkpx+O7/abqHQ6ydQlototfH1mTHurGujQBZS7D7xdhiMwo+fgQXl8IXlhsj02aF82LLu5kPwylG4YenJnlvD6ekz8qayR6kAnTll7OfKnjYby7TDwlKI9ihcdzXU+mGKDzxmzruJiclEIECakfAJxRRYZ5GA7+T7PSmj9N99Fta5D/cbJqW5o0RCinJP/H3Xg3CkBb5ys3F/0SKYPfuEuDrQbRxr0y7IyzhReTgWwSj877Nw+Uyoyh15n10HoX8Ils2ChhF8r9rYS4ACsjBCUg4HfPAqeHKnUVUJkCCBjo4Dx0niCowl15SEhIT6SuO2pUvh4ajO5sQA+0M6XYl8lidUUlntlGUFybVUs1yXFA0pJGIC0pFFrx3W1EGWG36wESqz4bLp8PJBeGE/3LISEikIxU7MbSRqxxHlGg+KAh9dBasaIDcXbn/UuD9jnH5kJiYmJmMiPGA1lwgnElNgvU10peAnvZBtgS+cwc9qXMeLwx87YXHA8NS8cQsEBuCR9YY4GIspNWC3n/CXUhTDZf14hdqvtkCxH6LtkEgC4xBYLhvMr4ScMdzdC7LBP8bSaBkzUdMvyWPH4P4H4Ibr4cMrTuxzjKNEiVLP6aG1Qgd8rQriOtzRAUci0N0p2Ktb2decjdc6yGTPEK45fTRbbEzBTwMZlGgBvtsheFWFf66GTpI8oYRYVu0mL2VjRhHs7IQHd8PsciNSB7BkknE727TGDJPZqhLjpuvw2cuhKPvsn9vExOS9gSSErptLhBOJKbDeJhwCfCrkTdAz3pWAPWHIsRlV/HY36BHoi5xZYK1YaNyklDQ2Qmkp2GyC7QfhhZ3w0aUQcELugvEn4VtUWHKGysPsjDMcAxud3bB1J0ytgcrK03OvyihHQxv1GF4LPN4DhyLQ3wtPtOk4UxYivihWp8aUulaKI33M1AtZ5qoAIMMiuCbPGAvQSYqX9Ag7SfENNYvLGxS0Pg27ArMyVbb1wP2H4ZJSo2BhdwcUeqA2E+56HF7eA1/8EGQGINs16lTHjUMBq4DmPqNxd0MplOe/9eOamJiYDEc3qwgnFFNgvU34VPhqFrzFHsuvM8UDXy6D/LSYKpkGfh1KArCzz+ihl3EGoRWNwpNP6qxZo2BzGJVo65ac2y/vvQfh8aegugzWrz9d4Nk4PdkpFAGPy0gWbx2E+w6FODQEezqdJCPg2TOArHIx+ZJDBLJfRO1oQoRrcbuM0JwQMDMgeYwIW9BZh4trCfAX+nmaY2jNxdy2U2WgX/KbEAQKQLOCxwJ3N4EMwbIcyLbC3U9AOAh/2gsRBb6+5ERPyTdLVvqSX2qEA50wrdjIddM009jdxMRkghAehGouEU4kpsB6GzlVXCUYQMGGhbHDHANp7yr/sJwjRXCS0/jsdPVfVxR+1wirCo1Gz2Phcgmuv14hEhX89+3g9cBXPn3icSknThDqunGs4cfbmUrRrkvW2AwFEopBcTl85uOQXwD/sxemBmDFGO7ph5rhP34JkRC4/PC3/ZKOhEruRb0kW10Q0Qm1ZOCb1cOmIzNYI/5GSfwAHdUrkJxInhpAZycJejVJqZ6i1qpQQy8tuk6nmkkk6edYkyFwqgZhQS4Ew9DXC1+cCRVuuG0LLF0NN8+DtgT0RMA+xjusb8j4N/MMTbOPs24GxFOGuEom4dZ7Yf50WDxzfONNTExMRkWG0OTz53oW7ypMgXUOaedRnBSRy4ox9/vjDmMJ7oa5o+8jkSRJkeOw8ok6yHNCAo2nacaGyjKKUThdLdntguZ2mFwHFyw6sb25PUY8maK6dGw/ib5B6B2EmhEMT48zMADf+D4snAsfWX9ie56iYEuLHCnhkp/Dsx3AEVg/Hy5YBuVpkaiho0sd67EnIacB3EaYLZmCh56A3j4MTwybgD47rTvyYIkCKmCTBLdnoaQSfO/o5/miU5BbkmK/a4jXCOME1pJPdsLCo0M6G4OStQVdzHX4KdCtNKXcXNMAXQL+rkJZKXxuOQwoUJABnSkolxCMgcMKHg/MOkP0EGAwApLxCyynzbiBYfvQMAlKRog2PrYNSrNhesn4jmtiYmIihUBXzJD4RGIKrHNIHqtQOXOd/bopZ27H0koPB2jifWIGhb4BNOKEyaSZIVQUpsgULix40iEkHcnj9FOBg8nlLqrLTz6eNfdJNNkNXD/meQ82wZ7DYwusrh5oaQaZdjxvbTX6BE6ZopCnQD8aLxJhc4ETsixQBC9uh6qpKVK+MBIfW+gmTJKVNj+ohnrZ1wm/2Aq9ig42HQZS4AWsDogCgxqUSLDoOKuGKChupvVIBbfJW7iiuZHwpFb2EsKGxI2DfQp0qJKk1UJXzEo46eESt4+5RTAYhNuiUJ0BuQ7oUSHPYfhkuS1QlQUfnwcuK3jGIa4AKsYwij0TQsDK+SM/5neOfw4mJiYmAAIPqrJkgo726wk6zjsbU2CdQxyML9kp33fmfbLxY6USgc4R/oJGnGrWkzVQTldK8DNXCr/Q+JLzRNZ4CkZNFw+os9CInPG886fDnDN4OdVWw29+csLE9Omn4dAh+MY3jByiYyR4UkRYOE1h42sWUKGoFmyZIZ6jkyLcTCJASuhQaIS0nu+Gjz4ILft0qAMeT0J3FLpU+IjdiFzVaeAGQkmK5h2lLKuFcFYWyxQL9VVFSGAxGVRhQ8VBhhpjsRPqhZU6LZ8/9mlYrENc5XCzyxeGSpjU5MVVBI91wNfq4Fv1xpsoGIO6Uewp3m6W1J3rGZiYmLzTkIRI8sK5nsa7ClNgvUtwYMOBDYmOhyISDJHS3XzlsJ0e4ILyINO8OqSbCA8SZik2Moblf+lS0qilKFdUHEoBAFGOEZeDeJmOOkpC1ngSrYc7xK9fD7HYiXF12Pk4AfInW9lfrLN5IE6+VZCleNh3yAVlFjKGWaHvbZNc9y9w1AUEJVgEFCvQpRjhG5GCLHCWhIj2qPCCm4YGnY/NrKCsPIMG4aGNCP3EqcSLEwsakkJhw2MX5NgthDTJZmc/B+1JOrDRZ02wcC6smQmb22F7B3hVcFjgT9tg6zH4/AWQNw4xbGJiYnK+IRHomEuEE4kpsN5lCBR8TCdODKdwcnGWztYYRAacqGiQCYNEeIrdAKxiOv60yPrvcJwfxYZY6dD4lScPgWBINvKHcCsvD07iB1l28kZwpm8JQ4FzdBd3DR0dsKY7iXo8EHVrPKulWKDYcAiFqdiJKzob/B0c9oaJxi2EDucTGHJyZQG40q/Unl647sNweJMEJBQkYJUdpuowXUC7Tm64g8CqVrxBnVd2L4SDFpYF55CzOU75FBv4oRAXhcPEpYpAkuAPdLKKXCapHj7s9NCHRj5WLicTVBAqtA1A5yDsbIHKLCPCWBR469WCJiYmJucSzez2PCpCiCVAjZTyV0KIHMAjpTwy1hhTYJ3nhFOwuR/qfZA1Rl7NwV7D/8ohoVFtIu7sIVcUcWuZwuNDGv/aKXk1qrNdb6NPOcIgA2SQi0gnvvfTDZYjTHVGyacKpERHI0wtL4Ym0R2FI1FJjoOTkuVTOmzvA2cOZI3SFmgz3QyS4CJK0CXs6IRWT4rNjji1WMhP/2raRJjnCTIoY6SscQbdklocdLGRArkYRZQjAN0D5AFJoCiOO2OQ8EMWPO+PEwpJ3NOCNP+1hFQqC7os5ORDlsfw+horl82CQEXBmr6+qSNUd3Z1wZIcuLAG7n/JsIi4YCosmiDH9nNJPAH2UVr+mJiYvLsReLCx6Mw7jos7Jug45wdCiH8G5gCTgF8BVuB3wOKxxpkC6zznUBjuaQKtGFaNkhSt6XD3LmO5qn8X9Mt6Zi7XWDzFEArTPJJkt2TbgM6/uY5Q5m9ngbCzWEaxib1IMYMDbKfBEWGNPovJShEi9ggx2cxGe4AFmSlqHGuIe7vZiZ0ZnOjQvLMbhkKQNUbFWg1+4mjEk/DTZ+GZZrhpiY1PVVnIFydC0hKBB8GAACElehh0by8h9RCatKOIcjIyoPq6QVpTQYJH3cQXg9wg4GCK0A8DYLdypKQKghKhuKkpgKuWgmoRzJ49skJt5ghxYlRQyy2UjVhteZwHHgKnHa6/Dq5fBta3+R10dABsKhSOs/JwvHT2wv/cBx9aC5MqJvbYJiYm5z86YeK8fK6ncb5yBTATeBVAStkmhDjjp7ApsM5z6rzwmSqoGsMtYeMhoxnwawk4ZIFIl4X2v1vIskJ1iYbfqvDJgMIfbJ30OTTsuooUR7CmHqNDyeTvqW+SaetnpqigUuQCAiy1xKSVeOI59ul1+KwuLlCzsXLyOliR5+SlwZc7oNANpcNeetnpvK+BlE6/HKIuw8vyYgXfMHHVo8HzAy6m2Apo8PZQlspElroodgxRr05BFUsBo5ry6Mp2nFJHBjUiLblE7rGBLwF+GyQVHMesLJwqcNvBWwezCmHNKO1+4sQ4wG50dLLIJQOjj9HhNvjLZrh+DbjThZ7BEASFYTcBJxo8h6Jw+xNwUQNMLYVYArqHoGQCWiKdSnvYWC6daIHldcHcqWd22zcxMXn3Yi4RjkpCSimFEBJACDFK19+TMQXWeY5NgYbA6I+398FzR2Bfh+FwcOV0eH4ntPQk2TjUyhMhsLvgA/44JZ4OjiUDbGrJ55nGGWR5FnFR9p/Zqh4iFqjkj5YNoLSD+zKw1OFMObmw/3ZmuKqo8XgJET7JnBMgz23cwFgufLoNpmeeLLCOE3X2MnXZDuboDfgcOa9vlxL+fTNs6FZYXufjDo8PYQPDtD0bTuk7WKV46dLDhOM+aLVCj4A8m9HoUQGyBa0+uPcayPRBmet0w9QthHmFGB/CTwU1xInj5UQmvtMOBZknR6h8HvjwOuiMwV0vwQfnGf5kFhWKM8GbFmLP7IN7XoZvrIPagtH/794MC4sm9njHcTnh0uVn59gmJibnPwIPdhZO0NH+Z4KOc97wByHE7UBACHEzcCPwv2caZAqsdzjP74GFuaCHwO6EoWz4P1dC1HmIXdZX2a+7aNUyaRG7qBEp2sKVuEOFtNgm0f5aPv3zXBTEOijMjdJuW8yTQuVygnjw4lBLqPL/E1W2SjRVZxOHcGKnkJwR52JR4HPTjSWsU0kSZYCtTLH5kejspovJ5KAgODgAv3kWIjHoVEHUnj5+OJd2FjEJnQ1PSnbeJol6gAIF4mDL17BmShKBBM+Ew3ykIINf0stCvNiQFODAjQUdw5xVIKjk9I7NBVlw5QhdI4rzoacJdAlHgxBwQLYTrlpgCC2AIj+8sgP+kgu1l58Ym0gYQs9qJsObmJicZ+iEiLLpXE/jvERK+X0hxGogiJGH9X+llH8/0zhTYL3DWT/f+GJfPMlo/vzD/XBkN3xyJXitTfjDbp4PlvGItYJAME5ndwHWeIpkxA5Spf/pTBLrbFzifIk25QU2sYB+BHVM4wIhUFxTsOABUuSQwILgMHGKsWHTExBrA2c5kg601O+xK8tRlRmvz+8IHTTTTT2FBNmDjww2Y6OTAbKYQz4BOiSUlENbEzSc2XeVYEQw16tyw/sl/y8p0WyC/AzB5qNJdmV2ExYK0aiLSYWDRPEzgMYRImyih4vIYRHZzMfNfMYV5T2NGaXg98Ptr4HXCh+qgB8/CtcshrlVUJQBJU4Y6jgxRkq49TdGpOgfPjL6sXt6oKsbpkx+U1MzMTExeZMIpLlEOCppQXVGUTUcU2C9gwmRoNHaxySy8DqtzHPAV0php4iTssbIEHGmObo5ppayX3HQrVpoi1ejCxXRryCTcSxlLmoH2lmZ9wwZrgGuoZEUk8jGSit/xEE+BVxCiiQJwnQxxJ/pZxnbmBcK421/CUo/j7QPIPVGhCgHTgisbgbopJ9uWcyjQ/MIDA5SGM+iaSDJgYoU+VmwKAAvXgk9Q5A1dltGAD7x+uEFP/qsIKVBTAMpLKz4s5OOdgduaaMyUUmpS+UT5NNOjGdRCUlAGE72SVLYGTucJCUcbIMsL2T5IJqAB3aAzwX9OvgssOkgHOiCu3bCrkG4ZS5c0QD9XSf6OQoBtRVgO0OVnssFftNLy8TE5G1GwY2DUdpDvMcRQgzB6/kxNowqwrCUcsxPa1NgvYPpJsw22snBRZDteEWAuSU1lJb0so3DWKnAaenjI8p2HtcVEm4LgcIMXK5JbPMk6G3uI5ATZK1/D3NTQ0SpxE+STtp5mVqmRy4iW7WQaY9hx0kB9fTQxlqc9KQO8rC1lnVZV+M7tB1RsRiL86sMiUwc6NjSv4RmUs0kSvh+LMbdaibB/eX0/rKA0mUenj3m5d71GnUWlS4lynOZQS4nizfystQ12NUND++HD8wQfLohg+9GYEo2eO2SnyUjTBYWFlmcNMTzeHLQSU0ApK2b/bSymnosY5jrNXXDzzZAdQF89lLoDsErzZDtgdwiaGoCNQz/eBls6TW8sI4cg0tXQU72yblf61ad+XpcLuNmYmJi8naiEybMlnM9jfMSKeXrWcVCCAGsBxacaZwZD3wHU4qf9dRRgJuEHiTZ/TcIHSKHfBawnFpmYcdDgTKNxZYqGqw2Lg30M935IkHLEF3RIg6+Ws3LWj1NzgpcrCOHTxCRCwilmnmwQ+Xmw4L7w5tIkaSFIdy4mYyLH0aX8rXecqpbarkDK1Ht56TYy2siQTPR1+doxYIfN9OEk0TYZiiOUoWmlgysmSqvRI2XYBxJCJ3UKUn0Y7F9O3z/+5Bvg2UVcHc7NAt46Rq490LwOyAsIYLEgUKV5gfNRkKHfPyUJst4qEVlIDH6OWwq5Hlgdtq6oCQDPrEYbloAS73Q0Wtc0pJa+PIi+EgD/OFB2Pj86blWPQOw4QVIJMd9iSYmJiZvC4aTuzIht3cz0uAhYM2Z9jUjWO9gVBQy082iG1Izoe85yOlCoYoAmahoZFFEBlORdJItqsl37COC4Hm1D3deL0mPjYbAbqLWeVgpR6GEJfp+5iVv5zueuezyr+Hr3bUssG1gqtZEwvFhnmIrpfZtZGW6ePjg5XwmtYTMyg1cyGHmRuagWMJIawpdWOjjYYSWgdo/n4u7svnFQQ9UpqBTpTipUJZ+BU7BRR3OMT2ohtPTB7m5MH8+FGQY9gIHWsCpQLYDVAEguFE6uXMf1JXCfB9Md4BLAbDjSdrZMwQLMiEwytLdzkYjSpWV9t8TAurSfmTTFPj8MlhSaCwFHm9E/5Fr4K5HYOd+qB+WPz8wBHuPwNIZYBsmvl7eY4iuZScXS5qYmJi8bSi4cTHvXE/jvEQIceWwuwqG6WjsTONMgfVuwZYJ1V8H1Y6OJKULft+Ww6LAxeR6YBV5xImykxYEkm8W9/B45kEy1GMkhqrZmJhLv82DhW1oSpxa61p6vIIqtZGQ5meHsoXV5OPBTT/9+EUULSWIJ+zQqnB0aBVH7SoV227F6e8hUlsDlmt5PtzCcz09/OTPK9CbJd5sBWtNitlJKH4askqF0awZxi2ugkPw41/DmqWS5ctBUQQq8Kmyk/f77V545IhgejF4rHBgCO5rg5vKoNBh2Dd8o+64GDsZTYPBCMydDAEPVBSevk+hx7i198BtD8KKWbBiNuTlGPlWgVOsKqpL4MvDEtyP52ftPWJUUJ6PAiulGXln3nEUH5iYmLxz0QkTYuu5nsb5yrphf6eAoxjLhGNiCqx3E6qdDgY4QDtzmYQ6TLK4sOPASjHVpDhKtvo3rvXa2RPt496XbOyc2keT08uNBVEuUGPssGRSqg4R1n0sztpDrZogpMbIJcl6FuPUPfzfrhxscYUpNljgWYWqbodJl4BN4Zc9fvZb/GzsuoreHVa0XgvsEwwVQ4FH5RKfQo9d0NUNk+ve2GV6PXD1xfDCs0GC3Spr147swvqzH8Luw7Di65BfbzjOW04RUyOJKzAS1x/dCp+7BOYPMynt7jWEUW72iW1CgCJO5FtZLHDl6pOPl0jCc3tgXg08uAn2t0GHhG9dBuuXGoLufOTJnfDSfvjq5eAepRWSiYnJOx8J7/rlvTeLlPKGNzPOFFjvMgK4qSIPp6JwQ/HJjykoVFGPRj4h2rFzgFx7kOzVHfgcRZQcO8r2RDH/4NQoJo9pIkVEValgOnAhkgQqHjKBhZZZrFMtWCdJvlaj0KTbmSwXoWQbKqOUFDu0CDoZ9A44oFGFQR11ts7kkEJXr8CqQm31G79GIaB+MjhUB2638YEwEIS/vQjL50JuFsRSUF8HliDU++HXT0B5LvxTPXQkYUMQ5rsMc9aRqMqH900F+yn57xs2GmLoxmtObMvPgm9ebwir0Ujp0Bs0XN4f2gkH2yHDAx4BmRPsyj6RTC0BuxWcZo9CE5N3NSpu3Mx5W84lhHAAzwJ2DB1yv5Tyn4UQFcB9QBbwCvBRKWVCCGEHfgvMBnqBa6SUR9PH+jrwcUADPiel/Gt6+0XAjwEV+IWU8j/T20c8xyjzvBVGTwyWUn5urOs0Bda7DAdWisgc8bEgIZzYsZKLjxuR/IWpSgnd1gEyUl005B7mE9YZ1OFAQydJmAKcqIY9+uvHGSLBE8oRXJZMBrsyeLlNcnGx8Q0cTIJdgbVeFV234OlTWZplZ6AWQpdqFEyTfCVbhSHIzYGsN9hOJpSCVwdhTgBqa0/0FhwYgu17YVq1IbD2DUByBvzDKphSCRsehKQGy+vhyRBsDEFKwvSw0Qon95QWMXkBqMyA790OV14I89LLd+tPiUylUvDIk1CQAwtmjT5vlx0+YHT74eYLoHPQ8MvKGqMF0qns7oQiHwTexuW64mzjZmJi8u5GI8KQ0Wrv7SAOXCClDAkhrMDzQoi/AF8CfiSlvE8I8XMM4XRb+t9+KWW1EOJa4HvANUKIKcC1wFSgEHhCiNetqn8GrAZagC1CiIellHvSY0c6x0i8pTVTU2C9R4gQ5Xk2U0wB9UxGYCXAZQDcbUvQIV8gwzYfV0KCKtkpGjlCB1UU0sDJYSY3VhbqxRzodqJHHXzpQfiZL8rlqxUOD9lwWgTTswXLPS6uqAKqjo8cv4V53xC4rPDqXqirgMx0u6D2OPy1B4qdUDnMzqC8CP7pFvC4ISolLluKksAA4Sydlx0DfPjCGp7vUdjTD7mDoB8CVwW8eMhoiXPB7NPnYLcBKuzrhoakEckJ+E/eZygML2+DgtzRBVYsAV2DUJK2bbh4+rifhtdJafD3RnhfOcw8S+1yTExM3tvoY1jWTCRSSgmE0net6ZsELgA+lN7+G+BfMMTP+vTfAPcDPx1ml3CflDIOHBFCNMLrmfqNUsrDAEKI+4D1Qoi9Y5xjpHn+5q1cpymw3iM4sDOJSjI5vZuvio3C2DZS8UfofzGGr+ajaDUZSHQ09NP2VxBogxnUuuGHz8G+V2Efdp5oTDH75m5KNDuHhvwUWmHeMBEU12B7EBp84BjjfXysG36wAWaVw44XYf0FsDztf1ftgi9XQGZ6yaovDJsPw5xyyE4vtf08kmRLZ4KtLRptT9kIt5WTvUuy+h+hQ4fIMdj3CrwahC+tg8YofOcoXJ0DtcPM3UsL4aJL4aHtsHQQKtKRnJYB6I3Zsv3SAAAgAElEQVTAlDzI8MNnrzMc2oeT0k60zvnLNti4Gz61BupGEUeRyNj+VxYVPjUfrG/P55+Jicl7DAU3XsYIw08wQggVY4muGiPadAgYkFKm0ru0AMc/MYuAZgApZUoIMYixxFcEvDzssMPHNJ+yfX56zGjnGGuuOcDXgCnA69moUsoLxhpnCqx3ORoaEcJ48VFF+WmPJ4jQyDPYVI0WcZSmVTb8+v2U8hmWMh3fCJV97RG45zBEEjDYCQwBUYhnW+jrd3DB1DYuTjqpt5+cuGNTIN9hLCGORaYHUlnQlQGf+TAU5Z14TIgT4grg1WNw7xYjx+nS9DLeHKtCyGnloVw3uFNYFOjZouLoh84iqJoM/1ULVTlG3tSABt0J6D3Fn+rRfXC4F25cDGXpVddIAv53EwzF4f31sKgcik9p6DwUhR89CqvrYeEkI/errOuEADyV/gF4bhNcfMHYfQpt5rvVxMTkLKERYZBtE3W4bCHE8OW1O6SUdwzfQUqpATOEEAHgQV6vJz8vuRv4PXAJ8EngOqD7TIPMj+x3OR20sI/XmMwsQggqyTnJuTwR6yHi6CEUbUc7GkQvyGEgu4uX4ltZ2TydotyXqfNdgxIXcKwRSipxqC50FaIWyHRCq56OcjXqtLwcY1dVAp84zExZy3AvWyGgbBz5Q14n/PuFRrw4+wzJ1bPKjNyqOeUntmUMWdA6VN7nk2yMQ8qiQI7Gxav7qLbkoGBE0Y4zzw8VLsg5Rdy81mFEqrI8oKQvw6KA126c0zvK3DZth+2bYGX646JpALpi0B+F7BEaK1hscMESswm0iYnJuWUCqwh7pJTjypiXUg4IITYCC4GAEMKSjjAVA63p3VqBEqBFCGEB/BjJ7se3H2f4mJG2945xjrHIklLeKYT4vJTyGeAZIcQZbe/PmsAao0pgJfD/ML55Q8D1UsrGszWP9zrZ5OGjke08S5BScvCScbzJ8dAA7t/fRd3KZQinQo94kaR00JLKwN+zmye2rMXjyuX/XOKEV59Eeez3sHI9yrJLIAcsGvzr9XDNQY2kriBKkuRM7mSrZueIPkRFV4r3V5xQIboOz7wCdjtMmwRui5EkbrefPu+sU8RLUkp0wC5Ojqhluk/PadreB1s6BJeV2+mOphj0xbBeYkfRFQ6FYesgFDkM8SalRBGCvBHE0k1zIZSAomF5VzYLfGoRhBOQM0qCulWF1iPw+GOweDpYlbQdhISHdxlLi6WZxn59QfjJ/bB2MbgzYCAJBQGoegPJ7yYmJiZvFRUXfma+LedKL7kl0+LKiZGM/j1gI/B+jCq/64A/p4c8nL7/Uvrxp6SUUgjxMHCPEOKHGEnuNcBmQAA16YrBVoxE+A+lx4x2jrE4vr7RLoS4BGiDUarJhnE2I1ijVQncBqyXUu4VQnwK+CZw/Vmcx3sOHUkbGgWo2HFQSz2FBHFTQAAj0UciCTt3Yl1ajyd3BrgWkcqowZv4CfWDu+kPx9niVEglM3lsCLbnz+aG6S34S2q5t99oQWNR4fFWDceKKJZcHZclhn1yirhqIxhz8UKfis2X4qJMlaQm+PUh+OVL0PgyCC/YdEkoAkW5kl98UJCTJ7B6wO8Cp1XSk4BSh0BV4N9jg/RJjR87M1DF6L+yesNw071xkkmFDa9YYYMLHDoVn21mY7SfQty47Q4G4vAKSfqk5IP2kUNR2W7jdioum3EbjWVz4T++CJlpYXZxPSybJNkZlNz9gmBunSAxAHMqYc9ReGUQ/voHcGZBlh+qLPCFNYZD/TuRllZoa4d5b0/Ft4mJyQSQIko/O9+u0xUAv0nnYSnAH6SUjwoh9gD3CSG+DWwD7kzvfydwVzqJvQ9DMCGl3C2E+AOwB8MA9NPppUeEEJ8B/oph0/BLKeXu9LG+Nso5xuLbQgg/8GXgVsAHfPFMg86awBqjSkCmJwdGmK/tbM3hvUozKe5liGvwUoEVPzn4yTllL42YpREmTSZKE51sptsexx1upiR+kKitmQ7bt5iUfzk7U/MZcAfoXv8xQkJny2HY8CedoUxJ1uIOZq98Fb8CyZypTLbk8/cBnT1HfPx3Y4oH+rvIz3ax++5M5FJJbrVGeL9E71aQmQpUQOMxyfI7dUgICIN3XYyGRYK+uMLlupUvTZFYLDHs6OhCjFnn0pUMMvuqzbQ35XPkxUlgtUBMofPZDPa4LeiKlalF0OmGUEyhyD3+3ofjRQhjyW/4/T6r5H+jSfYNWenbJLjxfRql2QoPbhZs2gNaAuo8UOeCSBAGwu9sgbXvoCmwTEzeaejj7KbxVpFS7oTTw2Xpqr/T+vVIKWPA1aMc6zvAd0bYvgHYMN5znIFNUspBYBBYMd5BZzUH69QqASnlJiHETcAGIUQUCDJKR2ohxC3ALQClpaVnc5rvOoqwcBUeisf47xVYyOI6QCHMMYIcpUvvRZMeirfFICtOXu/DtB7ZRvHeq1h4yYewdrkhIPnjd1VCSSBfEJkWwFeQxWzvZi6311Elgyy0VvKxwRS+zF4unXk3W11zieQvgF1Omi4OUvb5TqzbvBzrDJAMOJFhBXJ1nJUhoi/5CHsUmtxddEVz+UGrZLsjxE8qAhSoVqxn+ADI9AlSB+2kInbs/kHisQzQJJE7o7zYE6Dogwr7DsKKPHi2V6X2LRpoRhLwwGtGpOviSSPvMxgCnxB83Gvl2RbBoXZJezDO+qs02jO9FFeBtQUCSbioEgryoSL3rc3rXLJgnnEzMTF556DiIsCMcz2N85UXhBBHMRLd/ySl7B/PoLMqsE6tEhBCTMMIq61Ni62vAj8Ebhph7B3AHQBz5syZ+DDDuxgLghrOrByOx4JSR8HSpJC7KETL4TCbHnOR3RfEGmgj9oEqHnFN59tqnIiAT97oJrQhDsVWuFAHRcPuSjHVbqOGBxACLvNdz32F29jU9Bt0ZxFzbC/Tfk0u4c0+5ua+RJ81k+QyDz/IepX2g4t4Kt/DnqEoml/HszZFfUMvfXYdPSlI9Qtod5BdYcE1Do+WPMVL4NUZbI9qJP5ihfYYuAREosTvU9iBwpqb3FQG4IuZRkWjLqEjZlQ4Km/wB1z7ELzaBk4rrKgExymJ6skU/Pj34HYKMr0CnwP6pkA4YePFw3BVIcyZA/t3w6GDRgJ9Vf4bm8ObIaGBzbR8MDExSaMRoY9d53oa5yVSylohxDyMpclvHF/KlFL+bqxxb0sV4bAqgYuBBinlpvRDvwcefzvmYDIKiWN4LHdTUXIZqqWY4PNf5uhjz2JRoHxxmMEw1E85yl4xiChuoTVYj9NXQLQ7D3rsLMh8lkV6nBViAUpyC73795JIHCRr4x4cjU2Ey30UTI4zLXMn1qlwnf4rHrNcwjFZyXT1Tj5QF+BbU5dyTCrsicY56uigTdfpivrpj6r06lCkHKFZjWFjKkGi5OJDjBHJ+pcVLt7/4yhtjQmIhKHDByigamRICxXhk4XQC11wfwtcVQxLR4kcdadzHHNOMUstC8BVUyHDebq4AiNPbe5kw6T0lT0wzQ83XAQ9+RaKNfhYHjhssKAU490xjMZ2yPGBf4Q8sLdCTwRah6Ah78z7mpiYvDeQCDSzF+GoSCk3A5uFEN/FCAz9Bjg3AmuMKgG/EKJWSnkgvW3v2ZqDyegMEmYPLcxSMrB7pxPwTgJcVASWYhF3IfpB98KC+59GPNxCuHwpJd9ayxVf3cMT9wd5JZiByxthX0sD1/k+T9YuaHzew/6HHyTU9QBr//NKFtUo2Owv0KRVsK9rCpf6HmX1wd2k5szGKffTEM7F6ylGRWGSsDPJZWcAJykh6XdaiGdqbHKlWFuVxIeP12imkU4upoEAo6uOhbUqrT/zcEU0ykNPOcEuoROmLfZw08fsTCo5ef8su+GtlT1CNeNx2kmgcLrAsqiwpGL0cULAxYuMv+dNgZ/+HuYoMKcAmlvhUBu8lIQLC6D0lMpBj8MQZhNNtgsCZuNmExOTYVhwkkn9uZ7GeYkQwgdcgRHBqsLw7TpjIsTZjGCNViVwM/CAEEIH+oEbz+IcTEZBQcGKirBkgv+y17eXr7gAy08dZGTHsMcEu/1+Wh4JMvsLNczgYnJXN7K6/ll29T7HU80NPJa4kB/pX6bZ/0e+Ev0teatD7H8qhP/Vn9PgshLOctKyq5Lp/j3UDu1B869myh+rqZvWjDWzCHxVJ80rgBUEZFvgJ7WCtpROg3U6cRIc7U9RYPPgdY+vGd+Nn3Gyw6nRG9O4dl4JX7lGoSYwwo5WWFNmJJmPRv0Ygm689PSBV8DMSYarfX42JHVwdRj+WgCRpPG3TYX8s5jkbjF/qJqYmAwjRZReXjvX0zhf2QE8BPyblPKl8Q46m1WEo1UJPIih/kzOIV6czBnWY/DYE0+QikapWreOmu/9jPj9/8SBah9xLY7LaWPaNdegopJNGQF3MQ2xNn7SUo8+5OHA/gbuXa5wtbiVsoscLDgYA4+GVQO5xcbBvR5CK50k72xiV+F+kgNP4jg0j/xbPkr/c8+RWVODJ//0xKNcRZBrUxkixDPaJl7oraZClrG8ZnzXWJkLfr+KRVfJK4YNbVAchgW5OgVWgZJeZnykD3o1mOoCNLh3D8zIg7kFpx8zThQAO2+843J+jlFd6HDAc31Q6IBqN3yo0ni8NQh3vAoBO3x63plzpF7rhBw35JmeWSYmJm8Rc4lwTCrTzggjIoS4VUr52VO3m07uJgD4q6rQ43EA5Krrsay6DscLD9P2ya+SsDn5UVERjvoA1k9fSM2Nt5CdeQstd+VDUoLbRRQXG0tXM8XSxqJ1R7GHBGpzkug+FXUowSzyKClxsmWOpEKJk7P1UfbfaqOfchyBALXLl9P13HNUXHstnlOqRu3YKFbz+GChj8I38P4/EgaLA4Zi8N0dOtqghFbJksvjfLtGYbHupCUM12YZpm3JJDywC7a0G8toGRlw/wCs9sJsN6RIsoONSCSzWI11HIUEAMEIvHLIyKealhaHizLS5qPD94tDf9hIjI+nziywslxGUvz5QFQDp5k0b2LyjsWCk2ymnetpnJeMJa7SLB5poymwTAAIVBiJRJIYUe5AoRBr5GaW/3aAyDPw2gOwrzdGf1Oc0IFmyieXQDgLNQ7aDIk+ZOFW23fofyqD2mnbeUL9d5y9CZg0gw+U2qietZWuvGqirgzcXhs2h0pFcA/9zWXYPB6C+/YxdPgw4aam0wSWDRszmAYjNENOEEUjyWDcx5EIzPZLBpQYmdjR3QpFNXBU0ZH5CRxVCWJPuTmwzYqrQrJnEO7cBzdPkfRkRHngsMKdGx0oIfjcPDicgMMx+GMffPdRqChSuPIqN05HCuUN/NJ7aBPsaTH+/szFUJw98hLdpGzwh6DGOz7hVDBKb8O3m6NBuKcJbqiGAjO3y8TkHUmKKF3sOdfTeFdhCiyTEZC06dt5eOWV5MhWllv+yoJ2yD0a47HVXpQMB1mikEW/GeTVvRolWa2sL7yXrW0z+fttV7P5z8WULpjNH795Kys1N0LJgMhWrJW1zHTsQhcK4TIb7qEks9bcjLewEJlMkjV7Nv5Jo5hJAS8fMaI2NXkgZQKdg+yniaOJCF94Yg3N+x18c+0AgboulpHPuqJMPPMgUih53qOj9wiIKXzSo1KbgLuUGEoNeN0WHiRGPB9m5Vg4mLTwX/vgS3VQEIZ7NkDjNvA9JVi02Eld+RAiLbC62UeKOAU0jPFsMi77PkXAx5cbvRjPNvv2Jejr01mwwI7yRr0pTqGlB9whyHmLnmImJibnDsmE9iI0wRRYJqcgcGDjg7QpP0fqFqIiQDjgJZYRITG7EGpsPJ7K53Lp5oryCE37BygZasLqT3Kz7S6eeulqtLDKgC2PJ3BQmOoje/fT5Hgq0aZfjTz2fWwdgxxQq4lXOJlTFUBVEgjdTmZ9PVp/J7tf3sgr89Yw1+FmsjjxrV2WCe70XZ1DJOWvyRUruSPkJV7YgV3L5KGd3fygLkQhTpqbdbY+kGBBtqByrZV5RXYuXCy4Zxdc9jvJliMWpl+d4rNehYtjXvwv/RWPfpj/unQJDz9dwEt3FXD9dVCdKxnIkqxtkGRl9dGhJWjR4qyyOvCKAnRSYz6nV8yHLVmQ5zeiV2NRPoZ1Qm8fhCNQWjze/83ReeCBCD09GjU1VnJy3tra3qJymFdqJs6bmLyTseIkh6nnehrvVEb8lWoKLJPT6OYesthHaaqHZFyls7+QX193LTXedo7Y69jdUY4q4WaHHbX+WfqdzTgsfrxHK8nO6KHz6hxYmOS+3qv4U1BjzoKXueFQnJXWi/D2PUmntZvkVBsbbYvZO/hrViRD5NiSNPo+zS8R6Jl+huRhHunycZu/kGyHUcFXMKzpskI1VnETOVSQ8OygvOwohyNQkiVYIquxCif3vpAgFoPEAZ2L5lr4i1VFSUBfDAatOqpTMjVk5T8OR3iyWaG7bznFfbMRXUkUX4T+AFxRC+vmS45eDZe5VFKs4Kl4kieCgvl+8NuMSfVFDfPO/BESzn0uWPkGq5+TSVBVUIaJlgc2wLEW+ObnwfkWo1zXXuuiv1+Snf3WVZGigO0si6toHF54DWZWG/0aTUxMJpYkMTpN16QxEUK4pJSRER768Uj7mwLL5DSCWOnDyYCSw6DLw9BkH8XJXqRDEpI+vhPIJqVsReMIVxWvoGfgLrKjIaw1ZSTusuNwhHDm9jMQymSwJxcZ0DhWEqRE30Xd7PfjTASJas0EpZsMxz52HA5SOq2fJyMudrkXMlScRX5XMzlDWSiaAwpPt0gQwoqKsZz4VXsdX9MOklPpYZHMQ0WgoVE8r4fdjR7y8+0sKVfwJGCaB2baJUMVKb6UpZCnxfj6nhhNe1xox2wcyMtEf1qBLHDPlNy6U/Cf8xXq01/qFuyU9tgI/gU2zxWsTq8M7uuBoTjkV5821TfFffdBYSGsXHli20XLoW/grYsrgKqqd9Z6Xmc/PL7ZEKumwDIxOTuYS4QjI4RYBPwC8AClQogG4BNSyk8BSCl/PdK4cQksIUQtcBuQJ6WcJoSoBy6TUn57IiZvcn4RofT/s3fe8VHU+f9/fmZn+2az6T0kQAoQekdQQECQJvbeznKe3qmnd+rpNb/e99Tvld81Pb3z7O0Uu1gRAUGl996SQEjvyfb5/P6YIAmkAYEEnOfjMY/dnZnPzHsWdvKaz7shsNOoriKkgWKSxIhaHFo92Wo9MyIcSLIwkUCENZ2kmJ9AsJSgOZWrIwWvUYawBKnz28lK2crOnblU2auoGP4umG4luhoqX3+UPua38Nzcm0THAfaZMhhWsZZFMYOxOPwMCezkxxHjiY7pwKcG9CWCJ62DOBhpopdFoAiopJ7KPtu48ef96aVGYDIJpgCNUhKKk/xeWDAheT5cTNChQnIElCpo6yVYgErwlzfw2lYLmfss3H6TZLviI9Kv8pd/mtlTAFFjD9swNlWPYWgNrxcKCyE7+4j1AdhXBlmJesHS5kyYAK4jZsPSU7vGPXiyec7rY0tpiPtSBTGmrilDn5EI914GsYa4MjA4KajYiad/d5vRU/kzcB7wHoCUcr0Q4uyOBnV2ButfwM+Ap5oOvkEI8QpgCKwzEBUPFZSiksoBwqgygNAg01vILlsDYWsYMx6CqKhoCDUW1FjMwF9dcEVI5arqEtRQFZHJtSAEDTvdvBh0Y6pYRvaXqxnSawKNzpdwF21k254E+qeEsYhERigaGg3cFT2DNHMzcRUOQ/lBiEtu6TdrIt6kEt8syzAKNxMYTqTVhamZe3y5N8TXhLnXbsUuFC4xpVIkFTYVKVAOrAFkEFIUQllmnGEfX5gtJBWF2ZFaD7Vm1u3wIENgbeZ1F6LtQPbGRigrg6wsfb9DbCyEV76GH0+FPkfEXmVkdPIf6xg4FaUUQjLMW+HdrPO78Oz284sjVeUJkBjdZYcyMDA4giA+itne3Wb0WKSUhUK0uMuHOxrTWYHlkFKuOOLg7Uf2Gpy2eAKNRChgNcUTxQ4csp7hoQ2Yq0JMy3+VxfteZUBuLmuHnkt/0ygyqqLBnQAW3Xc1VrXzr4h4fmrey7b6bNzOKvr03kOlEuB5pYRxsyZxtWcQIS4jXCLIHOzHYlGpc7q51RQiChcxuFsatW8bvPc8XHQLpHfshxMIomk53bF2I3z6nsqsK0zYe+v/l+OElZvT4Y9LNHyrAH8QZADsNgiYyR1sQqsK8eZfTVz9kIuvwyZmXwMpAvL6QX4B+AOQ3WRSYwgCGniaeeBiYuCsVqqkDEiFH5wDvTqepDthygLwVglclACxJ9E7aBIKdzg87HCauCSh/VL0FQGwKeA0AhUMDHoEWqfynb+XFDa5CaUQwgzcSSfa/HX21lYuhOhDkxdECHExcPB4LTXo2aTWrSdBSWZL4RYCOTZMpjjSG3xUb2ggcVkVyz4TuLM3Ic76lmVXj8C0ejvJWwdguuU/4ND9WpMsdu4O2flnaBvZ8VsZkLyJaiLZXdebta4F9Fr6LGP35GCJFyhKEQy8jIjkvrRZ2ik5A6ZeDAnH5iMLBGHddsjupc8eWRVBWlOFTylhxXqIjICfjoE/7NNwREFtA2h2YF+Idb0lCaPqMK+IYEmZlfMcCqHxsC4EXglfLIGa2sMCa18D1AdhVCdEk9MKA9M63u8QSzZBYTlccXark3hs3QWbd8K8aXqAfHNizDA1BqJPQm/D5ggE05QUpqW0v583DH/ZDal2uCXj5NpkYGDQMWZsJJLb3Wb0VH6IHsieAhwAPgVu72hQZwXW7cDTQK4Q4gCwF7j6+Ow06OkIzw+xoHKekk9BwUoCqelER1xLbZWVtf95kLI6yYp9ELGvmIFZK9nvcuIPfkHczkIiBvZDUUARgjlOjT2OL0DzYxE+qsK9cEQ2YsHP74eP48nyx4l/KwXbWakoOQ3tG2V3Ql6HvTWPIv8gPP8+XDwFJo2Egf0Oi4+aOnhjAcRGw4VzFMqkwpVzVd7fYWX+RonfrzE8XI/mqcM7oQFviYd/fRrJjLEwcbzu4ht/bsvMwf6RunA7GVTUwcEq0CSthqJW18DBUtC0owWWIqB3K4VauwurAmOjIaGHVKI3MPi+E8BHETu624weiZSyHLjqWMd1SmBJKfcAU4QQTkCRUtYd64kMTh+kYkHzvkuv0EdEFVlx1BZiTlTIuvBciot/T8VvfklWOEhcfRDl6f2kDHdQH2/hsQ/cnFWpMXOS/uc/hMLZIpoVooD9ZLIhPIQJ4kv2hPtQZ/bw5cjBTN6aRFL8UJT0sR1YdXxkJsNtl0BGsv65ufDwuOH6iyDCBSgwaxyclQXbqgQz8gQ3jRJ8uKuWZ3cp7PsihS8dGrb0MA0rTHx+Hjy7BbJiIOuInoXiiFn2Hbtg5WqYNR0iTyBI+4IxsL8UPvkGJo8A+xHiZOxwGDG47WD7noQiYEY7Nb8MDAxONcLIImwDIcTj6DHnXuBjYBBwt5TypfbGtSuwhBA/bWM9AFLKPx2PsQY9G01bj6Z8TqMIsmdpAX1nB7B64xGRdsbedRfpjYWYPv0PlX4v5buhcWsjWTd7GTL7Wj6MuJbABjvP1Y3h3DQHI9JSSAwlsaPOjzns4TNlLiaXjyhRwWZrDj5fETMqgvRSTk70tapCXpP7bm0NBDUYFQVfbYXXv4XJQ2FeDry2FkSChlQF0R5BkgZDkgRfb0tn35cS9kuoEph6hbl+hm7r9SN0MeX1Suz2tmMXNm+B5Stg+NATE1hCwI5CWLQG+mdAeiK8sUzPQJw8EDbshmK/vt9lo4//PAYGBt8/zNhIouuSUs4wpkkpfy6EmAfsAy4ElgDHL7Dgu5CYHGAkTSmKwGxgxXGbatCjUZQMpGkQatJMYs8twJFrBtkHzANQrVaibr6fr/bWULN2JZU7tqOGwfSEZGLeEiojI/Dd9Sm3Th3IPXl/4vnEPDLUSgab+9HoiKFMBvCJfDb7l+N0l7Hr8lj272ik1ym4rj4OPe2jpB6u+Agqi2BlCYzMgtREyUZLiD2aiQtyTQTCYFJgVDZE1kFNOeCDe/PMXDYI9pdBcgzs3i158UWNW25RSElpXWSdNwUG5UFmxolfw9lDdHG1YKEeh+9zgNkEm/bAG1/C5VP1WTuAHftg/Ta44Fwwn+TYKwMDg9ObID4OsKu7zeipHNJKM4E3pJQ14khXRTuDWkVK+VsAIcQSYNgh16AQ4jfAhydirUHPRYgEVPOPUePBGQ+UvgqNX4CyDJJ+TLCxkYicXLIvvYwlv3iAio2bKAtA8toQ1sXfEtjgZUjOCu6vuwPTlkg23vcXprlNDEew5oCVR7/thTd5OvG5y7AlN7BOfINa6WJU9FWILsxiKanSZ3PiPfpnd5PIKAtDegoQhmmZ8O5BOOAV/KKfGSuwcC/0iwWHGT44AHfeK1izThARCz+bCo/8HT5aA7+4Aybl6hmCUVF6vSuLBd57T/88caJ+PocD+vTummsyq5AUC8PywGKGAU0PnD4/JMZAn+TDbtCyCtiZD8GQIbAMDAzaRyLQOMl1XE5fPhBCbEN3Ed4mhIgDfB0N6myQewIQaPY50LTO4PtAYD+oURCqglA1MdnZTHjwQTa/8QaOvlkIi0p5xUYWPBvm4PZispOhoQaG+zfjKBf4q59ks+0qHq4rp7LagmVQI9V18exdNA7PgEoi7PXsdu7gQMOvmW4ehcMyq0vMtlmOjocCiLPB+/Ng8zoI1EH1bqAMXHmCWj98vR8KS2BiL7guF5Zth4UBiDDDH56C+W+ALRpcZklpqWDqVBPffguvv67Xulq6FIYNOyywTgZDj2gZZrNC9hEZiWcNhzFDdMFVUgtRDjAJOFAOybFHFzftCWgSDvghyXJ0b8OD9bDyIExKhwgjON7AoEuxYCWFLmpFcYYhpby/KQ6rRkoZFkI0AHM7GtdZgfUCsEII8XbT5wuA54/PVIPTjph5UPkBuEaDLeO71bVTfe8AACAASURBVIqqEpWZSWRaGg0V2VR8/jaOSEHOGElNWZDiPUEKTQqufRXYPrmSmXmZfDvuPPJJJMnhJXAwi9LyBLR0BYs/wEciFta+xOzRcZhpO4goTBgfjTjbLuoAQGQ7RcSjLTChKSmxthZ8Piiv0psp/2wc/P0tWFQNV58L366C+r1gDYPDBpMnQGa65IYLJKpJ49lnTTzyB9iwCQb2l0Q7IS8P2i49empYuRVWb4OpY+Hpr2ByLiTa4d8fwnXnwYicbjWvVbY0wNNFcE0ijGxWCq0+AC9uhHUHNYr2weAkhbF53WamgcEZRwA/hezubjN6JEKIa5u9b77phfbGdTaL8HdCiI+ACU2rbpBSrj1WIw1OU+x9IeUuALTKSkREBMJsJmfOHJJHjmTbO+9gc1ZS30slou88tpQfpHDpUjR/iBAanss/IBTjY4gpTK/+lXwWPxkiFOomqBRrA/nIdy6W/WFwmng/cQ7Zux6jYcm3DL38JiyOo2sLVFHBAfIZxHBEF2S9uN368toC2F0AD/4Qbputx2Ct3QahBnjgQjirHyREw54CeOB/wK/pGYiFxVBWDZVeyRdrJHYkWf1h+nQTS7bAx1shNxqumKC79ZpTWgkuhy7c2qKqGrw+SE48tuuqa9CP/9UOSHBBbpJe3uGcwdArASrD+g3A3YNmspKtcI4Heh3xfeyphhWFGuuKfCzcrXFusoshWUdnUgLsKYK3l8EFZ+kuUwMDg44xXITtMrLZextwLnrfjxMXWEKIdPRGIm83XyelLDh2Ow1OV2Q4jPeFFzCPH49lxAgUk4lAqp/4O0YSsftcKvbVE5OdTYnPR9ywkdSXllKzezfevT7McVDXYCWpupjE2BLc4VoqFDdqOMRufyaq2c+W/DxsfWp5zeFhbuzzeKvntSqwYojFTWSXiKvP1ur97XZshT/8Eyb0g0BAn/3y+2HRcog1w0Xj9YxEgN7pkJUuyLoB7rhFYI+Abw7A1ucg3AAhCSUI1u2FuTdq1HiBfgp/fx3mPwzpCXr7HLMZnvkAhmbD9DFt2/jBZ7C3AB66u/UCo20xeQSMzYM/f6rPpT35FYh4uHMkxDngH1WgmoLc4DZh6SHp2dFmuCj+6PU50XDDcHjcJQjHqFzbr3VxBVBaA/klegyeIbAMDDqHBSup9OluM3okUsofN/8shPAAr3U0rrMuwg85XF7HDmQC24EBbY4wOOMQJhO2Sy5BiT7cFK6SfGooJrHPLOb++2UAQj4fNYWFVO7Yyp4//gD8jTSUQk79PorqU0gMVhKwKmRZ9uCqDjLItYZ/7L2HeksE8bZCis3JBPtZCbs3AEeXOhcoWOiaIByPEywmeGUZlIYgJuqwiDGZIMoN8WpLYaOqkNsX/H5BUpLeXWd4Pwj+UrBxP8zrA/5YhS/WhqlZ5wepwX7Bqjwn9/0RfnghbNgMFsWPUuknb2pTIa42mDZRd10ei7g6hN0Gd02D38yHD1bD6DGwNx0OFME50VDn8FOLhVhOYv+cLsCqwsw+CmMybDgE2Nv5LkblQHocdNCpx8DAoBkB/OSzt7vNOF1oQNdB7dJZF+HA5p+FEMOAHx2fXQanM6aUlj1QMhlDEC923NA02WR2OKgvKaGxohpX3/Hkf/Ul1Q0B1i2Fik8P0O9VB7uzB9AQcjG4di110TbO7f0JK6tG4wzU47LUsSx6OAmm3Zzs/r4js8EXgJFDYfgAmNRH7yutqvryo6uODpQXAm649vBnVYUbz4dp9VBYJxgWLwhrsGcH3BsOQDgEVQq2YidRql5pfcZk+P3v6tmy2c/M82ykJrUtcBLiTuwa316np778cBJYYuHDbRAugol5cN5QV4fjexIxpo7j2hRFD+I3MDA4FgSyh8xk9zSEEO9zeJLJBPQD/tvRuONqsyqlXCOEMEoZGqCxAihA4yIUDkeVx/XrR3SfPtRNn454+WW+fOghqsv1kgXFDxaSeVUIMbE/5UkJKH6NPNsmHEmNCLOfBi0Kt3SQZL/0lFyDzQL3zYNX3oFL7oFf3QY/bGoE1YlSJ9+R6tIXAEyQN8DEz+9ReP2/kJbqYNY8+MltYNd7YvPwbyPJzw+Rm2tm794wbrcgJqbzN7hwWI8ZS0/WswjboqJBj/2aOAR6J0AgDD4vRB+HtvrgE0iI1wWpgYHBmYMFC+kdT8p8X/lDs/chIF9Kub+jQZ2NwWpe0V0BhgFFx2SewRlJiF2EKURSB7RM2zNZLNQUFpKQl0dUVhbVhYWEAj5iC3xov95L0mMmlLx+5KzcThVOyibHY/dXokQ7qFYcmGU0hH2gthMB3kWYVRg+GMYNaykeVmzUtw3t1/748noIhCDZ03L9Y4+5eeyx1sekpKikpKjU1mr87tFGrA4Tf3rUgbWT3s9d+fD3F+GymTB+xOH1gSB4A4ezKK8dA5UNkBHbzM1ob3msr9fqpSnmTgZrO97C5CTdjWpgYHBm4SfAXvK724weiZRysRAigcPB7js7M66zM1jN8+FD6DFZ8ztvnsGZgESjmt3YicOGriQcXIakHhN6ipsXjV0ESEJFcgDLhER6MYHe0yZQtO1FGhatoOqlT9nmA9vnVvK+Xk9oVTlaXzuT9r7DprvOJl+kME6Jw1y0Cg58AyPuhKZWOkU1ej0nexcWzlxbALVemJAFb/6t2fVK+HS5HsfUkcB6+iuo98OD08F5jOFhERGC5N5W9pcqlFdBSiezBdOT4dLzIe+I7hbPLITCCrjpXEiMhH+/Dgmx0HtO28dathaKSmH8MEhuJcj8EMMGdc62U4WUcKAY4qLptDA1MDBoHaMXYesIIS4F/g/4Ej1n6G9CiJ9JKd9sb1xnBdYWKeUbR5zwEuCNNvY3OAMJUMd+lhBNP1IYB4BCBM319zd4WUQj1pCJpKDCeFst8SKEsFcQOXQfytD+OG+8g8GNvdDsFTQWrqVOfkOMBFFhY575UTThxdLgh9KPICYPH5WUsha3lkowEEfYdoJBSUfw3jqo9cGgVIhqNgknBPzocr0xcUeM7wP1vvaF3+4C+HIlXDwNIps9sgghuPdHFiqqOy+uQBd+E0a2XCelPnsVDuszamFNr/Je39D+sa6dC9V1kNS1X+1JZ8deeOJlOHskXDS9u60xMDh9sWAlg4zuNqOn8iAwUkpZCtBUyf1zoEsE1gMcLaZaW2dwBmMlkt6cj4W2OxbHY8KBICwFO8MxjJYllIm9JJCBnWs5wCZszloiGiqo1PYQGBRJsuXHVH/6GRFjxqAKFSldyPoixIHVUBKgkZVUZQap9pcwSESBNhM4q8uu67px0OBvKa4OEd3J5sxnZ3W8T3GZLrLqGloKrCVLA2zYGOamG22AwO+HffsgM1NvvXMsCAG3TIXqBkhtCvT+6fW6m7M94mP0pasIheCTpdAnHXJPYuZ3dKR+jozUk3cOA4PvA7qL0Ki81AbKIXHVRAXtpX430e5tVwgxAzgfSBFC/LXZJje6q9Dge4aTpHa3D8BGbywsVKtYph5kqRCklkWwt14ie6VzrcwiYsszyI3PEJ0zlShtAMJai+v2H4FJ/+9YPX8+vm3biLvmRkyFb6Ba8jFrG6mXDkLhAah164D+YO+aYKCMJiHiC4DVrIuU6lpdBHU2yN0bApup/f3HDYNBuRDRTMgFAnDpVSGqKjXGjA4yYriF1Wvg+Rfglpth+LDOnb+4WH9NTASXXV8O4bA3xWX5j64dVV4JC5frMVwpibow8gX04qer8qFvHFQFwR/Wi6V2lgavHtfV4D25AisuBn5y3ck7voHB94mw4SJsi4+FEJ8ArzZ9vgxY0NGgjmawioBVwBxgdbP1dcDdx2GkwRnI0ib304Qm4WAL7aB/8AW2WgezQESS4yrC4vcQ9O6loXAPEY59kNMXUiMQ+5+HvZEQnQ6RR4i3yHRI/xUKfyY6uBY1WEZN5RpiDvhAlMO4n3SZyGr0w/97D6YM1otwPv0mXDMbcjJg4VJY9BVMPxfOGXv02NoA/GUzzEqDoe2UBxCipbgCfYZq2rmCrds0+vTW48yys2D2LOh9DAk9S5fr5bYuvbj17fOX6D0I772s5fpd+fDJEr0ifUoiLPga1u2An18NvWMh0g6Vwc7bcYjICPjpjeC0d7yvgYFB96NXcjcEVmtIKX8mhLiIw66Tp6WUb7c3BjoQWFLK9cB6IcTLUkpjxsqgVUrCR6yQtSRo1QyRm9hAH3bbE5iX8CWycTfe6PWwdSyk9IHIHIieDJmhFuLKc9FFEAohzHpAk4MfIuudVJd/TOiAj9KQg3jFD+EAR6E1QOO7YJ8Kps4HFNnMel2ozAQ9juqcEXo80uZdcMujkF8JT3wBK9+E7CNmchwqTEqCdBeEwxJTJ2o1NeeFZ+3s229HNg2LjYVZMw9v/3AxZCbrM0GHsgADAThYCukpunCbM/Po4zZnaFbrVc2H9Ic7b9DdbAAD+0CEQ88krFfh9TqYEwGO47jvRns63sfAwKBnYMVCb9K724wei5RyPseY3NeRi/C/UspLgbVCCHnkdillD8spMugOLnYfsUIdiUW1kSZeITHcyLqQC3vdJlJte4mprkPxHUQUzQfTLsh4BOQe6kKr8KpRrEQwWiQSa26qWqqFUaQDk3kofvtnrMxReSV6FNfWj2V67SbwFUJssxoFwgqmdBDtdHpuBUWBcc0yBaeP119LqiCjLxQdAEc27Ko6WmCpCoxPhMWLNRYvlvzkJwoeT+dFVn0DPPUqZGXAjZe03FZQCO+8q9fqmjgOLmrKBFy7CeZ/CHfdrPco7CiDLrfZfXPZVli1Ey6bAIlRMDDnsGszM1lfABwCepnB0r09qw0MDE4BfgLsosPSTt9LhBAXAo8B8ehZhAKQUsoj//q1oCMX4Z1Nr7NO2EKD7w1CCFQGEs0dFFQEUKsli5dPZXbma7yWORDfMCdJoUbSY0IM8P4Ja3URfgqoTJiJ3TSZAr5Bkk5cOAM+fRJCAcR5cyhI6s1eanErxawNhZm+ezHYYo8QWCrYx3fZteSkw7/uA58CqTEQ2YqQkVKycGGQTz9VMJkEplb6pYZCki1bNLKyFOz2lorF6YBLZkBsFNTWgbtZAPzGzbB5fZiYqBCLNRMXzdF/sv2z9SzC46nyvmYXbCmEgjKIiYAnP9Jjz26e1rIdj0OBsYaLz8Dge4HEKNPQDo8Ds6WUW49lUEcuwoNNb38kpbyv+TYhxGPAfUePMjAAUfcl8eYUJuzrw6Lir9gak0Q45mwq42NpVCPI1rZTbz5IpOljFHcksbVOrOFiYhU7n4VrcMkqLhPp2M1lmAMW7KSSoV7Dbu0lomWAaa4D0O8OUE5uDz0hIOvodogt2LkzwIMPlVNZ6eCG66MQAp55BkaMgMGD9X2KiiSvvhrk6qvNDBjQUoEJAcPyYMtOePIVuPkycMSBNQD7ymH9jgYay+uxCj/nT45j+nQXES4Y1F8fHw7D8wtAk3DDTFoVeM25bAKMLYOBGXpZB39QH2tgYPD9xYaFPhjpuG1QcqziCjpfpmEqR4upGa2sMzBo+qutF7q9Mq8vq6J34nE34He4cOfXEeluQImzkhuuhbAfxVRGo0Xga1zBfaGJrFVG0MtWgtNykEkTXRCaSIrFzgD641CSqcZCKglg1ytiaoRR6EBVdEAgBJ9thIFpkH6MfexiYkzMnWMlOtrKnDng9eplFlJTDwustDTB7bdbSEpq298WHwMjB4LXCu8UQOJB+PPz0BhwgGrCX1/O44+XM2SIjcREFSnhgyV61t/OInBaoLgGEiJBbfo6tu3UG0UPb7JjUxWkO2FEs7ISP5mtz3cfTzPp7mbvPti+A6ZM1ntCGhgYHB8+guw0GrS0oMk1CLBKCPE68A7gP7RdSvlWe+M7isG6Db2pc28hxIZmmyKAZcdlscGZjxAQewsA0cBtfYfyfvALYrcKzi5YzlfnX8N1TMHpfYxAYDfegAO/oy+NQdglI9CCGhsqMni8wUvfvFl8oSVzg6aRpFjpwz2ECWFq+q9bRRmb+JZ+DCMWPXhII4SCqgu9ut1g8eiuxHao98HXO8GiHo/AUvnFL1oOuv/+wz0H9a9EkJrafjBTbDRcOF2fTYqM0PVj4y3wyfvwzhsBpMlGYqLA75cEgxIQrNkOOw9AUi8YNxL+32dAEkTnwHXRsHYjVFTqAiuswbflYDGBu9nEn7ULq+KfarbvgOXfwNjRENnJmmUGBgato53gg+oZyOxm7xuBac0+S6BdgSWkbNs3IISIBKKA3wP3N9tUJ6WsPGZTj5MRI0bIVatWnarTGXQxYYJILYC67G/I+MHU52RhJQ5LYD1opWjWKfhFERXhZM4prSQYCHCwOBl3VZgf5ZXTO7GOgSYrySKZEqrpTypWdFVQTXmTwBpODInUU0AxHxPLODz1btj8Z3ClwYCfdmClXpzTZTs8+9NVSClpbASn82iBVVqmLwP6tayhJaVk/36NmBgFh0Pw1VdBtm3TOOcclRdf9DNunMr06Xr19/JaCGggTfD8csiPhdRcmOcGewiK/bAjCDclQWQHtbqas3Cxnq04Y2oXfREngVAIGhvB3W6oqYHB6YcQYrWUckTHe3YNOSPy5FOr/tslx5okBpxS27sbIcQDUsrfH7m+oxisGqAGuKLpIPGADXAJIVxSSqPsq0GH7GApXqWOIWPvIGQKUMQrRDEYpyUZC4Mw48EesJNSks9TtliWBVfxsS+dMtVMoRbHEFMBpUKlnCDF1BBPJLVaBGtkAyMUB2rJDJ45KPCZNL6ocnDW8DRucXrxWDMgKg8ienfKTs+xJR52yJo1sHw5ZGQE+dvfAtx1l5UZM1pOF325DFavhQd+Cp5mMzDbt2s89ZSPYcNUrrnGyvjxZsaPh+JijaIihaKmmfwYD0hPOY5wCLVxPyM8ObzxRYCVDRbqo52UoBC2QE6ULsLEMbjRSsp0gdWTUVVDXBkYdAU+gmznYMc7GrTGJegTUS3o1O1WCDEb+BOQDJQCvYCtwIAuNNDgDMVNAlZcCNWBGSepzAFCFPMmLvqTwCzw1VN24ACfil7k9k1izkDJt41BMmNK8AsLxdQxmT6kE0eUjOAZzccG/FiQ1NfaebEUtrlCKDYrBfv6kNvfwjVmF2T/4KRe25tv1lNbK7nxxoijtuXnw5494HRKNm4M8MjjYaZMicTcTGNNm6Q3UPYc4d7yeAQJCQppaS0Do+rqIDJSoaFBUBuCd4p9vLOolD3lASqjXMSGNlO4cjC82MAzQ+z6L3ZxkINzVWbfayLHBfMLIEKFazL0EhNtcWUbRUsNDAzOTIwswuOmVb9AZ59nHwHGAJ9LKYcKISYBV3eVZQZnNilH6HAn6WiEiGUKtqa4KdwxxIyaxMU1YRZHRJFuqiA70s87WhkV4RAWU4BSKknnAFZxLucJK6MxMwkLNRmw2wm19X7qRB0eTy1OUYBG2gkHv3dEVJQJVdVa3TZzJowdCx6PheKKMEJVkVISCoGq6r9HT+TR4gogMVHh/vv1IK6aWr2Uw/btAcrKNabONTN6oEJxAN7fY2HBi+mE9knCPhtlUyshVYV1VvgGPQiuRrLuRT9/nOlgTl8obARVQH0IPJ1MwgwGYcUGyMnUY8UMDAzOLGyYyemgFZpBm7Qaa9VZgRWUUlYIIRQhhCKlXCSE+H9daJzB9wwFlUhaNtozKTDYZeLdZQ7WUkLu2AMcDEbjVJwMsrtJxUyYACAZbTpckCrOAo+kwP04yJcSTezCogWYTwVni2gSOv3f/Ng599y2C0VZLHpvQBA8cK+DqirJ737XiM/n5957I4mL61j8VVTBn/8FCTHwyQKNSLePtBEWRg0RxGmQvBL8K63g1yAIvpciIDIENQEIKlBm03s8xkjCZtDqIVcDlwXe3gJOM1zaiXno4nKY/wmcfw5MOYY+26tW67kGI7830RgGBqcnXkJspbTjHQ1a44RmsKqFEC5gCfCyEKIUaOgqywwMDpFfDq98Gkl1yIktvZzkuGqC+xPZEqvRz5LBWEc/lDaCtF2YGCDchLWJaI1/xaEKIm1XndoLaIevv4ZXXxUUFfkQQvLoozEdjnE5oG8G+Bth2EAzF8xREW4FqwU+3QRb6xVsw8G3SoEgeqBVWRDdk28BIkBEkzlG47eZYArDqxt0MeuMArsK1XXgOdrD2YLURLjjakiKP7ZrTk05tv0NDAy6D611nWDQMW+0trKzAmsu4ENv8HwVEAk83DV2GRjobA6FWOzUyD7LjKnCSlTVUMyhAMsKXazaFMN/LBZ+GQc3DpJU2YJEYSKqFRegSURgUlJIU9KhEzEF9Y3g8+uV1I8VKWHZFoiLhJwOavSNGye48kp47jmFyMjO1UZQVVj+tcYnnwXJTtYYe5adxRtgbC5MGggEwbsHtu3SKK8NoT/3VAFeQAOTgJhIyuoFZi/kJkG0BVYvCTI8Poi7t41nPlK459L27RACeh9HmzJ9Bs/AwKCnY8NMLsYPtjWEEHHAzUAGzXSTlPLGptf/bW1cpwSWlLL5bNXzx22lgUEbHKyF27dL1jTCACTTRwuC9S5WlMBXdUAYfNXw532wvz5A7OQ63CjcjAfbkSJKOMFxe6fPvXs/1NTB2cOP3W5fAD5ZCxkJHQssjwceesjOVVdZSE3tXGxYTQ2UFkt89QH8vhCb19uINAtG5+hZj72sYC+Fi2cIFi4MsnNnEAgDIUCFsB1TjiDpHJX1JVDQAH99KUTpQni7PsTk8V5u+5WTD4vhnFhQwmBVO64GD1BcCku+hfEj9X6IBgYGpy8+gmyhrLvN6Km8CywFPke/wXaKjgqN1tF68FanGh0aGHQGbwB+vwT2+U1YBGxrVOjzLQwYWE++pRhbdDyhsEqoUqFom41NNpigQUCRtB5efmwMztZnoo4HuxXumAmODpotH0JVBVlZnY8JKymBh+5V8PuspKTYefNNQYxdb9IMYLHCqJGCyy81cfvtDubO9XPgQAyaZiOsmUnJtDAuz8zPLhZs9oGvPoRzcy1UqdgdYUaOMNOQDAsOQowCHy+CIRlwweiObVu9CT76Um80PfcIgdXYqM962Y1ehgYGpwUSQdjIImwLx5HtAjtDR3WwOojMMDA4PoLsI4ikhgTyC+w0Nggi7QoiDAkO6BOGqJCftF4Hqai34LdZce+ppzjYhykpZqYoThJRcXTRDaGzxTdbI+E4XIvNCYUk//hHmP79BVOnHp46qq2FZ5/VSzb84hd6ut9dd7W0NTkFJk0TZGWB1aoyaVIMdXUa/QfGs3RNkB9dL8nMhL5pGkOdgieeaGTWZMl1z1uJi1NITVWpDEJ2AyRLSI+C2hL93G43BIMSkwmUVgLfxg3Tq8APzTv6mj5dCGYVZs44se/meAiHobAUUuMOt88Jh+GLL6BvX8jMPPU2GRj0dOyo9OcYgyy/P3wghDhfSrngWAYZ3bsMTjmBiq28Wf8bKtNs7JLD2LpnKgv3pCFNTq7MURjhAJ8XrF4nt3gy+IvFirehisbaeG47D+6eoGDhzJkaURRIStJdiNu2BVi/PsgFFziIiBBcfDFENHvMObLfXmERLFoOA7IhIQ6GDFFwOBTGTgBLlJURwyUP/NyLzxdi2DCYNcvG2LEWFi70sX59gKgoE717O4jKtHDXc2FysxR6R5koyoGNGxv4wQ8OMnmyk1/+LgmnFdyOw+eO8sCUCa1f08QJ3dfbcMs+ePp9uGkWDO6rr2togCVL9D6RhsAyMDgaLyE2U97dZvRU7gR+IYTwo6cTdcqLZwgsg1NOzWMPMmzXYt7946X0jl/JKlc6M6ZsZnHZ2YQa47ErgoMNkDVgNRnW7SywzEKJyGGdE4a79V56ZxKKIrj0Uv2n+OMf17LgYz+eGDPnTbEwatTR+/t8+uLx6DFTsydCYjzs2AG7dsGMGZCVCalJsKdQMHSoYOdOyX//G+DZlwU33Grl0w/9bNkSxOGQTJxq5ZUPwoSKJXu/aWDXThdut8KSEgiHJaGQZNFmSI+D8TmduyaPp+u+n2OlVyLMPQsymrkt3W64+25wubrPLgODnozhImyb4/XmGQLL4JRj7pVNeMtihr7zJZ45gqUV5zAp/ls2ukaTVAP5ZdA/Bcb0isJLKg4RgVmFCVHgx4vEijgNbgQ1NUH27/eRm+vCZOqcD/KyyxwcKFeJT2j7p/ncc7B/P9x5J7z2mj5TNGI4xMRIBg4UqCp4ffDx59AvCx661055hZmBI4IU76vnV/eHuOF2OzEJCpOnOdhRbMe5NkRNlcBukzzwYIjRYxSuvcrJv1/L5MOvYFSKRmavo79zr7fnxVm5nTBl5NHro40CqQYGbWJHJY9j7HT/PUIIEQVkobcLBEBKuaS9MYbAMjjlOK6+n5L/vIr7F9tIKujDq1Pfwx9O50rlBYr6X4mrpoy00UMQSjaUK+AOgwVCBNnCctLpR8yhCvBNaFr3uaTaYuHCChYurODuuzPo27dzjQ7Hj7cyfnz7EfNpaXpQvssF11wDK1bAL38pWbchwLixknXrbcTGQrRbFxsACJWzp7h49+VqTBYvt92UxKsLI2iMhNsmwU+vM3HPrwOsW2/jib8HePZZL6lJbryaiTdfbGDbQslbb0XSvAjZpq3w0htwxYUwuJU4LAMDg9MHLyE2UNndZvRIhBA3obsJU4F16J1tvgYmtzeuh/1JMvg+YIn0kHfr/UQOy8Mz9iqEcxTO5AvwRLjpbyogfdNb+Ao+Q9v1Dnz7Z9j1EQAqZvowmEjivjtWGMmH6zTG3Ac/ewUCoe66qqMZM8bDvHkJpKbaOt4ZqKgIU1racQbwzJlw2236zFFuLuTkgDsSHA7IP6BwzkSIjgXhAG9TduTKLTBjtpX3Pozilgfi+fMfYMl/Jef0h70VsLNM4X8fstG/F6iqhln18dprfrzVXhoK89m4sYI1a1rmbFosYDbTorfiIcKavhgYnEQwvQAAIABJREFUGJweSPRehF2xnIHcCYwE8qWUk4ChQHVHg4wZLINuIe7mHxJz3Y1889e/YqvxMPycqcBUCIepSY5iXfgpUsq8WPunkRzR77tyou4jprAXUsXnkX4O1iWyplBQXKdnwp0IGpJSvMRhw3QCN4vkZBvJyZ0TV8Gg5IknagkE4Oc/jyQiovPnHT0aRo8WVFVbeeo5yMvTY7ImjobEWAiFoLYBauoFF59rxe6BL0OSgF+SGCd442soLQ9iLivj7jtc3H2HlU3bolA0E6NGqVx/fTTr1ztYtEhpEROW3Qd+e1/rM4cvLNEbSV9zdqcvw8DAoBtxoDKQjrtLfE/xSSl9QgiEEFYp5TYhRIcRqYbAMugWhBCYrFaShg5FdTRLTTOZCEZH0BiKptYVjcMei0Zmmy2bM7FxWYbKFT8XBIHUpsbJmgYHqyDODZbOFU3/jjJ8LKCQKaSQzqmJijaZoG9fMz6fxGY7tpoRFRXw7ruSs8+G++86PPbC8/Sg9988LPnBDYJevXQxNDwXkqMFGakCRYFJ2UGW1XnZUAE+X4C8oW52VVm56nxw2uHRR1PZtAm2bg3w/vsaM2davyvb0JZbNi8N1DMsGcHA4EymkTDrqepuM3oq+4UQHuAd4DMhRBWQ39EgQ2AZdBuF+EmYMgnnEfIpmnRGqzfhVGMwYUbFRgVevmY/E0gnksMxSlk4yBJA75bH/r/58NxnMK4f/ONHYOtkIVCAWGxMI4UEHB3v3EUoiuCKK45PzO3YofHoo3UsWWLh2WdtiGaFsgoLA6xc4WX2TCeZmfrP/Yuv4fPlcPcNkBwv+fbzChwWwT8fiSYhwYLJBOm94cNNMGuQXkR14ED46KMA9fWSqVOt2GyHzh3kscequeYaBxMnHo4zG9mnpY3VPj1uLKqHBcQbGBgc5gx1750wUsp5TW9/I4RYhN4u8OOOxhkCy6BbCCNZTA3DcNKflgHgCirRZBw15liKrTd4oaERDpZBRS2kxHU85hAmBKmnaOaqK8jNhdxcjT59jg56CgbB57USahabNmyALjgTY8FsFsyYEYHVKlrEim0ug73lMCIDEjywvQomTzOzdqWP/y4NYo+zcPFg+OX/1PL2x5K6hvoWAqs5UsKTayCkwQPjdNehQeuUlzeybFkB553XF5vNuD0bnDrsqAzCSLVtjhDCLaWsFUI0/2I2Nr26oP2sAOMXbNAtmBBcRCzWznRv37uSGHcis2OyOn38X12pt3yJckHSGR5WEBWl8PbbHj5eDlv3Cvo3m80bP97Cww9L9joFH30MD0yC5AQ44IWPt8HsgTB2rBMp9Vgtny+M3a5wxShBfgmU75NsUSVvlgu8q30se6KKlL4+Lv91AlJCTn8Ladsl11zb9q1ECBiTDEHNEFcd0dgYpKiojmAwbAgsg1NKI2HWUdPdZvQ0XgFmAavRn/Gb/8GSHOU7aYnxCzboNuydnY4uWA8J9RCT1uljqypc2BRgLSU8vRY8Ibh4eM8r59Ae5bXw+lIYnAnj+x9ev/cgfL4a4jyQlwKFBwRrtumzRKmxeo/Cej/kl4IjUXD99QHCxRqrzzHz0XMmthbDvgqY3h+sCrzzDixY4GPr1iLmXuUk+eo6Cr9K4bePB/HtDGCLDTDxkTjOvVwjQjMzK1P/Hm+82kW1X8Mn2g+4mpRxUr+mM4b09Ehuu62VIl4GBqcArTMPvN8jpJSzml6Pq/+DIbAMej4TbjwhVfRtTYDfLZPYy6wMi4O+GV1n2smmqAI25eu99A4JrGAIXvxM7wO4uwg2b4ZgFcyeAR9+Avffq+F0SS69IsjXa73s2yUJ7wlDZYCl72vs3h3H5SNsBEL6MUBvxxMdDRERgpBQeOo9C8vnmwhLC2KIxLdZsmN+mNyhYXaVmnhpPpwzBOITBDExJhxHJEv6QvCfrZAXA2e3LFlmYGDQA3FgYgjd2IKhByKEGNbedinlmva2GwLLoOdzglNOUU7J7Exw2SGhhxUqlhJef1ePlbrqoqMvtV8a3DodkqNbjglrejNlIWDEEBidAwdL4fNvQmyrEyhFIS6phbz4MJuWCagREJKkxjWwZImNeRfZ8DTrojVlCkyZYkPKDP74H0HhaxAu0EABWW+HEVb8tQpP/bUcTYYp39OLPz9iIjY6wG9/52bmpJZPvpqE2qAutAwMDHo+DYRZQ213m9HT+GPTqw0YAaxHdxMOAlYBY9sbbAgsg56BvxQssSC63n+XY7by95kd71ddB05b64UzTxZSQkGRHv8UCunFO5tjVmHIEV5+ixkumwgLvoUBGTBhEPoMkgzjMtdjc9nJ7qVw2y0qVquVwcNCrPomQHSUwqRJkew8aOP3z8Dtl0N6kn7M/PzDM1jDB8AfkuGHDytUq+BJURkQL0iu0fimRKW+3oa/USUUkDTUBThQcvR1Oczw86EtCr8bGBj0aISRRXgETUVFEUK8BQyTUm5s+pwH/Kaj8YbAMuh+fEVQ8CQkXwWu3G4xYc9+ePotyOsDV3dCjHUVigJ3XK/P+BwprgA0TSIELUovAPTP0JfmeDwm/v6oi8/WmMjNELjdsG4LLFlv5oc3mumfpWf5hTfAgQq+c+tVVMATT8Dw4XDppXDWMMmHCzRG9A1TGzTzu/sUAmFYvtnErx7szf4CicOpYDKFyeobQ22txsGDkJTUMg7LEFcGBqcPDkwMxd3xjt9Pcg6JKwAp5SYhRL+OBhkCy6D7scRD4iVg73XKTvnCWnjyS/jlEDh/ku5qU4T+eqpxtFFuq7FR8vjjQaQM88ADVmy2o58u/U0uOKsKUkrS00zcmnn4InqnwcXToVfK4TFjBunLIaKiYM4c6NX09T/6aD2/+U01UkpwOcm/OJqIPEFjKkSnKORm6/v5fCZ+dHcVL/y7jKRElQ0bMoiKOiyywmG9dZH9GGqQGRgYdA+NhFlNfXeb0VPZIIT4N/BS0+ergA0dDTLmAw26H0UF9yAwnboqlMtqYJ8HdtfpnzNT4P4b4LJpJ+d8mgbrdoI/oH+WnSjqFQhIli71s2hRgOXL/a3us7UctlXo7xctauCPf6zA7z9cD8sdAWePAqcDCouhvJVCzYoCZ50Fqan6Z7s9hJRBwAeNjXz2QZh1rweZFR0mvlmpq/V7JYsOWAm7orBYNMzmlur0vZXwf+9AY+umnxAFBVBp9KU1MOgyJAINU5csZyA3AJvRexLeCWxpWtcuxgyWwWmDbGxEtDXd08SH+2BANGS0MtNd1gBby2BwIjw8Fi7vC+OTDm93n8Taov6gXloh0gb/+7sQW3aEuedOMxfObfsZx+NRSEz04fVqxMe33tNwYPzh92lpZnw+vXhoaanGV18FmTzZjMejn+NAKXgiILaDXo133x3Jxo11rFsXZvLkaPLyYP16iTji4TYlTnDNPBv2ySreSo2XX67i1lsPFx3LiIewBMtJuMu4XHxXTd7AwODEcaIw/DQqsHwqaepD+E9ggZRye2fHGQLL4LRA219IeP7rqFdeh4hruyx7pBWsbTxAvb8dVuyHKi/M7QcJqSfJ2FawW2He2XpJhTffEtQ0wjtv1XPh3PZjHi680IHbLcjLa93PZmqmz7KyrGRl6fvt3Rviyy9D5OSYvhNYh9yCBw+GcTgEkZGtiztVVbjookj69fNz22027HaFyZMV0tNbzlAt/PAg7oogsQOTeORPJnKtddx0UzQmk77f0N76cgivVzJ/fpiJE02kpp6YLzb6FBac3rXLy9KldVxxRWyrbloDgzOBBjRW0tDdZvRIhBBzgP8DLECmEGII8LCUck574wyBZXBaIOITME2ZrgcMtUPzGakjGZqoi6sB8W3vc7LJzoapM8KsXhXgwnkd//wuuqhl+xlNg7oGiIxof9zw4SrJyQopKS0Fwe7dAebOLaawUOXxx23cequuVAJB8AXB3TRBOHduJLNny++aOvfuLfD7Nf71ryISEy3MmxdHIKARCmnU1JtIyfAQY4rkmRc1LpytEBtztICSEnw+PXCf06igYSgk8flkp9y6BganK4dchAat8mtgFPAlgJRynRCiw+KjJ01gCSFswBLA2nSeN6WUvxZ6OtQjwCVAGHhSSvnXk2WHwZmBsFgQA/JO6BhDk/XlRCgp04VC4nGKNLMZXnnGQjBoxm4/dpFRUAzLN8BlU8HUzr1QVQVpaUfvkJBgwulUqa3V+Pe/a74TWOv26lXfLxl/eF8hYNs2Hx6PicREM4GARnFx4DuhceONKQSDGrc/rPDVZ4IlBUFe+E8D997TwP/8ysItt3iw2w/b4HAIbrrp9Humy811kJt76hp/Gxh0B04URtB6P9GuRgiRBrwAJKC3nHlaSvmXpp5/rwMZwD7gUillVZNu+AtwPtAIXH+oyKcQ4jrgoaZDPyKlfL5p/XDgOcAOLADulFLKts7RgclBKWXNEdncHT5yncy7nR+YLKWsF0KYga+EEB8B/YA0IFdKqQkhunE+wcDg2PjPqxAKwy/vPv5jqKougNpjx44QmzeHmT7d0kKIpSdCVET74qo9XC4TixcncN99RVx44WE/25BMyE2F0moorYG8XlBWFuKpp8qIjreQNiqBC8eq3HNPGhaLbk9BQYjnnqslzeXCXKXiD2sQrqfOH+LBB8tISzPTe4Cb7HRwnLr8BQMDg+OgAY1v8Z6q04WAe6SUa4QQEcBqIcRnwPXAQinlo0KI+4H7gfuAGUBW0zIaeBIY3SSWfo1eBFQ2Hee9JsH0JP+fvTsPj6M6E/3/rareu9WttbXvsrXaeF8wFhgb7LCaPUBCSICwZCCQkJmEmUlmfvfemczMTWbInUx2CMwACQkQlmCWYIMwxjbY2Mb7Lsm29q3VavVa5/dHCWzZWq2WZcvn8zz12K6urjrVskqvznnPe+AeYANGgLUCWNV3zoGuMZQdiqLcBmiKokwBHgLWDXeT45ZQIAyfpcWa+zYB3I8xdqn3Hdc8Xm2QpHi7dgVc94Xxv866dRGee66HNWv650So6tDDg00EODZMHoXNpvH447lcfLHn830WszE8uPUQvLrRCCKTk01ceaWHJUvc2MzGoJ7bbcJmM6I7TQOrVWFqQYzvPRzmnnuCWO0qTreVG25IoLzKwavvGb1uA9neAC9sM9ZMlM4eQo6FnrcEaly2Ya8jRMNnPVBCiG5gF5ANXAs81XfYU8DKvr9fCzzdF1esBxIVRckElgNvCyHa+4Kqt4EVfa+5hRDrhfEf+umTzjXQNYbyIFCJ0XH0LNCFMZtwSOPaX68oioaxCnUJ8FMhxAZFUYqBWxRFuQ5oAR4SQuwbz3ZIUrxUTD1TV/KzZUsTP/qRg+pqOy7X8S6r+vowzz/fzhVXeCgv7981FCJGbAxXvWQaLCgFk2bke/n9Oikpgi8uOPXYvDwzjz2WzJ/+FMbt1nn44URyCowfzt99VMVigW9mQcIgow5v74XdTVCaBlVD5M5JZ040qvOLXxxg8eI0pk+X69KdT5yozCNuXc2piqJ8fMK/fymE+OVAByqKUgDMxOhpShdCNPS91IgxhAhG8FV/wtuO9O0bav+RAfYzxDWGUtG3mfq2a4FrMJbMGdS4BlhCiBgwQ1GUROClvvLyViAohJijKMr1wBPA4pPfqyjK14GvA+Tl5Y1nM6VJqhsdFwrKKBKqwxGjN2eiVVbaKC21k53tPKWKe2trhNraCA0NkVMCrDyGyX4fhtlkbACRiGDHjl6EgJkzjRykQEDn9df9zJplpajImLG4cqUFv1+nsTHKvXdpuFzq51Xphyp9cf00OJgNJWfZ+pDnM01TmDEjkYwMWQPjfONH8CFx605uFULMGe4gRVFcwAvAw0II34nPur58qXHtTh3FNZ4BHgW2A/owx37ujGScCiE6FUVZgzEGegR4se+ll4AnB3nPL4FfAsyZM0f2WUuj0k6M/6Gbq3BSxMgiph118EwN3HRh//ICE2Hu3AR+9zsnkYjA6ezf5X7BBQ6++10zXu/4RoJWq8q3v53er4BoW1uM9et7sdmUzwMsn0/wwx9GWLcuyP33m7jllpElyuYnG5t09lAUhUWLBi+DIk1uujhzZUj6crNfAJ4RQnwWEzQpipIphGjoG+b7LIXoKEbu9mdy+vYdBS45af+7fftzBjh+qGsMpUUI8eqIb67PeM4iTMPIvO9UFMUOXAb8C/AnYAlwCLgY2DtebZDOXx5ULsdB5iimHQthbPpZEs6fHFgBHDkSw2xWyMoaYOHCcWCzqUQigqNHI2Rnm8jNNfOtbyWTknL8cxUCLBaFsjILc+cO/EgJhYzjZHFQSTo7OVFZQHy+QX8+zOt9swJ/A+wSQvz4hJdeAb4C/LDvz5dP2P9XiqL8DiPJvasvQHoT+CdFUT6r33M58D0hRLuiKD5FURZgDD3eAfy/Ya4xlB/0LZXzDhzv5jshMBzQePZgZQJP9eVhqcDzQojXFEVZCzyjKMojgB+4exzbIJ2nNBTKGF0QUpUPP7gFbGcmdjktPh9YrYL0kWQNxEFrq87Pf97Jyy83c++9adx9dwpZWf17zjwehUcfNaOqZhyOgYdjt2wx1ia88MIz0WpJkkbLj+BDPXymLrcI+DLwqaIoW/r2PYYR9DyvKMpdQC1wc99rr2OUaNiPUabhqwB9gdT/Aj7qO+7/E0J8tojWAxwv07Cqb2OIawzlq0AZxmS9z4YIBcdH4wY0bgGWEGIbRuLayfs7gSvH67qSNBZnU3DV0gJOZ//FoCsqzmwhwJdfjvLGG4JDh8KsXt3F3XenDHicyzV0ntvMU54EkiSdVQTE9DMzRCiEWMvg1YaXDnC8AL4xyLmewMjlPnn/x8ApxROFEG0DXWMYc4UQpaN8j6zkLklnK4+Hz5PFJ8qSJRpZWQk0NmZSWXn6M4wGu49wGPbuh6klE3+vknQ+c6KyUBl4Sa7R+nVcznJWWacoSoUQYudo3iQDLEk6S50NAUdRkUZRkcZ777l56qkI+/aFuPlmy+eJ7wcPhti9O0R1tbNfKYmRqq2DZ5+Hu+6A4gmeWCBJ5zM/gg+ikYluxtlqAbBFUZRDGDlYCkbH2sSVaZCkyUzX4VAdZKZP7krl0ajgjju6qKuL8pvfuqis1Jgxw4Q/Ai+/6mPn9iBpaRpz545+mY3iIrj/bsge4xJGkiSNkQA9JhczH8SK03mTDLAk6TQdrofHn4QrlsCKSwY+JqLDS41gVeHqdFDPnTWOP2cyKfh8vUAUXbdjs6mEYvCLA1Bc7SYvx8LUqac3+0hVITdn+OMAfv1H49fGu248vq+pBWqPwZxpxrlOV5cPYjokn1BbMxYT7N0bITNTIzFRLoIrTW4uRWGRFp/SL7+Ny1nOHkKI2tN5nwywJOk0ZaXDNcugaojUx7YIbOwCswJLU8F1jn7H3X+/g9deC/LwIw7KyoxI5upsyLZbcc6MT97GZxqbjfUaU/tqZDU0w+9+D3WtUDWt/7Gbd8Daj6G0EDzu07/mL56DYAgee8C4NsCuXRF+9rNupk83c++9Yzi5JJ0D/Lrg/fBY1oGQTnaOPu4lafw10MZmdlFBEYWcOoZls8GyU9Yg6C/dArdnGT1Y52pwBfC3f5uM1+ujrTXAoUMqhYVmpo6taPyg/AGw9uWfHT4G//Es9ByDSxfArTf2P3bJQphVOXxwFQ4LGhshN5dTKuMDzJ8BofDx4AogI0OjrMxEVdVZkAwnSeNMoKCfoVmE54tz+JEvSePLh58AQTrwDRhgAXzaCuEYWKwQEjArsf8woKLATM+Abx03ra3Q2gZTp4xt2OxEL74Y4Ic/7CI5WcPl0rj//tENJXR1G2sSjqQ9JQXH/+52woxSWHgDFOeeeqzNCrYRFB7/5BPBSy/pPPywRkbGqa9fPN/4UwhBV5dOYqKG06mx8joP2VlQf8RYQmig90rSZOBSFBab4xMS/E9cznLukwGWJA2imBwScZHEqd0jUSF4Ixzifw4qhLuiZHjDOJQkvFbIcwxwsjPo9y/A7j3wyIOQnydobIyQmWlGHUMCWH6+Rnm5GZNJofaYlU93wrSKkb23tQP+3zNw03KomjK66yZ74NbL4YknoK4ENDfUNcL1S6G1GV58EW6/HTKHWSy6slLB4dBIGyYYO3AgwtNPd3Pbl9y8+Bczxw7D0TCs+iCGIxJl9dMKM6bLHi1p8vHrgprQWbKMxSQh+wOl85afGFsIEBhk7U4TGumkYBlgLcN2dN7Xgqx3BNnQAymew5hDOlv1bo5xxqohD2jeHFg4D9K9sGFDD//yL01s2RIY1Tl6e3UefdTPj3/cQyQiqK528Pzz6TzzbCbuJCuhUawJm5gAV10M+ac5U1DX4ehRePJJeO4l2LQT6hqMGlrd3UaF+OG4XArTpilo2tBBZlaWiauuctAV0tjdAHlFsGkNhI4qdHQKbr0nRlvb6d2HJJ3dFIjFaZMA2YMlnad0fNTyEqsoAmYzg9F1O6Wh8lXVSUBXCWaGyTgwlY8OxfiVP4x/SYhbSR3ReXoi0B2GjNFXOBjU3NnGBpCTY6Gy0kZGxuiG9P7xH4P85Cc6qalRLrrIxLx5VtLSjMfFY98aXXtMJpg7bfjjBmO3wx13wL/+KyysgrxiKMkzhgd/8IP4DYMCOBwq8+bZCUfg7+6GBCv85SWIdEOHrqFaFPw9kDJwQXtJOncJQJZpiCsZYEnnnSgR9vM+gm1cTQKFjG4WnI5AV3TSomZmZcBcl53dn0CyplPqsHHRCL+tdAG/2AotvfDVKpiaNPx7Risvz8K9944gSekkUauKww35+cbw4ETr6ACzGdw2WLbw+P54BlfRKASC4HaBxQxT840Fqv/uB7C2QcUSU7lmNuTnje68dXVBNm3qYt48D9nZk2u1a79fsG2bYNYsBZtN9lycy1wqVMfpa/hMXM5y7pMBlnTeCdJNC90EyKOAXGyMPO8gQpT1bKWbHqyRcv7clUptBP5+gUJ1pUqq2zmqcXet7+Cz7ffG2VdYufQKM5dMU2hpCbN/f4SSkomrpjp7NiQmGrMAx8uajbD2E/jOV8HV16GpKJBVDkcCMDcDphaP/rzPP99EW1uY+voQDz00yujsLFdbCy+/rJOZqVFYONGtkcbCr0PN6DIJpGHIAEs67zhJopi5fMrHHOEoGZTgYWTdR9346cSHFSt2WyOKNZUXgjrWOpVH8pRR9aioCnx9GvgjkDbBifEnu2mOgqZqdHbqXHllG3V1On/4QyrLl9tZtxM6/fCFuUYAciaYTFA66qVWR6e8yOglc5zUyVSUCA/OPP1h3NJSB+vWhZky5Sz7IsdBeTk88ohGevpEt0QaMwHIMlhxJQMs6byjoJBFMQKVMGESBpglOBg3CXhJppNupqjZVCQEWR/q5pedTsp3OVhZNbqIw242trONqW9UsLtb0NlpJRAQvP12jOXLYe9RaOyA5bNBO2H0MBYzhtmsfSOuEXR20E0xThJG+Kg5cMAoMzF//ujbXFcX4+mne7nzTjs5OaMf1szJMLaTqQqUjyHn6qqrUlmyJAnXGS6EputiTDNHR0JVFVm6YpJwqVAdp98B5BChQQZY0nkrm4HHNIJ00sQ20qjAcVKyugmN+Vzw+b9vcAd5xy/oaFf54CBcXd4/6DjX5eaq/OxnLt59N8p3vmMMEd6+xFhW5rP71HWIROG/fgsvvgr33wG33QT1hFhFGzdjGnGA9dHHcPAgzJkz+s/RaoWMDBWr9ezKBVIU5YwGV0IIVq06zAcfHKW8PJlbbinFbJ5E/ymlceHXocY/0a2YXGSAJUkn6aKeRragYT0lwDrZdM3GHxKCrCv6lLLCFDTtNJJ0zhKxGHR0Q+oJ6/EpisLVV9u4+mrj3+EYfNQKxW7oOAKvrtf51NTOJzss7Pi5gOYATXVujjY72GxSMM3zsOaonYQLwZs8fBuuWwmRyOkFqenpGnffPfmG4UaroyPIBx8cJSfHxY4dbdTW+igpGYcZFNLkIocI404GWJJ0kmRKUFBpZBdWkuiIFvF+JMZ1VjPJAyRZdap1eGzNNNFKOYWoZ13K+si8swne/AjuvQamDpJMvrkNHt8JV+dC8zp4fr3gWNRJh0ODaTpsVNm3vY2/frgLFBOmexLY1K6Q7IDrLx2+DVbr8SFG6fTY7SacTjNHjvgxmVTcbvmBSsNzaVDtis+55BChQQZYknQSM3ZSKSdELx/h493YYV4MW/lAWPiVw4tG/yGobPLw48NL5jkbXAGkJUJ6EiQMMlkwFIM3W8FkN2pQtXrgU5eG1mwmrAjotoBZA00xCkcRJPqsRoPDxBs5JtIcCosXnNFbOi/Z7WbuuWcae/d2kJubgNcre/Wk4fljUNM10a2YXGSAJUkDUFHJZQ6HqGexSafG3Mqnipl9kXpKTSUoSiL0bKB+16u0B7MpWTQNm3JuL6Eyc6qxDUZRwKpBjgvS7LB0FjxZq9OUreNU/QT2OxGqGSIJQAegQ2eMqRVwtGty5aad7dLSHKSdbVNTpbObgEEWtZBOkwywJGkI1eTSpERoE36iOvxTpJ4nHb1o6lzw/Q6zdS9K62p276oio2Im+dyFyrkdaA3GosJ9hdAWhgKnUSg1Yyo0tyv0djgQEYzKnA4nWtjFvHlebr/dRsE0jT3HFObMmOg7kCRpMC4Nqkc+oXpIcojQIAMsSRqGC5VbzH/GpwsqKETVZgMmOsyCT3NUep02XvjYRW5RjO+ba7Boyya6yeMm0WJsYIwE/lW+yt8eELSuVaFRASvQo1JQmMzSZU6CvUHqtge45RoHCBNtbfFbZqa52UjMH26hZ0mShuePQU3nRLdicpEBliQNw6lqfNt8K1ECpCiFKH09VL2uXOyxrbzhvZidM2dyOGqlzbSZTCZvgHWyuysgtlPjxwEIeaH5sMCRLPj+3zsJh4I89lg3ZjM0N4eZOz+F9z8QPPbXCgkJYy+l8OsnIBSEv/87oxCpJEljIGceBxP2AAAgAElEQVQRxp18LEnSCCQpp06r85qvZbtmpVnNwpadyX2WECnanWe8bRMhEoU/boSuAFw8B0QUygtg/fsCt0vhi7eY+Y//CKHrEXRd59AhE0ePBunqDvH22xrXX5/Q73w9UYgISBxF0dXll0M4PH7BlUBHOYcnLUjSaLg0qE4c/riRkEOEBhlgSdJI6a2gt4E2FRQFk1bCYu1eShr/QnbjWqyld4ApFaKtoFhAi1NCw0BN0WHNGpgyBfLitLydrkN7O6QOXfoLgGYfbDkMER3mFcN9XzT2zyg3lguyWODOO63s3CkIBmNkZTl4+uk2IpEoWRnOfgFWZwR+Wge9OtyVDYUjzM2ePWv09zgSkRiEdR3F7MOOB4Wzq3CpJI0Hfwxq2ie6FZOLDLAkaaRCL0FsFzi+C4oXADtmiloPE+0+SChUh9UchY7/AsUKyd8G1VjYThAlzG7MFKMy9kWTIxHYuNGY2TfWACsQiNLcHGPHDoUtW8zcf79C8jBFQdM9MLfY6MEqOWEdOo/neDDi9Zr59a+zAXjiiQ5SUiK0tRn7T+SLQnfM6MFqjdCvvv5RH7T3Qn4CuOM0KS4QgJ5eSBskF0xVQFNUrDK4ks4ncogw7mSAJUkjZV4KWhko/X8yB6dex/aAgs+yn1Q9Qq7WhUkk4FTE5wNMOl30shqVBFTyx9wUqxW+9S1jceKxqq8P84//2ILdbuLLX/aSlDT8SU0a3DiC9QIDIYUEB9x0k4epU401DRcv7h9g5trgOq/xG/T0E0YOhYBfbYYP9kG+AhdmC25cBG732IKeP7wKazfAI/dBcQFYT5r0qanGRhyDq2BQR9MUzGYZsElnJ5cJquNU8F8OERpkgCVJI2UqAopO2b1W8/M0eXSuD/NReyoZF93O7QkH+ZLSRAYFAGik4OZuVPp3w8SIoqKeVq5PvCqev/66zpo1vXg8zfzoR16UOMUA9c3wXy/DjdUwu1TjkkucAx6nKDB/gNwPRYGqVNi4Cz75RPDsbsHzSYJVf1JRTrORdXVQsxrq2+DRH8LtX4TbLz+tU41YNCr493/vJC/PxO23j9+wsSSNhT8KNW0T3YrJRWZwStJY6UFS1S4+ODyLxoY8du4o5e3mUmz0DyhODq4ihNjMKg7yyZls7SlKS02UlVm5/vpU3O74/c6V6IKZUyBzFGUZWnrg8AlTxW+ohHtmAbpCugpzZnHawRUYpR30KIgwTMkffEmgeNI0mD3bSkXF5KyPJk0Snw0RxmOTANmDJUljNtuchludyq48B7Wh/bjTzNybdpCQ3k5IFFDXU0ieU8V6UiVzFQ0nidhIGPjEZ8gVV9jIzHRz8KDC3r1h6usjLFnixGQaW1dWggNuvmR073l2OzR0w2OLwW01erGumw9LysDXrZKXM7Lz6Dp0d4PH03//nDnwg+/Dp7ug+kLwnIEOJUVRuPzygXvvJOls4TJBdZxq1MkhQoMMsCRpjJJIZ56WzuOXbuCY2I9bs6AToie6jtpgiN/WX0tO+BIem9k/t0nDRAWLJ6jV/amqEUy9/36APXvCzJhhIy3tzD8eLi2A1l5wndTZk+gxtpFas0bw+OMRvvc9WLjw+Ml83fCbZ2HmtJEHV8GgznvvBZkyxURRkeyFkiYnfxRqmie6FZOLDLAkKU4K1Ol4ySfo8+By91CvOYmYYiQ52rAJHxCnXw/HwQUXJHPBBcm0t8fo7IxNSHAFUJoCxTFjJt9YFBQIUlIiRCL978PpgEXzoKhg5OeqrY3xwgsBZs2ycN99MsCSJim5FmHcyQBLkuLgUC2EwnY8yXZ+txoun21Hd8zHnvEBt+dl8MdVHq5dDd+/HmaPcJhrIiQnayQnT9yqzL95D1q74ZvLwT2GahbFxSq/+Y3j8565z2garFg6unOlpWmoqpPMzIldrfqjj0Ls2BHl+uvtuN0yfVaKL5cJqtPicy45RGiQAZYkjVEwCL9+xqgq/sCd8JdX4MVn4KJLsvju3Tfws3dUfvCfAmw6rxzT+fu/OcDXkjIpQM4oO5nLCt3Bz8okjM3JwdXpstkUZs82U1g4fiUWXn4NbFZYftnxfe0dYLeBvS/Q3LQpwvbtERYtssgAS4o7fxRqGie6FZOLDLAkaYysVpg/C3oC4E4ANQrJHqieDQkOlYMHAbuAMCi53bSHOomSMdHNPit9aZFR/0odZfzQ3mkEKI6x13A9hcOhcM894/uo7A0Yifk9PUZifyQKP/4FlBTBnTcbx9x8s4NLLolRXCwf29I4kEOEcSe/UyVpjBQFrllhVFd/7jn41+8bS9iYTLD5MNiSQc2LYLf5KC+O8M9eH3Z9G6hnR4L72URROK06XH94HQpyYHl1/Ns0HiIRWLUKpkwRlJTAFSsEv/xlgC/8SGHPbjNz5uhcfYONipLj70lOVklO7h959vTAnn0wtQRcrjN8E3EQiei8/norCxcm4vVOfH6bzwddXZB7Bsp3nG1cZqhOH/64kZBDhAYZYElSnGgaZGVBSsrxBYgPt0CEGAsWbSVx/iHerV/KK8EObrQexBQIgHUmaN6Jbfg5KKbDu/shHIOlU+CGFUYPFsDeI/DWWlhYDrMvODPt0QU0hyDZApYR9L75/bBuneBPf+plxowYV11l5/XXI6xfL4gIM6+vAosd7vva0Of5aDM89SzceiNcPsrcsrOBpil4vRbs9rNjyPOVV2DHDvje98B9no3g+yNQc2yiWzG5nB3/qyVpElBVWLYMvCfES5dVweVVOr42D/u2VKDbFD4MLMGifwECf4bwrolr8Bh0R+FgrxFYDCQag+au8bv+wTZ4cw+8sxf2NENqMrj6Sk2t2gj/9t/wv39sDDeeCe+1wo/3wbN1Izs+KQluvFGQmKijKAr5+Sr/8A9O5l/mwFmg4sm0kJgiEMPcQEUZfGEZTKsc+HUhBE8/HeXJJ6Pog32xJpCqKixcmEhCwtnxu351NVx7LezdG6SuLjLRzTmzPhsijMcmAbIHS5LGVYIdLq80s6O1hNfDMZalxPjnFCsKKeB5BExZE93E0/JcM+wMwN2ZUDVADc1Vm+HX78GllfDNL8T/+qlOSLQbgVzaSUNjX14Gmg/yM09vuPF0BKIQFUbgORLd3YKXX9YBG0uWqGiaSnW1hRenwfsboK5W4GtuZPVqG0uXDr5AnDcNbr5h8OsIAe3tAl0/c8HmuSwvD2w2wZIlvQSDvbzxRjJTptgmullnhMsM1ZnxOZccIjTIAEuSxpnTBn+3XOPvOGmav7lwYhoUB7k2aIhA0iBPkNYg7G+HxBH26IxWkgO+fYkRNNhOWps62Q1/def4XHcwy7xQ6ITsESbZW62QmqqgaZCcfDwKTEuC61cY97V6tY38/LH9cFdVhQceML5ImiYXmh6JhASFK66w0doaJTX1/PkR6Y9AzZGJbsXkcv7875Gkc0VPG9g8oI3t23PNmggbNkS5/34rHk98swG+kAwrkgbvIVo5C8qyYFZeXC/bj7Xv46kPQk07LE2BDCu8dxQaA3DLlPG79ilt0aBiFDk7FovCAw+oRCLGLMWTKYoyZM/VaFgsMrAaDbsd/u3f7MA4TEk9m322FqEUNzLAkqQ4CQSibNjQQjisM3t2CqmpRu9DAwE+pYPFpGMf4FtOYIzdKCjQdQzefRzKLoPSZZ8fE4tECHZ24kwbeSXAaFQQjY7f0NBQw29JTlhYPD7XPdlmH/ylHdItRoAV0SEYhd4QxGKw/RA0tMG1F42+/MN46gkqPPcGXLcEMlInujXS+c5lhurs+JxLDhEaZIAlSXGg64Knn95PfX0ATYNNm1p56KEKNKuZ7x+KUZcUoDCtiykDLJdTQz0RYiyjEOyJkDsHUvpHJ3tfe436tWuZ88ADJBePLHK57DILl14qJtXQUHcENrRChQeyHMa+i5OM4KrKBVt8sCUKu+vgwf8DgSYdrQC8JSp7nHD/NPBYYEsdrN4Nt8yF7AE6ivYdgVc/hJWLoGic0uRcdphbCYkTu9a3JAF9Q4T1E92KyeUs+n1Oks5dfn+EI0d6yM93kpPjpLc3RnNzkI+OCl58z84nm9I4pu8B4BhraKDm8/d6ceClL1Pc4oBZN0Fq//wsp9eLIy0Na8LofhpPpuAK4KNW+P1heO2EXJFEM2QnCDp0wS8b4e0uwZNBne7ZUWJpgnCdUULh2T1Q322851Ar7G+BhkFmOtY2w/6jUB+HxW+FEKxe3cuWLaF++00mmFNxvLzEYNraYuzcGR57QyRpKJ8NEcZjkwDZgyVJceFwmHC7LTQ392KxaCiKQmKihZaYQtCuEtjj5Jt7c3jnniiq3YJywu825QwzPhTuIm/hAvIWLRrnuzj7VSTCglSY3/eRRdFZQws10SiZXan8cbdOJDGKHnVAjgJ7Y6BZmJYN8zKgb9SWFVUwLQcKT/joW33Q2QMFXqieBoXpkDeCwovrNhjDpQvnDfx6b69g9eog6ekaM2YME00NQFXBbB7+OEkaC5cFquNUYFUOERpkgCVJcWAyqdx5Zwmvv36EYDDGDTfkk5xsZfceCDQrENLY2pZDc49KuX3gQClMjFUcJg07F0aSYOsfiPWGabO/jyPzKlzZN53huxq5IL2oqFgYfQAxGhl2uOuE5PU6mthNGykWK8LUhlk1Y0/tItqViCo0Mi8KktqeibvTCM5cKgRDYLdCyQn1yjp74KerIBCC6gq4cg4UjzAfxeeHk/sJIxGdYDBGQoIZh0PlG99IwGY7vd7EpCSNpKSJXWhamvz8YaipnehWTC4ywJKkOPF67dx5Z/+pa9fmgTsMvrCCdy6UnZqC9blOwrxFB8nBdnqf+DVK+X6KNu7ho/YyKh9upGKc2z8We/gUKzbKmD6m8+wLgC8GM10w3FrNUSLUsZlcHKQoVXQnh7i9upPNoRC6eyuZtnaaP15G+2aIJcLcFPj3GsiywV0XG+fwB8GkQU8QgmEwa7CvAZ5fDwVpMLdo+FpaK06ooL5+PXzySYSmpsNEImGWLElj2TIvmZkmunXBR+EoZSaNhDgtRC1JcSNnEcadDLAkaRwl2qH+EYW2iCDXMfQPaw9W5uElNSZozW2ha7YVtfAiXM+r5HR7IU5FAMdDCRVoJ9f5GqWwDr9tgoAOqWYYrgSUCTPTmIsVGy7cdBAi5m5mgW5lVddeDjR62BhR6OkGInDpkzBlOrh0WNABPX74p3cgzwOPr4Sr58CRNth6DLobYXMteN1GoDUYIY5/TX0++MUvdF55ReD1JvHQQz7WrGlm8eI0Dh1S2J8b44/hKJcJuDXR9PmMxve2Q2MHXDPP6FmTpIngskB1fnzOJYcIDTLAkqRx5raC2zp8j4UVla+QB05ou/rbHGUrpa7FWL+4B5LPUM2D0+Rk9CsN6wg2c4TDtFFECjPUHC7yKLSGwXtCzlGQEFYsRhmLk6RwfJwvCStXkwsq1IXTeKs5TLDeCUVAMwRV6AqCLwrv1oMWgIIUY7kfXcCF5aDrsO9V+KyyxVAVLj7cCG+tga/dDrk54HRCYSGYzQomU4w3VweIKE6W3xEjQejc+JiJYz6ND9pUdnbD5ZVQPQtWb4WuAMyfCgVxWmxXMkSjgr17g5SV2VBlr+GQ/CGoOTjRrZhcZIAlSWdQLa200s0M8tGGmMSbQgEpFBjzfPMWnrH2nUnt9LCXZlJwsptm8knmC8n91905QB07OUAemVyg9wWZqhmhN0LoWTAtRjHPBT0AohM0o6bCPal2OjPt/KYD/BqE1Bj5KVFSHVZ8vVCzF64pgaoMmJkJWt+XQlXhK4th7V7IT4GCIeYftHeArxv8Pca/NQ2+/32VBx9UeWetm/96xs62Q07auhVMHbDpBYhNNXO0Cxo3w8frIT8D7lwKbd2QN0hP2Zp3Yft2+PKXIDFxDB/4eSgSEXR16USjQhZcHQm5jmBcyQBLks6gY3TSQCdV5AwZYI2aHjD+VB30NDdjdjqxOAdYJHAMIugIBJbPhgJ7m8GaDOrpPUbMaKgodBNCRcGMRl0P/OkIWFW4JjfGflsz7fTiFJ1w5KdGNnnOg6C3gH4Q9BxgLvT8FqKHwfVVMJfj1OAfpsJXvLCrPUYNuylJD3GnbRZ7WuHVPZCdDJcMUGk+Jxm+uGD49lcvhqximFrUf7/HA90JdtrT7Pi6MMY+PQqddSqhLGh2AQIO1MK7HwruvFGhMGPw6+zdB4dqoaNDBlijZberLFwY3++Dycplheo4rd4lhwgNMsCSpDNoLoVEiGElfvPuo0TROn5LLNJMrMvEzqe2gXcalV++mgOvbaC3O8yc++5HNZkItLZi9XjQzGZikQh1NTUkFhWRVDjwk1UgOBz1IT5azT6PTmtpKZcd6MabPhXCtSB0cAwRHQzBg51qijlGF9l4SBA2/vMQmFQ4Fg7x7fp6MqY0IQgS0QWNpny8up8yESWq5WG1PUiGmodA0KX5MYsIzpM+18JEKEzUKCMNMxomoDIdQlb4/TEo7oXck1ZECcagK2YULz1ZZ9CoEp/hgv/+CA63QUsELi+HRh/sa4XsBNjRCJlJsNcCEbOK0w3mMjBnga8B1IMQiOns3iPYdkijMs/oARvIbV80gqu801x26NChMEJAUdEANyRJffxBqNk/0a2YXGSAJUlnkAkN0xiTwT93YB1NbTW8ODuGJcFCV6iBQ6YCVn7tMI4/vMPWfb+H0jC+1wL83++/xXzLTDoO15FaUcH8Bx+kp6WZna+8RNYFswYNsD7cvYFfJxxElAW46c21dBRFqduxm4T2PXQtSCGNqiHvJiZgqFqnWXjIwsNeHxwJG8vcqKYIR9iCP9aAT6+nQc/gAM0ccS7ArlrIibZwobkHp6ZzEUFKELTYXVjFbJxqyefnjrCPXl7CxhUUUdXvugUOWOGFtAFijpdb4eNu+GYO5JyQaN8ZhJ9sgt4o3FYBoaiRuxWKGK8/8wkc7YJ8D6zeDu2dkOAEEYU8N2h+SN4Fn66H3DS4eIbCnHkKP3oRvnUdXFB0alsAEhKM7XS0tUX51a86AXjkkWTS0uQjXxqCHCKMK/ndJknnKL2jls7mTXSEc2kTSTQ/k0Ld4hKmZFZSNddH1ARTQ228ft/NqPU67/9qLSmftNF+4ADFy5ZRN+MTQv/biimaTQz9lCHLIBHeim4hFg2S7+yke0kqWdYdFMy/md7kNoLsIogfJ3YCuqAxBoUmUOgFvYeXehP5J1+Q5J5u/su8keKClYPeyzOHjWVwbsqDFxoP4wlu5ZakF2gngY1iDnVqHq3Ch6o4UQ7uRzebSCjO5BB+ykgiW1mJSenfFaXTik4HOi2nXC/ZAksGya8qtoM/Bp6Tno6BiLFFdSNZ/msXwo6jsL0ODjZDXiK0+CHFAV4HRAMgeqFDg93bwaGBYofyJJhWAA/dqRBV4bVtsHHPqQFWRwes+xAqK6CgYNCPbkhOp0p6ugkhwOWSC3eMRTgMtXVQWGBU4Z9sXFaojtNcGjlEaJiE/00k6fzQMfMi6mK9KF2dWGMBKnIySXm/B7u9CU3JpjDvE5La/KQm+altzaTMkkp2dSkzv1SA5hZ8snMVerkLNfIMPVSxgP6VNcNEOVqVQTs6c/BwZXolKgJHVhUCnVqyWM9hSunlWCCX13rgb5KhIPoa0ch2nox9k712EH4X74QLGerZfXMedIZhTuK7TEl8kV3BHXT2QHF0P5uCM4nYbcQiViymCLbWNoK+DnzFKVRiLCTo4NQFAy3MQyMPjdFNzZvjNraTeV0wuwjMMZibCVYTuG2wrd4o5XDDNFheCu/th/QcCHVCfSdEk0A4IWiDnAvgJ3dAbroxJPjE61CaCV+Ye+r11n0Iq96AnbvgO98e1S18zmZTeeihJIRAzqIboy3b4LdPw333wPRpE92a+PMHjckfUvzIAEuSzlEJWhol2gI6/L08rTSTrWrcXVVOSiiXjPlOdPN/0JUFV/iTKLEo1N/yJcyJXjIqO9j16hZcuxpJKklC0WcNUAAB3NjJYypRwlSTieuEQEVBpY4AMRRe1A/xapeJ3k4Pd1kdFCQswqRNZWbEzq5AlCrVxMrcaUSJESGGnVPH5SoTQQidWGQNDnqpcS6kwZnA/MiHON/sYv71c5mRkkCnHuErudNJ8bpx4CIbB3Tth1gQkvsPAypomBhhOfYR2BWAmjDkWGFl35OzIgu+cyVkJhozEN022HzUmBHorIJMG2zoMCrEe9PhumooOCEWNKtQngs5A/SmVVYYwdXCESTcD0VRlGGLpUrDm1IM110LBXGqFXVWkoVG40oGWJJ0jrJgo5AKsgpipLXVkjM7lSnpbiAXIXrQA+k49V4UVx64byWv4niwMeWKXrKXXoTVlkyQBBL6lriJRSK07dlDYkEBMZeNlSIdh+4mWzUZM/iaD4HVAZ50qkQu/x49yBY9mW6XD1+jg2ORVtCyQctmFhEsLpUHtIO4RSpraMBHgPmUkiVc6Pp2VHU6KDEEHahKBprpLqx0ka8E8PVuwbRH47K2o+zhdd7sLeWoz4vHY+dv7CkkCmPNRyI+YzxunKWbId8KlSdMSlNVyD8pOLpxOmysg6/ONfK2flUL8xJhfjJM8fY/9ssrBr9eQcHp91xJ8ZeUBJctHf64c5XLCtVThj9uJOQQoUEGWJJ0jrMqGktS+yfwKIoTzfYI4AA1FZT++Tdmux2PvRSAEwumt+/fz8b//E8qbryRomXLeKbewabuJi5L7OCbBz7EtGctnW4z23Pn0VQ2g0athAZFUCliLMvoYWmqC6Hr7H3tNco0DcsXvDylb6UkOIWIzUpIROkggCIE5vCb7I/+N5vUFKYdFSxiNk1TUwjoddzeqdD152cJrd2GJRP825txZ13Hnw/MZL/JR0N+L/ekmig3mUlIvQBrvCYODMFrgUdHsBhuZaax6cKo9P4NJ+Q6wXNCx53PZwwRDlRJQxew3QclTnAM8YRu7gSrGTyyCoEUB/4g1OyZ6FZMLjLAkqTJSisY9VsSCwqouu02vJWV9MbgiTqdIE602Ed0Z+0hXJaNrS3EkZxefOHd2D7I5+aLkvmGJUbjc7+lee4sciqmcXTDBkJqA91LwWJT2GfZywXNN7A/1cSfQzvY3ukm5lxElbaD9Fg7rxSUsaD5GIti7XT1fMT2P66jaO/HmNvgwJQqDuZmsS08BT2q0Lk/ledbVTzTN5CQ4+dQOIM79akst4++mvxYRGPw7GqYXgQzTkowO9YLvzwEy7xw0Qk9XFEB/90SRPO3UBlNZ1qRBctJT2F/FN5uAmcWFA9yS6EI/PRV8CbCN66O731J5zE5izCuZIAlSdLnzHY7BdXVgNH7UunpYVcsyPwjG+iZ4WSHpwxfRzIRi0a+pZYjNhvfs4ToOPBzjlW+zU73WyTbvkL7vdM4vK6X3Of3ELqpGAdBLO3/i+qmKJl5TqqP2NhYMod91gIayaRHdbIu3Y49VEOi3sHhW5bR8acE5lh3YK4KcpAcPFo3wh+DkI1wc4zs/S+TZTnMVvdN3NqewpWxP/DzphpI/y7Yk3B6vcPc7XGRSIz6eh8FBYkjTgZfdQQ2dUNpuP/+ul6IxCAijLUNH/wVeDMguQBWzobVgRYCWh2vtaos3ZbNDeWQZDGquXsTwW2Gv+oL2Da0gt0RYyMhMqMqaz+wkXAIvvU1WDoTEh39ry2EoKUFUlJAG6o+hiSdxGWD6qnxOZccIjTIAEuSpAEpCqwqMtO45dccztjHNk8Bna1N7I+WkhU7wq5gFZFCHz/aH+OIUomlsJj5CTXUs4dku59wYSpNu7uY0bQPV24juwpLmPlJHXPe30sDXhpK0+lW8+gJTyG/4zDztz2Dy95J3bwSXL09vH7VLXx68CCxZJW0PXvRldkI3Yla0YUedvA9299zh+85Ljr2AS2NGbjdR3ml7BjZj87lkO1KSr9xD6GCSjLVNKY6hy5RsG9fO88++ylf//ps8vI8I/p8anuN2WTzT8hbaQvDk0fhgjQ/C9wmtm+y8ef10B6CaK5RpDQ7MYHNbaWYXIl84Ife7ZDQAF063LsCnDawarCrC35zACoyBJtSu6Cnlz1HizCvhmsvheqqU9u0davg6ad1rrxSZenSwQOsnh5BVxdkZhpJ8P1fi2G3q6iqwoYNfiwWlZkzHYOcaeza2yPouiA1VRZCnUj+XqjZNdGtmFxkgCVJ0qBsziSS5y7F3B5Fre1FmNexN6+YtqiXxk/TSck20dkKLSIVc0KYWkch6aodj6uVozlp2NpTCezTKM3UaBFpPJs+nSvefBK1p4X78zdjSbmAF373Cok1TzJlgY/1CYWY50TZpE+n0ZbB1tzpmN1BClLKCR9z4lGbyHPtgxzBp7ULeENciienmQf3/hvupz5i130XEb1nHj0X/46/Tl/E7mvy+Gfl30lPuRNPfhlCCIgGEHoXWNNQ+yq/T5mSzJ13ziAnZ4D6DCIKykmPSiG4b+qpy08nm+HydMFPD/cSbjahdtmoWAAZdvAD104DmyMRfxjaLLC9G9o+hDfXwfe/ZgRXnylywVeLYKrbxG3WRLYE3eRcDcF5UFrCgDweBY9HITl58K+pEIJf/SpKQ4PgS1/SmDbteP5aLCZ44okGqqsTueACFxs2BHA4jABr926dffsEV16poQ4Qr/p8UX7+8yMsXZrE7NkjC1IBmprCRKPndoC1caOO3Q7Tpp3jtcbkLMK4kgGWJElDsv3lA0Tte1j1BlIdVpIXtXDQPYWODAdTRQ8XX6xwWziLnco+cizLOMgWWlSNyqRd+BbZOLhtBvPf2MaH8xcRarCTYnbQcqCNhI37UWf+lOJohNo9R2hq0ym71Yfo3MAORzkFx/bhibTTasqmoHYzSbYcsnPryA3WYgv10GH3kpjUwBFnBseWr+T6jBChbe1cdGgbG0ygdvXQsj2PpxbNw1t3O1OaGmh6P4X3YmXk3ZpJTqqNev+VVGkXUpYco1yMFT0AACAASURBVKDYidJ5CJGQZZQ1iDRBTw0ivIugNQtVnU2reQeH9Z10taZQlHg5hUmz+xVobQ0LfnhAZ29XMr4XFI69K6AY6BS4E2KEY2b+42tQnAJlCnii0DQdFk+FrjY4XA8FfYn0Vg0W9i0AXUeUohSdEmCosl6FhQo/+MHgCf/7asHtVGhuVujuFiiKIBw+vhCypilcfXUKXq8R7Hz966lEo4K3346yZo0OKBzzqcyapjB3ev9zKwpYLCom0+iCjPLycztLPxoVvP66IDkZpp3D9bFcNqguj8+55BChQQZYkiQNyRI0o0XyKWk6iM3WS9uqD3j7hqswqSpmezflipvLrTnoZBFG5/ccptZkwaN3YjoUwp1iw9ZqYc7bW0hSOvFmNWPttBJsa8MRzMBdZaX01nya364jsbGTorXdPJT/Ezb+WwxneSY1SxaT/PirTN90kJRXLqRkx37sR/1kXVdHy1oz3SVphAtzWF2xgq5iC49Ev0Ru1qfMfuc5uqdmcrClip0pM9lqqiLLtJtwaoSfRFfSuz6B+VN38s21lcRCguVX/plLAutZscvHgbJp5PQcIdW/j93pc7EEnqXJ9QpV63bysvte1MwYP2towq80kKW6KfVZaPnEwp5O2ORX6NRB9wA2Ae/rkC7w1Sq8dAhe2gDeefC92yAPKE6F6Unw+1ehp9f4zHsjRg6co69TJ49T18oJhXRUFczmkQU04Qj87Bkjb+tYg0aCS2Xdul6+850e7r3XwfLlTmw2KCgwKuI3NcNLr6rkZMR45WWdLVsE06crZObqlOSfGsQlJJh4+OHTXDDxHGYyKTzwgIo5fsuLTgh/L9TsmOhWTC4ywJIkaWiXfRnt2DzcNZ1ktGzBlCoIY2WmtZV/ySlgAR568dFNC16KWckS2rVyRHAfmaWzMGthAtsEM976v+xWLbgvyaVXN9Na10ui51ISkztpam0jITuf8qPbUbUIygv1RLZ6OdTSSPaWF1DfrafXAd4/fcS2t6J4UgSJ296mNyOfrLWphK/eT6sjjwNTl1Hflceqi1aSk34D/6fxHp5v+RK217cRvLWCT+csYvru7WBSOLh7Cm0dqQQ6E4iFzbyxYQWdi12stXZTG8jFnVROZWMOQoMEbQn+XhvdeS7eaqombVc3B+0l+Mw2dmiCD4NBjmxWiRwwQ0yBJIx6C1UCulTYGwV0sOnQZaZ5p8J/T4dLSmFlptH788Vrjn/kb+0z3n5d5eBfllhMoOuM+Ad7YwM074G1+yASVcjK0mlpibF7T4ynnvITCFiYN89ESYnRm9XeAXv3Q0qSgsMBFougpUVQlAdJIx8BPC94vZNkQoGcRRhXMsCSJGloCYlQOh+l4JfQ9hvmpuTzaizMhbYU3Bg/aW0kYMIKvT14jh7CU1QBScdrc4UqbmPHe3XEYgFsJZeTENmCr+0DLOkhPM5cGt5KIRbdjUVdQnTHBtSQICE3F/3IEYK7uxBWB+YkH1uejdB5GFLSoOSfXXSmJOMLJbDbM5cuSyo9HVYObKnEZvJz0DyLH81+HGVahGcunIFmE2SnHuNdlrG1fRbCZqel0wUzfeAzEXaD3qWx1zGF5Q1/4bXeK2iNpHDoT4VYS7ppO5SJ2RImeNQJ3hiK2YopBq4uhWOhENFaBXqAdkAAFgXqBSQAJgFRBcJRaDJDGmz+Dbi+AYdcsH9PmIoCcGYJFBQuLuyfj1TXDuEYlKQd3+dwDF37Kxg0/rTZIBAQbNums/+ASnOHQsQNyekqX17uIKrYmTFdsGKFGc8JgVPZVPjetyElWaV6kZm2Np2DBwUXXzxJggmpP4HMwYqzcQuwFEWxATWAte86fxRC/OCE138CfE0IcWaL10iSdFpM1lS8WY/iRUE5aWFoBQUzVqjfDWtehNSHIDHl89cTS6aQc80dJJWUQHY2ppRdKB+l0rh+CoopEYu5kfwHV2COtBD5n/c4Wm+iZd8OVLudUFsbAAGnSn61RtexCG0tkLeliYxFWRxutjGj8wM+nrWcHa0XgM9CUPHALhOhq+2YzBbq/dnEorC9cTb+Ix7ItICmgltgyQ2Rq+/Eu2M/j9if4ict3yLZ30Fl86e889ZyunwZEE2BoIXYUQX2AEsUEjSB97CAdkFHsw06FciLgM8MnTooOrhUaNLBpEIVoFshKQoVGtRAzbOCa5fBJXtDrHXHsFzlIz0nwrRmMyZdYW6hkZD11m7o6oVvj7CSeFcX/PSnRqX5b3wD1q3TeeutGGVTdcpcZnpNcHG1whVLTVg1mFbFKYnxigJpfTW80tMV0tM1KipO53+OdC5w2aF6iB7T0ZA5WIbx7MEKAZcKIfyKopiBtYqirBJCrFcUZQ70rdIqSdI5QxmuYnpxFaRlfR5cRQIBtj79NPXr1iEUhUWPPkrjJ5+w4/nn0WMxzI7dXPjoo1gcu+j1+4lecgWHPtnK3jVP0dsXWGGChfMU2rIEMy/T0Fpi+CqS2Xjblbyw7i56XA5KUzdB0EnmgQYO+73EOi0QUGhsTGNR8TqCMTu72kqhwwX+CPz/7d13eB3lmfD/7zNzetEp6tUqVrEly03GxjYyzQYHAw4lBBJSgLB5QwhsCmmb3SSb7Jtc2eS3m5C8CZtASAIkS0xCb6HYGDC4YFvulm1JttXbkU4vM78/RuDe8JHk8nyu61yWxmdmnjMq59Zz33M/EQ3yk8bM1XYPA/Yirg/9AVNCULl3NwmHhYmuPfRfuI5tb9QxUOgg3OGDCxMwbEPNUyktVJjUk+LpHhUydbBpCDvoAWA9qPk6KR9Gq/wbBKbJMZLbbLhr+nHVhukN5GCtjtBtV3jiPT+BpE74eTclFRF+/92/UmnrB74MwMemQ0o/+a9TNAqhkBEkxWJQXAwZGYIrrxRccQX0DUKWzwjAPnrtqX4XSOeiYARWbBzvUZxbRi3A0nVdx7gzGcA88tCFECrwE+AW4KOjdX5JksaBqoLvQB4r0NbGYEsLtsxMBnfvxubxkFlZiTM3l/3vvIPNZ/ydtfPZZ0nF46hWK10DKbJq6wj19ZMKhUAHMc9FQ3KYUMRF4XUxbHPy2VPkI/vdAXYoNex9uhJvZT+DLxbChRrYBbihp7uSl3b7cGX2YGmHTH8zl8beYFdWOUMuF0ObMxgMZuFJhHh7xhJudHybqdnvscddQm5bDwU57WQt6eEp01WwQ+BqHCZosqEUBAi7g7S1Jsiu0miNlkFYQR8SEAd8CqlhBXoAF5gKNYpmxqE6Tk59L/F9IXyLe8ldOETOTjMvJhoJCwGqoK3Vxv9sv4afX2Csr9ihpbCqAr/lwKyhrnPcBZxzc+H66yEry3hkZqp873vKB81HczKPva90ntKRNVhpNqo1WCPB1FpgIvBLXdffEULcAzyl63rH4Q3uDtv3TuBOgJKS8+/OFEk6F7jy8rB6PIjhYeZ+7Wt4Rn6W/RUV+CuMduVaMkn+9OlEAwHypk/HlZtL+eWXo0Wj9G5cg0cfpvP5YbIvMWO39DJw9Qws7WH6Y1kkqkG8bIIWJ4Mb7RBTwRWDSpNR96SniEdTdA8XQ8pOR6QEfV6C/15/O+EZftZeMJW9ndXYBxIsU66iy5pF6VArWfE+rNUxwhEr+1JFKDtsuCaFUISZnEk9ZNs68YsIe3dOpW2GCfJ0GFBAFTAHeEaDch1eBuqhokynKiODDjWBPzmBqtlR6he04NIH6AyW4FsKD/8VInGBuxqa2q18f7OVWQPQZEkQbFP4P/UqpUUCl0vh8Wdg1lQoO8avRk2DlGa8X/7zj1O0JHV+8nmViTKwko7BZYfGNLWZkClCw6gGWLqup4BpQggv8DchRCNwI3DxSez7APAAQENDwylMjkuSdKaweb3MufdeEuEwjsyjv7srJhN1H//4B587/H7iwSBzv/Y1zFteo3vZ/bSoEXZ5XQz9bYiqRSnqmnaxbJqNNeG5pGYr0KFCn4AIsMoBZRqYwOSLc3XB8+wcmMgmayNEBY+Fb+HCrBep3tBMiX8n1oJhonmFVDqb6bIWUyaGydjazuasYnJ8A2x0zSfPpPLl6Ta68hV6on3kmhxs6DfRfHUQPZZppAEHBCSBuA4ewKfhnNtPos4OkQxe3aIR6rAAFrbUWrhwyUQyhZ+WeB6fWQyfvwgeekPwxhaF3q0m/vycYKUVZpda2b9T58t/CHHrTSY+9Sk7DfWQP7ISUDSqo6pgNh/4g1VRoLoqxS9/A7/5mY6wJbHv17n/h2Y8GcZEoyQdLBiBFevH5lxCiAeBJUC3rut1I9v8wF+AUqAF+Jiu6wPCmIn5b+AjQBj4jK7r60b2+TTwLyOH/YGu6w+PbJ8J/B6wA88B9+i6rh/rHKP1Osek7ayu64PAa8AlGLNZzUKIFsAhhGgeizFIkjQ+zHb7MYOrY7G4XFQuXkzxl35I4X/+Ddfcu9n3rA3vjCwywmFaSwqIxqy4CocRXgHVSXAroALbgP0KdAqyXPvY6q8iT3QbM1q9ZvRhJz+Z+UP+65pv8DX/A6wKzmdbdhFXmF+mtK+L0kQYtXQhu4MXk7U6xNRt25hr3UtpaZz2lM6LATd/aHGyxzJM0goWNYzoxAiuOoAW3bgby5ak4L69lF65j11BCLUJ6AQGoG2ryhNrFd7Rw8yeFGB2LkytgE/Pg+oMhc6/W+l53MLed+HNVwTP/0Vj5euwenUSAKdNY9myOF1dGsuX66xefejfoJGIxhNPBIiGEyhRHXUoxavPwN3fgj8/edpf0nOSpukMDsbGexjj5/27CNPxOLHfA1cetu0bwCu6rlcCr4x8DrAYqBx53An8P/ggIPs3YDZwAfBvQoj3a7v/H/C5g/a78gTnGBWjeRdhNpDQdX1QCGEHFgI/1nU976DnBHVdP8aiD5Ikne9Us5nCuVdQOPcK2t6p5skvfhVlex8OPUnVf65lcmYh6zrmoxYOEu4pRH9TNQrY39YhHzr1EqLFCp2OUtiToNjXSoc5F7N9iETIxrTgemb0r+TV6UsJDjgx+0PsM4cp7n+Nq/syKZg8g7LQdi7L+juDtk+jCxsZvTrdXUmE08lQnxWvtR89I0hPeR4M62ADhynCnBs2olULUp1e9ro0ku2K8SetrmMKCkSvA1dvJuV5BzqZ107U+dk9sGGvoDMMl18Aza1gEgKrXWHOHGOWau9ejbffTlFdrTB3rorpsN/kdrvCVVdlEBUqmjPBYJvKpCkmIinIHHkLam6GSATq6o5fzzVahofB5Rqfcx9NJJJkaCiB12sd76GMC5cDGqem51gnShHqur5CCFF62OZrOZDdehh4Hfj6yPY/jNR1rxJCeIUQ+SPPfVnX9X4AIcTLwJVCiNeBDF3XV41s/wOwFHj+OOcYFaOZIswHHh6pw1KA/9V1/ZlRPJ8kSeewHU/up39NlP73IMcBH7/jUaJzHLj9YYJWO9uuh/79uehDAlGioPt1yFYZ7vExw7uRZPVubrU9xLrcGTwbvRp9v40l3mdYkViIvSPCrng5lowYef4++rr30VtSwHWp7eQ7ikjGBW+pClkiQW8syUDEx8DWDOL9Zoaa/OTd1M3gQJL5F6wkqyST/Lq9kN/CVq2WoMXO5CzYHoJ4NIUajZPtdFDiUsgiRZhBwFgDcVDTUB3Q9GsVAfQHoHsYlq1XuLDExqX1RtJh+nSVnByFggKByXT0CCU728Q9f4aXuxSWlulcd4vAZwWrCQYH4cHfQzIJX/wCnKjMVdeN1g9eb3q+lp2dcP/9cNNNZ87yMk6nGafzLG/HfhqCYVjx3rgOIVfX9Y6Rjzs5sChUIbD3oOftG9l2vO37jrL9eOcYFaN5F+FGYPoJniN7YEmSdFIKSnKZUSiwpHTyJkNuX5zP7nqM8uI2nlSuwTwxgOObKvEdVkQygVsNY68MoOak8GS0kxFL0REspNa2jae0q+nzZPKA+jm89n66B0oonrCT7P4eRFQQtQkuir5Hb0snU8q9KO5rycHCVj2MLWuYHFOI7o5MzOVxrEIjN+HlxrooC7J6aHJ7ifq9BLUqpibz2DfgYl2zoMwOEa9OKpiioQYmZ+pc7c9nZ6egwwl5LsgZKY4aHGkSmumF5hjscQrUoMr0APi9xpqBeYWC7ggUHOW3aDvDPDPYxMa9HbR4P8KyQSs7HtapLBV893LYuhUsZsjKBN9JNMzp6IAVK4w7E9OxJIzXCwsWQEHB6R9LSpP03kWYJYRYc9DnD4zUVZ/cUIx6qVGtvR6Lc8hO7pIknfFS8TidrzxL2OqjNqefNg3a/gwldw4wv2ctb029kBxnH6GhTFzFSZTubu6IPspboo587wC95hy05jhbMqdSpO/GZkpg9kbRIy5smUH0XrDFNRJOE3HFQv5vX2PWrD3snziRpH8hAX8tN5LBdIud/c4gQUuKCu+7DO9PMXh9Lb9yFDDZorFXsTJMC1FqWawUk2ty8Kt22BozVtAJhMxE95nZ6dL4xtUafSETv3sPBlKwuApKIvDGRkhlGZ3bF+dBVSVcXwGRvdDTbwRYAKs74fkW+OpM8NoOvV4DROnStlFXt5F1O2eyY18Jrb9NYisx0fy2IDcILivcfhu4R5Y5jMd1mpp0amsFNtuhs2IFBbBkSXqCKzC6yy9cmJ5jHcumTUbzVBnEnRyXAxqPOyVy8h6BXl3XG05xty4hRP5Ih4F8oHtk+36g+KDnFY1s28+hN8wVYaT89o98fPjzj3eOUSEDLEmSznj7Vq1i/Rsb0CIR9u01Fi52WmF2IUTnO2hIbWB3YDKbehyYzGbiQ15+OvR1rENhlm79HTVFXSwYXM0rky4nYMnANzRIW3wCLms/usmMVQTQkgk2DczggupVbLj1Fhb846dMLbiE9VlT2cFWrqSWajWDH3vK2EQcMZxkeVUKs1uhRFXQRD8pWpmKkwJK8WDU8lxeCzsHYN0QbF0JynbQKgSOlEqeE2qKoDcCewKwYzd09sDO1bDxHXgkDl+9D+6+GZhgXIvWdti6B+Y2gM8GnqOUDFWTSbZ2Ia9Nzya7MsTQTyBWqBDrTrGhrY+FszpRu+owmw/cTtjbC88/r5OdLY6aMnSdZfkGq5UjatOkYwuGYcWaEz9vFD0FfBr40ci/Tx60/YtCiD9jFLQHRgKkF4H/OKiwfRHwTV3X+4UQQ0KIOcA7wKeAX5zgHKNCfvtJknTG69m2DWG1EeofgISxLR6Dtc+DcCWZWLmbPclyNJMFbbeApBlbTphSbS83vLWMgQkFrF87mX8q+R0/KfoimtBRYylc7iH2tNThKArSHizG2hslmO3gotwVDC9PYJntYgLGHZAeHPw1mqI5pXCf043VU8/H0EmiY0VBJ49CltBGgDb2MoWJDEVhSx/cNh92PQxsBy0GlXWC5YNwjQteeQumTYQ+Ffx58NYzsPU1QIfBBHz7j7A+AP4cuPMiKMky2ixkWGHyMeqxTSgs8ZTwejCAT4ugeqDZrKAPaZR5m7hq6ZP0Rv6V7eYsnmzRycuPkpcvuOFjCnv3WujpgZkzx+RLO2oqK8d7BGeZMVyLUAjxGMbsU5YQYh/G3YA/Av5XCHE70Ap8bOTpz2G0aGjGaNPwWYCRQOrfgdUjz/v++wXvwBc40Kbh+ZEHxznHqJABliRJZ7yCmTPJb2ggo72dfWtG/szWdfpbU/gfbcX1EStFNXu4SUzmR9sBZxSbO8Gl/Y9R/HoTWdpOmvqy+e0NH6Us0YZJ19iV1Y85GSVq9dMVyUFr8hL3RXjrpcvIaAxSxIvsjM5hspZBoWLk5SaZUjhECoEOWFERGOXoIFBwU8Eg64kxgI7O+nbB05shQ4UrqmFLJkTtMM8MN2bC2jeh+WEIFcAn/9noX9U8CDgx3kpSMBzQeOQ9gc8hqJ0EldlQdBKluUVmJ//jm47mE9h+qPCH5+C91KsUR3/Hax0N/K1ZUOTv4rLsdaxQZuKyqCy7O0XfZg9ZmpXtm8QR6xNK5y6XAxrTFFSfxF2ENx/jv45YbXPk7sG7jnGcB4EHj7J9Dcbqn4dv7zvaOUaLDLAkSTrjFcycybRPfpK9q1ah2u0kwmEykjFMHevJ88Bgno+LspP8MN/E/70QNjzzMhvv/xmWQDNvb4eSOUkWiEdZ/YM55P68mkm52/h7wMHN9kdY7r+MPw/eglYk6B8qRCSjdAfy6XhgERWvfouWt3+Gw2vFX1vFRLPAbH6XFqxUMeeoY53DFJ5pgvt2C4Y94HSk+PvrSS6cG+DO+0x8++t+vvKvsLkZSvNgQiHs7oSnngctF8yFkAqCthMwa5CnkXVlggrdRsp/9DsG3+iDlgh8svDQtgdurCRJMWQOcPu1GaRm5PGTLTeybHgqTqUDMRxmftZztAXtvD4wi8GYALuZ3sEkv30QvvplM8qYdEs8M2ka583rD4ZhxeoTP086eTLAkiTprFB97bXkz5xJV1MTJpsNp30nYe1RusL9JLFzt1rzwXMrGqYTuvwKzJ7rKFvQytD2Vwj/bBcT83Umtq2hW/FQbGlhe1sNtbYtNNjeJVUK69bP5hMzH2Vm/FWGLbnEC2txTfaxSvk7fTyNlQoKceHS/ESUJL1PPUdGVRWemgPn3t6p8tPlsLMT1BQsnBNiysVhqmsSbIz1YC52kNps44UW+KgXvvTP8MSzkDMBHGWg2cBaBv6rYHCCTt68OEV+wadU8ByjA7tqLGN4VLvYx0Z2Mp9pOHOmoHdUMNm8CseELi4LxQhvr6J5dQ3Ta6y09JhIoYNL8L0fx/jTX5M8/mcr1aXnSZRxmEeXwcQymHOq5dpnozFMEZ4vZIAlSdJZQSgKnpKSD9YzTDIF+/5BMjftoKViKlZx4BY3V14ec7/6VQCCXW8T2bcP4VQJtGwl72dDNM8sZ81dMaZpm/iT4xN4hnqoGVzP0Aw7cbOCo0unNZiPpfI+lg/8CxHTIOvFVHy2Nex7r5sdGdPJTiT41pQpWLOyDhnn+k6YVgT7ukAbgBfftGO6zETncJIMdZiSOp1YJRTOgbJSuGwSLJxt1FWF49A5A0o8sCEK6yIq88w2hvYKciuP3ZFzrh/mHr5Ri4BiJxsfE8gnAyeRFERiFhYW51KV28+UgYuZc08+gRD8pL6dOZ9s5dcrL2DnHoh16ezdGWPzBguhAEybcv7M5oAxe1VeDJXl4z2SseFyQuOs9BxLrkVokAGWJElnJRM5ZGTfgz67l2pLBpaRZp2Hc2XOQjjm0eWL0pMME5oWY09hCQM9WeTo3XS1qlzYsYarOx5n55Mh2j9zCbn/sYtf/ccTPBjoYvi/vsOSnz9BKOFi2GqjdGiISFeCgVQc19KyI843pwTebYFCN6gJiJYIwt4U3TGFEqePmkk2PFNg0JtgrkfBbDowLeWwQPlI3dNMJ0w0wePvKXgCMOewom0NnSaGKcaGH8th/xmD4SfBeQl+Uy4NWi3RBGQ54LvzTJiUWqCWRAZcOR96B+H62k1k1+8iq34OP/ylRiAGOW7Bsr9pZOaqfP0eKDxPWh48+EiK//5FCrtJ8OSy86P5aDAEK94Z71GcW2SAJUnS2ctiR1iKOe7iJiYTjk/eiW/WXDq9axj2QOWLf6H87Z/TohZw3cr/j8A/OtgxIUjZxucpEjHcFZfizl1OWabKpqJ63vxjI9ff9AhoJrr8FSxu2ki9bsW4M/xQ5X745DzYIsCXgMJyE163i4GkTpbFwfZyndhwkkhRN2+rNi7Q/SjvF04ltkHsOXB8CpQsTApUFcGFR5lZCJPibQbQ8R4ZYClWcF4GqjG79vYOeGcH3LsETAfNQpnN8MD3jE7tqrIQ9Mu4NE/Qcb3KxQ1Odm1NkZ2noAO5OafyhTk7rd0Mm3fCbx6IsWNbikw/mNPV/OtMJ1OEaScDLEmSznlCVfFNnko9JYhkHEtlJj1rd9H36kNob+7CFE3QvBKKF82lYcJClKpcZguNt3p0okkVa6mdSZtW4ZhspzLz04jWjVDkOOb5LBaj9cJ0D9S4YVNfjLxkH2ucuexMgFtAQdLJJk3lH6KPOeYIGRQDSdCDoBvvdE4zNBbBUARe2QMLJoJ9JJZyYeK6WCG796sEC8B1WLNRTNkffDipEDJdR0/xHdgmQKgU5MDXbjc+nzX1zHuLiMd14nEdlyv9+cpEAmIJuHwBdO1LUTtJxWTSgTNkwcRR5HJC4+z0HOuRX6XnOGe7M++nR5IkaZS48Bm/9aYspWAKdMTdzJjUT2DvXioXLyYZDrP+y3dRmBji0iorC2+bzvpLClhrmo4tHqSnbx431nyC6Jeuwew4doBV74Gv2sBnBrMC0195mK3N24ld8VkyiifRHNLY1eSg3xXFMXk322jh07oPjzoRXN85YgXkcAJ298Hs0gMBFkBbu4kHXoc7GmHuxGO/br/beJyseFxnzZoEHp/KtnaV4hy4YGTNwGRSJxYDp3Psg45wWONXvxokENC47TYPZWXpnV2aM8146DfY+eTNFv7+d514PK2nOGMFg7DirfEexblFBliSJJ23pt92G/FQiB3PPEPu1KkoisJbipnuUJyK/XE8a95i2vwKCpRmBqblE3p8gMgX2rD7y9B2N6OZTCglpUc9do4V4qEQA11d+LxZlLj2ketyk20VWLqCLAtquBQHM9QcgjjYHHyeuZGtkHk3qIcuEJiXAV+46MhzTCqAOxuhtvDI/zuW7n7I9BhF9cfS3Jzij3+MgNmEfYITqwXqJoIidH79a42eHp2Pf1xhypSxrXoPBjV6e1MkEjqdncm0B1jvE0IwaZKJmhodIc792asPpG8tQgkZYEmSdB5TTCZsHg/1n/jEB9sm3/MVNvzyZ2hqL+p7cWJD+8m6qJCmpkuhYpDhyO+x8z1jHRZx/ABjcM8e9r71FjNuvx3nRVfz0WSKUEcrtcEAwhEic0olGSY7YVQEJkDlaOmolkHId4H1sN/YTivMOWzmavVOeOk9uO1yyD9Ko9AtLTC5FHKOuaYPlQAAFpVJREFU00S0uFjhoosseLNUNnVAeSHYbcaiz11dOqoq2LFDZ8qUQ/d75x3o64OPfOS4l+VDy8kxccMNbgYHU0yfftzKu1Oi6zp/+lMPFovgppsOpFbPp+DK5YTGC9NzrEf+Jz3HOdvJAEuSJOkgc7/yFSbPn0DilR+z+88baNkRZNpdrax0pyhdu4W2ns2oBV8m8xgzV++LMESgNsikqo8hVBVUlb6Na2l5/nnqvnQvOSkzb8ZDxJJhCk0qda6rwLWEYEjw3HNdzJ/vpaDAynAM/rIZrqqEuoMKzVcFwatCjf3Q8wYjxiOWOPq4Lp5x4mvgdivccotx4INjpbw8aGwU7N8PF110ZHA5MGCsaTiaGhoOLzY7fboO/f3J83rtwmAQVqwc71GcW87jbydJks4GSaIMsw8v5QjGJiXlnbGY2OZNlNX7yJz6Fl2F2WSvfZfUgmzYvZ23Nn2dqyf9BEyHFjYNRCCWNNohRNUQ/WI/uZYDU0zZ9fW4i4rotjlo71MoxkRlxM01bhXLyGzY8HCMpqYQlZUOCgqsuK1w2zTwHxZIbY5CnunIAOviKTCrElyHbU8HRREsXnzs3OKVV6b/nKOtszOFqgruuit/vIcy/mSKMK1kgCVJ0hktSj89bMBFAWaOXVieVmYnlo/fR1/8YQKRxbh3VWLvephED2xZGcG0tJW+6C4yXdMACMXhiS2wfWT2xm6GpTX5zMhdgnmkiURLJE5TfJCG3CAx7QnKbYvYHi9hmk3FclDgmJ9v5ZvfnIDDcSCQyXYeOcTPZh793jYhRie4et/69UmWL09y7bUKpaWWE+9whtuyJYXNBnPnnv2v5XS4XNB4RLfaD+eR36fnOGc7GWBJknRGc1HARK5BYWz7EQmHg6rb70BPpTBZLRRYX2DrqyvZsh98bgv9Ti+ZI89dtgWa+6DYYwQ4kQQ82gR3O6zkuSGahPlPRghqKnNmDvFvVVuZ7KunQzezRcvAFvNSdFBJ0cHB1bEoaSgP0nQjSDuVUqNkEpqa4jidOjfdZELTwOM5e1u8X3rp+R1YvS84DCveGO9RnFvO3p8KSZLOG0cLrvRYhORfHiC5cf2onVc1mzHZbCAUNNfNRPvmU1V3F/Wlv6JUFAFGWnBHLxRmHAhU7GajoeeGTuPzpuEYyYww2flddJJA0a2ktGwKdAsre6z8shPaYqc+vsE4JD9kWkfT4Rf98NTwqe3X0GDi5z93cfPNbhIJaGnReOmlFLquH/Hczs44X/nKDvbsiX64QUpjS0/TQwLkDJYkSWcpfddqIqv+neTAAnz1fxr18/nr5jLziyXYfT4suD7YHksa/x4+C2RRYWgkaJqUobC4uo0hWysNmdtwsYbliev5uGki/xAKHQo4FehmEIAcvHQHoHcIagqP3iC0JQS/2WP03Lq5+NRfjyKg0gK5J/kuoGmwZh1MrjF6YPX2psjONpFMwrJlCUpKTNTUqIftoxGPh9G05KkPUBpTLhc0HqUVyIfxyOj/OJ4VZIAlSdJZSUyoI3HDJ1CqLxub8wmBp/jISCbLYcxYhRPgOGiibTgGNSPrQLtUM/e7mxlIbMJhKSTOzTSqleQrgs/mGCuUxESAN9iAK+oiZ8DHH9eU09oDX1oME49Sf60K42E9jTzER06h+SgYS+soCoTDOps2xSgsTFFVZeWOO8zk5R2ZZywosPGLX0z78AOUxkwwCCtWjPcozi0ywJIk6awknH7883403sPApMJHa+CRJiMtaFGN4Ko6C2J2nW+ugql+jWK/E7+YQCGfRShW8hTYSg+b6eYSytjNGhLsY+o2P2x5gYa6r+N1mMjxHP28xQ74ds3pBVjJJPzxWWicARUnmAVTFJg+9YPPmDLFxrp1UQoKzJSUHDoITdNR0lEkJklnMRlgSZIknabJuXC3w6i5GopBdSbs1zQuezrJUFJAj2DWJ+aQ8oe5KfIOtzn34uUaOmmjm272EWOIQQQmtMkNUDqHBV4TC4AkGmFSOI5Sh2Y/cS38camq0aE903vq+2ZlqcyYYcNmO0pj1JYYPp8Jn+/EbzGdnTFWrgywcKEfj0e+JY0XlwsaG9NzrEceSc9xznbyu1mSJCkN8tzGA+BbLyb58VYN/DFsHRqpbIU1b3qw5Jr41+AkSi/bzCXWzZh4jTpiDDGMwnQmMwOXJceYBhuxgj10MMwiKsnmKP0aToMQMKv2w++flXX0CK+oyILFcuyptVTqwFI9ra0x3nxziPp6lwywxpFMEaafvItQkiQpjd5rDfDWw//B9NaH0IoVJk7fhKeqF3tpCIRCfKuTvybzeCG6mxYtxZt6LS2pRohewD9C+bysDZI4qOOj/sFtWem7PatvEH76R1i9+cPtv2dPnFjs2LcvHi+40nX47W/h3XeNzxsa3Nx3XzFut8LDD7czMHCMFvTSGNDS9JBAzmBJkiSllYUYPnc3PtsAe5Vmiqa3kdU1wJrAbHzuLgZme1jVUcGq3W7srgsIlZswhUz0rsrD4k8xtzBI9eQEZSajMdYCyomRxMmx+zWt7oOOKCwpOLn+WJEY9A7CwNCHe43DwxrxuI71QywHKATMm2csuwOgqoL8fCu7doVpaYkSiWj4fMc/hpR+MkWYfjLAkiRJSqNJE5zc8wM/T7WE2GIfwG5O4c3rZm48gLmyheZIDh3N+ehtHjpNYA1FCRdZ0boh1Kayt81Ob2kHyYx1FDAPJ7mYjhNcAWwbhj1BWJQHtpOoyyrKhW/dBs4P2fG9vv701gOsqzvw8YoVYTo7k9x4o5vvfKdMFsePk2BQZ8WK1HgP45wiAyxJkqS0UpiaXUTAaSUSbqa4fzvr1KlMLt5FIJ5EV/dgFlBVtI1NsSwy+gPstFcQ0B0QShDLGMJq28Uw+wiyHye5JzzjTcUQ104uuHqfO73lXB/ad78bZM+eODNnmqmoGMU1fqSTIAOsdJIBliRJUhop2PFxJ9c6wNH1Ft0v/oZISQ+1c/oYGHJjUSZQ3/UcnQ2zecrnZt26HPQeK57abtwd/fxwwVpqLEsJUoiFQiLEiSUstEYh3wI5R0nLWdRD6uI/lFAIYnHwpzE9FwzBUy/ClZcc+7jf+IaLzZuj5OfLJWvGk5EiTE9ZtkwRGmSAJUmSNBoiu7g4+RhtNR1MLh5gZ08lWa5e8oY0ZgQnsPOvz3PJTCcdMxZR1r4W73Azi4tWs9B/EZZokAxbFS+wkf6EzovNU9gTNjHdAd8sg6LjZOg2RDWeCcAik0KGCtUn2YLhtZWwrx0+/9n0vHwAmxUmV4HroNmy1tYkQghKSoyIcNEiB4sWjdEi3tIxGSlCeYNBOskAS5IkaTSoTpK2GJYyL6bMIFu0yYR77ORu6sC/chO5thgVHS9yaxIeyvKwOzSV6pwNDA+8ToY2gFL+ZRxYeXVI47WAgklo2OIaOxI6KVucBBqlODEddjP4a6YQvU5BU7sLmzj5AGvBXIikeclAkwmm1R267Z57wpjNgscfP8U28tIo05EpwvSSAZYkSdJosOQRD1lRV7vYVFxKpXs3vs37aX+sj76KRUx0e5iQkSK5602yG6aSl2pjT7SEQO9aCtxlKAguZRKbohpWW4KwrhNxpljhDLGNQXRgNpk0kHnIaa8z2VFdkDvx1IbrdhuP0XbXXVZMJqOQ/S9PpfjbEwkWXwI33WTFZhPEE7CrHSqLjC750thwuQSNjekJCWSK0CADLEmSpFHiENfTt6udsuEU+0tCsGWIqVNjDC+t5nV9kOiaLSyfWkfby4VMbNxNvzObfwneyre6d6DkJsi1JLiiQLA9s4sS7GC1kCcUghhNDOMjPYd0XWc3KSagUiJGfq2foV0OFy40isi2tcCt34XEgMpfHwtSV6cyc6aZ3R3wh5fgc1dBecG4DvW8YqQIY+M9jHOKDLAkSZJGibn6YpKzQrQ88ygtL23Bsq6F0i+Uor+yhVX3Xcw/JlYxaPaQX9yNpTdBS045G2IN5HouIBUKM8PSyY2ilM/bsijCihcTKXQ2IIihMQ2jcjwBdGoauYqCi7OjzUFzO1gyBIleHYdVZcIEY9wTC+HOJTDhxDdPSmmlI5uEppcMsCRJkkaRv6ICl/CgvLKLsnqFUMkFRCqm8+yAm8Goh3fXzmTppGXUercwPJiNaRAKKvKpsiao1PIwCTMezYxDU3hsNcyvEszMOjQtaBGCeeqRd+EldZ1WXSNfKDjE6AVesViKrVuHqKvzYDKd3NTZ3Dp46qeCgTbBlFrnB8vumFQoyx+1oUrH4HIpNDam505OmSI0yABLkiRpFJkdDrzFxdzxyqsMuRV+nJqJ4t9OxN7Nbr0UrGa2v1HNoitf5V9feI2hgS6+Nfw1HrMIqmpiTLHFGRYuPmMdJIaNpHbyfRQ2ppL8IRHnYpOJpeYP0Xb9JFmtKsXFjpMOrmIx8Lrg0pkCZh65iLU09oJBjRUrIuM9jHOKDLAkSZJGkaekhDn33sur3/42T7duYc33v0J+f4Bai5dNAT/u3hTTJyeZv6sG8/5+/hQv4+WNgN3E1oCP1319xPcIzDdtZdaFbWjiekJYadcT9BBnhsjAhjH70xyB/+2FK30wwwW5ikqZolImRr9aPDPz5AK4YBD+78/A40lx4zVQVqayYSes2Qaf+YixCPTb2yArAyplDdYYkncRppsMsCRJkkZJKpFg57PPMPj0Xdg7O8h+AZb8UwYvlXwKpb2LO4oSrJ+6jyF3mLXZhTwY+QQrN+ZAbwohUuiDZgKxTLDrPL6ziuikDt4Nh6h27uLFOOianX9zqNSLDABaY7AvDs1RI8AqVBTusZ7esjbpZrFAIAT3P5jkgfujNG3MINMjKMkFRTEWg357B+hBnatmQHW1QNeN/5NGj5EiTM/3ikwRGmSAJUmSNEr6tm9n+ff/nUBvgJwKL4U/yKO3Xye7tpe39bk8tVug76tAebKAop/+hoKa56iobqB5bQ0TaGHv8gpSbhPmhQEmuhysDV9O0ZDCBHsbmbhwi37ysQMZJEgxJwPyLCoTRi8beMq6+yDDqROP62gmhUAEPnc7PPSkwq79Jr7//TA/+IGT/l6j87vbBZeWppg3L8W/mwSPPWVmOAIfWzTer+TcFgymWLEiNN7DOKfIAEuSJGmUZFZVMeNzn+OpVa8QzIpxWVUL706rp8C8F0cojGi1oLt0dLfOy7suYbpjAzl97eyxl5P0qvhregjsc1ERb2VWdiYJPYdLM9/Dbu5mqSkfB3F8IhsdnX/QhEOxssAxabxf9geCIXjpDRhoD9HblaD+Kg87+hVCZp3w5Qo8a+bHPw3ReLGdV99VWHolzJsFWX6BN0NgscD0SRCNj814e3sj9PdHmTjRe54uOi1ThOkkAyxJkqRRolosNPyfz7OnNMaO/jd4efpHCKYc6BaVij+tIznHxP4tpehBK32/9vHUwqXEexwwMUaNp4ldlROpLt5MeWkHV9iLqRZ5ZCiTGSKbAZFDiloyMRZILsKP9RR+pSdTsK8XirOMuqdTEY1q2Gwnztm5nNA4CzZvUsnyweW1gkYNXtoNIgC6S8WaaeaPTwtuuxNmVhn7hZwK331R8OkagWUM36XCYZ1lywbJzIxx5515Y3fiM4CRIkzPCuAyRWiQAZYkSdIoGtJDeGubqBkOUvPa73h9xsUIh0oyL0lD5jZ+qn6bpMvO4PJCcJnAasycrEg2klMxhDurg4XePvaoZsroxUkRwzh4kXbMCO6gGBsqUyg5pXFt3AMPvQz/tBjqSk9+v3A4xUMPdbF4sZ/y8hPX7PQPQGm5nUlXHNh2S71gwrdVHvpHis1/1dnaEqHfYsdmM167wwS5ToE6hnVX4XCK3/62h2RSUFNz/r01GinCofEexjnl/PsukiRJGkNOrFQqxexf/wo9KwcpWP0maoWXjHiMYEUOyf90QIYAzQoBHdzAepX4sJ+y8hQ3+HuZZLKwBwV1JIXjw0wVTpyoWI/Ssn11M3QMwKKpYDtGa6PyPLj6AijJObXXY7crLFzoo7Dw5HomHb4WIcBwEIrd8J0lcPvfgEGNrU8mGMw34/UIdm6GikzGNMCy2xWuvtpPYaGFgoIzqIhtTMkUYTrJAEuSJGkUmYSZsqLv4plSz87ND1BQuYT2h/4Zz4V2Yq9ZoVuFwST4TSASWPICzKpZTSA4i/+aDDXicrp4m0xKyMPoW2BF4SqOHhlpGjy5GoYiUFMIVSOtDnQdhqLgMTKKeF2waOapvx4hBFVV9g9zKT7wxhpY0wR336oQi9tY/Q68s1ahb0Dn5z8S7O0EIaD2tM5yaoQQzJp1/i5AbaQI0/P6ZYrQIAMsSZKkMeCvv47Z9deRisXo272a1WSS7LmDrEgzvfYJUJpCmRZkSk0TF89fTqbJxVaRQQ4eonSSSQEmTtyUU1HgutnQOQATsg9sX94Mz26GW2dBfeGx94/HdTZuTFJba8JuH51C78ZZUF8D0QgoitGGAR38PuN8n146KqeVjsNIEQ6O9zDOKTLAkiRJGkOq1cqr1gfp2huhov1Rrszbzf9WfZu4zYpmc9McqmWz6CZPzyKHBA7yyOImLHgAiKGxmjDlWCk4RsA1o/zIbVYTmFWwnKCgvadH55lnEvh8ChUVo9Og1O0yHiD45f0KL74ERcWCj91wPt65d6aQjUbTTQZYkiRJY6y+Hp4PWGktmczGkgkk7DZoF7AeAtl+HA3TqPaZmWrOxa86AMcH+3aQ4AWGmIWDa/Ge9DkvLIMZRWA9wSRYYaHCvffa8HrHpgBqyhSFKVPG5FTSCckAK51kgCVJkjTGLrhc5WW/wu+1KVizAziejhDCBe2ATSF/Zw1dXsGqDKirPHTfIizcgo+8k0gXHu5EwdX7xiq4ks4cLpdKY+PJB+zHI2uwDDLAkiRJGmM5dvhMtSA22MWm+D52+2oIDbnBqjMtDhafIMsKrUHQdDi456UJwWROr8hckg4XDCZZsaJ/vIdxTpF/pkiSJI0xRcDMbPh5ZQk3MIUyNRd3CrxRnXvmweX5xvOuKzk0uJKk0ZVK00MCOYMlSZI0bmzYuHeyjVvLwGMBk6J8sETLpfnjPDjpvGKkCP1pOZZMERpkgCVJkjSOFAHZH9Swy+kqaXwYKcKe8R7GOUUGWJIkSZIkIdN76SUDLEmSJEk6z7lcJhobs9JyLJkiNMgAS5IkSZLOc8FgghUrusZ7GOcUGWBJkiRJ0nlPdnJPNxlgSZIkSdJ5zuUy09iYm5ZjyRShQQZYkiRJknSeM1KE7eM9jHOKDLAkSZIk6bwnU4TpJgMsSZIkSTrPGSnC9HS3lSlCgwywJEmSJOk8Z6QI9433MM4pMsCSJEmSpPOeTBGmmwywJEmSJOk8Z6QIC9NyLJkiNMgAS5IkSZLOc8FgnBUr2sZ7GOcUGWBJkiRJkgRo4z2Ac4oMsCRJkiTpPOdyWWhsLE7LsWSK0CB0XR/vMZyQEKIHaB3vcQBZQO94D+IsJ6/h6ZPX8PTJa3j65DU8PSe6fhN0Xc8eq8EIIV7AGFM69Oq6fmWajnXWOisCrDOFEGKNrusN4z2Os5m8hqdPXsPTJ6/h6ZPX8PTI63fuU8Z7AJIkSZIkSecaGWBJkiRJkiSlmQywTs0D4z2Ac4C8hqdPXsPTJ6/h6ZPX8PTI63eOkzVYkiRJkiRJaSZnsCRJkiRJktJMBliSJEmSJElpJgMsSZIkSZKkNJMBliRJkiRJUprJAEuSJEmSJCnN/n88PkCv9JSm+gAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 720x504 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PTHlIIzBJ-Ok"
      },
      "source": [
        "**What does this tell us?**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ObEWuz52KHW4"
      },
      "source": [
        "# Correlation Matrix\n",
        "iris_corr_matrix = iris_df.corr()\n",
        "iris_corr_matrix\n",
        "iris_corr_matrix['target']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Q7EVef8icGJj"
      },
      "source": [
        "from pandas.plotting import scatter_matrix\n",
        "\n",
        "scatter_matrix(iris_df, figsize=(12,8))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ao553L4-cfVu",
        "outputId": "84164868-dc61-40d9-ce10-7458c1eaf1fd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 328
        }
      },
      "source": [
        "## Build a scatter matrix for housing here for median_house_value, \n",
        "## median_income, total_rooms, housing_median_age "
      ],
      "execution_count": 53,
      "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>longitude</th>\n",
              "      <th>latitude</th>\n",
              "      <th>housing_median_age</th>\n",
              "      <th>total_rooms</th>\n",
              "      <th>total_bedrooms</th>\n",
              "      <th>population</th>\n",
              "      <th>households</th>\n",
              "      <th>median_income</th>\n",
              "      <th>median_house_value</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>longitude</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.925208</td>\n",
              "      <td>-0.114250</td>\n",
              "      <td>0.047010</td>\n",
              "      <td>0.071802</td>\n",
              "      <td>0.101674</td>\n",
              "      <td>0.059628</td>\n",
              "      <td>-0.015485</td>\n",
              "      <td>-0.044982</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>latitude</th>\n",
              "      <td>-0.925208</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.016454</td>\n",
              "      <td>-0.038773</td>\n",
              "      <td>-0.069373</td>\n",
              "      <td>-0.111261</td>\n",
              "      <td>-0.074902</td>\n",
              "      <td>-0.080303</td>\n",
              "      <td>-0.144917</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>housing_median_age</th>\n",
              "      <td>-0.114250</td>\n",
              "      <td>0.016454</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.360984</td>\n",
              "      <td>-0.320434</td>\n",
              "      <td>-0.295890</td>\n",
              "      <td>-0.302754</td>\n",
              "      <td>-0.115932</td>\n",
              "      <td>0.106758</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>total_rooms</th>\n",
              "      <td>0.047010</td>\n",
              "      <td>-0.038773</td>\n",
              "      <td>-0.360984</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.928403</td>\n",
              "      <td>0.860170</td>\n",
              "      <td>0.919018</td>\n",
              "      <td>0.195383</td>\n",
              "      <td>0.130991</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>total_bedrooms</th>\n",
              "      <td>0.071802</td>\n",
              "      <td>-0.069373</td>\n",
              "      <td>-0.320434</td>\n",
              "      <td>0.928403</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.881169</td>\n",
              "      <td>0.980920</td>\n",
              "      <td>-0.013495</td>\n",
              "      <td>0.045783</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>population</th>\n",
              "      <td>0.101674</td>\n",
              "      <td>-0.111261</td>\n",
              "      <td>-0.295890</td>\n",
              "      <td>0.860170</td>\n",
              "      <td>0.881169</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.909247</td>\n",
              "      <td>-0.000638</td>\n",
              "      <td>-0.027850</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>households</th>\n",
              "      <td>0.059628</td>\n",
              "      <td>-0.074902</td>\n",
              "      <td>-0.302754</td>\n",
              "      <td>0.919018</td>\n",
              "      <td>0.980920</td>\n",
              "      <td>0.909247</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.007644</td>\n",
              "      <td>0.061031</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>median_income</th>\n",
              "      <td>-0.015485</td>\n",
              "      <td>-0.080303</td>\n",
              "      <td>-0.115932</td>\n",
              "      <td>0.195383</td>\n",
              "      <td>-0.013495</td>\n",
              "      <td>-0.000638</td>\n",
              "      <td>0.007644</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.691871</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>median_house_value</th>\n",
              "      <td>-0.044982</td>\n",
              "      <td>-0.144917</td>\n",
              "      <td>0.106758</td>\n",
              "      <td>0.130991</td>\n",
              "      <td>0.045783</td>\n",
              "      <td>-0.027850</td>\n",
              "      <td>0.061031</td>\n",
              "      <td>0.691871</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                    longitude  latitude  ...  median_income  median_house_value\n",
              "longitude            1.000000 -0.925208  ...      -0.015485           -0.044982\n",
              "latitude            -0.925208  1.000000  ...      -0.080303           -0.144917\n",
              "housing_median_age  -0.114250  0.016454  ...      -0.115932            0.106758\n",
              "total_rooms          0.047010 -0.038773  ...       0.195383            0.130991\n",
              "total_bedrooms       0.071802 -0.069373  ...      -0.013495            0.045783\n",
              "population           0.101674 -0.111261  ...      -0.000638           -0.027850\n",
              "households           0.059628 -0.074902  ...       0.007644            0.061031\n",
              "median_income       -0.015485 -0.080303  ...       1.000000            0.691871\n",
              "median_house_value  -0.044982 -0.144917  ...       0.691871            1.000000\n",
              "\n",
              "[9 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cO1fNb-bH8Xe"
      },
      "source": [
        "## 4. Create training and test sets\n",
        "\n",
        "We want to be able to evaluate our model and \"fake\" real world data. Sklearn provides a function to split our dataset up into two sets, one for testing and another for training.\n",
        "\n",
        "test_size: proportion of the dataset that should be in the test set\n",
        "random_state: makes sure you get the same sets each time you run it\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xJ0Ykdf7H_dP"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "iris_train, iris_test = train_test_split(iris_df, test_size=0.2, random_state=42)"
      ],
      "execution_count": 85,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Dz-AFO_mceB2"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-ZxQQYJ0Mn16"
      },
      "source": [
        "## 5. Prepare Data for Model (Feature Scaling)\n",
        "\n",
        "The scales of the features vary drastically. If we change the weights of all the features by 1, its going to affect some way more than others.\n",
        "\n",
        "The Standard Scaler scales down the range of all features to be between -1 and 1 with the mean at 0. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iYj5W0NNMmvF"
      },
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "scaler = StandardScaler()\n",
        "iris_features_train = iris_train.drop(['target'], axis=1)\n",
        "iris_features_train.head()\n",
        "\n",
        "scaled_iris_features_train = scaler.fit_transform(iris_features_train)"
      ],
      "execution_count": 91,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vs-kfNkyefvp"
      },
      "source": [
        "### Scale down all features of the housing_df test set here."
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Lrw6BHyXOHLd"
      },
      "source": [
        "## 6. Select and Train Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "06s9gMG8OLE_",
        "outputId": "cf50a813-24f3-465a-86c4-97e28cf8d125",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        }
      },
      "source": [
        "from sklearn.linear_model import SGDRegressor\n",
        "\n",
        "lin_reg = SGDRegressor(eta0=(0.01), max_iter=10000)\n",
        "lin_reg.fit(scaled_iris_features_train, iris_train['target'])\n",
        "\n",
        "print(\"Predictions: \", lin_reg.predict(scaled_iris_features_train[:5]))\n",
        "print(\"Actual Values: \", iris_train[\"target\"][:5])\n"
      ],
      "execution_count": 92,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Predictions:  [-0.22098592 -0.08840147  1.28925782 -0.04920673 -0.1040756 ]\n",
            "Actual Values:  22    0.0\n",
            "15    0.0\n",
            "65    1.0\n",
            "11    0.0\n",
            "42    0.0\n",
            "Name: target, dtype: float64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_APSVK32QXts",
        "outputId": "151f90f7-f381-486e-8484-75024c7ca555",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "from sklearn.metrics import mean_squared_error\n",
        "preds = lin_reg.predict(scaled_iris_features_train)\n",
        "\n",
        "mse = mean_squared_error(preds, iris_train[\"target\"])\n",
        "np.sqrt(mse)"
      ],
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.23982819913154396"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 93
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vYe-Pe1bRTvi"
      },
      "source": [
        "Can we get better by tuning hyperparameters? What should we make them?\n",
        "\n",
        "Lets search!\n",
        "\n",
        "Sklearn's gridsearch allows you specify different values for the hyperparameters you want to try. When you train the model, it'll train many times using each combination of parameters.\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LOH-rNroioqR",
        "outputId": "437942cf-482a-4dfa-b220-67657845781c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 326
        }
      },
      "source": [
        "from sklearn.model_selection import GridSearchCV\n",
        "\n",
        "param_grid = [\n",
        "            {'eta0': [1,0.01,0.001,0.0001], 'max_iter': [100,1000,10000]}\n",
        "]\n",
        "lin_reg = SGDRegressor()\n",
        "\n",
        "grid_search = GridSearchCV(lin_reg, param_grid)\n",
        "\n",
        "grid_search.fit(scaled_iris_features_train, iris_train['target'])\n",
        "print(\"BEST ESTIMATOR\")\n",
        "grid_search.best_estimator_"
      ],
      "execution_count": 94,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "BEST ESTIMATOR\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_stochastic_gradient.py:1187: ConvergenceWarning: Maximum number of iteration reached before convergence. Consider increasing max_iter to improve the fit.\n",
            "  ConvergenceWarning)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_stochastic_gradient.py:1187: ConvergenceWarning: Maximum number of iteration reached before convergence. Consider increasing max_iter to improve the fit.\n",
            "  ConvergenceWarning)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_stochastic_gradient.py:1187: ConvergenceWarning: Maximum number of iteration reached before convergence. Consider increasing max_iter to improve the fit.\n",
            "  ConvergenceWarning)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_stochastic_gradient.py:1187: ConvergenceWarning: Maximum number of iteration reached before convergence. Consider increasing max_iter to improve the fit.\n",
            "  ConvergenceWarning)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_stochastic_gradient.py:1187: ConvergenceWarning: Maximum number of iteration reached before convergence. Consider increasing max_iter to improve the fit.\n",
            "  ConvergenceWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "SGDRegressor(alpha=0.0001, average=False, early_stopping=False, epsilon=0.1,\n",
              "             eta0=0.01, fit_intercept=True, l1_ratio=0.15,\n",
              "             learning_rate='invscaling', loss='squared_loss', max_iter=100,\n",
              "             n_iter_no_change=5, penalty='l2', power_t=0.25, random_state=None,\n",
              "             shuffle=True, tol=0.001, validation_fraction=0.1, verbose=0,\n",
              "             warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 94
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZayHAYBilHch"
      },
      "source": [
        "## 7. Test on test set\n",
        "\n",
        "Now we want to see how well our model generalizes to new data."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PQ7Hg4CWRZBR",
        "outputId": "0892ca08-4a1d-415f-eac9-904941acf9e8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        }
      },
      "source": [
        "iris_features_test = iris_test.drop(['target'], axis=1)\n",
        "\n",
        "scaled_iris_features_test = scaler.fit_transform(iris_features_test)\n",
        "\n",
        "lin_reg = SGDRegressor(eta0=(0.01), max_iter=10000)\n",
        "lin_reg.fit(scaled_iris_features_test, iris_test['target'])\n",
        "\n",
        "print(\"Predictions: \", lin_reg.predict(scaled_iris_features_test[:5]))\n",
        "print(\"Actual Values: \", iris_test[\"target\"][:5])\n"
      ],
      "execution_count": 106,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Predictions:  [1.0919364  0.04932578 2.28194537 1.15655419 1.35086509]\n",
            "Actual Values:  73     1.0\n",
            "18     0.0\n",
            "118    2.0\n",
            "78     1.0\n",
            "76     1.0\n",
            "Name: target, dtype: float64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jpmcIyeYmJ5-",
        "outputId": "0522fe69-2b21-4ae8-d7c9-f44eb01e68f6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "preds = lin_reg.predict(scaled_iris_features_test)\n",
        "mse = mean_squared_error(preds, iris_test[\"target\"])\n",
        "np.sqrt(mse)"
      ],
      "execution_count": 108,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.2361282997649308"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 108
        }
      ]
    }
  ]
}